Chapter 3: The Epidemiology of Fractures and Dislocations

Charles M. Court-Brown

Chapter Outline

Introduction

In the seventh edition of Rockwood and Green26 the epidemiology of fractures presenting to the Royal Infirmary of Edinburgh in a 1-year period in 2007 to 2008 was analyzed. This hospital is somewhat unusual in that it is the only hospital admitting orthopedic trauma from a large well-defined population living in the City of Edinburgh, Midlothian and East Lothian, in Scotland. It does not deal with pediatric fractures but it does act as a secondary referral center for complex fractures in the South East of Scotland. For the purposes of epidemiologic analysis these fractures have not been considered and the data in this chapter comes exclusively from the catchment area of the hospital. This provides an accurate analysis of the epidemiology of adult fractures in a developed country and the results will apply to other areas with a similar socioeconomic status. Obviously this does not include all parts of the world, but it is likely that our data is applicable to many areas. In this edition of Rockwood and Green the structure of the chapter has been altered. In the seventh edition of Rockwood and Green26 the fracture epidemiology of the Royal Infirmary of Edinburgh was compared with that of the R Adams Cowley Shock Trauma Center in Baltimore, USA in the belief that there would be a significant difference in fracture epidemiology between a large hospital dealing with all orthopedic trauma and a specialized Level 1 trauma center dealing mainly with severe trauma. In this edition adult fracture epidemiology has been examined more closely and the different factors that affect it have been defined. In addition to discussing the epidemiology of adult fractures in a specific year the epidemiology of adolescent fractures has been examined as has the epidemiology of open fractures. The epidemiology of dislocations has not received much attention in the literature and an analysis of dislocations over a 1-year period is presented. 
The chapter will be based on the principle that fracture epidemiology is affected by gender and age, but is also influenced by a number of social and medical comorbidities which, as yet, have not been precisely defined. However it is likely that these comorbidities can be examined by looking at the effect of social deprivation on fracture epidemiology. This will also be undertaken in an attempt to provide a more complete epidemiologic analysis of fractures in adults. 

History

Surgeons have treated fractures for several millennia but until the advent of radiographs the diagnosis of different fractures was based on knowledge of the human anatomy, clinical signs, and educated observation. However a number of surgeons did analyze fracture epidemiology in some detail a good example being Malgaigne70 who analyzed 2,377 fractures in the Hôtel Dieu in Paris between 1806 and 1808 and 1830 and 1839. He analyzed these fractures according to age, gender, seasonality, and the location of the fracture. He observed that fractures were most commonly seen between 25 and 60 years of age. He understood the importance of calculating incidence and showed that while patients aged 25 to 30 and 55 to 60 had a similar prevalence of fractures the numbers of 55- to 60-year olds in the population was less than 50% that of 25- to 30-year olds. He stated that there were very few fractures in patients aged >60 years but there were very few people of that age left alive. 
Malgaigne70 stated that fractures in males exceeded fractures in females by a factor of 5:2 but that the factor varied with age. In children of 5 years or less females presented with twice as many fractures as males. The ratio changed with increasing age so that between 15 and 20 years there were eight male fractures for every female fracture. The ratio continued to change so that between 70 and 75 years of age the fracture rate was similar in males and females. Thereafter fractures were more commonly seen in females. Malgaigne did not agree with the previously stated view that fractures were more common in winter because cold weather made bones more fragile, and he actually recorded that fractures were more commonly seen in spring. 
A review of the epidemiology of the fractures that Malgaigne70 treated showed that 46.9% were upper limb fractures, 52% were lower limb fractures, and the remaining 1.1% were spinal or pelvic fractures. There was a high prevalence of humeral fractures (14.4%), femoral fractures (10.1%), and lower leg fractures (33.1%) and he recorded that 5.3% of the fractures were proximal femoral fractures and that 1.3% involved the proximal humerus. He, and his contemporaries were obviously aware that different fractures occurred at different ages and he observed that it had recently been stated that fractures of the diaphyses tended to occur in adulthood while intra-articular fractures occurred in old age. He thought that this observation was essentially correct and he stated that fractures of the “cervix femoris and cervix humeri” occurred in old age and that women sustained a large proportion of the fractures of “the carpal extremity of the radius.” 
More exact analyses of fracture epidemiology were undertaken by Stimson95 in the Hudson Street Hospital, New York between 1894 and 1903 and by Emmett and Breck37 of El Paso, Texas in three time periods between 1937 and 1956. To be strictly accurate these analyses cannot be compared with data from Edinburgh, Scotland as they simply report the workload of their particular institutions. However both hospitals dealt with a wide range of different patients and conditions and they treated many thousands of fractures. Thus their results are of interest and there are, in fact, no better epidemiologic analyses to compare with modern data. 
Table 3-1 shows a comparison of the prevalence of different fractures treated by Stimson and his colleagues95 and by Emmett and Breck and their colleagues.37 Both groups treated both children and adults and for comparative purposes their results have been compared to those of the Edinburgh Orthopaedic Trauma Unit in 2000 as both adult and childrens’ fractures were examined that year. It is of course difficult to compare all the fractures accurately. This is particularly true of forearm fractures where proximal radial fractures are often combined with forearm diaphyseal fractures. In Table 3-1 all of Emmett and Breck’s data on forearm fractures has been combined as it was difficult to analyze their assessment of the individual fractures. It is also difficult to distinguish between isolated fibular fractures and those fibular fractures associated with ankle fractures, tibial diaphyseal fractures, and proximal and distal tibial fractures. The overall figures do however point to changing trends in the epidemiology of fractures which reflect massive social, health, and economic changes in society. 
Table 3-1
The Prevalence of Fractures in Three Time Periods Over the Last 100 Years
Fracture Prevalence (%)
1894–190395 1937–195637 200023,85
Clavicle 5.9 6.2 4.3
Scapula 0.7 0.7 0.2
Proximal humerus 5.7* 2.6 4.8
Humeral diaphysis 5.7* 2 1
Distal humerus 5.7* 5.2 2.5
Proximal ulna 1.1 21.2* 0.8
Proximal radius 9* 21.2* 3.8
Radius and ulnar diaphyses 9* 21.2* 2.3
Distal radius and ulna 11.2 21.2* 22.2
Carpus 0.2 2.4 2
Metacarpus 9.7 4.2 10.5
Finger phalanges 19.3 7.6 11.2
Pelvis 0.7 2.5 1.2
Proximal femur 4.7* 6.6 8.9
Femoral diaphysis 4.7* 2.5 0.9
Distal femur 4.7* 0.6 0.4
Patella 1.7 1.8 0.8
Proximal tibia 10.4* 7.3* 1
Tibia and fibula diaphyses 10.4* 7.3* 2
Distal tibia 10.4* 7.3* 1
Ankle 10.6 8.8 7.7
Tarsus 1.5 3.5 1.6
Metatarsus 2.8 4.1 6.4
Toe phalanges 3.1 4.4 2
Others 1.5 4.6
Fracture numbers 8,982 9,379 7,760
X
A review of Table 3-1 shows that in 1894 to 1903, when motor vehicles were extremely rare and the life expectancy of both males and females might reasonably be expected to be about 50 years of age, there were many fractures that one would regard as high-energy injuries. There was a high prevalence of fractures of the scapula, tibia and fibula, and ankle and it seems likely that many of these fractures were work related as there would have been little workplace legislation. The prevalence of fractures of the tibia and fibula was particularly high in 1894 to 1903. 
Unfortunately Stimson95 did not separate proximal humeral and proximal femoral fractures from other fractures of the humerus and femur. But it seems likely that there was a much lower prevalence of fragility fractures in those days. This is confirmed by observing that the prevalence of distal radial fractures has doubled in the last 100 years or so and this must be largely because of the increase in numbers of elderly people in the population. Table 3-1 shows that the prevalence of proximal humeral and proximal femoral fractures has certainly increased since the period around the Second World War. 
Another major difference in fracture epidemiology over the last 100 years or so is the decreased prevalence of fractures of the hand. Stimson found that 29.2% of the fractures that he treated involved the hand compared with 23.7% in Edinburgh in 2000. It seems likely that this is related to a safer work environment nowadays. It is also likely that Stimson’s estimate of the prevalence of carpal fractures is an underestimate as radiographs were in their infancy at that time. 
Table 3-1 shows that the epidemiology of fractures is changing and there is no doubt that it will continue to change. Recently many authors have commented on the increased incidence of fragility fractures but nothing is new and we should remember Malgaigne’s70 observation over 160 years ago that fractures of the proximal humerus, proximal femur, and distal radius were more common in the elderly and in women! 

Fracture Incidence

It is surprisingly difficult to analyze the incidence of fractures accurately. In many parts of the world there are no facilities to allow accurate analysis of what is a common medical condition. However even in more affluent areas there is remarkably little accurate information about the incidence of fractures. One might think that the analysis of all fractures in a specific population during a specific time period would be relatively easy, but in many countries orthopedic trauma is treated in different types of institution with severe trauma being treated in Level 1 trauma centers, or their equivalent, whilst less severe trauma is often treated in community hospitals or by surgeons in private practice in the community. Thus few hospitals see the whole range of orthopedic trauma and as there is usually little communication between hospitals, it is often impossible to accurately analyze the incidence of fractures. 
For this reason a number of different types of methodologies have been used to try to define fracture epidemiology in both adults and children. Table 3-2 shows the results of several analyses of fracture epidemiology in the United Kingdom,23,31,32,56,85 Norway,87 and the United States.41 The difference in the results is striking! All the studies in Table 3-2 include both adults and children, but different methodologies have been used and this is likely to account for the wide variation in results. 
 
Table 3-2
Fracture Incidence Reported in Various Studies
View Large
Table 3-2
Fracture Incidence Reported in Various Studies
Incidence (n/105/yr)
Study Years Country Overall Male Female
Donaldson et al.31 1980–1982 UK 9.1 10 8.1
Johansen et al.56 1994–1995 UK 21.1 23.5 18.8
Court-Brown and
Caesar,23 Rennie et al.85
2000 UK 12.6 13.6 11.6
Donaldson et al.32 2002–2004 UK 36 41 31
Sahlin87 1985–1986 Norway 22.8 22.9 21.3
Fife and Barancik41 1977 USA 21 26 16
 

To obtain the overall incidence in Scotland in 2000, the adult fractures reported by Court-Brown and Caesar23 have been combined with the children’s fractures reported by Rennie et al.85

X
Donaldson et al.,31 in their early study, examined a geographically well-defined population in England and looked at both the inpatient and outpatient fractures in the area. They observed that they might be missing some toe and spinal fractures, but they felt that they had missed relatively few fractures. A very similar methodology was employed by Court-Brown and Caesar24 in the sixth edition of Rockwood and Green. They assessed all of the adult fractures treated in the Royal Infirmary of Edinburgh in 200023 from the same catchment area as in this study. The fractures admitted to the pediatric hospital in Edinburgh in the same year were also analyzed83 and the incidence of fractures in the whole Edinburgh population is shown in Table 3-2. Given the 20-year gap between these two studies it would seem that both studies had very similar results. 
Table 3-2 shows, however, that other studies have produced very different results. The studies by Johansen et al.56 in Wales, Sahlin in Norway,87 and Fife and Barancik41 in the United States all record similar fracture incidences and it is interesting to observe that the diagnoses of the different fracture types were usually taken from the records of the local emergency departments. Many of these patients would not have been seen by an orthopedic surgeon and the diagnosis would have been made by an inexperienced junior doctor. This is in contrast to the Edinburgh study where all diagnoses were made by orthopedic trauma surgeons. In the United Kingdom much of the fracture data used in epidemiologic analysis is taken from the General Practice Research Database.63,67 The diagnoses made by nonorthopedic surgeons, in the emergency departments of different hospitals, are relayed to local family physicians where they are recorded and then analyzed to produce epidemiologic information. This may lead to an incorrect estimation of the number of fractures in a community, particularly in those fractures that occur in areas where soft tissue injuries are relatively common such as the hand, wrist, ankle, and foot. An example of this problem is seen in the study by Johansen et al.56 of fractures in the combined pediatric and adult population treated in South Wales in 1994. They stated that the overall incidence of fractures of the hand and foot, ankle and finger, thumb and hand were 2.41/1,000/year, 1.42/1,000/year, and 4.41/1,000/year, respectively. The combined pediatric and adult figures from Edinburgh in 2000 24,85 for these fracture combinations were 1.3/1,000/year, 0.9/1,000/year, and 3/1,000/year carpal fractures, respectively suggesting that inexperienced doctors will overestimate the prevalence of fractures, many of which are seen on an outpatient basis. A good example of this is the “? scaphoid” soft tissue injury that is often documented as being a fracture. 
The third type of methodology highlighted in Table 3-2 is that used by Donaldson et al.32 in a later study. They asked patients to complete a questionnaire regarding whether they had had a fracture in a given time period. Table 3-2 shows that this methodology produces results that are about 300% greater than one would expect. If the incidences suggested in this study are applied to the United Kingdom population there would be about 2,200,000 fractures annually in the United Kingdom which is simply not the case. This problem is obviously methodologic as many patients will be told that recurrent or continuing pain may be secondary to undiagnosed fractures by family physicians, physiotherapists, nurses, osteopaths, or other paramedical professionals without there being any proof that this is the case. 
Other methods have been employed to try to estimate fracture incidence. In countries with privatized medical systems insurance records have been used. These may be very large but they tend to present an unbalanced view of the population. Brinker and O’Connor11 examined a large privately insured cohort of patients but the average age was 29 years for males and 28.7 years for females which is not representative of the population. However it explains why 57% of their population presented with fractures of the forearm, hand, and foot. In a cohort covering the whole population one would expect a figure of about 42% (Table 3-3). In a similar study Orces and Martinez80 examined the incidence of wrist and forearm fractures in the United States. They looked at the records of a large number of emergency departments and concluded that the incidence of wrist and forearm fractures in males and females ≥50 years of age was 78.2/105/year and 256.9/105/year, respectively. It seems likely that these values should be much higher and ours are 154/105/year and 642.6/105/year. Clearly one must examine the whole population to get accurate figures. Bradley and Harrison8 examined inpatients in Australia to provide fracture incidence figures but as about 55% to 60% of fractures are treated on an outpatient basis accurate figures cannot be obtained using this method. 
 
Table 3-3
Epidemiology of Fractures Treated in a 1-Year Period
View Large
Table 3-3
Epidemiology of Fractures Treated in a 1-Year Period
No. % n/105/yr Average Age (yrs) ≥65 yrs (%) ≥80 yrs (%) M/F
Distal radius/ulna 1,221 17.5 235.9 58.4 41.8 18.1 28/72
Metacarpus 781 11.2 150.9 33.6 8.2 3.1 80/20
Proximal femur 753 10.8 145.5 80.7 90.6 63.7 27/73
Ankle 713 10.2 137.7 49 23.8 6 46/54
Finger phalanges 696 9.9 134.5 41.6 13.6 5.8 60/40
Proximal humerus 478 6.8 92.4 66.3 55.6 23 31/69
Metatarsus 465 6.6 89.8 44.6 17 5.2 37/63
Proximal forearm 378 5.4 73 45.6 17.2 5.8 46/54
Clavicle 257 3.7 49.7 44.5 21 9.7 70/30
Toe phalanges 248 3.5 47.9 35.7 3.9 1 59/41
Carpus 194 2.8 37.5 38 7.7 1.5 64/36
Pelvis 119 1.7 23 75.6 74.8 58.8 30/70
Femoral diaphysis 82 1.2 15.8 70.2 67.1 39 48/52
Humeral diaphysis 70 1 13.5 56.8 42.8 20 47/53
Tibial diaphysis 69 1 13.3 42.3 8.7 0 71/29
Calcaneus 65 0.9 12.6 41 9.2 3.1 74/26
Proximal tibia 59 0.8 11.4 54.5 30.5 11.9 52/48
Forearm diaphysis 55 0.8 10.6 48 27.3 16.4 69/31
Patella 49 0.7 9.5 64.8 55.1 28.6 41/59
Distal humerus 48 0.7 9.3 58.5 56.2 29.2 42/58
Distal tibia 42 0.6 8.2 41.7 17.7 4.4 67/33
Fibula 41 0.6 7.9 46.8 14.6 2.4 46/54
Scapula 37 0.5 7.1 54.8 32.4 16.2 76/24
Distal femur 36 0.5 7 67.3 52.8 38.9 17/83
Midfoot 28 0.4 5.4 39.4 7.1 0 61/39
Talus 12 0.2 2.3 30.1 0 0 83/17
6,996 100 1,351.7 53.2 34 17.3 47/53
 

The numbers, prevalence, incidence, and gender ratios are shown together with the average ages and percentages of patients ≥65 yrs and ≥80 yrs of age.

X

The Incidence of Fractures in Adults

In this edition of Rockwood and Green a further year of inpatient and outpatient fractures presenting to the Royal Infirmary of Edinburgh has been prospectively analyzed. In the sixth edition24 all fractures presenting to the hospital in 2000 were analyzed. In the seventh edition26 a year of fractures between July 2007 and June 2008 was analyzed. In this edition a further year of fractures between September 2010 and August 2011 has been analyzed. The analysis has been confined to patients aged 16 years or older and the 2001 Scottish census44 has been used to calculate fracture incidence, this being the last census that was undertaken. During this year soft tissue injuries and dislocations were not studied prospectively, but the chapter includes a retrospective analysis of dislocations which presented to the Trauma Unit in a 1-year period in 2008/9. 
During the study all fractures were analyzed, including 104 spinal fractures that presented to the Orthopaedic Trauma Unit, but these spinal fractures have not been included in the epidemiology tables as in Edinburgh spinal fractures are also treated by neurosurgeons with spinal cord injuries being treated in the National Spinal Injuries Centre in Glasgow. However, spinal fractures have been included where they occurred in association with other fractures. All patients resident in and injured in the defined catchment area have been included as have those injured elsewhere but resident in our area. Patients resident outwith the area have not been included in the analysis. In this analysis the effect of gender, age, and social deprivation on fracture epidemiology has been studied and because of this fractures in males and females, and in the age ranges 16 to 35 years, 36 to 64 years, and ≥65 years, will be presented separately. Fractures in the very elderly will be discussed in Chapter 20
The overall epidemiology of the fractures that presented during the year is shown in Table 3-3. It shows that there were 6,996 fractures during the year, giving an overall incidence of 1,351.7/105/year. The average age was 53.2 years and 53% of the fractures occurred in females. Overall 34% of the fractures occurred in patients ≥65 years of age and 17.3% in patients ≥80 years of age. 

Gender and Age

The importance of gender and age to fracture epidemiology has been understood for many years. In their classic epidemiologic study Buhr and Cooke13 highlighted the fact that men have a bimodal distribution of fractures and women have a unimodal distribution with a significant progressive increase in fracture incidence in the postmenopausal years. This is demonstrated in the overall fracture distribution curves shown in Figure 3-1. Analysis of the Edinburgh data shows that males between 16 and 19 years of age have a fracture incidence of 2,506.4/105/year which falls to 937.4/105/year in males aged between 50 and 59 years and then rises to 6,860.5/105/year in males aged 90 years or more. In females the equivalent values are 792.1/105/year, 1,398.5/105/year, and 7,769.8/105/year. This illustrates the considerable difference in the incidence of fractures between males and females. 
Figure 3-1
The overall age and gender fracture distribution curves.
Rockwood-ch003-image001.png
View Original | Slide (.ppt)
X
The difference in fracture epidemiology between males and females is further highlighted by comparing Tables 3-4 and 3-5. Basic epidemiologic data for males is shown in Table 3-4 and for females in Table 3-5. In males the average age is 42.4 years and only 17.1% of fractures occurred in males ≥65 years of age. Fractures of the metacarpus, finger phalanges, distal radius, and ulna and ankles comprised 52.1% of all male fractures. In females the average age is 61.8 years and 48.8% of fractures occurred in patients ≥65 years of age. The obvious tendency for fragility fractures to be much more common in females is highlighted in Table 3-5 which shows that fractures of the distal radius and ulna and proximal femur make up 38.5% of all female fractures and if one includes ankle fractures and proximal humeral fractures the most common four fractures in females account for 57.7% of all fractures. 
 
Table 3-4
Epidemiology of Fractures in Males
View Large
Table 3-4
Epidemiology of Fractures in Males
No. % n/105/yr Age (yrs) ≥65 yrs (%) ≥80 yrs (%) Multiple Fractures (%) Open Fractures (%)
Metacarpus 624 19 255.6 30.3 4 1.8 9.2 0.6
Finger phalanges 418 12.7 171.2 37.6 7.9 3.1 6.2 8.5
Distal radius/ulna 340 10.4 139.3 44.2 15.9 5.3 6.2 0.6
Ankle 330 10 135.2 42.4 14.2 3 3 0.3
Proximal femur 205 6.2 84 78.5 87.8 54.6 4.4 0
Clavicle 180 5.5 73.7 38.2 10.6 3.3 6.1 0
Proximal forearm 175 5.3 71.7 39.6 8.6 2.9 11 2.3
Metatarsus 170 5.2 69.6 36.6 6.5 1.8 8.2 0
Proximal humerus 149 4.5 61 59.7 39.6 16.8 9.4 0
Toe phalanges 146 4.4 59.8 37.1 1.4 0.7 5.5 8.2
Carpus 125 3.8 51.2 32.9 4.8 0.8 6.4 0
Tibial diaphysis 49 1.5 20 41.6 10.2 0 6.1 24.5
Calcaneus 48 1.5 19.7 37.5 4.2 0 32.6 6.2
Femoral diaphysis 39 1.2 16 63.9 51.3 30.8 18.9 5.1
Forearm diaphysis 38 1.1 15.6 40.3 10.5 7.9 7.9 7.9
Pelvis 36 1.1 14.7 65.2 55.5 36.1 22.2 2.8
Humeral diaphysis 33 1 13.5 51.5 30.3 18.2 3 0
Proximal tibia 31 0.9 12.7 49 25.8 9.7 12.9 3.2
Scapula 28 0.9 11.5 48.5 17.9 0 44.4 0
Distal tibia 28 0.9 11.5 35.2 7.1 0 29.6 17.9
Distal humerus 20 0.6 8.2 46.1 30 20 35 15
Patella 20 0.6 8.2 52.4 30 30 10 10
Fibula 19 0.6 7.8 38.3 15.8 15.8 0 0
Midfoot 17 0.5 7 36.5 0 0 30.8 5.9
Talus 10 0.3 4.1 28.5 0 0 40 10
Distal femur 6 0.2 2.5 55.3 50 50 0 0
3,284 100 1,345.3 42.4 17.1 7.9 5.6 2.8
 

The numbers, prevalence, incidence, and gender ratios are shown together with the average ages and percentages of patients ≥65 yrs and ≥80 yrs of age. The prevalence of open fractures is also shown.

X
 
Table 3-5
Epidemiology of Fractures in Females
View Large
Table 3-5
Epidemiology of Fractures in Females
No. % n/105/yr Age (yrs) ≥65 yrs (%) ≥80 yrs (%) Multiple Fractures (%) Open Fractures (%)
Distal radius/ulna 881 23.7 322.2 63.9 51.8 23 5.6 0.9
Proximal femur 548 14.8 200.4 81.6 91.8 67 6.2 0
Ankle 383 10.3 140.1 54.7 32.1 8.6 2.9 1.6
Proximal humerus 329 8.9 120.3 69.3 63.2 26.1 7.6 0
Metatarsus 295 7.9 107.9 49.1 23.1 7.1 12.9 0
Finger phalanges 278 7.5 101.7 47.6 20.5 9.3 7.9 4.9
Proximal forearm 203 5.5 74.2 51.1 24.6 8.4 7.9 1.5
Metacarpus 157 4.2 57.4 46.8 24.8 8.3 9.4 1.3
Toe phalanges 102 2.7 37.3 33.6 2 0 3.9 4.9
Pelvis 83 2.2 30.4 79.8 83.1 68.7 12 0
Clavicle 77 2.1 28.2 79.8 45.5 24.7 3.9 1.3
Carpus 69 1.9 25.2 47.1 13 2.9 7.2 0
Femoral diaphysis 43 1.2 15.7 76 81.4 46.5 4.7 0
Humeral diaphysis 37 1 13.5 61.6 54.1 21.6 0 2.7
Distal femur 30 0.8 11 69.6 53.3 36.7 13.3 3.3
Patella 29 0.8 10.6 73.4 72.4 37.9 0 3.4
Proximal tibia 28 0.8 10.2 60.6 35.7 14.3 14.3 0
Distal humerus 28 0.8 10.2 67.4 75 35.7 10.7 0
Fibula 22 0.6 8 54.8 13.6 4.5 13.6 0
Tibial diaphysis 20 0.5 7.3 44.1 5 0 5 10
Calcaneus 17 0.5 6.2 51 23.5 11.8 11.8 0
Forearm diaphysis 17 0.5 6.2 65.1 64.7 35.3 5.9 0
Distal tibia 14 0.4 5.1 54.7 42.9 14.3 7.1 7.1
Midfoot 11 0.3 4 43.9 18.2 0 27.3 0
Scapula 9 0.2 3.3 74.5 77.8 66.6 22.2 0
Talus 2 0.05 0.7 37.5 0 0 50 0
3,712 100 1,357.5 61.8 48.8 25.4 4.4 1.2
 

The numbers, prevalence, incidence, and gender ratios are shown together with the average ages and percentages of patients ≥65 yrs and ≥80 yrs of age. The prevalence of open fractures is also shown.

X
Tables 3-4 and 3-5 also contain an analysis of the prevalence of patients who presented with multiple fractures or with open fractures. Table 3-4 shows that 5.6% of male patients presented with multiple fractures. These will be discussed in more detail in the sections on individual fractures, but Table 3-4 shows that more than 20% of males who presented with fractures of the calcaneus, pelvis, scapula, distal tibia, distal humerus, midfoot, and talus had multiple fractures. Table 3-4 shows that at least 10% of fractures of the tibial diaphysis, distal tibia, distal humerus, patella, and talus in males were open. In females 4.4% of patients had multiple fractures (Table 3-5) but it was only in fractures of the scapula, midfoot, and talus that more than 20% of patients had multiple fractures. There were fewer open fractures in females and only in fractures of the tibial diaphysis did 10% of patients present with open fractures. 
The importance of age in the epidemiology of fractures in males is illustrated in Tables 3-63-8. Table 3-6 shows the fracture epidemiology for males aged 16 to 35 years with Table 3-7 dealing with males of 36 to 64 years and Table 3-8 with males aged ≥65 years. In addition to the numbers, prevalence, and incidence of each fracture the percentage of open fractures and patients with multiple fractures is also shown. Each table also gives the commonest two causes of each fracture. 
 
Table 3-6
Epidemiology of Fractures in Males Aged 16 to 35 Years
View Large
Table 3-6
Epidemiology of Fractures in Males Aged 16 to 35 Years
No. % n/105/yr Multiple Fractures (%) Open Fractures (%) Causes
Metacarpus 470 29.1 517.7 7.9 0.4 71.4% db/assault, 12.8% sport
Finger phalanges 237 14.7 261.2 5.5 6.3 44.7% sport, 35.9% db/assault
Distal radius/ulna 141 8.7 155.3 5 0 45.4% sport, 29.1% falls
Ankle 139 8.6 153.1 2.2 0.7 52.5% falls, 33.1% sport
Metatarsus 100 6.2 110.1 5.4 0 48% falls, 27.8% sport
Clavicle 98 6.1 107.9 3.1 0 51% sport, 23.5% falls
Proximal forearm 90 5.6 99.1 10.2 4.4 32.2% falls, 30% sport
Carpus 87 5.4 95.8 4.6 0 35.6% sport, 33.3% falls
Toe phalanges 78 4.8 85.9 0 11.5 50% db/assault, 40.6% sport
Calcaneus 26 1.6 28.6 39.1 7.7 76.9% fall height, 11.5% low falls
Tibial diaphysis 21 1.3 23.1 0 14.2 57.1% sport, 19% falls
Forearm diaphysis 19 1.2 20.9 0 0 57.9% sport, 15.8% db/assault
Proximal humerus 15 0.9 16.5 0 0 26.6% sport, 26.6% falls
Distal tibia 15 0.9 16.5 33.3 13.3 60% fall height, 13.3% MVA
Fibula 11 0.7 12.1 0 0 63.6% sport, 9.1% MVA
Humeral diaphysis 11 0.7 12.1 0 0 36.4% falls, 36.4% db/assault
Distal humerus 9 0.6 9.9 22.2 11.1 33.3% MVA, 22.2% sport
Proximal tibia 9 0.6 9.9 0 0 6.6% sport, 22.2% low falls
Talus 9 0.6 9.9 44.4 11.1 33.3% fall height, 33.3% sport
Midfoot 6 0.4 6.6 20 0 50% sport, 33.3% fall height
Femoral diaphysis 5 0.3 5.5 40 0 60% fall height, 20% MVA
Scapula 5 0.3 5.5 40 0 60% MVA, 20% fall height
Pelvis 4 0.2 4.4 50 0 50% fall height, 25% MVA
Patella 3 0.2 3.3 0 33.3 66.6% MVA, 33.3% sport
Proximal femur 3 0.2 3.3 33.3 0 33.3% MVA, 33.3% sport
Distal femur 2 0.1 2.2 0 0 50% sport, 50% MVA
1,613 100 1,776.7 4.9 2.2 32.9% db/assault, 30.3% sport
 

The numbers, prevalence, and incidence of the different fractures are shown as are the prevalence of open fractures and patients with multiple fractures. The two commonest causes of each fracture are shown. (db = direct blow).

X
 
Table 3-7
Epidemiology of Fractures in Males Aged 36 to 64 Years
View Large
Table 3-7
Epidemiology of Fractures in Males Aged 36 to 64 Years
No. % n/105/yr Multiple Fractures (%) Open Fractures (%) Causes
Finger phalanges 146 13 127.3 5.8 14.1 47.4% db/assault, 21.5% falls
Distal radius/ulna 145 12.9 126.4 6.9 1.4 55.2% falls, 20% sport
Ankle 144 12.8 125.6 3.5 0 81.3% falls, 9.7% sport
Metacarpus 129 11.5 112.5 10.2 1.6 48.1% db/assault, 27.9% falls
Proximal humerus 75 6.7 65.4 8 0 65.3% falls, 10.7% sport
Proximal forearm 70 6.2 61 10.1 0 45.7% falls, 22.8% MVA
Clavicle 63 5.6 54.9 9.5 0 41.2% MVA, 33.3% falls
Toe phalanges 63 5.6 54.9 15.4 11.1 73.1% db/assault, 11.5% falls
Metatarsus 59 5.3 51.4 12.7 0 55.9% falls, 13.6% MVA
Carpus 32 2.8 27.9 12.5 0 62.5% falls, 12.5% MVA
Tibial diaphysis 23 2.1 20.1 8.7 30.4 52.2% falls, 17.4% MVA
Proximal femur 22 2 19.2 4.5 0 68.2% falls, 18.2% sport
Calcaneus 20 1.8 17.4 22.2 5 55% fall height, 25% low falls
Scapula 18 1.6 15.7 52.9 0 33.3% MVA, 22.2% falls
Forearm diaphysis 15 1.3 13.1 20 20 53.3% falls, 20% MVA
Femoral diaphysis 14 1.2 12.2 30.8 14.3 42.9% MVA, 28.6% falls
Proximal tibia 14 1.2 12.2 7.1 0 35.7% sport, 21.4% MVA
Pelvis 12 1.1 10.5 33.3 8.3 33.3% MVA, 33.3% falls
Humeral diaphysis 12 1.1 10.5 0 0 41.7% falls, 25% db/assault
Patella 11 1 9.6 18.2 9.1 45.5% falls, 27.3% MVA
Midfoot 11 1 9.6 37.5 9.1 45.5% fall height, 27.3% falls
Distal tibia 11 1 9.6 27.3 27.3 27.2% MVA, 27.2% falls
Distal humerus 5 0.4 4.4 60 40 40% falls, 40% MVA
Fibula 5 0.4 4.4 0 0 20% sport, 20% falls
Talus 1 0.09 0.9 0 0 100% fall height
Distal femur 1 0.09 0.9 0 0 100% falls
1,121 100 977.5 6.2 4.7 45.5% falls, 17.1% db/assault
 

The numbers, prevalence, and incidence of the different fractures are shown as are the prevalence of open fractures and patients with multiple fractures. The two commonest causes of each fracture are shown. (db = direct blow).

X
 
Table 3-8
Epidemiology of Fractures in Males Aged ≥65 Years
View Large
Table 3-8
Epidemiology of Fractures in Males Aged ≥65 Years
No. % n/105/yr Multiple Fractures (%) Open Fractures (%) Causes
Proximal femur 180 32.7 465.8 3.9 0 92.2% falls, 3.9% low falls
Proximal humerus 59 10.7 152.7 13.6 0 94.9% falls, 1.7% low falls
Distal radius/ulna 54 9.8 139.7 9.3 0 94.4% falls, 3.7% MVA
Ankle 47 8.5 121.6 4.3 0 83% falls, 6.4% sport
Finger phalanges 35 6.4 90.6 13.8 3.1 59.4% falls, 18.7% db/assault
Metacarpus 25 4.5 64.7 41.2 0 72% falls, 12% sport
Pelvis 20 3.6 51.8 10 0 90% falls, 10% MVA
Femoral diaphysis 20 3.6 51.8 5 0 80% falls, 15% pathologic
Clavicle 19 3.5 49.2 10.5 0 63.2% falls, 10.5% MVA
Proximal forearm 15 2.7 38.8 20 0 80% falls, 6.6% MVA
Metatarsus 11 2 28.5 18.2 0 63.6% falls, 18.2% db/assault
Humeral diaphysis 10 1.8 25.9 10 0 100% falls
Proximal tibia 8 1.5 20.7 37.5 12.5 50% falls, 12.5% fall height
Distal humerus 6 1.1 15.5 33.3 0 66.6% falls, 16.7% fall height
Carpus 6 1.1 15.5 0 0 100% falls
Patella 6 1.1 15.5 0 0 83.3% falls, 16.6% low falls
Scapula 5 0.9 12.9 20 0 40% falls, 20% fall height
Toe phalanges 5 0.9 12.9 0 0 80% db/assault, 20% falls
Tibial diaphysis 5 0.9 12.9 20 40 60% falls, 40% MVA
Forearm diaphysis 4 0.7 10.4 0 0 75% falls, 25% sport
Fibula 3 0.5 7.8 0 0 33.3% falls, 33.3% db/assault
Distal femur 3 0.5 7.8 0 0 100% falls
Calcaneus 2 0.4 5.2 50 0 50% fall height, 50% low falls
Distal tibia 2 0.4 5.2 0 0 100% falls
Midfoot 0 0 0 0 0
Talus 0 0 0 0 0
550 100 1,423.2 5.7 0.7 83.8% falls, 4% MVA
 

The numbers, prevalence, and incidence of the different fractures are shown as are the prevalence of open fractures and patients with multiple fractures. The two commonest causes of each fracture are shown. (db = direct blow).

X
Tables 3-63-8 show that age has a significant effect on fractures in males. Between 16 and 35 years of age fractures of the metacarpus and finger phalanges account for 43.8% of all male fractures and if one includes distal radius and ulna fractures and ankle fractures it can be seen that these four fractures comprise over 60% of all fractures in males aged 16 to 35 years. Some of the high-energy fractures that are commonly associated with young males such as those of the femoral diaphysis, distal tibia, and hindfoot are in fact relatively uncommon, but when they do occur are associated with a high prevalence of open fractures and multiple fractures. The relatively low incidence of femoral diaphyseal fractures shown in Table 3-6 may be a surprise to some surgeons, but in recent years it has become clear that the femoral diaphyseal fracture is essentially a fragility fracture occurring mainly in older females. This is shown in Table 3-11
As Figure 3-1 shows middle-aged males have a lower incidence of fractures and Tables 3-63-8 show that the incidence drops from 1,776.7/105/year in the 16- to 35-year-old group to 977.5/105/year in the 36- to 64-year-old group. They show that the incidence of metacarpal and finger phalangeal fractures falls markedly and there is also a slight drop in the incidence of fractures of the distal radius and ulna and ankle. The only fractures to show an increased incidence in middle-aged males are those of the proximal humerus, scapula, proximal femur, and femoral diaphysis. Much of the literature suggests that scapular fractures are simply high-energy injuries associated with motor vehicle accidents,20 but Tables 3-7 and 3-8 show that this is not the case. Hindfoot fractures and fractures of the distal tibia decline in incidence in middle-aged males. 
Despite the lowered incidence of male fractures in middle age, Table 3-7 shows that there is a higher prevalence of open fractures and more patients present with multiple fractures. Unsurprisingly the fractures that are associated with a high prevalence of open fractures and multiple fractures tend to be high-energy injuries. A high prevalence of open fractures is seen in fractures of the tibial diaphysis, forearm diaphyses, femoral diaphysis, distal tibia, and distal humerus. The fracture associated with the highest prevalence of multiple fractures in middle-aged males is the scapula fracture, although fractures of the calcaneus, forearm diaphyses, femoral diaphysis, pelvis, patella, distal tibia, and distal humerus are associated with multiple fractures in at least 20% of middle-aged male patients. 
Table 3-8 shows that there is a considerable change in the epidemiology of fractures in older males. Fractures of the metacarpus and finger phalanges are much less common and fractures of the proximal femur, proximal humerus, distal radius, and ulna and ankle account for over 60% of all the fractures in this age group. Fractures of the femoral diaphysis show a higher incidence, although none were associated with a high-energy injury. Pelvic fractures, humeral diaphyseal fractures, and patellar factures are more commonly seen than fractures of the tibial diaphysis and hindfoot which are relatively rare. Overall the fracture incidence rises to 1,423.2/105/year which is in 16- to 35-year-old males. However, the incidence continues to rise with increasing age and in the 80+ male population the fracture incidence is 3,302.7/105/year. The prevalence of open fractures is very low in this age group, although it is important to note that 5.7% of patients still presented with multiple fractures. This is related to osteoporosis in this elderly population. 
A review of the causes of fractures in adult males shows that in younger males (Table 3-6) direct blows, assaults, and sports injuries account for almost two-thirds of all fractures although in 12 (46.1%) fracture types, motor vehicle accidents or falls from a height were one of the two main causes of fracture. In middle-aged males (Table 3-7) almost half the fractures were caused by falls from a standing height, although 17% were still caused by a direct blow or assault. Fifteen (57.6%) of the fracture types had a motor vehicle accident or a fall from a height as one of the commonest causes. In older males (Table 3-8) 83.8% of fractures were caused by falls from a standing height, but it is interesting to note that motor vehicle accidents were still the second commonest cause of fracture. 
Tables 3-93-11 illustrate the different fracture epidemiology in females of different ages. Table 3-9 shows that the overall incidence of fractures in females aged 16 to 35 years is 664.4/105/year, this being 37% of the incidence in equivalently aged males. However, finger phalangeal fractures remain common in young females, although the incidence of metacarpal fractures is only 13% of that seen in males. Fractures of the finger phalanges, distal radius and ulna, metatarsus and toe phalanges comprise 56.1% of all fractures seen in young females. The prevalence of open fractures is very similar to that seen in young males, but only in tibial diaphyseal fractures were more than 10% of the fractures open. Only 3.8% of the young female patient group presented with multiple fractures, although the spectrum of multiple fractures was not dissimilar to that seen in young males (Table 3-6). 
 
Table 3-9
Epidemiology of Fractures in Females Aged 16 to 35 Years
No. % n/105/yr Multiple Fractures (%) Open Fractures (%) Causes
Finger phalanges 97 15.5 102.8 3.5 5.4 37.8% db/assault, 34.1% falls
Distal radius/ulna 94 15 99.6 3.2 1.1 72.3% falls, 9.6% sport
Metatarsus 91 14.5 96.4 12.2 0 67% falls, 12.1% db/assault
Toe phalanges 70 11.1 74.1 0 5.7 55.2% db/assault, 40.9% falls
Ankle 69 11 73.1 1.4 1.4 78.3% falls, 15.9% sport
Metacarpus 65 10.3 68.9 1.5 1.5 50.8% db/assault, 26.2% falls
Proximal forearm 53 8.4 56.2 1.9 0 79.2% falls, 10.2% sports
Carpus 19 3 20.1 0 0 84.2% falls, 5.3% sport
Clavicle 17 2.7 18 5.9 0 29.4% MVA, 29.4% falls
Proximal humerus 10 1.6 10.6 0 0 60% falls, 20% sport
Tibial diaphysis 7 1.1 7.4 14.3 14.3 57.1% falls, 28.6% sport
Humeral diaphysis 6 1 6.4 0 0 66.6% falls, 16.6% sport
Midfoot 5 0.8 5.3 25 0 80% falls, 20% sport
Distal humerus 4 0.6 4.2 0 0 75% falls, 2050% stairs/low fall
Calcaneus 4 0.6 4.2 25 0 75% fall height, 25% sport
Forearm diaphysis 4 0.6 4.2 0 0 75% falls, 25% stairs/low falls
Distal tibia 3 0.5 3.2 33.3 0 66.7% fall height, 33.3 falls
Pelvis 2 0.3 2.1 50 0 50% MVA, 50% sport
Proximal tibia 2 0.3 2.1 50 0 50% fall height, 50% falls
Femoral diaphysis 1 0.2 1.1 100 0 100% fall height
Patella 1 0.2 1.1 0 0 100% falls
Fibula 1 0.2 1.1 100 0 100% fall height
Distal femur 1 0.2 1.1 0 0 100% fall height
Talus 1 0.2 1.1 0 0 100% sport
Proximal femur 0 0 0 0 0
Scapula 0 0 0 0 0
627 100 664.4 3.8 2.1 56.2% falls, 18.6% db/assault
 

The numbers, prevalence, and incidence of the different fractures are shown as are the prevalence of open fractures and patients with multiple fractures. The two commonest causes of each fracture are shown. (db = direct blow).

X
 
Table 3-10
Epidemiology of Fractures in Females Aged 36 to 64 Years
No. % n/105/yr Multiple Fractures (%) Open Fractures (%) Causes
Distal radius/ulna 331 25.4 273.7 4.3 0 89.7% falls, 3.9% sport
Ankle 191 14.6 157.9 1.6 1 88.5% falls, 5.8% low falls
Metatarsus 136 10.4 112.5 9.9 0 89.7% falls, 4.4% db/assault
Finger phalanges 122 9.4 100.9 6.7 4.5 41% falls, 39.3% db/assault
Proximal humerus 111 8.5 91.8 5.4 0 88.3% falls, 5.4% low falls
Proximal forearm 100 7.7 82.7 7.1 1 80% falls, 12% MVA
Metacarpus 53 4.1 43.8 12.2 0 45.3% falls, 39.6% db/assault
Proximal femur 45 3.5 37.2 6.7 0 77.8% falls 8.9% pathologic
Carpus 41 3.1 33.9 9.8 0 87.8% falls, 4.9% sport
Toe phalanges 27 2.1 22.3 20 6.1 85.2% db/assault, 14.8% falls
Clavicle 25 1.9 20.7 0 3.7 44% falls, 32% MVA
Fibula 18 1.4 14.9 0 0 77.7% falls, 11.1% MVA
Proximal tibia 16 1.2 13.2 6.2 0 37.5% falls, 18.7% sport
Distal femur 13 1 10.7 18.2 0 100% falls
Pelvis 12 0.9 9.9 25 0 75% falls, 16.6% MVA
Tibial diaphysis 12 0.9 9.9 0 0 66.6% falls, 8.3% MVA
Humeral diaphysis 11 0.8 9.1 0 0 90.9% falls, 9.1% low falls
Calcaneus 9 0.7 9.1 0 0 44.4% falls, 22.2% fall height
Femoral diaphysis 7 0.5 5.8 0 0 85.7% falls, 14.2% sport
Patella 7 0.5 5.8 0 0 85.7% falls, 14.2% low falls
Distal tibia 5 0.4 4.1 0 0 80% falls, 20% fall height
Midfoot 4 0.3 3.3 50 0 50% falls, 25% low falls
Distal humerus 3 0.2 2.5 0 0 100% falls
Forearm diaphysis 2 0.2 1.7 0 0 50% sport, 50% low falls
Scapula 2 0.2 1.7 50 0 50% falls, 50% fall height
Talus 1 0.1 0.8 100 0 100% falls
1,304 100 1,078.3 3.9 0.8 78.7% falls, 9% falls
 

The numbers, prevalence, and incidence of the different fractures are shown as are the prevalence of open fractures and patients with multiple fractures. The two commonest causes of each fracture are shown. (db = direct blow).

X
 
Table 3-11
Epidemiology of Fractures in Females Aged ≥65 Years
View Large
Table 3-11
Epidemiology of Fractures in Females Aged ≥65 Years
No. % n/105/yr Multiple Fractures (%) Open Fractures (%) Causes
Proximal femur 503 28.2 865.2 6.2 0 96.8% falls, 1.8% low falls
Distal radius/ulna 456 25.6 784.3 7.1 1.5 95.6% falls, 2.9% low falls
Proximal humerus 208 11.7 357.8 9.2 0 93.8% falls, 5.3% low falls
Ankle 123 6.9 211.6 5.7 2.4 95.1% falls, 2.4% low falls
Pelvis 69 3.9 118.7 8.7 0 97.1% falls, 2.9% low falls
Metatarsus 68 3.8 117 20 0 91.2% falls, 4.4% low falls
Finger phalanges 59 3.3 101.5 18 3.6 72.9% falls, 15.3% db/assault
Proximal forearm 50 2.8 86 16 4 94% falls, 4% MVA
Metacarpus 39 2.2 67.1 17.6 2.6 92.3% falls, 2.4% low falls
Clavicle 35 2 60.2 5.7 0 91.4% falls, 5.7% MVA
Femoral diaphysis 35 2 60.2 2.9 0 88.6% falls, 5.7% pathologic
Distal humerus 21 1.2 36.1 14.3 0 100% falls
Patella 21 1.2 36.1 0 4.8 95.2% falls, 4.8% db/assault
Humeral diaphysis 20 1.1 34.4 0 5 85% falls, 10% pathologic
Distal femur 16 0.9 27.5 12.5 6.2 81.2% falls, 12.5% low falls
Forearm diaphysis 11 0.6 18.9 9.1 0 90.9% falls, 9.1% pathologic
Proximal tibia 10 0.6 17.2 20 0 70% falls, 20% low falls
Carpus 9 0.5 15.5 11.1 0 88.9% falls, 11.1% db/assault
Scapula 7 0.4 12 14.3 0 100% falls
Distal tibia 6 0.3 10.3 0 0 83.3% falls, 16.6% low falls
Toe phalanges 5 0.3 8.6 0 0 80% falls, 20% db/assault
Calcaneus 4 0.2 6.9 25 0 100% falls
Fibula 3 0.2 5.2 63.3 0 66.6% falls, 33.3% MVA
Midfoot 2 0.1 3.4 0 0 50% fall height, 50% sport
Tibial diaphysis 1 0.06 1.7 0 100 100% falls
Talus 0 0 0 0 0
1,781 100 3,063.3 5 1.2 94.3% falls, 2.9% low falls
 

The numbers, prevalence, and incidence of the different fractures are shown as are the prevalence of open fractures and patients with multiple fractures. The two commonest causes of each fracture are shown. (db = direct blow).

X
In middle-aged females (Table 3-10) the incidence of fractures is slightly higher than that seen in equivalently aged males (Table 3-7). The incidence of fractures in this age group is 162% higher than in the younger female age group, but the incidence of fractures of the distal radius and ulna rises by 275% and the incidence of ankle fractures by 216%. As with the younger female age group, most fractures are low-energy injuries and there is a very low prevalence of open fractures in middle-aged females, although the prevalence of multiple fractures is the same as seen in younger females. 
In females aged ≥65 years (Table 3-11) the incidence of fractures rises by a further 284% to 3,063.3/105/year. Fractures of the proximal femur, distal radius and ulna, proximal humerus, and ankle account for 72.4% of all the fractures and the incidence of pelvic fractures increases by 1,200% compared with middle-aged females. It is interesting to monitor the increasing incidence of fragility fractures between Tables 3-9 and 3-10. The incidence of established fragility fractures of the distal radius and ulna, proximal femur, and proximal humerus rises very quickly, but the increased incidence of fractures of the humerus diaphysis, distal humerus, proximal forearm, femoral diaphysis, distal femur, patella, ankle, and pelvis should also be noted. 
Predictably 94.3% of fractures in older females are caused by falls from a standing height with only 2.8% of fractures in this group of patients being caused by mechanisms other than a standing fall or a fall from a low height. The prevalence of female fractures caused by standing falls rises by approximately 20% in each age group shown in Tables 3-93-11

Social Deprivation

The other factor that undoubtedly affects the incidence of fractures is social deprivation. There is good evidence in the orthopedic literature that social deprivation correlates with musculoskeletal pain,99 high-energy lower limb trauma,69 Perthes disease,83 and outcome after hip arthroplasty.55 There is also evidence that social deprivation is associated with fractures in children,10,91,93 adolescents,75 and young adult males.72 In adults it has been shown to be implicated in fractures of the tibial diaphysis25 and hand,52 but it has become clear that social deprivation is an important factor in determining the incidence of many fractures in adults.27,29,76 
A study of the effect of social deprivation on the incidence of fractures was undertaken using the fractures treated in Edinburgh between July 2007 and June 200829 and analyzed in the seventh edition of Rockwood and Green.26 In Scotland social deprivation is analyzed using the Carstairs score,15 this being a Z-score created from each postcode which is based on overcrowding, male unemployment, household status, and car ownership. The Carstairs score has been successfully used for the analysis of deprivation in many branches of medicine including orthopedic surgery.3,33,38,51 Using the Carstairs score the population can be divided into deciles with decile 1 being the most affluent and decile 10 the least affluent. Decile 10 contains the least affluent 10% of the population. Figure 3-2A shows the distribution of the population of the catchment area of the Royal Infirmary of Edinburgh according to the social deciles and Figure 3-2B shows the incidence of fractures within the different deciles in both males and females. It can be seen that there is a significant difference between the distribution of the population and their fractures. Statistical analysis shows that there is no significant difference in the incidence of fractures in either males or females between deciles 1 and 8, but there is a significant difference in deciles 9 and 10 and it is clear that the effect of deprivation on fracture incidence is seen in the most deprived 10% of the population. 
Figure 3-2
The population numbers in the different social deciles in the Edinburgh catchment area (A) and the fracture numbers in the same area (B).
Rockwood-ch003-image002.png
View Original | Slide (.ppt)
X
Table 3-12 shows the incidence of fractures when deprivation is taken into account. Once the figures are adjusted for age it can be seen that in males the overall incidence of fractures in the very socially deprived is about 4 times that of the rest of the population and in females the equivalent figure is about 3.5. The difference in incidences is statistically significant for both genders. Table 3-12 also shows which individual fractures show correlation between fracture incidence and significant deprivation. 
Table 3-12
The Effect of Social Deprivation on the Incidence of Fractures
Fracture Incidence (n/105/yr)
Males (1–8) Males (9–10) Females (1–8) Females (9–10)
Distal radius/ulna 120.7 475.3 254.3 749.8a
Metacarpus 188.1 748.8a 44.1 153.1a
Proximal femur 80.8 219.7a 194.7 451.4a
Ankle 105.5 327.3a 103.3 345.5a
Finger phalanges 159.6 457.3a 78 219.8a
Proximal humerus 54.6 197.3 106.9 341.5a
Metatarsus 61.3 139a 78.8 251.2a
Proximal forearm 58 206.3a 58.4 172.7a
Clavicle 71.3 228.7a 24.5 78.5a
Toe phalanges 23.8 94.2a 18.8 47.1a
Carpus 48.5 179.3a 16.7 78.5a
Pelvis 18.5 58.3a 22.9 82.4
Femoral diaphysis 14.7 35.8a 13.9 66.7
Humeral diaphysis 11.9 44.8a 10.2 31.4
Tibial diaphysis 21.8 67.3a 5.7 7.8
Calcaneus 13.3 62.8a 4.1 7.8
Proximal tibia 8.1 89.7a 13.9 39.3a
Forearm diaphyses 20 62.8a 4.9 15.7
Patella 6.2 22.4 11 39.3
Distal humerus 4.3 40.4a 11.4 19.6
Distal tibia 12.8 35.9 6.5 11.8
Scapula 6.2 44.8a 8.6 31.4a
Distal femur 5.7 9a 7.3 31.4a
Midfoot 4.3 35.9 5.3 27.5a
Talus 7.6 17.9 4.1 7.8
1,078 3,918.9a 1,096.9 3,317.1a
 

The population has been divided into deciles 1–8 and deciles 9–10 (see text).

X
The commonest fracture types in deprived males are those of the metacarpal, distal radius, and ulna and finger phalanges which constitute 55% of all fractures in the deprived population. This compares with 43% of the more affluent population. Table 3-12 shows that the incidence of hand fractures is significantly higher in the deprived group than in the less deprived group, but it is salutary to note that while fractures of the hand and carpus constitute 35% of the more deprived group, they still constitute 30% of the less deprived population. Presumably testosterone and alcohol are an issue in all sectors of the male population! 
It should be noted that proximal femoral fractures in males are only the sixth most common fracture in the socially deprived and are in fact less common than clavicle fractures. Further analysis shows that 15.3% of the most affluent male population in Edinburgh presented with proximal femoral fractures compared with only 8.6% of the least affluent. Analysis also shows that deprived patients present at a younger age and have a shorter life expectancy. It would seem that even in an affluent city such at Edinburgh, that many of the older socially deprived males do not live long enough to have a proximal femoral fracture.27 
In females the incidence of distal radius and ulna fractures and proximal femoral fractures rose by 295% and 232%, respectively in the very deprived and Table 3-12 shows a similar increase in other fractures. The biggest increase in incidence in female fractures is in fact in the femoral diaphyseal fracture where there was a 480% increase in incidence in the very deprived population. The reason for this considerable increase is unknown. In males the proximal tibial fracture showed a rise of 1,107% with distal humeral fractures, midfoot fractures, and scapula fractures showing increases of 940%, 835%, and 723%, respectively. It would seem reasonable to assume that the overall effect of social deprivation is caused by a number of medical and social comorbidities which affect both males and females and will cause an increased rate of fractures, but in males more aggressive behavior may account for the greater difference in the incidence of a number of fractures. 
There is evidence that a number of diseases are related to social deprivation and it has been shown that fracture incidence is affected by factors such as a rural or urban domicile,74,88 education,50 occupation,35 type of residence,39 marital status,39 and smoking and alcohol.4 There is also evidence that bone mineral density (BMD) is affected by social deprivation.9 In recent years there has been interest in the effect of ethnicity on the incidence of fractures17,94 and a number of authors have pointed to the different incidence of proximal femoral fractures in particular in different parts of the world.30,59,94 It has been pointed out that in the United States, African American and Hispanic males who present with hip fractures are younger than White males.94 Pressley et al.84 drew attention to the paradox of African American males in the United States having a higher BMD, but also a higher incidence of fractures, than White males. It is likely that these findings relate to deprivation. Ethnicity is difficult to study in Scotland, but in other areas of medicine the relationship between ethnicity and deprivation has been shown14,98 and it would seem likely to be important in the epidemiology of fractures. 

Fracture Distribution Curves

The earliest fracture distribution curves, based on age and gender, were proposed by Buhr and Cooke.20 They analyzed 8,539 fractures over a 5-year period in Oxford, England, and proposed five basic curves. Their type A curve affected young and middle-aged men and they referred to it as a “wage earners” curve. This is equivalent to our Type B curve (Fig. 3-3). They suggested that this occurred in patients who presented with fractures of the hand, medial malleolus, metatarsus, foot phalanges, and spine. Their J-shaped curve affected older males and females and obviously it described fragility, or osteoporotic, fractures. It is equivalent to our type F curve (Fig. 3-3). They stated that fractures of the proximal humerus, humeral diaphysis, proximal femur, and pelvis together with bimalleolar ankle fractures had a J-shaped curve. Buhr and Cooke’s third curve was an L-shaped curve that affected younger males and females and was equivalent to our type C curve (Fig. 3-3). This was said to occur in distal humeral fractures, tibial diaphyseal fractures, and clavicular fractures. They also described two composite curves with either a bimodal male and unimodal female distribution or a unimodal male and bimodal female distribution. These are equivalent to our type D and G curves (Fig. 3-3). They said that these curves described fractures of the proximal and distal radius, femoral diaphysis, proximal tibia and fibula, and the lateral malleolus. 
Figure 3-3
The eight fracture distribution curves.
 
See Table 3-13 for list of distribution curves for different fractures.
See Table 3-13 for list of distribution curves for different fractures.
View Original | Slide (.ppt)
Figure 3-3
The eight fracture distribution curves.
See Table 3-13 for list of distribution curves for different fractures.
See Table 3-13 for list of distribution curves for different fractures.
View Original | Slide (.ppt)
X
Later studies produced similar distribution curves. Knowelden et al.64 analyzed patients in Dundee, Scotland and Oxford, England who were >35 years of age. They showed that fractures of the proximal humerus, pelvis, and proximal femur all demonstrated an osteoporotic type F curve (Fig. 3-3). It is interesting to note that they had a type A curve (Fig. 3-3) for femoral diaphyseal fractures, but they recorded that the highest incidence of femoral diaphyseal fractures occurred in the elderly. Donaldson et al.31 constructed four curves for proximal femoral, proximal humeral, distal radial, and tibia and fibula diaphyseal fractures that are very similar to the curves shown in Figure 3-3. Johansen et al.56 constructed eight curves covering different body areas, these being the hip, spine, upper limb, pelvis, forearm and wrist, ankle, hand, finger and thumb, and foot and toes. These are also very similar to the curves shown in Figure 3-3
Analysis of incidence in different fracture types shows that there are eight basic fracture distribution curves which are shown in Figure 3-3. Most fractures have a unimodal distribution affecting either younger or older patients. Some fractures, however, have a bimodal distribution whereby younger and older patients are affected, but there is a lower incidence in middle age. If one analyzes males and females separately the distribution curves shown in Figure 3-3 can be constructed. It should be remembered that the curves shown in Figure 3-3 are diagrammatic. The relative heights of the peaks of the curves will vary, but the overall curve patterns remain appropriate for all fractures. 
A type A curve is often thought of as a typical fracture curve with a unimodal distribution in younger males and in older females. Generally the younger male peak is higher than the older female peak, although this is not the case in all fractures. An example is the metatarsal fracture where the younger male peak and the older female peak are at a similar height. Type A curves are seen in fractures of the scapula, distal radius, tibial diaphysis, ankle, and metatarsus. In type B curves there is also a young male unimodal distribution, but fractures in females occur in relatively small numbers throughout the decades. Type B curves are generally seen in the hand and affect the carpus, metacarpus, and fingers. However, they are also characteristic of femoral head fractures. 
In type C fractures both males and females show a unimodal distribution. These fractures are rare after middle age. These fractures tend to occur in the foot and affect the toes, midfoot, and talus. In type D fractures there is a young male unimodal distribution, but the female distribution is bimodal affecting younger and older females. Generally the second peak starts around the time of the menopause. Type D curves are seen in fractures of the proximal forearm, forearm diaphyses, and tibial plafond. 
Type E fractures are the opposite of type B fractures. They show a unimodal female distribution affecting older females with a relatively constant, lower incidence of fractures in males throughout the decades. The type E pattern is seen in pelvic fractures, distal humeral fractures, and distal femoral fractures. This may be surprising to orthopedic surgeons who see young male patients with these fractures after high-energy trauma. However, if the complete epidemiology of these fractures is analyzed across the community it is apparent that the high-energy injuries are relatively rare compared with the lower-energy injuries seen in later life. 
Type F fractures are the opposite of type C fractures. In type F fractures both males and females show a unimodal distribution affecting older patients with the incidence being higher in females. This pattern is characteristic of fractures of the proximal humerus, proximal femur, and patella. There is some variation regarding when the rise in fracture incidence occurs. Generally it is earlier in females than males and usually occurs around the time of the menopause in proximal humeral fractures and patella fractures, but somewhat later in proximal femoral fractures. 
In type G fractures females show a unimodal distribution affecting older females and males show a bimodal distribution affecting both younger and older males with the incidence being higher in younger males. This distribution is seen in calcaneal and clavicular fractures. It is also now seen in femoral diaphyseal fractures. Type H fractures are unusual in that both males and females show a bimodal distribution. This fracture pattern is seen in fractures of the humeral diaphysis, tibial plateau, and cervical spine. 
One can use the system of eight curves shown in Figure 3-3 to define other fractures. Although Table 3-13 shows that ankle fractures have a type A distribution, analysis of the different types of ankle fractures shows that only lateral malleolar fractures have a type A distribution. Medial malleolar fractures have a type D distribution and suprasyndesmotic ankle fractures have a type C distribution. Both bimalleolar and trimalleolar fractures are fragility fractures showing a type E distribution. Similarly, proximal forearm fractures have a type D distribution when they are all considered together, but further analysis shows that radial neck fractures have a type A distribution, whereas radial head fractures have a type H distribution. Both olecranon fractures and fractures of the proximal radius and ulnar have a type F distribution and should be regarded as fragility fractures. The distribution curves for different fractures are listed in Table 3-13 which also shows the distribution curves for the different fracture types. 
 
Table 3-13
The Distribution Curves Shown in Figure 3-3 Applied to Different Fractures
View Large
Table 3-13
The Distribution Curves Shown in Figure 3-3 Applied to Different Fractures
Fracture Location
Clavicle G
 Medial A
 Diaphyseal G
 Lateral A
Scapula A
 Intra-articular A
 Extra-articular A
Proximal humerus F
Humeral diaphysis H
Distal humerus E
Proximal forearm D
 Radial head H
 Radial neck A
 Olecranon F
 Radius and ulna F
Forearm diaphyses D
 Radius A
 Ulna H
 Radius and ulna A
Distal radius/ulna A
Distal ulna A
Carpus A
 Scaphoid B
 Triquetrum A
 Hamate B
 Trapezium B
Metacarpus B
Finger phalanges B
Pelvis E
Acetabulum G
Proximal femur F
 Head B
 Neck F
 Intertrochanteric F
 Subtrochanteric F
Femoral diaphysis G
Distal femur E
Patella F
Proximal tibia H
Tibia and fibular diaphysis A
Tibial diaphysis B
Fibular diaphysis A
Distal tibia D
Ankle A
 Medial malleolus D
 Lateral malleolus A
 Bimalleolar E
 Trimalleolar E
 Suprasyndesmotic C
Talus C
 Neck C
 Body C
Calcaneus G
 Intra-articular B
 Extra-articular G
Midfoot C
Metatarsus A
Toe phalanges C
Cervical spine H
Thoracolumbar spine F
Fracture Types
Periprosthetic F
Open G
Multiple A
Fatigue C
Insufficiency F
 

The curves of different fracture types are also shown. In this section the term “Multiple” applies to multiple fractures and not to multiple injuries.

X

Changing Epidemiology

There is no doubt that fracture epidemiology is changing very quickly. This is due to multiple factors which reflect a massive change in the health and economic status of many countries. There has been a great deal of literature dealing with the increasing frequency of fragility fractures which is thought to be secondary to the improved health and longer life expectancies of the older members of the population.5,48,57,61,62,73 However, the change in fracture epidemiology is much broader than this and reflects the important industrial and road safety legislation introduced in many countries since the Second World War. The changes are well illustrated in Table 3-1 and by comparing our epidemiology for 2010 and 2011 with that of Buhr and Cooke13 who analyzed over 8,500 patients, between 1938 and 1956, in Oxford, England. They prefaced their paper by pointing out the changes in health they had encountered. They stated that smallpox, diphtheria, the enteric fevers, and rickets had practically been eliminated and had been replaced by new viral diseases, radiation hazards, and the diseases of degeneration. They noted that in the elderly population cardiovascular degeneration, strokes, diabetes, osteoarthritis, and fractures presented problems as pressing as the great infections of a few decades earlier. Buhr and Cooke13 clearly recognized the problems that osteoporosis presented in the 1940s and 1950s. They pointed out the prevalence of proximal femoral fractures, particularly in women, and they noted that in nonmalignant pathologic fractures a third were caused by “senile osteoporosis.” 
However, when one examines their results it is clear that the situation has changed dramatically. They defined five fracture distribution curves. Their type A, wage earners, curve was the equivalent of our type B curve affecting younger males. Their type L, prewage earners, curve was the equivalent of our type C curve affecting younger males and females, and their type J, postwage earners, curve is the same as our type F curve affecting older males and females. They also stated that there were two composite curves with bimodal distributions affecting males and females. These were the same as our type D and H curves. 
They categorized 22 different fractures and a comparison of their distributions with ours indicates considerable social and medical differences between the two time periods. They classified carpal, metacarpal, finger phalangeal, medial malleolar, tarsal, metatarsal, toe phalangeal, and spinal fractures as being type B fractures. Table 3-13 shows that the intervening 50 to 60 years has altered things considerably and only metacarpal and finger phalangeal fractures still have a type B distribution. Surgeons now see many more females with the other fractures. 
Their type F osteoporotic fractures were similar to ours. They felt that the humeral diaphyseal fracture was an osteoporotic fracture and if one compares the epidemiology of humeral diaphyseal fractures with distal radial and ulnar fractures shown in Table 3-3 it would seem that they were correct. They listed distal humeral and clavicle fractures as type C fractures and again a review of Table 3-11 shows that surgeons now expect to see many older women with these fractures. Thus the main difference between the two time periods is the numbers of fractures in females that are now seen, but were not seen in the 1940s and 1950s. This must reflect the changing role of women in society and the fact that successful medical and surgical treatment, including joint arthroplasty, now allows older women to sustain fractures that they could not have sustained in the past! 
Other studies indicate that the increased incidence of fractures that is now seen is mainly because of a significant increase in the incidence of fractures in females. In a study carried out between 1954 and 1958 in Dundee, Scotland, and Oxford, England, Knowelden et al.64 examined the incidence of fractures in patients aged >35 years and found that the incidences of the fractures listed in Table 3-3 in males and females in Dundee were 1,017.3/105/year and 921.3/105/year, respectively with the equivalent figures in Oxford being 811.4/105/year and 871.5/105/year. The equivalent incidences in Edinburgh in 2010 to 2011 were 1,062.5/105/year and 1,711.6/105/year. Thus in 50 to 55 years the incidence of fractures in Scottish males has risen by 4.4%, whereas the incidence of fractures in females has risen by 85.7%. The increased incidence in males is less but the spectrum of fractures in males has changed considerably over this period. There are now many fewer industrial fractures and many more fall-related fractures and consequently there are less hand and foot fractures in males. 
The fact that the epidemiology of fractures has continued to change in recent years is illustrated by reference to Table 3-14. This shows the incidence, average age, and the prevalence of standing falls as a cause of fracture in six fractures treated in Edinburgh in three time periods. The results are taken from patients ≥15 years in three study years, over a 20-year period, using prospectively collected data from the same catchment area. Four of the fractures are the classic fragility fractures of the proximal humerus, distal radius and ulna, pelvis and proximal femur, and the other two fractures are the principal diaphyseal fractures of the lower limb, the femoral and tibial diaphyseal fractures. Table 3-14 shows that there are considerable differences in the epidemiology of all the fractures, but the changes vary between different fractures. It is quite clear that the overall incidence of proximal humeral and distal radial fractures has risen over the last 20 years in both males and females. This is not the case in pelvic fractures or in the overall incidence of proximal femoral fractures. However, Table 3-14 shows a marked increase in the incidence of proximal femoral fractures in males in the last 20 years, presumably because more males are living longer. 
Table 3-14
The Incidence, Average Age, and Prevalence of Fractures in Patients ≥15 Years Caused by Standing Falls for Six Common Fractures Treated in Three Time Periods
Incidence (n/105/yr) Average Age (yrs) Standing Falls (%)
Overall Male Female Male Female Male Female
Proximal Humerus
1993 47.2 28.7 63.8 56 69.9 76.1 89.2
2000 65.1 41.4 86.3 57.4 68 71.3 84.7
2010/11 92.4 61 120.3 59.2 68.9 73.1 90.9
Distal Radius/ulna
1991 158.3 87.3 221.5 42 64.2 54.4 83.5
2000 201.5 131.5 264 38 63.3 39.2 81.2
2010/11 235.9 139.3 322.2 43.7 63.4 50.6 91.3
Pelvis
1991 21.6 16.3 26.3 46 73.6 28.9 73.9
2000 17.6 11.1 23.4 50.4 77.8 29.6 84.4
2010/11 22.9 14.7 30.4 64.7 80.4 56.4 91.6
Proximal Femur
1991 143.8 57.4 220.8 71.9 80.2 86.6 91.2
2000 133.7 74.1 208.8 74.5 82.6 83.4 95.3
2010/11 145.5 84 200.4 78 81.1 88.3 95.3
Femur Diaphysis
1991 8.9 8.6 9.2 39.5 62 25 54.2
2000 10.4 7.8 12.8 35.9 78.4 35 80
2010/11 8.3 8.6 8 63.4 75.6 42.8 86.4
Tibia Diaphysis
1991 24.4 37.2 13 32.8 60.7 16.1 52.9
2000 18.5 24.6 11 35.8 49.5 18.3 50
2010/11 13.3 20.1 7.3 41 43.6 36.7 65
X
The comparison of the incidences of femoral and tibial diaphyseal fractures shows considerable differences in epidemiology. The femoral diaphyseal fracture has not changed in incidence, but the incidence of tibial diaphyseal fractures has decreased markedly in the last 20 years. The overall incidence has declined from 24.4/105/year to 13.3/105/year with a decline seen in both males and females. The difference is probably because the femoral diaphyseal fracture is essentially a fragility fracture, whereas the tibial diaphyseal fracture is not. 
Examination of the age of the patients and the prevalence of fractures caused by a standing fall shows that there has been no major differences in proximal humeral or distal radial fractures in the last 20 years. However, the average age of patients presenting with pelvic fractures is rising in both males and females, although the rise in male age is more dramatic. This is because there are now fewer high-energy pelvic fractures and more lower-energy fractures in a male population which is living longer. This would seem to be confirmed by the increased number of pelvic fractures caused by standing falls. The average age of males presenting with proximal femoral fractures is also increasing, although no other differences were noted. 
The differences noted in pelvic fractures are mirrored in the femoral diaphyseal fracture where there is clearly an increasing average age and a higher prevalence of fractures caused by standing falls. The same is seen in the tibial diaphyseal fracture, although this fracture is unique in that the average age of females presenting with tibial diaphyseal fractures is falling. 
The tibial diaphyseal fracture is a very good example of the changing epidemiology of a nonfragility fracture. A review of the incidences of tibial diaphyseal fractures in Europe at different times shows that there has been a considerable decline. The literature is complicated by the fact that different age ranges of patients are often assessed and that children and adolescents are sometimes included. However, by calculating the incidences of equivalent age groups trends can be shown. Knowelden et al.64 demonstrated an incidence of tibial diaphyseal fractures of 17.3/105/year in patients aged >35 years in Dundee, Scotland, in 1954 to 1958. The equivalent incidence in Edinburgh in 1991 was 18/105/year and in 2010 to 2011 it was 12.3/105/year. Similar trends are also shown by examining the Swedish literature.36,103 Emami et al.36 compared the incidence of tibial diaphyseal fractures in 1971 to 1975 and 1986 to 1990. Their incidences in patients ≥20 years of age were 40.6/105/year and 31.9/105/year, respectively. In Edinburgh in 1991 the incidence of tibial diaphyseal fractures in patients aged ≥20 years was 24.7/105/year. It is therefore clear that the incidence of tibial diaphyseal fractures has fallen since the Second World War and continues to fall. 
A review of the epidemiology of tibial diaphyseal fractures in Edinburgh between 1990 and 2007 has shown a progressive decline in both males and females. In males the incidence fell from 43.6/105/year to 25/105/year and in females it fell from 15.8/105/year to 6.2/105/year. Analysis showed that there was a statistically significant decline in incidence in males aged 15 to 34 years and in both males and females ≥65 years. A review of the open fractures presenting between 1990 and 2007 also showed a statistically significant decline in incidence in both males and females with the greatest decline in incidence being seen in females ≥65 years. There was no decrease in incidence in Gustilo type I fractures, but there was in Gustilo type II and Gustilo type III fractures, with Gustilo type III fractures showing the greatest decrease in incidence. 
A review of the causes of tibial diaphyseal fractures showed that there was a decline in standing fall–related fractures in females ≥65 years and in both sports-related fractures and motor vehicle accident fractures. The data showed that there was a significant decline in pedestrian tibial fractures in males aged 35 to 64 years and in females ≥65 years. 
The various studies that have been quoted show that there has been a significant decline in tibial diaphyseal fractures since the Second World War.36,103 Initially this was presumably related to industrial and workplace safety legislation, but more recently much of the decline must be due to a decline in motor vehicle accidents associated fractures. However, there is also a decline in sports-related tibial fractures, and in young males, the overall decline may simply relate to a more sedentary lifestyle. In older women the reduction in pedestrian fractures is presumably related to the reduction in age shown in Table 3-14. Tibial diaphyseal fractures are unusual in that there is a declining incidence in elderly women and presumably this relates to the fact that tibial fractures are not osteoporotic fractures and the causes of these fractures are changing. 
The declining incidences of different fractures have had an effect on the distribution curves. Buhr and Cooke13 defined tibial fractures as having a type C curve, but as females became more affected, it was changed to a type A curve. Tables 3-93-11 and Table 3-14 show a decline in elderly females and the distribution curve appears to be continuing to change and it may be that in the future a new curve will be required with a bimodal distribution in males and a unimodal distribution in younger women only. 
Much has been written about the epidemiology of fragility fractures in the last 10 to 20 years. The assumption is that fragility fractures are increasing in incidence, but it is surprisingly difficult to know if this is actually the case and, if so, is it true for all fragility fractures or only for some. A good example of the confusion is seen in the proximal humeral fracture literature. There is no question of doubt that these have increased in frequency since the Second World War. Knowelden et al.,64 in their analysis of patients >35 years in Dundee, Scotland in 1954 to 1958 quoted an incidence of 44/105/year (32.8/105/year in males and 52.2/105/year in females). Table 3-14 shows the incidence of proximal humeral fractures in the population of Edinburgh over an 18-year period, but if patients >35 years are analyzed, the incidence of proximal humeral fractures in 1993 is 69.5/105/year (39.9/105/year in males and 94.1/105/year in females). In 2010 to 2011 the incidence was 136.6/105/year (88/105/year in males and 178.2/105/year in females). There has therefore been a progressive rise in the incidence of proximal humeral fractures in Scotland in the last 55 to 60 years. A review of the Swedish literature also shows a rise in incidence between 1950 and 1982,5 although the incidence in the early 1950s was slightly higher than recorded by Knowelden et al.64 However, Kannus et al.62 in a study of proximal humeral fractures in females aged ≥80 years in Finland between 1970 and 2007, showed that the incidence in this age group was 88/105/year in 1970 and 304/105/year in 1995 but there was no further rise and in 2007 the incidence was 298/105/year. It is interesting to note that the comparative figures for Edinburgh females ≥80 years were 285.5/105/year in 1993 and 497.7/105/year in 2010 to 2011. Thus our incidence in the 1990s was not dissimilar to the Finnish incidence, but our incidence of proximal humeral fractures continued to rise. Given the similarities between Finland and Scotland there is no obvious explanation for this difference but it seems likely that the incidence is rising. 
The fracture that has received most attention in the epidemiologic literature is the proximal femoral fracture. There is no doubt that this fracture has also increased in incidence, but there is considerable variation in its current incidence throughout the world and there is also debate as to whether the incidence of proximal femoral fractures is now declining in developed countries.19 The earliest incidences that are available are from Rochester, Minnesota in the United States where Melton et al.73 studied the incidences of proximal femoral fractures in six time periods between 1928 and 1992. They looked at the incidences of proximal femoral fractures in the whole population and found that in the period 1928 to 1942 the incidences in males and females were 17.3/105/year and 46.9/105/year. They documented a rise in incidence in both males and females until 1963 to 1972 when the incidences were 69.3/105/year and 125/105/year, respectively. Thereafter the male incidence continued to rise until 1983 to 1992 when it was 82.2/105/year, but the female incidence plateaued so that it was 115.2/105/year in 1983 to 1992. However, Melton et al. also published the incidences of proximal femoral fractures in Olmsted County, Minnesota74 and they stated that in males and females ≥35 years of age in 1989 to 1991 the incidences were 142/105/year and 219/105/year, respectively. The incidences appear to be different, but it may be that during the 1983 to 1992 period the incidences were rising. Knowelden et al.64 stated that in 1954 to 1958 in Dundee, Scotland, the incidences of proximal femoral fractures in males and females >35 years were 43.4/105/year and 105.4/105/year, these being lower than the incidences in Rochester, Minnesota in a similar period. The current incidences of proximal femoral fractures in Edinburgh in males and females >35 years are 131.7/105/year and 233.4/105/year, respectively. This is about the same as in Olmsted County in 1989 to 1991. It is difficult to know why there should be a difference, but it may reflect better male health in Olmsted County. 
Table 3-15 shows the incidence of proximal femoral fractures in different parts of the world at different times.2,6,66,86 All of the studies looked at patients ≥50 years of age and all were carried out in Caucasian populations or were age adjusted for the Caucasian US population. The results encapsulate many of the problems that exist in defining hip fracture incidence. The Scandinavian studies6,86 both show much higher incidences in males and females than in other parts of the world, but the two studies disagree about whether the fracture incidence in declining in Sweden. The incidence in Japanese males is very low2 compared with the incidence in females and in general the incidence in males is highest in North Europe.6,86 
Table 3-15
The Incidence of Proximal Femoral Fractures as Reported in Different Parts of the World
Incidence (n/105/yr)
Country Males Females Comments
Sweden6
1993–1996 390 706 Declining incidence except in females ≥90 yrs
2001–2005 317 625
Sweden86
1987 (rural) 710 Females only
1987 (urban) 750 No change in incidence
2002 (rural) 600
2002 (urban) 690
Hong Kong66 180 459 All 1997–1998
Singapore66 164 442 Age adjusted for US population
Malaysia66 88 218
Thailand66 114 269
Japan2
1987–1988 59.2 245 Age adjusted for US population
2004 115.2 453.7
Scotland
1991 134.5 514.3
2010–2011 224.7 494.4
 

All studies2,6,66,86 have reported on patients aged ≥50 yrs and all were carried out in Caucasian populations or age adjusted for the Caucasian US population.

X
There are many other studies of hip fracture incidence, but the age ranges are often different. Chang et al.18 showed that the incidence of proximal femoral fractures in Australia in patients ≥60 years was 329/105/year in males and 759/105/year in females in 1989 to 2000, which is not dissimilar to the Swedish results in Table 3-15. Kanis et al.59 undertook a systematic review of hip fracture incidence worldwide and standardized all the results with United Nations age data. They found a very wide difference in incidence from approximately 20/105/year in South Africa to approximately 575/105/year in Denmark. It is highly likely that in many countries data collection is poor, but the paper certainly highlights huge differences in the incidence of this fracture. 
Clearly there are many other fractures in which the incidence is probably changing quite quickly, but these fractures serve to highlight the enormous changes in fracture epidemiology that have occurred in the last 50 to 60 years. There is no logical reason why the changing epidemiology should not continue in the future and surgeons should be aware that a number of fractures which are now not regarded as fragility fractures will probably be so in the next few decades. 

Variation in Epidemiology

It has already been pointed out that the epidemiology of fractures varies widely. Some of the variations are undoubtedly accounted for by the different methods used to collect and to diagnose fractures. However, despite this there are significant differences in the incidence of fractures in different communities. These differences have mainly been studied in fragility fractures and the literature is consistent in pointing out that the population of Scandinavia6,12,59,61,86 has the highest incidence of these fractures. The reason for this is unknown. However there is evidence that the incidence of fractures varies with racial type,17,97,101 domicile,46,48 season of the year,54 and social deprivation.25,75 The importance of social deprivation has already been discussed in this chapter, but clearly the reason for the variation in epidemiology is more complex. Scandinavian countries are relatively affluent and one would expect a lower incidence of fractures in more affluent countries. However, other factors such as life expectancy will clearly play a part in the epidemiology of fragility fractures. It seems likely that social deprivation accounts for some of the variation in fracture epidemiology attributed to ethnicity, particularly in the United States. Pressly et al.84 pointed out the apparent contradiction of young Black males in the United States having a higher BMD, but also having a higher incidence of fracture. This is likely to represent deprivation. In a recent study Cauley17 pointed out that despite increased bone density, Black women in the United States were more likely to die after hip fracture, had longer hospital stays and were less likely to be ambulatory at discharge from hospital. 
It has been suggested that there are racial differences in fracture incidences. This has mainly been studied in the Far East with Wang and Seeman101 suggesting that in the Chinese population bone cortices are thicker and there is more mineralized bone matrix. However, it seems likely that racial differences, like most things in medicine, are multifactorial and will involve life expectancy, deprivation, and other social and medical comorbidities. 

Open Fractures

In this study year 1.9% of the fractures were open. Further analysis shows that 66% of the fractures were Gustilo47 type I fractures, 19.7% were Gustilo type II fractures, and the remaining 13.6% were Gustilo type III fractures. A review of the previous two editions of Rockwood and Green24,26 show that in 2000 3.1% of the fractures were open, while in 2007 to 2008 2.6% were open. There also seems to have been a decline in the prevalence of Gustilo type III fractures with 22.8% being recorded in 2000, 19.9% in 2007 to 2008 and 13.6% in 2010 to 2011. It therefore seems likely that there is trend toward fewer and less severe open fractures. This would seem to be confirmed by examining the incidence of open tibial fractures in Edinburgh over the last 20 years. In 1991 34.7% of the tibial fractures were open giving an incidence of open tibial fractures of 8.5/105/year (13.8/105/year in males and 3.8/105/year in females). In 2010 to 2011 20.3% of the tibial fractures were open the incidence being 2.7/105/year (4.9/105/year in males and 0.7/105/year in females). In 1991 42.9% of the open tibial fractures were Gustilo type III but by 2010 to 2011 this had fallen to 21.4%. 
Tables 3-63-11 show the prevalence of open fractures in males and females of different ages, but given the relative infrequency of open fractures, particularly in older patients, it is difficult to analyze them meaningfully. For this reason all of the open fractures presenting to the Royal Infirmary of Edinburgh over a 15-year period between 1995 and 200928 have been analyzed. As many open fractures are associated with more severe injury, the injury severity score (ISS) for each patient was analyzed, in addition to creating a musculoskeletal index (MSI) which is the sum of all fractures and severe soft tissue injuries such as ligament disruption, dislocations, nerve damage, vascular damage, and tendon injury. All were given the score of one and the total used to provide an assessment of the degree of musculoskeletal injury. 
In the 15 years of the study 2,386 open fractures were treated, giving an incidence of 30.7/105/year. They occurred in 2,206 patients with 2,079 (94.2%) presenting with a single open fracture and a further 127 (5.8%) presenting with between two and seven open fractures. The average age was 45.5 years. Analysis showed that 69.1% of the fractures occurred in males with an average age of 40.8 years and 30.9% occurred in females with an average age of 56 years. In males 10.2% of the fractures occurred in patients aged ≥65 years and 2.1% in patients aged ≥80 years. The equivalent figures for females were 42.9% and 18.6%, respectively. 
The overall male and female fracture distribution curves for open fractures are different from those for all fractures shown in Figure 3-1. The curves for open fractures are shown in Figure 3-4. This shows that in adult males the highest incidence of open fractures occurs between 15 and 19 years and that there is an almost linear decline with increasing age. The incidence of open fractures in males aged 15 to 19 years was 54.5/105/year compared with 23.3/105/year in the 90+ age group. In females there is a unimodal distribution rising from 9.2/105/year in the 15–19-year group to 14.6/105/year in the 50–59-year group. Thereafter there is a rapid rise in incidence to 53/105/year in the 80–89-year group. There were insufficient fracture numbers to calculate fracture curves for open fractures of the scapula, proximal radius, radial diaphysis, carpus, and proximal femur. Table 3-16 shows the fracture distribution curves for the different open fractures. It is clear that the overall fracture distribution is different from that of closed fractures. In open fractures it is younger males who are most affected and not infrequently they sustain their open fractures as a result of high-energy injuries. In females Figure 3-4 shows that the open fracture distribution curve is not dissimilar to the overall fracture distribution curve of females (Fig. 3-1), but there is one significant difference. The overall female distribution curve shows a marked increase in incidence in the 50–59-year group in the postmenopausal period. In open fractures this increase occurs one decade later and is therefore probably not just related to osteoporosis, but to increasing overall patient frailty which affects the soft tissues as well as bone. It is also true that the ability to avoid dangerous situations is compromised in older patients. A review of the open fracture distribution curves shown in Table 3-16 indicates that in open fractures which are commonly caused by high-energy injuries such as fractures of the pelvis, femoral diaphysis, distal femur, patella, proximal tibia, distal humerus, and proximal ulna the distribution changes to one which highlights the increased frequency of open fractures in younger patients. Thus the distribution curve for femoral diaphyseal fractures changes from a type G curve to a type B curve showing that open femoral fractures are predominantly seen in young men. Other fractures such as those of the distal femur, patella and proximal tibia change from a curve showing a unimodal distribution affecting older patients to a bimodal distribution where younger patients are affected more commonly. In lower-energy open fractures it can be seen that there are a number of changes as more elderly patients are affected. Thus fractures of the metacarpals and finger phalanges change from a unimodal curve affecting younger patients to a bimodal curve where elderly females are also affected. In both distal radial and ankle fractures there is a change from a type A curve in all fractures to a type E curve in open fractures emphasizing the frequency of open fractures in elderly females. Further analysis shows that 73.3% of open distal radial fractures and 20.8% of open ankle fractures occur in females aged ≥80 years. There are a number of fractures in which the distribution curve does not change. These tend to be fractures which mainly occur in young patients anyway. There is no change in the curves for the different foot fractures, although whilst closed talar fractures affect both young males and females, the open talar fracture seems mainly to occur in young males. 
Figure 3-4
The age and gender distribution curves for open fractures.
Rockwood-ch003-image004.png
View Original | Slide (.ppt)
X
 
Table 3-16
Fracture Distribution Curves for Different Open Fractures
Fracture Distribution Curves
Upper Limb Axial Skeleton and Lower Limb
All Fractures Open Fractures All Fractures Open Fractures
Clavicle G C Pelvis E C
Proximal humerus F H Femoral diaphysis G B
Humeral diaphysis H F Distal femur E A
Distal humerus E G Patella F A
Proximal ulna F H Proximal tibia H A
Ulna diaphysis H D Tibia and fibular diaphyses A A
Radius and ulna diaphyses A G Distal tibia D D
Distal radius and ulna A E Ankle A E
Metacarpus B A Talus C C
Finger phalanges B A Calcaneus G G
Midfoot C B
Metatarsus A A
Toe phalanges C C
Modes of injury (Open fractures)
Crush injuries A Direct blows or assaults B
Falls from standing height F Falls from height C
Road traffic accidents G Falls down stairs F
Cutting injuries B Sport C
 

The overall distribution curves are included for comparison. Distribution curves for the different modes of injuries that caused open fractures are also shown.

X
The basic epidemiologic data for all open fractures treated between 1995 and 2009 is shown in Table 3-17. This shows that almost half of all open fractures involve the fingers and that open fractures of the fingers, tibial diaphysis, distal radius, toes, and ankle accounted for about three quarters of all open fractures. Table 3-17 also shows that a number of open fractures are very rare with ten of the open fractures averaging less than one per year in a very busy trauma unit. In 15 years there were no open proximal radial fractures. 
 
Table 3-17
The Epidemiology of Open Fractures
View Large
Table 3-17
The Epidemiology of Open Fractures
No. % Age (yr) ≥65 yrs (%) ≥80 yrs (%) M/F
Finger phalanges 1,090 45.7 43.9 13.4 4.2 79/21
Tibial diaphysis 267 11.2 43.3 18 6.7 67/33
Distal radius 184 7.7 67 67.4 30.4 23/77
Toe phalanges 170 7.1 41.9 11.8 1.8 66/34
Ankle 126 5.3 56.7 42.9 14.3 43/57
Metacarpus 104 4.4 34.8 7.7 4.8 90/10
Proximal ulna 51 2.1 47.9 29.4 7.8 69/31
Metatarsus 49 2.1 42.2 14.3 8.2 80/20
Patella 46 1.9 36.5 10.9 4.3 72/28
Radius and ulna 44 1.8 40.9 20.5 6.8 74/26
Femoral diaphysis 43 1.8 31.8 4.7 2.3 77/23
Distal tibia 31 1.3 48.1 22.6 3.2 58/42
Proximal tibia 29 1.2 47.7 24.1 10.3 59/41
Distal femur 25 1 40.6 20 12 48/52
Ulna diaphysis 25 1 43.2 16 0 68/32
Calcaneus 18 0.8 43.7 22.2 0 78/22
Distal humerus 18 0.8 48.5 33.3 11.1 78/22
Humeral diaphysis 16 0.7 51.3 37.5 12.5 75/25
Proximal humerus 12 0.5 56 25 8.3 50/50
Clavicle 9 0.4 44 11.1 11.1 78/22
Pelvis 7 0.3 40.9 14.3 0 86/14
Talus 6 0.3 31.3 0 0 83/17
Radial diaphysis 5 0.2 44 20 0 80/20
Midfoot 5 0.2 28.2 0 0 100/0
Scapula 2 0.08 29.5 0 0 100/0
Proximal radius/ulna 2 0.08 71 50 50 50/50
Proximal femur 1 0.04 45 0 0 100/0
Carpus 1 0.04 20 0 0 100/0
Total 2,386 100 45.5 20.3 7.2 69/31
 

The number, prevalence, and gender ratios are shown as are the average ages and percentages of patients ≥65 yrs and ≥80 yrs.

X
The severity of the different open fractures is shown in Table 3-18. The Gustilo grade has been used to define the severity of the fracture and the ISS and MSI have been used to define the overall injury suffered by the patient. Overall 75.9% of patients had an isolated open fracture with no other musculoskeletal injury with 81.3% of open upper limb fractures being isolated compared with 62.8% of open lower limb fractures. Overall 26.8% of open fractures were Gustilo type III with 18.6% of upper limb fractures being Gustilo type III compared with 42.6% of lower limb fractures. The overall average ISS was 7.2 with the average ISS of patients with upper and lower limb fractures being 5.1 and 11.1, respectively. Overall 7.2% of patients presented with an ISS of ≥16 with 2.5% and 13.8% of patients with upper and lower limb fractures having an ISS of ≥16. 
 
Table 3-18
Severity of Open Fractures
View Large
Table 3-18
Severity of Open Fractures
Fracture Severity Injury Severity Score Soft tissues
Isolated Fracture (%) MSI Average ISS ISS ≥ 16 Gustilo Type III (%) Principal modes of injury
Finger phalanges 84.5 1.3 2.9 0.5 24.9 55.4% crush, 31.5% cut
Tibia and fibula 71.9 1.7 13.5 15.3 44.6 46.1% mva
Distal radius 78.3 1.3 10.9 6.5 2.2 71.2% fall
Toe phalanges 67.6 1.8 3.3 3.5 17.1 45.3% crush
Ankle 86.5 1.3 12.6 5.5 47.6 54.8% fall
Metacarpus 45.2 1.7 5.7 1.9 10.6 55.8% direct blow or assault
Proximal ulna 82.3 1.5 11.3 5.9 13.7 43.1% fall
Metatarsus 22.4 3.6 7.6 0 57.1 40.8% crush, 36.7% mva
Patella 58.7 1.9 9.1 19.5 30.4 58.7% mva
Radius and ulna 86.4 1.2 10.8 6.8 4.5 43.2% fall, 25% mva
Femoral diaphysis 37.2 2.7 18.1 39.5 65.1 53.5% mva
Distal tibia 83.9 1.4 13.1 12.9 45.2 51.6% fall height
Proximal tibia 48.3 2.1 14.3 20.7 58.6 51.7% mva
Distal femur 25 2.7 18.6 40 72 80% mva
Ulna diaphysis 76 1.5 12 8 16 28% fall, 28% direct blow or assault
Calcaneus 22.2 2.7 15 50 77.8 72.2% fall height
Distal humerus 72.2 1.5 13.6 11.1 44.4 33.3% fall, 33.3% mva
Humeral diaphysis 62.5 1.8 17.5 37.5 18.7 50% mva
Proximal humerus 91.6 1.1 10.2 0 8.3 41.7% fall, 33.3% mva
Clavicle 77.8 1.7 6.4 11.1 0 33.3% fall
Pelvis 57.1 2 19 42.9 0 42.8% fall height
Talus 50 3 10.2 33.3 50 50% mva
Radial diaphysis 40 1.6 12.4 20 20 40% fall
Midfoot 20 5.8 14 40 80 40% mva, 40% fall height
Scapula 100 1 13 50 0 50% direct blow/assault, 50% fall height
Proximal radius/ulna 50 2 25.5 50 0 50% fall, 50% fall height
Proximal femur 0 2 10 0 0 100% mva
Carpus 0 6 8 0 0 100% cut
Total 75.9 1.5 7.2 6.5 26.8
X
Tables 3-17 and 3-18 show that open lower limb fractures tend to be more severe than open upper limb fractures. Open fractures of the femoral diaphysis, distal femur, patella, proximal tibia, tibial diaphysis, distal tibia, talus, and calcaneus tend to be associated with the highest ISS and MSI and the highest prevalence of Gustilo III fractures. The majority of these fractures were caused by high-energy injuries such as motor vehicle accidents or falls from a height. The highest prevalence of Gustilo type III fractures is seen in fractures of the hindfoot and midfoot, although the incidence of these fractures is only 0.4/105/year. 
Analysis of the mode of injury showed that the commonest cause of open fractures was a crush injury with an incidence of 93.8/105/year. Eighty-three percent of these open fractures were finger phalangeal fractures. The second commonest mode of injury was a fall from a standing height with an incidence of 59.4/105/year. The average age of this group was 64.4 years and 60.9% were ≥65 years of age. Open fractures following falls tended to be isolated and not particularly severe. Only 0.7% of patients who had an open fall-related upper limb fracture had an ISS of ≥16 and the average MSI was 1.2. Similar values were seen in open lower limb fractures where 1.1% had an ISS of ≥16 and the average MSI was 1.1. However 3.2% of the open upper limb fractures secondary to a standing fall were Gustilo type III in severity, compared with 31.1% of the open lower limb fractures. 
It is often assumed that motor vehicle accidents cause the majority of open fractures, but this is not the case. In this study motor vehicle accidents caused 15.9% of the open fractures, giving an incidence of 48.8/105/year. The average age of this group was 40 years of age and 14% were ≥65 years of age. Further analysis showed that 26.1% of these patients had an ISS of ≥16 and 50.7% of the fractures were Gustilo type III in severity. 
Analysis of the multiple open fractures showed that 5.8% of patients presented with multiple open fractures. As one might expect, open lower limb fractures were associated with a higher prevalence of multiple open fractures. This was a particular problem with open fractures of the talus (50%), distal femur (44%), calcaneus (33.3%), patella (23.9%), proximal tibia (20.7%), and femoral diaphysis (14%). In the upper limb 12.5% of patients with humeral diaphyseal fractures, 9.1% of patients with distal humeral fractures and 3% of patients with distal radial fractures presented with multiple open fractures. Again this illustrates the greater severity of open lower limb fractures. 

Multiple Fractures

Orthopedic surgeons will be aware that although most fractures present as isolated injuries, patients may present with more than one fracture and there are certain accepted patterns such as the association between calcaneal and spinal fractures in a fall from a height or the association between proximal femoral fractures and distal radial or proximal humeral fractures in elderly patients who fall from a standing height. It is often assumed that multiple fractures are the result of high-energy injuries but with increasing aging of the population it is likely that surgeons will be called on to treat an increasing number of patients who have multiple fractures. 
Overall 4.8% of patients in the study year presented with multiple fractures and Tables 3-4 and 3-5 show that males had a higher prevalence of multiple fractures. The numbers of multiple fractures varied between two and eight with 77.3% of the patients with multiple fractures having two fractures, 16.1% having three fractures, and the remaining 6.5% having four or more fractures. Table 3-13 shows that multiple fractures have a Type A distribution curve and this emphasized by comparing the basic epidemiology of multiple fractures caused by falls from a standing height with those caused by falls from a height or motor vehicle accidents. Analysis shows that 5% of patients who were injured as a result of a standing fall presented with multiple fractures. Their average age was 63.5 years and the gender ratio was 25/75. In those patients injured by a fall from a height or in a motor vehicle accident 19.7% presented with multiple fractures. The average age was 41.6 years and the gender ratio was 83/17. The multiple fractures associated with each individual fracture are shown in the individual sections dealing with each fracture type. 

Fragility Fractures

The importance of osteoporotic fractures has been highlighted by many authors but in a recent study Cauley et al.16 compared the absolute risk of fractures with the risk of different cardiovascular events and breast cancer in women aged 50 to 79 years. They found that the projected number of women who would experience a fracture exceeded the combined number of women who would experience invasive breast cancer or a range of different cardiovascular events in all ethnic groups except Black women. They found that the annual incidence of fractures was greatest in White and American Indian women and lowest in Black women. 
There has been some debate as to which fractures are osteoporotic fragility fractures. Traditionally four fractures have been regarded as osteoporotic or fragility fractures, these being fractures of the proximal femur, distal radius, proximal humerus, and the thoracolumbar spine. However it is self-evident that other fractures commonly occurring in osteopenic or osteoporotic bone should also be regarded as fragility fractures. Buhr and Cooke,13 in 1959, indicated that humeral diaphyseal fractures, bimalleolar ankle fractures, and pelvic fractures had a type F distribution and they also demonstrated that proximal radial, femoral diaphyseal, proximal tibial, and lateral malleolar fractures had a bimodal distribution with a significant proportion of the fractures occurring in older women. Other workers have also suggested that there are a considerable number of fractures which should be regarded as fragility fractures.23,57,64 
Kanis et al.58 defined osteoporotic fractures as occurring at a site associated with low BMD that also increased after the age of 50 years. On the basis of this definition Johnell and Kanis57 proposed that vertebral fractures, all femoral fractures, wrist and forearm fractures, humeral fractures, rib fractures, pelvic fractures, clavicular fractures, scapular fractures, and sternal fractures should be regarded as osteoporotic fractures. They also suggested that fractures of the tibial and fibular diaphyses should be regarded as osteoporotic fractures in women. 
If Table 3-3 and Figure 3-3 are examined a list of the fragility fractures that may occur in osteopenic of osteoporotic bone can be drawn up. These are shown in Table 3-19. Table 3-3 shows that there are a further seven fractures where patients have a higher average age than that of patients with distal radial fractures, this fracture being widely accepted as a fragility fracture. If these fractures are combined with fractures that have Type E or F distribution curves and with those patients over 50 years of age who present with fracture types A, D, G, and H an estimate of the true scale of fragility fractures in a developed country can be obtained. All humeral and all femoral fractures, with the exception of the very rare femoral head fracture, should now be regarded as fragility fractures as should many long bone metaphyseal fractures. Based on the fractures shown in Table 3-19 and the patients who presented with Type A, D, G and H fractures and were over 50 years of age, Court-Brown and Caesar23 estimated in 2000 that 30.1% of male fractures and 66.3% of female fractures were potentially fragility fractures. They also pointed out that in a large Orthopaedic Trauma Unit 34.7% of outpatient fractures and 70.4% of inpatient fractures were potentially fragility fractures. This illustrates the scale of the current problem. It seems likely that the problem will increase and with increasing aging of the population other fractures will be regarded as fragility fractures and will be added to the list shown in Table 3-19. The fact that Table 3-5 shows that only eight fracture types in women currently have an average age of <50 years illustrates the potential problem facing orthopedic surgeons in the future. 
Table 3-19
A List of Fractures which Should be Considered as Fragility Fractures
Proximal humerus Femoral diaphysis
Humeral diaphysis Distal femur
Distal humerus Patella
Olecranon Bimalleolar ankle
Proximal radius and ulna Trimalleolar ankle
Distal radius Pelvis
Proximal femur Thoracolumbar spine
X

Mode of Injury

In this edition of Rockwood and Green the modes of injury have been divided into eight basic categories, these being falls from a standing height, falls from a low height, including stairs and slopes, and falls from a significant height, this being defined as above six feet. The other modes of injury are direct blows, assaults or crush injuries, sports injuries, motor vehicle accident injuries, pathologic fractures, and stress or spontaneous fractures. The epidemiologic parameters for these modes of injury are shown in Table 3-20. Gunshot injuries are very uncommon in Scotland and none were treated during the study year. In 0.4% of the patients the cause was unknown, usually because the patient was intoxicated. 
 
Table 3-20
The Epidemiology of the Different Modes of Injury
View Large
Table 3-20
The Epidemiology of the Different Modes of Injury
Prevalence % Incidence n/105/yr Average Age (yrs) Multiple Fractures (%) Open Fractures (%) M/F (%)
All Males Females ≥65 yrs ≥80 yrs
Fall (standing) 62.5 836.4 62.3 54.3 65.7 38.9 20.6 1.5 0.5 30/70
Low fall 4.2 57 51.7 48.2 55.2 27.1 10.8 6.8 3.1 51/49
Fall (height) 2.3 31.6 36 37.5 30 8.1 2.5 33 10.6 88/12
Direct blow/assault 13.6 182.6 33.3 31.1 40.1 3.6 1 5.7 5.8 75/25
Sport 11.1 149.2 31.3 30.4 35.5 3 0.3 2.1 0.6 82/18
MVA 5.2 69.6 42.6 41.7 45.8 10.2 3 17.4 6.4 78/22
Pathologic 0.4 4.8 67.3 63.5 70.3 60 24 0 0 44/56
Stress/spontaneous 0.3 2.7 49.9 44.5 54 21.4 21.4 0 0 43/57
 

The prevalence, incidence, and gender ratios are shown as are the percentages of open fractures and patients with multiple fractures. The average ages and prevalence of patients ≥65 yrs and ≥80 yrs are also shown. Low falls include falls down stairs and slopes. Direct blows/assaults include crush injuries.

X
The commonest cause of injury is a fall from a standing height which accounted for 62.5% of the fractures treated during the study year. These are commoner in older patients and are the most frequent cause of fragility fractures. The other common causes of fractures are direct blows, assaults or crush injuries which cause about 14% of all fractures and sports injuries which cause about 11% of all fractures. It is accepted that sports injuries are a combination of falls, direct blows, crushing injuries and falls from a height, but they are conventionally grouped together as sports injuries and this has been done. Both direct blows and assaults and sports injuries tend to occur in younger patients with only 3% to 4% of patients being ≥65 years of age. They are more common in males. 
Motor vehicle accidents are often perceived to cause the majority of fractures, but Table 3-20 shows that this is not the case. In 2000 7.2% of all fractures in Edinburgh occurred as a result of motor vehicle accidents but it is now 5.2% and there is little doubt that the decline in motor vehicle accident fractures is partially responsible for the lower incidence of fractures such as those of the tibial diaphysis. The United Kingdom has one of the lowest mortalities from motor vehicle accidents in the world, but, as yet, does not have a formal trauma system such as seen in the United States, Germany, and other countries. This confirms the importance of accident prevention. 
It is possible to construct age and gender curves for modes of injury in the same way as can be done for individual fractures and the eight curves shown in Figure 3-3 can be used to describe modes of injury. 

Falls from a Standing Height

This is the commonest mode of injury and Tables 3-63-11 show that most fragility fractures occur as a result of standing falls. There seems no doubt that fractures as a result of standing falls are becoming more common and this is confirmed by comparing the 2000 study year data documented in the sixth edition of Rockwood and Green24 with the current study year. In 2000 51.3% of fractures followed a standing fall or a twisting injury, compared with 62.5% in 2010 to 2011. However, the average age of the groups was the same. It seems likely that the incidence of fractures caused by standing falls will continue to rise. 
Further analysis of fractures caused by standing falls shows that 40.7% of male fractures resulted from a standing fall, compared with 85.4% of female fractures. Table 3-20 shows that the average ages were very different and that about 40% of the patients were ≥65 years. A review of the incidence of fractures following a standing fall shows that the overall incidence in males and females is 530/105/year and 1101.1/105/year and that the incidence of fractures in the ≥65-year age group is 1182.6/105/year and 2880.9/105/year, illustrating that falls from a standing height is a particular problem in older patients. Overall fractures caused by falls from a standing height show a type F distribution (Fig. 3-3). However, falls from a standing height can cause fractures in younger patients. The data shows that in patients aged 16 to 35 years standing falls caused 22% of fractures in males and 55.2% of fractures in females. In the 36- to 64-year groups the equivalent figures are 44.5% and 78.3% and in the ≥65-year group they are 81.6% and 94.2%. As one would expect Tables 3-63-11 show that there is a relatively low prevalence of open fractures and patients who present with multiple fractures as a result of falls from a standing height. 

Falls from a Low Height

In this edition of Rockwood and Green falls down stairs have been combined with falls from a low height and falls down slopes. Table 3-20 shows that these accounted for 4.2% of fractures in the study year and that they often affect older patients with about 27% of the patients being ≥65 years of age. The average age of males and females injured in low falls was 44.8 years and 54.4 years, respectively. The overall incidence of these fractures was 53.7/105/year in males and 51.6/105/year in females. If the population of patients ≥65 years of age is considered, the relative incidences are 44/105/year and 89.4/105/year indicating that there is only a small rise in incidence in females ≥65 years of age. Overall fractures that occur as a result of falls from a low height have a type F fracture distribution curve. Table 3-20 shows that there are other differences from fractures associated with falls from a standing height. Low falls are associated with a higher prevalence of open fractures and multiple fractures indicating that these falls are associated with higher-energy injuries than falls from a standing height. 

Falls from a Height

Fractures caused by falls from a height are relatively infrequent accounting for only 2.3% of the fractures seen in the study year. They mainly affect young males and only 8% of the patients treated during the study year were ≥65 years of age. They therefore have a type B distribution (Fig. 3-3). The overall incidences of these fractures are 47.9/105/year in males and 5.9/105/year in females. This is the lowest incidence of fractures in females from any mode of injury, except for pathologic or stress fractures. There were only ten females injured in a fall from a height during the study year and none were aged ≥65 years of age. It is also interesting to note that the 117 male fractures occurred in 76 patients and that only ten (8.5%) occurred in patients ≥65 years of age. 
Table 3-20 shows that falls from a height are associated with a higher prevalence of open fractures and multiple fractures than motor vehicle accidents. Unsurprisingly 70.6% of the open fractures were in the lower limb and 83.3% of these fractures were in the distal tibia, ankle, or foot. The number of multiple fractures varied between two and eight. Classically falls from a height are associated with hindfoot and spinal fractures and this was seen to be the case with 32.7% of the fractures being in the calcaneus. Although spinal fractures have not been included in the overall epidemiologic analysis, review of the data shows that 14 (11%) patients presented with a total of 28 thoracolumbar spine fractures, of which 32.1% were thoracic and 67.9% were lumbar. The commonest area to be fractured was the T12 to L1 area with 5 (17.8%) T12 fractures and 10 (35.7%) L1 fractures being seen in the study year. 

Direct Blows, Assaults, or Crush Injuries

This mode of injury, like falls from a height, is commonest in young males and therefore has a type B distribution (Fig. 3-3). It is the second commonest mode of injury after falls from a standing height. It is accepted that it covers a number of different causes of fracture and that the patient’s history may not always be accurate or honest. With that proviso, further analysis showed that 50.8% of the fractures followed a direct blow, 29.1% occurred in a fight or assault, 11.3% as a result of a crush and 7.3% as a result of a twisting injury. It is not surprising to see that 48.1% were metacarpal fractures and 27.5% were finger fractures. These are common fractures in adolescent males and Table 3-12 shows that they are commonly associated with social deprivation. Indeed analysis of the incidence of metacarpal and finger phalangeal fractures in deciles 9 and 10 of the Edinburgh population showed it to be 959.5/105/year. The overall incidence of fractures caused by direct blows or assaults in males and females was 291.7/105/year and 85.2/105/year, respectively. 
Table 3-20 shows that direct blows and assaults are associated with a relatively high rate of multiple and open fractures. This may be surprising to some surgeons given the low energy involved in most of these injuries. The fact that 5.7% of patients had multiple fractures relates to the fact that 7.6% of males had multiple metacarpal fractures. Interestingly, no female presented with multiple metacarpal fractures following a direct blow or assault. Four percent of males with a finger fracture following a direct blow or assault had multiple finger fractures, compared with 3.1% of females. Analysis of the open fractures caused by direct blows or assaults showed that 68.7% were finger fractures. 

Sports

Sports injuries occur in a very heterogeneous group of patients who are injured by falls of different types, direct blows, and even motor vehicle accidents. In addition, there is an association between stress fractures and sport. In general sports fractures have a type C distribution affecting younger males and females, although as Table 3-20 shows many more young males are affected. Predictably there are very few patients ≥65 years of age and there are relatively few open fractures or multiple fractures. 
It is self-evident that the epidemiology of sports-related fractures will vary throughout the world depending on the degree of affluence, availability of resources, and the popularity of different sports. Thus, analysis of sports in the United Kingdom will not include sports such as baseball, American football, ice hockey, or cross country skiing. However, many sports are universally popular and during the year football accounted for 39.5% of the fractures, rugby for 13%, cycling and mountain biking for 11.8%, and the different winter sports for 10.1%. The overall incidence of sports-related injuries in males was 258.9/105/year with 51.2/105/year being recorded in females. In the 16- to 34-year-old male group the incidence of fractures related to football was 594.9/105/year with 212.2/105/year, 106.1/105/year, and 72.5/105/year being recorded in rugby, cycling and mountain biking, and winter sports. In the 16 to 35 female group the highest incidence of fractures followed winter sports at 29.2/105/year with the incidence following football and rugby both being 20.6/105/year. Either ladies do not indulge in mountain biking or they are better at it than males as there were no fractures! 

Motor Vehicle Accidents

The assumption is often made by the lay public that most fractures are caused by motor vehicle accidents. However, Table 3-20 shows that this is not the case. It is likely that the number of fractures caused by motor vehicle accidents will be greater in other parts of the world, but in the developed world improved automobile design, speeding restrictions, and improved alcohol legislation have caused a reduction in the incidence of fractures. Overall motor vehicle accident fractures have a type B distribution with a unimodal peak in young males and Table 3-20 shows that the overall incidence of fractures is 69.6/105/year. The prevalence of multiple fractures and open fractures is second to those associated with falls from a height. Further analysis of the incidence of fractures following motor vehicle accidents shows that the overall fracture incidence in males is 115.1/105/year and in females it is 28.9/105/year. Obviously the incidence of motor vehicle–related fractures will vary with the precise involvement of the patient. Table 3-21 shows the basic epidemiology of the different types of involvement in motor vehicle accidents. 
Table 3-21
The Epidemiology of Fractures in the Different People Involved in Motor Vehicle Accidents
% n/105/yr Average age (yrs) M/F (%) Open Fractures (%) Multiple Fractures (%)
Cyclist 46.7 32.5 40.2 81/19 6 8.6
Motorcyclist 21.7 15.1 41 90/10 10.3 21.3
Pedestrian 17.5 12.2 48.8 62/38 3.2 24.4
Vehicle driver 10.2 7.1 47.2 76/24 8.1 34.8
Vehicle passenger 2.2 1.5 29.6 50/50 0 16.7
 

The prevalence, incidence, average age, and gender ratios are shown as are the prevalence of open fractures and multiple fractures.

X

Cyclists

Table 3-21 shows that the highest incidence of fractures as a result of motor vehicle accidents occurs in cyclists with an overall incidence of 32.5/105/year. In males the incidence is 55.7/105/year and in females it is 11.7/105/year. Overall there is a type C distribution affecting younger males and females, but clearly the incidence in males is about five times that of females. A review of the fractures sustained by cyclists indicates that they are mostly upper limb fractures with the commonest fracture being the clavicle (20.8%), followed by the proximal radius (16.7%) and the distal radius and ulna (12.5%). Ninety percent of the open fractures caused by cycling occur in the upper limb. 

Motor Cyclists

Motor cyclists present with the second highest incidence of motor vehicle accident associated fractures at 15.1/105/year (Table 3-21). The overall distribution curve is a type B curve indicating that these fractures are most commonly seen in young men. The incidence of fractures in males is 28.9/105/year, compared with 2.9/105/year in females. Obviously many fractures in motor cyclists are very severe and may be fatal, but of the fractures that present to hospital the commonest in the study year was the distal radius and ulna at 17.9%. This was followed by the clavicle (14.1%). However, a review of the open fractures shows that 62.5% affected the lower limb with the patella (25%) and tibial diaphyses (25%) being the most common. 

Pedestrians

Table 3-21 shows that pedestrians tend to be older and to present with fewer open fractures, but there are more patients with multiple fractures than in motor cyclists. At first sight this might seem unlikely, but review of Tables 3-63-11 shows that an increased prevalence of multiple fractures correlates with increasing age and osteopenia. The overall incidence of fractures in pedestrians is 12.2/105/year, but it is higher in males (16/105/year) than in females (8.8/105/year). The commonest fractures seen in pedestrians were metatarsal fractures (23.8%), finger phalangeal fractures (11.1%), and tibial and fibular diaphyseal fractures (11.1%). The patients presenting with tibial and fibular diaphyseal fractures had an average age of 38.3 years and the gender ratio was 86/14. Overall pedestrians show a type G distribution of fractures with bimodal peaks in males and a unimodal peak in older females. 

Vehicle Drivers

Surgeons may be surprised that vehicle occupants present with the lowest prevalence of fractures in road traffic accidents (Table 3-21). Overall 12.5% of the fractures occurring in motor vehicle accidents were in vehicle occupants and this in fact is higher than in the 2007 to 2008 study year presented in the sixth edition of Rockwood and Green24 when the figure was 11.2%. Vehicle drivers, like pedestrians, have a higher average age and about a third of patients will present with multiple fractures. The rate of open fractures is also higher than that seen in pedestrians. The overall incidence of fractures in vehicle drivers was 7.1/105/year, the incidence in males and females being 11.5/105/year and 3.3/105/year, respectively. The overall fracture distribution curve for all vehicle occupants is a type H curve with bimodal distributions in both males and females. This is also the case with vehicle drivers, but there are too few vehicle passengers to calculate a curve accurately. The two commonest fractures seen in vehicle drivers were the metacarpal fracture (16.2%) and the clavicle fracture (13.5%). 

Vehicle Passengers

Fractures in vehicle passengers are surprisingly rare, which presumably is a testament to improved car design, seatbelts, and airbags. Only 2.2% of fractures in motor vehicle accidents occurred in vehicle passengers and Table 3-21 shows that they were in younger adults. The gender distribution was equal and there were no open fractures. The commonest fractures were spinal fractures (37.5%), followed by fractures of the distal radius and ulna (25%). All of the patients who presented with multiple fractures had spinal and distal radial fractures. 

Pathologic Fractures

Table 3-20 shows that 0.4% of fractures in the study year were pathologic, being caused by the presence of metastases. Not unexpectedly, this group was older and there were no open fractures or multiple fractures. Further analysis showed that 76% of the fractures occurred in the femur with 57.9% of the femoral pathologic fractures being proximal in location and the remaining 42.1% being in the diaphysis. The remaining fractures occurred in the humeral diaphysis (8%), proximal humerus (4%), radial diaphysis (4%), spine (4%), and the distal tibia (4%). Pathologic fractures have a type F distribution. 

Stress and Spontaneous Fractures

The remainder of the fractures that were seen during the study year were stress or insufficiency fractures where there was no obvious cause. Only 0.3% of fractures were stress or insufficiency fractures, giving an overall incidence of 2.7/105/year. These fractures have a type H distribution with the stress fractures tending to occur in younger adults and the insufficiency fractures in the older population. In this group of patients 63.6% of fractures were metatarsal fractures with an average age of 34.4 years and two (18.2%) occurred in the proximal femur with an average age of 82.5 years. If stress and insufficiency fractures are considered separately, stress fractures have a type C distribution and insufficiency fractures a type F distribution. 

Other Modes of Injury

During the study year 1.6% of patients either did not know or were not admitting to the cause of their fractures. The average age of this group was 39.1 years and the gender ratio was 57/43. It is interesting to note that 47.6% of the fractures were of the metacarpal or finger fractures where direct blows or assaults are common! 

Gunshot Injuries

Civilian gunshot fractures are less common in Europe than in the United States. The North American literature suggests that they have a type B distribution and most commonly occur in young males.49 In the last edition of Rockwood and Green26 an analysis of gunshot wounds treated in the R Adams Cowley Shock Trauma Center in Baltimore, Maryland, was undertaken. In this very busy Level 1 trauma center 6.5% of all fractures were caused by gunshot wounds in 2007. The data confirmed the type B distribution with 93% of the patients being male with an average age of 28 years. There were also racial differences with 83% of patients being Black and 15% being Caucasian. The average ISS was 16 and the mortality was 5%. Analysis showed that 7% of patients had an injury to their central nervous system, 30% had a thoracic injury, 33% had an abdominal injury, and 22% had associated spinal injuries. Table 3-22 shows the distribution of the gunshot fractures. The most common fractures occurred in the tibia and fibula diaphyses, pelvis, and hand. 
Table 3-22
The Numbers and Prevalence of Gunshot Fractures Presenting to the R Adams Cowley Shock Trauma Center in Baltimore in 2007
No. %
Tibia and fibular diaphysis 24 9.8
Pelvis 23 9.4
Hand 23 9.4
Forearm diaphysis 16 6.6
Lumbar spine 15 6.1
Femoral diaphysis 13 5.3
Scapula 11 4.5
Distal humerus 10 4.1
Distal femur 10 4.1
Cervical spine 9 3.7
Skull 9 3.7
Thoracic spine 9 3.7
Proximal forearm 9 3.7
Humeral diaphysis 8 3.3
Proximal humerus 8 3.3
Foot 8 3.3
Proximal femur 8 3.3
Face 7 2.9
Acetabulum 5 2
Clavicle 5 2
Proximal tibia 5 2
Ankle 4 1.6
Calcaneus 1 0.4
Distal radius 1 0.4
X

Specific Fracture Types

Clavicle

Table 3-3 shows that clavicle fractures account for about 4% of all fractures. Overall they have a type G distribution mainly affecting younger and older males and older females. This is shown in Tables 3-63-11. If clavicle fractures are subdivided according to their location within the clavicle, fractures of the medial and lateral thirds of the clavicle have a type A distribution while middle third clavicle fractures show a type G distribution. Table 3-23 shows that fractures of the medial third of the clavicle are rare and it should be remembered that in younger patients these fractures may involve the proximal physis. In this study year there was a similar prevalence of fractures of the middle and lateral thirds, whereas in the 2007 to 2008 year published in the seventh edition of Rockwood and Green26 there was a higher rate of fractures in the middle third. This is likely to be coincidental, but as lateral third fractures tend to occur in older patients it is possible that they will become more common in the future. 
Table 3-23
The Basic Epidemiologic Characteristics of Clavicle Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Medial third 2.7 51.6 86/14
Middle third 48.6 37.9 77/23
Lateral third 48.6 49.6 62/38
Common Modes of Injury
Fall (standing height) 40.5 56 54/46
Sport 25.3 28 91/9
MVA 22.2 41.1 74/26
Associated Fractures
Scapula 2.3
Spine 1.9
Metacarpal 1.2
X
Table 3-23 shows that standing falls, sport, and motor vehicle accidents are the commonest causes of clavicular fractures. However, analysis of the different fracture locations shows that 56.8% of lateral third fractures resulted from a standing fall, compared with 24.8% of middle third fractures. The commonest cause of middle third clavicle fractures is sport (34.4%), followed by motor vehicle accidents (29.6%). Cycling or motorcycling are clearly important causes of clavicular fractures and if one combines all types of cycling, it is apparent that cycling causes about one-third (33.6%) of middle third clavicle fractures and 25.7% of all clavicle fractures. Tables 3-4 and 3-5 show that open clavicle fractures are very rare and few patients who have clavicle fractures present with multiple fractures. Table 3-23 shows that when patients do present with multiple fractures, the other fractures tend to be spinal or upper limb fractures with 2.3% of patients with clavicle fractures also have a coexisting scapular fractures. 

Scapula

Tables 3-33-5 show that scapular fractures are rare and are more commonly seen in males. Much of the literature relating to scapular fractures has come from Level 1 trauma centers and has reinforced the view that scapular fractures are invariably high-energy fractures, mainly involving the scapula body.20 This is in fact not the case and scapular fractures actually have a type A distribution affecting younger males and older females. Table 3-5 shows that the average age of female patients with scapular fractures is in fact 74.5 years and that 77.8% were ≥65 years of age. Thus in females, scapular fractures are unquestionably fragility fractures. 
Scapular fractures are divided into fractures of the body and neck, glenoid, acromion, and coracoid. The epidemiology of these fractures is shown in Table 3-24. Fractures of the acromion and coracoid are clearly very rare and fractures of the glenoid are in fact commoner than fractures of the body and neck and are often associated with a dislocation. Table 3-24 shows that patients who present with fractures of the glenoid and patients who present with fractures of the neck and body have a similar average age. Table 3-24 also shows that the commonest causes of scapular fractures are standing falls, motor vehicle accidents, and low falls. However, further analysis shows that 66.6% of glenoid fractures were caused by low-energy standing falls and falls from a low height, whereas 41.7% of body fractures resulted from motor vehicle accidents or falls from a height although a further 33.3% of scapula body fractures followed a standing fall. The average age of this latter group was 61.5 years. 
Table 3-24
The Basic Epidemiologic Characteristics of Scapular Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Acromion 8.1 52 100/0
Coracoid 2.7 71 100/0
Glenoid 56.7 56 67/33
Neck and Body 32.4 50.3 83/17
Common Modes of Injury
Fall (standing height) 40.5 64.1 47/53
MVA 27 43.9 100/0
Low fall (stairs) 13.5 56 100/0
Associated Fractures
Clavicle 16.2
Spine 10.8
Distal radius 8.1
X
Open scapular fractures are incredibly rare and none were seen in the 15-year study detailed in this chapter. It seems likely that open scapular fractures associated with blunt injuries are essentially unsurvivable. The fact that most scapular fractures in younger patients follow high-energy injuries and that low-energy scapular fractures are associated with advanced age means that many patients who present with scapular fractures, will also have multiple fractures. Tables 3-63-11 show that multiple fractures are mostly commonly seen in males aged 36 to 64 years. Table 3-24 shows that the overall distribution of the fractures associated with scapular fractures is not dissimilar to that of the clavicle with upper limb and spinal fractures being most commonly seen. 

Proximal Humerus

These are common fractures accounting for about 7% of all fractures (Table 3-3). Tables 3-4 and 3-5 show that the incidence in females is twice that seen in males. Overall they have a type F distribution mainly affecting older males and females and Table 20-8 shows that in patients aged at least 80 years of age, they are the third most common fracture. Table 3-25 shows the epidemiology of proximal humeral fractures. They have been divided according to the OTA classification43 into unifocal extra-articular (Type A), bifocal extra-articular (Type B), and intra-articular (Type C) fractures. About two-thirds of proximal humeral fractures are unifocal extra-articular fractures and only about 8% are intra-articular fractures. 
Table 3-25
The Basic Epidemiologic Characteristics of Proximal Humeral Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Unifocal, extra-articular 64.2 65.2 37/63
Bifocal, extra-articular 27.4 67.4 18/82
Intra-articular 8.4 65.3 27/73
Fall (standing height) 85.4 68.4 27/73
Low fall (stairs) 5.4 61.1 27/73
Sport 4.2 45.4 65/35
Associated Fractures
Proximal femur 46.2
Distal radius 17.9
Scapula 5.1
X
The majority of proximal humeral fractures occur as a result of a fall from a standing height, although in younger patients sporting injuries can cause proximal humeral fractures. As with other fractures around the shoulder, 30% of sports-related proximal humeral fractures were caused by cycling. 
Table 3-14 shows that proximal humeral fractures are increasing in incidence. In 1993 the overall incidence was 47.2/105/year and in 2000 the incidence of proximal humeral fractures was 65.1/105/year, whereas in this study it was 92.4/105/year. The incidence in both males and females has increased similarly. In males it has increased from 28.7/105/year to 61/105/year and in females it has increased from 63.8/105/year to 120.3/105/year. 
As one might expect with a low-velocity fragility fracture, open proximal humeral fractures are extremely rare and there were none recorded in the study year. As with other fragility fractures there is an increasing prevalence of multiple fractures with increasing age. This is shown in Tables 3-63-11. Tables 20-9 and 20-10 show that the prevalence of multiple fractures continues to increase in advanced old age. Table 3-25 shows that the two commonest associated fractures are the other common fragility fractures of the proximal femur and distal radius, but about 5% of patients who have multiple fractures with a proximal humeral fracture will also have a scapula fracture. 

Humeral Diaphysis

Humeral diaphyseal fractures have a type H distribution curve with a bimodal distribution in both males and females. However, it is likely that the distribution of these fractures is changing. Analysis of these fractures in 2010 to 2011 shows that there was a low incidence in young females (Table 3-9) and with increasing aging of the population it seems likely that humeral fractures will show a type G curve in years to come. Table 3-3 shows that patients who present with humeral diaphyseal fractures have a very similar average age to those that present with distal radius and ulna fractures. It also shows that there is a very similar prevalence of patients aged ≥65 years and ≥80 years and that humeral diaphyseal fractures should now be considered as fragility fractures. The age of the population who present with these fractures accounts for the fact that 71.4% of humeral diaphyseal fractures follow a standing fall. 
Table 3-26 separates the humeral diaphyseal fractures according to their location within the humerus or whether the fracture was periprosthetic or not. Fractures in the upper third of the humeral diaphysis tend to occur in older patients with 77.3% being caused by a standing fall and 18.2% by a direct blow or assault. Fractures of the middle third of the humerus tend to occur in younger patients, although 71% were still caused by a standing fall, with 12.9% being caused by sport, of which 50% were caused by horse riding. Fractures of the distal humeral diaphysis occurred in the youngest group of patients with 64.7% following a simple fall and 17.6% following a direct blow or assault. 
Table 3-26
The Basic Epidemiologic Characteristics of Humeral Diaphyseal Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Upper third 30 67.7 57/43
Middle third 40 52.4 46/54
Lower third 24.3 45.8 41/59
Periprosthetic 5.7 78.3 25/75
Common Modes of Injury
Fall (standing height) 71.4 62.7 38/62
Direct blow/assault 11.4 40.4 87/13
Sport 8.6 31.2 83/17
Associated Fractures
Distal radius 1.4
X
Table 3-26 shows that 5.7% of the humeral fractures were periprosthetic with 50% occurring around a shoulder prosthesis and 50% around a previously inserted plate. These occurred in older patients and all followed a standing fall. There was only one pathologic fracture during the study year, but it is likely that these will become more common in the future. 
The fact that humeral diaphyseal fractures are mainly low-energy fractures occurring in older patients means that the prevalence of open fractures is low and only one open fracture was seen during the study year. The prevalence of multiple fractures was surprisingly low for a fragility fracture and only one humeral diaphyseal fracture was associated with a distal radial fracture. 

Distal Humerus

Distal humeral fractures in adults are comparatively rare with Table 3-3 showing that only 0.7% of fractures in the study year involved the distal humerus. Overall they have a type E distribution which may surprise some surgeons as much of the literature has concentrated on high-energy intra-articular fractures in younger patients. However, Table 3-5 shows that in females the average age is 67.4 years and 75% of the fractures occurred in patients aged ≥65 years. Thus distal humeral fractures should be regarded as fragility fractures. 
Table 3-27 shows that the majority of distal humeral fractures are actually OTA43 type A extra-articular fractures with only 14.6% of the fractures being type C intra-articular fractures. Even in this group it is interesting to note that there is a relatively high average age of 66 years. In fact 57.1% of type C fractures followed a standing fall in patients with an average age of 76 years and the remaining 42.9% were caused by motor vehicle accidents or falls from a height, but even their age averaged 52.7 years, suggesting that many of these fractures occur in fitter older patients. 
Table 3-27
The Basic Epidemiologic Characteristics of Distal Humeral Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Extra-articular 50 54.1 54/46
Partial articular 35.4 61.5 29/71
Complete articular 14.6 66 29/71
Common Modes of Injury
Fall (standing height) 72.9 65.8 23/77
MVA 10.4 37.4 100/0
Sport 6.3 43.3 100/0
Associated Fractures
Proximal ulna 20
Proximal radius 20
Distal radius/ulna 20
X
In recent years much of the literature about distal humeral fractures in the elderly has concerned the argument regarding replacement or reconstruction of type C fractures, but these are relatively rare. The commonest subtype seen is the A2.3 simple extra-articular metaphyseal fracture which accounted for 25% of all the distal humeral fractures seen in the study year. The average age of this group of patients was 74.3 years and 91.7% were caused by standing falls. 
Table 3-27 shows that 10.4% of distal humeral fractures occurred as a result of a motor vehicle accident. All occurred in cyclists and 16% were B1.1 partial articular fractures. Analysis shows that 40% of distal humeral fractures caused by motor vehicle accidents were associated with multiple fractures and 40% were open. Tables 3-4 and 3-5 show that open fractures are more commonly seen in males than females, but that multiple fractures are fairly common in both genders. Table 3-27 shows that patients who present with distal humeral fractures and who have multiple fractures tend to have fractures in the proximal forearm or distal radius. 

Proximal Forearm

Proximal forearm fractures are relatively common with Table 3-3 showing that they account for 5.4% of all fractures. They are evenly distributed between males and females and overall they have a type D distribution. However, a review of the four basic types of proximal forearm fractures shown in Table 3-13 indicates that the epidemiology of proximal radial fractures is somewhat different from that of proximal ulna fractures and fractures of both the proximal ulna and radius. Radial head fractures have a type H distribution and radial neck fractures have a type A distribution. The remaining two fracture types, the proximal ulna and the proximal radius and ulna, have a type F distribution and should be regarded as fragility fractures. 
Proximal radial fractures accounted for 74.3% of all proximal forearm fractures. Overall 63.3% occurred as a result of a standing fall, 12.8% from a motor vehicle accident and 12.5% as a result of a sporting injury. The average ages of these groups were 46.4 years, 40 years, and 29.6 years, respectively. Table 3-28 shows that patients with proximal ulna fractures had the highest average age of all patients who presented with proximal forearm fractures. Overall 66.6% of proximal ulna fractures followed a fall with the average age of the patients being 62.5 years. A further 13.1% were caused by sporting injuries, with the patients’ average age being 34.8 years. Further analysis of the proximal ulna fractures showed that 92.9% were olecranon fractures and only 7.1% were coracoid fractures. Patients with coracoid fractures have an average age of 34.5 years and clearly coracoid fractures are not fragility fractures. 
Table 3-28
The Basic Epidemiologic Characteristics of Proximal Radial and Ulnar Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Proximal ulna 22.2 54.2 50/50
Radial head 59.5 42.4 44/56
Radial neck 14.8 42.4 44/56
Proximal radius and ulna 3.4 50.1 38/62
Common Modes of Injury
Fall (standing height) 64 50.4 30/70
Sport 13 31.6 76/24
MVA 10.8 40.4 76/24
Associated Fractures
Distal radius/ulna 20.5
Carpus 10.2
Bilateral proximal forearm 10.2
X
Fractures of both the proximal radius and ulna are relatively unusual with only 3.4% of proximal forearm fractures involving both bones. Analysis showed that 61.5% occurred following falls from a standing height and that they had an average age of 54.5 years. 
Tables 3-4 and 3-5 show that open fractures are relatively unusual in both males and females, but 11% of males and 7.9% of females had multiple fractures. Table 3-28 shows that 10% of these were bilateral proximal forearm fractures and that the other two commonly associated fractures were distal radial and carpal fractures. 

Forearm Diaphyses

When all fractures are considered it becomes apparent that forearm diaphyseal fractures are relatively uncommon. Table 3-3 shows that they accounted for only 0.8% of the fractures seen in the study year. Table 3-29 shows a similar prevalence for isolated ulnar and radial diaphyseal fractures. This is different from the 2007 to 2008 year documented in the seventh edition of Rockwood and Green.26 In this year there were more radial diaphyseal fractures, although the prevalence of fractures of both forearm bones was very similar. Table 3-29 shows that the patients’ average age and gender ratio was similar in the three types of forearm fracture. However, the distribution curves are different. The overall distribution curve for forearm fractures is a type D curve affecting younger males and females and older females. However, isolated ulnar fractures have a type H distribution, while isolated radial fractures show a type A distribution. There were few females with fractures of both the radius and ulna, but in the last edition of Rockwood and Green26 it was recorded that fractures of the radial and ulnar diaphyses had a type A distribution. This is still likely to be the case. 
Table 3-29
The Basic Epidemiologic Characteristics of Radius and Ulnar Diaphyseal Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Ulna diaphysis 43.6 45.8 67/33
Radial diaphysis 40 49.5 68/32
Radius and ulnar diaphyses 16.4 46.9 78/22
Common Modes of Injury
Fall (standing height) 49.1 60 52/48
Sport 27.3 32.2 93/7
MVA 7.3 40.7 100/0
Associated Fractures
Metacarpal 25
Finger phalanx 25
Proximal tibia 25
X
The common causes of these fractures are listed in Table 3-29, but if one analyzes the three fracture types separately it is apparent that the causes are very similar with 50% of ulnar diaphyseal fractures, 50% of radial diaphyseal fractures, and 44.4% of both bone fractures occurring as a result of a standing fall. A further 25% of ulnar diaphyseal fractures, 36.4% of radial diaphyseal fractures, and 11.1% of both bone fractures occurred as a result of a sporting injury. The main difference was that 33% of both bone fractures were caused by motor vehicle accidents compared with 4.2% of ulnar diaphyseal fractures and no radial fractures. 
Analysis of the fracture morphology shows that there were no OTA43 type C fractures and this probably relates to the diminishing prevalence of fractures associated with motor vehicle accidents. In fact 83.6% of forearm diaphyseal fractures were OTA type A simple fractures and 16.4% were OTA type B wedge fractures. However, the average ages were very similar at 47.2 years and 48.3 years respectively, although the gender ratios were 33/67 and 22/78 respectively, suggesting that the type B fractures were associated with higher-energy injuries. This is confirmed by examining the prevalence of fall-related fractures which accounted for 52.2% of type A fractures and 33.3% of type B fractures. 
Tables 3-4 and 3-5 show that all the open forearm diaphyseal fractures occurred in males with 66.6% occurring in motor vehicle accidents. Table 3-29 shows that the prevalence of multiple fractures was relatively low and most were in the upper limbs. 

Distal Radius and Ulna

Distal radial fractures are the commonest fractures to be treated by orthopedic surgeons. Table 3-3 shows that there was an incidence of 235.9/105/year and Tables 3-4 and 3-5 show that the incidence is considerably higher in females than males. Tables 3-63-8 show that in males the incidence is highest in 16- to 35-year-old males although it rises to 231.7/105/year in males ≥80 years, whereas Tables 3-93-11 show that the incidence in females increases with age. The incidence rises in the very elderly and in females ≥80 years of age the incidence is 1174.4/105/year. 
A considerable number of studies have been undertaken to look at the incidence of distal radial fractures in different parts of the world. However, it is quite difficult to compare the results of these studies. Different methods of data collection are used and there seems little doubt that some studies have concentrated on hospital admissions thereby missing the majority of patients who are treated on an outpatient basis. Authors have also used different age ranges, but even when similar age ranges are extracted from the papers there are large differences in incidence which are difficult to explain. Most studies come from developed countries with similar standards of living and medical systems, but despite this there is considerable variation in the incidence of distal radial fractures. Sakuma et al.89 recorded the incidence of distal radial fractures in a small population in Japan in 2004, the population being aged between 8 and 91 years. They recorded the incidence as 108.6/105/year. Analysis of the overall adult and pediatric incidence of distal radial fractures in Edinburgh in 2000 showed an incidence of 277.6/105/year and it seems unlikely that the Japanese incidence will be that much lower, particularly when one realizes that the Japanese have the longest life expectancy in the world. 
Analysis of fracture incidence in patients aged ≥35 years varies from 89/105/year and 368/105/year for males and females, respectively in the United Kingdom in 1997 to 199879 to 104/105/year and 295/105/year in Rochester, Minnesota, USA in 1989 to 199174 with the equivalent for our study year being 131.7/105/year and 441.2/105/year in males and females. A recent study from Texas, USA80 of forearm and wrist fractures occurring between 2001 and 2007 in patients ≥50 years of age gives the incidence of forearm and wrist fractures as 78.2/105/year and 256.9/105/year in males and females, respectively. This compares with 136.9/105/year and 631.8/105/year in Edinburgh and 141.6/105/year and 676.7/105/year in South Sweden in 2001.12 It seems likely that the social conditions in Edinburgh, Texas, and Sweden are not too dissimilar and as has already been pointed out the changes in incidences in apparently similar populations may simply reflect the methods used in collecting the data. 
Table 3-30 shows that the majority of distal forearm fractures are distal radial fractures with only 1.7% being isolated distal ulna fractures. Both have a type A distribution, but more males present with distal ulna fractures and the average age is less. This is reflected in the fact that while 80.8% of distal radial fractures follow a standing fall, 42.8% of distal ulna fractures are caused by a direct blow or assault and only 28.6% follow a standing fall. Given the low-energy nature of the majority of these fractures, it is not surprising that Tables 3-4 and 3-5 show that very few are open. Analysis of the open fractures showed that 80% of them were Gustilo47 type I and the average age of the patients was 64.2 years. 
Table 3-30
The Basic Epidemiologic Characteristics of Distal Radial and Ulnar Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Distal radius ± ulna 98.3 58.2 27/73
Distal ulna 1.7 41.5 48/52
Common Modes of Injury
Fall (standing height) 79.9 63 18/82
Sport 9.5 32.3 80/20
Low fall (stairs) 3.8 50.3 34/66
Associated Fractures
Proximal femur 22.7
Bilateral distal radius/ulna 14.8
Metacarpus 13.3
X
As one would expect with a fragility fracture, Tables 3-63-11 show that the prevalence of multiple fractures increases with age. Table 3-30 shows that the commonest coexisting fractures in patients who present with a distal radial fracture is a fracture of the proximal femur. However, ten patients presented with bilateral distal radial fractures. Analysis of this group showed that the average age was 65.4 years, 90% followed a standing fall and the gender ratio was 10/90. Thus bilateral distal radial fractures would seem to be a sign of increasing frailty. 

Carpus

Carpal fractures are relatively common and accounted for 2.8% of all fractures during the study year. Table 3-31 shows that scaphoid and triquetral fractures account for about 95% of all carpal fractures and that fractures of the other carpal bones are rare, although fractures of the hamate, pisiform, and trapezium were seen during the study year. Overall the carpal bones have a type A distribution curve, although fractures of the scaphoid, hamate, and trapezium tend to occur in younger males and have a type B curve. Triquetral fractures are somewhat different. They show a type A distribution curve involving younger males and older females. 
Table 3-31
The Basic Epidemiologic Characteristics of Carpal Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Scaphoid 72.7 33.6 66/34
Triquetrum 22.7 50.8 57/43
Hamate 2.1 29.7 75/25
Pisiform 1.6 34 67/33
Trapezium 1 43 100/0
Common Modes of Injury
Fall (standing height) 58.8 43.5 48/52
Sport 18.6 25.8 92/8
Direct blow/assault 9.8 28.7 84/16
Associated Fractures
Distal radius/ulna 38.5
Proximal radius 38.5
Metacarpus 7.7
X
Table 3-31 shows that overall about 60% of all carpal fractures follow a standing fall, with most of the remaining fractures being caused by a sporting injury or a direct blow or assault. Analysis of scaphoid fractures shows that 56% followed a fall, 20.6% were caused by sporting injury and 12% by a direct blow or assault. The equivalent figures for triquetral fractures were 63.6%, 15.9% and none, suggesting that the older ladies with triquetral fractures have a rather more refined lifestyle! It is interesting to note that 5.7% of scaphoid fractures occurred in road traffic accidents compared with 11.4% of triquetral fractures. 
There were no open carpal fractures during the study year and Tables 3-4 and 3-5 show that relatively few patients presented with multiple fractures, although when they did Table 3-31 shows that about 80% were in the distal or proximal radius. 

Metacarpus

Table 3-3 shows that during the study year metacarpal fractures were the second most common fracture after distal radial fractures with 11.2% of all fractures occurring in the metacarpus. This is different from the 2007 to 2008 fractures examined in the seventh edition of Rockwood and Green26 where proximal femoral fractures were the second most common fractures seen. The change may point to a reduction in the incidence of proximal femoral fractures as has been suggested by a number of authorities, but it is not known if this is in fact the case. It may also be related to increased deprivation which plays a significant role in the incidence of metacarpal fractures.29 
Tables 3-4 and 3-5 confirm the considerable difference in the incidence of metacarpal fractures in males and females. The incidence in males is about 445% greater than the incidence in females and metacarpal fractures are by far the commonest fractures seen in males with about one-fifth of all male fractures being in the metacarpus. Table 3-4 shows that they are associated with the second youngest average age of all male fractures after talar fractures and that only 4% occur in the ≥65-year group. They therefore have a type B distribution. 
Table 3-32 shows that metacarpal fractures are least common in the thumb metacarpal and become commoner as one moves toward the ulnar border of the hand such that 17% of metacarpal fractures occur in the ring finger metacarpal and 60% in the little finger metacarpal. The average age is highest in patients who present with thumb metacarpal fractures and more females present with fractures of the ring and little finger metacarpals. 
Table 3-32
The Basic Epidemiologic Characteristics of Metacarpal Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Thumb 5.8 39.7 87/13
Index 8.6 32 88/12
Middle 8.1 33.6 89/11
Ring 17 33.7 76/24
Little 60.6 32.4 78/22
Common Modes of Injury
Direct blow/assault 58.1 26.8 88/12
Fall (standing height) 23.8 48.8 59/41
Sport 10.9 28 89/11
Associated Fractures
Two metacarpals 62.7
Three metacarpals 25.3
Distal radius/ulna 11.9
X
Figure 3-5 shows the prevalence of metacarpal fractures according to their location within the metacarpus. The three commonest sites for metacarpal fractures are all in the little finger metacarpal with 36.1% of all the metacarpal fractures being “Boxers fractures” situated distally in the little finger metacarpal. There were 283 such fractures during the study year and Table 3-3 shows that if this fracture was regarded as a separate fracture type, it would be the ninth most common fracture to present to orthopedic surgeons. The commonest metacarpal fracture outwith the little finger is seen at the base of the thumb, although distal fractures of the index and ring fingers have similar prevalences. 
Figure 3-5
The prevalence of metacarpal and phalangeal fractures in the hand.
 
They are divided into basal fractures, diaphyseal fractures, and fractures of the distal metacarpals and phalanges. The overall prevalences in each finger are also shown.
They are divided into basal fractures, diaphyseal fractures, and fractures of the distal metacarpals and phalanges. The overall prevalences in each finger are also shown.
View Original | Slide (.ppt)
Figure 3-5
The prevalence of metacarpal and phalangeal fractures in the hand.
They are divided into basal fractures, diaphyseal fractures, and fractures of the distal metacarpals and phalanges. The overall prevalences in each finger are also shown.
They are divided into basal fractures, diaphyseal fractures, and fractures of the distal metacarpals and phalanges. The overall prevalences in each finger are also shown.
View Original | Slide (.ppt)
X
Table 3-32 shows that 58% of metacarpal fractures are said to follow direct blows, although the prevalence may, of course, be higher. The mode of injury varies between metacarpals. In thumb metacarpal fractures 31.1% were caused by motor vehicle accidents and 31.1% by standing falls, with a further 20% following a direct blow or assault. In the little finger metacarpal 64.2% of the fractures followed a direct blow or assault, 23.6% followed a fall and 2.5% occurred as a result of a motor vehicle accident. The difference in the patient population is highlighted by comparing the average ages; 26.5 years for a little finger metacarpal fracture caused by a direct blow and 48.5 years for those fractures caused by a standing fall. 
Tables 3-4 and 3-5 show that there is a very low rate of open fractures in both males and females. There is a similar prevalence of patients with multiple fractures, but most of the multiple fractures involve multiple metacarpals, although 11.9% of patients who presented with multiple fractures had distal radial and ulna fractures. Analysis of the 708 patients who presented with metacarpal fractures showed that 42 (5.9%) presented with two metacarpal fractures and six (0.8%) with three metacarpal fractures. By far the commonest location of the two metacarpal fractures was in the ring and little finger metacarpals with 26 (61.9%) of the double metacarpal fractures involving these two metacarpals. The average age of patients who presented with multiple metacarpal fractures was 39.4 years, compared with 31.7 years for those with single metacarpal fractures. The gender radios were 85/15 and 79/21, respectively. The causes of multiple metacarpal fractures were similar to those of single fractures with 56.2% of the multiple metacarpal fractures following a direct blow or assault, compared with 59.7% of the single fractures. 

Finger Phalangeal Fractures

Table 3-3 shows that fractures of the finger phalanges account for about 10% of all fractures that are seen. They are the second most common fracture in males and overall they have a type B distribution because of the high incidence in young males. Tables 3-93-11 show that there is a fairly constant incidence of finger fractures in females of different ages. Table 3-33 shows an analysis of the epidemiology of fractures in the different fingers and as with metacarpal fractures (Table 3-32) it is apparent that the little and ring fingers are the most affected. The average age and gender ratio of the patients who present with fractures in the different fingers is not dissimilar and all have a type B distribution. 
Table 3-33
The Basic Epidemiologic Characteristics of Finger Phalangeal Fractures
Finger Phalanges Prevalence (%) Average Age (yrs) Male/Female (%)
Thumb 18.4 44.3 63/37
Index 9 38.9 58/42
Middle 14.3 39.3 64/36
Ring 25.8 38.9 60/40
Little 32.5 42.4 56/44
Direct blow/assault 39.1 39.1 61/39
Fall (standing height) 29.5 52.7 37/63
Sport 23.8 29.1 82/18
Associated Fractures
Other finger fractures 55.8
Distal radius/ulna 13.9
Metacarpus 9.3
X
The causes of finger phalangeal fractures are very similar to that of metacarpal fractures, although more occur as a result of a sporting injury. Predictably the average age of the patients injured by direct blows or sports injuries is lower than those injured by standing falls and more are male. The prevalence of fractures caused by direct blows or assaults is higher on the radial side of the hand with 34.3% and 33.9% of fractures of the little and ring fingers being caused by direct blows or assaults, compared with 47.4%, 50%, and 43.4% of fractures of the middle finger, index finger, and thumb, respectively. More fractures of the little and ring fingers were caused by falls and sports injuries. 
Figure 3-5 shows a breakdown of the different fracture locations within each finger phalanx. It shows that 23.4% of finger phalangeal fractures are basal fractures of the proximal phalanxes. The average age of patients with these fractures was 56.3 years and 51.9% of them followed standing falls. A further 11.5% of fractures were diaphyseal fractures of the proximal phalanges. The average age of this group was 41.2 years and the most common cause of these fractures was a direct blow or assault which caused 39.3% of the fractures. Distal fractures of the proximal phalanges accounted for only 4% of the finger phalangeal fractures. The average age of this group was 41.6 years and 35.7% were caused by falls. 
Analysis of the middle phalangeal fractures of the index, middle, ring, and little fingers shows that 16.3% of all the finger phalangeal fractures were basal fractures of the middle phalanges. The average age was 39.1 years and sport was the commonest cause (40.4%). Only 3.4% of phalangeal fractures were diaphyseal fractures of the middle phalanges. The average age of this group was 34.7 years. All followed a fall. Only 1.7% of finger phalangeal fractures were distal fractures of the middle phalanges. Again, all followed a fall, but the average age was 50.2 years. Fractures of the base of the distal phalanges of all five digits accounted for 21.3% of all the phalangeal fractures. The average age of the patients was 37.6 years and as one might expect with distal phalangeal fracture, a direct blow was the commonest cause accounting for 41% of the fractures. A further 8.4% of phalangeal fractures occurred in the diaphyses of the distal phalanges. The average age of this group was 43 years and 76.4% were caused by direct blows. The remaining 9.8% of phalangeal fractures were distal fractures of the distal phalanges. The average age of the patients was 35.5 years and 85.7% were caused by direct blows to the tips of the fingers. 
Tables 3-4 and 3-5 show that open fractures of the phalanges are relatively common with the highest prevalence being seen in 36- to 64-year-old males. Unsurprisingly, the commonest site of the open fractures was the distal phalanges where 25.3% of the fractures were open. 
Tables 3-63-11 show that in younger patients 6% to 9% of patients present with multiple fractures, although this rises with increasing age. The majority of the multiple fractures are upper limb fractures and Table 3-33 shows that about 55% of the other fractures are other phalangeal fractures. Eighteen (3.1%) of patients had two phalangeal fractures and seven (1.2%) had three phalangeal fractures. The average age of patients who presented with multiple phalangeal fractures was 55.4 years. The gender ratio was 50/50 and 50% sustained their multiple fractures following a fall with 41.6% occurring as a result of a direct blow. 

Proximal Femur

There is considerable debate about the epidemiology of proximal femoral fractures in different parts of the world and this has already been outlined in the section on changing epidemiology. Table 3-15 illustrates some of the issues relating to the incidence of proximal femoral fractures in different countries. It highlights the different incidences in different parts of the world2,19,66 and the higher incidences in Scandinavia.6,59,86 As has already been discussed, it seems likely that some of the differences are attributable to different data collection techniques but there are also considerable differences in life expectancy, deprivation, and other medical and social comorbidities in different parts of the world. The worldwide differences in proximal femoral fracture incidence have recently been examined by Kanis et al.59 Table 3-3 shows that proximal femoral fractures were the third most common fracture after distal radius and ulna and metacarpal fractures. The overall incidence was 145.5/105/year, but Tables 3-4 and 3-5 show that the incidence varies considerably in both males and females and Tables 3-63-11 show that they are the least common fractures in the 16- to 35-year age group but are the commonest fracture in the ≥65-year age group. Table 3-14 shows that the incidence in males has risen in the last 20 years as has the average age of males who present with proximal femoral fractures. 
Table 3-34 shows that in the study year the majority of proximal femoral fractures were subcapital intracapsular fractures. The average age for subcapital and intertrochanteric fractures was similar although the average age for patients with greater trochanter fractures was lower and there were more males. Virtually all proximal femoral fractures are low-energy injuries and most are caused by a standing fall. Tables 3-4 and 3-5 show that some patients do present with multiple fractures and Table 3-34 shows that these are usually proximal humeral or distal radial fractures. Open fractures are extremely rare and will only be associated with high-energy injuries. 
Table 3-34
The Basic Epidemiologic Characteristics of Proximal Femoral Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Greater trochanter 0.9 64.8 57/43
Subcapital 59.1 79.9 30/70
Intertrochanteric 39.2 81.2 22/78
Periprosthetic 0.8 78 67/33
Common Modes of Injury
Fall (standing height) 93.3 81.1 26/74
Low fall (stairs) 2.7 78.5 35/65
Pathologic 1.5 69.4 36/64
Associated Fractures
Proximal humerus 41.8
Distal radius/ulna 41.8
Pelvis 4.7
X
Overall proximal femoral fractures have a type F distribution and this is true of both intracapsular and extracapsular fractures. The very rare femoral head fracture is associated with hip dislocation and has a type B distribution. The section on dislocations indicates that four hip dislocations were admitted during the study year of which one (25%) had an associated femoral head fracture. 

Femoral Diaphysis

There has been a considerable change in the epidemiology of femoral diaphyseal fractures in the last 100 years (Table 3-1). It was the femoral diaphyseal fracture that was largely responsible for the change in management of severely injured patients in a number of countries in the 1960s and 1970s. The mortality associated with nonoperative management of young people injured in motor vehicle accidents was unacceptable and new fracture techniques were adopted and specialized trauma centers were set up. As a result of this, it is still felt that the femoral diaphysis is mainly a young person’s fracture, but this is simply not the case and in fact the femoral diaphyseal fracture is a fragility fracture. The change in epidemiology of femoral diaphyseal fractures is emphasized by comparing the age of the patients treated in our catchment area in 1991 with today’s patients (Table 3-14). 
In the previous two editions of Rockwood and Green24,26 femoral diaphyseal fractures have been shown to have a type A distribution affecting younger males and older females, but the situation has undoubtedly changed and it is likely that these fractures now have a type G distribution. Tables 3-63-11 show a higher incidence in younger males than younger females, but the incidence rises quickly with increasing age and Table 20-8 shows the incidence of femoral diaphyseal fractures in both patients aged ≥80 years. The incidence of femoral fractures in males aged 20 to 29 years in 2010 to 2011 was 8.5/105/year but it would seem that this incidence is probably declining in many developed countries and it may well be that in years to come the femoral diaphyseal fracture will have a type F distribution. 
Table 3-35 shows that femoral diaphyseal fractures have been divided according to whether they are subtrochanteric, diaphyseal, or periprosthetic. It shows that 69.5% of patients sustained their fractures in a standing fall, but if one simply looks at the subtrochanteric fractures, 76% were injured in a standing fall with a further 12% having a pathologic fracture. In the diaphyseal fractures 58.1% of the patients had a standing fall and 11.6% presented with a pathologic fracture. In the periprosthetic group 92.9% were injured as a result of a standing fall. Given the popular misconception that femoral diaphyseal fractures are mainly high-energy injuries, it is interesting that only 9.8% of the fractures resulted from motor vehicle accidents with 62.5% of them occurring in motorcyclists. The average age of this group was 47.6 years and all were male. An analysis of the periprosthetic fracture showed that all were around hip replacements and that 42.9% were in the proximal third of the femur with the remaining 57.1% being in the middle third. 
Table 3-35
The Basic Epidemiologic Characteristics of Femoral Diaphyseal Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Subtrochanteric 30.5 74.1 36/64
Diaphyseal 52.4 64.6 49/51
Periprosthetic 17.1 78.2 64/36
Common Modes of Injury
Fall 69.5 77.6 35/65
Pathologic 9.8 65.1 75/25
MVA 9.8 47.6 100/0
Associated Fractures
Pelvis 33.3
Ankle 22.2
Proximal humerus 22.2
X
The changing epidemiology of femoral fractures means that open fractures are now seen less frequently and there were none in females during the study year. Tables 3-63-8 show that in males the open fractures occurred in the 36- to 64-year group. Predictably multiple fractures associated with femoral diaphyseal fractures were much more common in males aged 16 to 64 years (Tables 3-63-8). Table 3-35 shows that patients who presented with multiple fractures most commonly had a pelvic fracture and two (2.4%) patients presented with bilateral femoral diaphyseal fractures. 

Distal Femur

The distal femoral fracture should be regarded as the classic fracture of elderly ladies! Tables 3-33-5 show that they very rarely occur in males, but when they do it is usually in older patients. These fractures therefore have a type E distribution and should be regarded as fragility fractures. Nowadays, an increasing number of these fractures are periprosthetic and Table 3-36 shows that in 2010 to 2011 27.8% were periprosthetic fractures. This compares with 15.4% in 2007 to 2008. Review of the periprosthetic fractures shows that 70% occurred around knee prostheses with the remaining 30% occurring around long-stem hip prostheses. All periprosthetic fractures occurred as a result of a standing fall. Open fractures and multiple fractures are very rare. Tables 3-63-11 show that there were no open fractures or multiple fractures in males. Two (5.5%) older female patients had bilateral distal femoral fractures and only one (2.8%) female patient had an open fracture. 
Table 3-36
The Basic Epidemiologic Characteristics of Distal Femoral Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Distal femur 72.2 63.8 15/85
Periprosthetic 27.8 74.4 20/80
Common Modes of Injury
Fall (standing height) 86.1 70.5 13/87
Low fall (stairs) 5.5 79 0/100
MVA 3.6 26 100/0
Associated Fractures
Bilateral distal femurs 50
Ankle 25
Proximal humerus 25
X

Patella

Patellar fractures are relatively rare with only 0.7% of the fractures in the study year occurring in the patella. Table 3-3 indicates that they are fragility fractures with an average age of 64.8 years and 55.1% occurring in patients ≥65 years. They have a type F distribution. Table 3-37 shows that about three quarters result from a standing fall. There is a small male cohort in which patella fractures are caused by motor vehicle accidents. Analysis shows that 66.6% of these resulted from motorcycle or cycle injuries. In females 93.1% of patella fractures resulted from a standing fall. Intra-articular fractures accounted for 87.8% of all patella fractures. 
Table 3-37
The Basic Epidemiologic Characteristics of Patellar Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Extra/partial articular 12.2 47.2 67/33
Intra-articular 87.8 66.8 37/63
Common Modes of Injury
Fall (standing height) 75.5 70.9 27/73
MVA 10.2 37.2 100/0
Low fall (stairs) 4.1 61 50/50
Associated Fractures
Acetabulum 25
Distal tibia 25
Clavicle 25
X
Tables 3-4 and 3-5 show that open fractures of the patella are more common in males and 40% of the patella fractures caused by motor vehicle accidents were open. There was only one open patella fracture in a female injured in a fall. All the patients who presented with multiple fractures associated with a patella fracture were injured in motor vehicle accidents and Table 3-37 shows that most of the associated fractures were located in the lower limbs. 

Proximal Tibia

Proximal tibia fractures accounted for 0.8% of the fractures seen in the study year. Tables 3-4 and 3-5 show a similar incidence in males and females, but an older average age in females. Proximal tibial fractures have a type H distribution with bimodal peaks in both males and females. The overall average age is 54.5 years which is similar to that of the distal radius and it is possible that in years to come, proximal tibial fractures will be regarded as fragility fractures. Table 3-38 shows that the majority of proximal tibial fractures were partial articular fractures, although the highest average age is seen in the complete articular fracture group. Table 3-38 shows that while standing falls in older patients were the commonest cause of proximal tibial fractures about a quarter were caused by sports injuries and occurred in younger adults. A variety of sports were involved but 33.3% of the fractures were caused by soccer. 
Table 3-38
The Basic Epidemiologic Characteristics of Proximal Tibial Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Extra-articular 25.4 57.5 60/40
Partial articular 62.7 51.4 54/46
Complete articular 11.9 59.6 29/71
Common Modes of Injury
Fall (standing height) 33.9 64.9 30/70
Sport 25.4 43 80/20
Low fall (stairs) 15.3 52.8 33/67
Associated Fractures
Metacarpus 25
Proximal humerus 12.5
Femoral diaphysis 12.5
X
Open fractures are relatively rare and there were none in females during the study year with the only open fracture being seen in a 67-year-old male. Multiple fractures were more commonly seen with 13.6% of the patients who presented with proximal tibial fractures having multiple fractures. The majority of multiple fractures (75%) resulted from high-energy injuries but two patients, with an average age of 75.5 years, sustained multiple fractures from a standing fall. 

Tibial Diaphyseal Fractures

The changing epidemiology of tibial diaphyseal fractures has already been discussed in the section on changing epidemiology but a review of Table 3-14 emphasizes not only the declining incidence but also the increasing average age in males and the decreasing average age in females. It has also been pointed out that the declining incidence in of tibial fractures in older females will, in time, alter the distribution curve although the distribution curve has been kept as a type A. 
Tibial diaphyseal fractures now account for only 1% of the fractures treated by orthopedic surgeons. Rapidly changing epidemiology means that the modes of injury are changing. In 1988 to 1990 37.5% of tibial diaphyseal fractures treated in our catchment area were caused by motor vehicle accidents, 30.9% by sports injuries, and 17.8% by falls from a standing height.22 Table 3-39 shows that the situation is now radically different with 44.9% of patients now sustaining their tibial diaphyseal fracture as a result of a standing fall. However, sports injuries are still fairly common and further analysis shows that soccer was responsible for 44.4% of the sports-related tibial diaphyseal fractures in the study year. Open fractures of the tibia have always proved challenging to treat and they have been relatively common. As has already been pointed out, the prevalence of open tibial fractures is decreasing, although 20.2% of the tibial diaphyseal fractures in this study year were open. Tables 3-4 and 3-5 showed that there were fewer multiple fractures associated with tibial diaphyseal fractures than many surgeons might expect. This is because of the change in patterns of injury. Table 3-39 shows that when patients present with multiple fractures they tend to be in the lower limbs. 
Table 3-39
The Basic Epidemiologic Characteristics of Tibial Diaphyseal Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Simple 65.2 43.3 71/29
Wedge 24.6 42.1 70/30
Comminuted/Segmental 10.1 40 71/29
Common Modes of Injury
Fall (standing height) 44.9 49.1 58/42
Sport 26.1 29.3 83/17
MVA 15.9 44.2 91/9
Associated Fractures
Ankle 25
Midfoot 25
Femoral diaphysis 25
X
There is an important subgroup of tibial diaphyseal fractures, those with an intact fibula. In the seventh edition of Rockwood and Green26 it was noted that they comprised 21.9% of the tibial fractures treated in 2007 to 2008. In the current study year only 11.6% of fractures had an intact fibula. These fractures have a type B distribution and generally occur in younger males. 

Fibula

The isolated fibular fracture has received little attention in the orthopedic literature. These are fibular fractures that are not associated with a tibial diaphyseal fracture, a proximal or distal tibial fracture or an ankle fracture. They are relatively rare and accounted for only 0.6% of the fractures seen in the study year. In the seventh edition of Rockwood and Green26 they were defined them as having a type B distribution, but Tables 3-4 and 3-5 show that they actually occur in both younger males and older females and that they should be redefined as having a type A distribution. 
There are two types of isolated fibular fracture with 65.8% being proximal fractures at or adjacent to the fibular neck. The remaining 34.2% are diaphyseal fractures. Table 3-40 shows that a fall from a standing height will cause fibular fractures in older females and 68.4% of the fractures caused by falls were in the fibular neck. This is very different from the sports-related fractures where 77.8% were fibular diaphyseal fractures which were probably caused by direct blows. All of the fibular fractures caused by motor vehicle accidents were located in the proximal fibula and 60% occurred in cyclists or motorcyclists. There were no open fractures and few multiple fractures, although those that did occur were in the lower limb. 
Table 3-40
The Basic Epidemiologic Characteristics of Fibular Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Proximal fibula 65.8 49.2 41/59
Fibular diaphysis 34.2 42 57/43
Common Modes of Injury
Fall (standing height) 46.3 53.9 16/84
Sport 22 32.8 100/0
MVA 12.2 50.6 40/60
Associated Fractures
Ankle 2.4
Proximal tibia 2.4
Femoral diaphysis 2.4
X

Distal Tibia

Distal tibial fractures receive a lot of attention in the orthopedic literature, but they are comparatively rare, accounting for only 0.6% of all fractures. They have a type D distribution affecting younger males and older males and females. However, Table 3-41 indicates that the OTA43 type B partial articular fracture tends to occur in younger patients with the OTA type A extra-articular fracture occurring in older patients. 
Table 3-41
The Basic Epidemiologic Characteristics of Distal Tibial Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Extra-articular 33.3 58.9 50/50
Partial articular 45.2 31.3 68/32
Complete articular 21.4 36.8 89/11
Common Modes of Injury
Fall (standing height) 38.1 56.8 38/62
Fall height 33.3 28.5 79/21
MVA 11.9 34.6 100/0
Associated Fractures
Calcaneus 33.3
Talus 22.2
Ankle 22.2
X
As with other fractures, distal tibial fractures resulting from a standing fall tended to occur in older patients, with the higher-energy fractures resulting from motor vehicle accidents or falls from a height occurring in younger patients, most of whom are male. Seven (16.6%) of the distal tibial fractures were isolated posterior malleolar fractures. These patients had an average age of 29.4 years and a gender radio of 83/17. Four (57.1%) occurred following twisting injuries and three (42.9%) as a result of sporting accidents. 
Tables 3-4 and 3-5 show that open fractures are relatively common particularly in males. Further analysis of the open fractures showed that 83.3% occurred in males, 66.6% resulted from a fall from a height and 66.6% were OTA type C complete articular fractures. Tables 3-63-11 also show that multiple fractures are relatively common particularly in males and females aged 16 to 35 and in males aged 36 to 64. Table 3-41 shows that when present multiple fractures are commonly seen in the hindfoot and contralateral ankle. 

Ankle

Ankle fractures are very common with Table 3-3 showing that they account for 10.2% of all fractures. Tables 3-4 and 3-5 show that the incidence is very similar in males and females, but the average ages are somewhat different with younger males and older females presenting with ankle fractures. They therefore have a type A distribution. However, different ankle fractures have different distributions curves. Lateral malleolar fractures have a type A distribution curve, whereas medial malleolar fractures have a type D distribution curve. Suprasyndesmotic fractures have a type C distribution curve and both bimalleolar and trimalleolar fractures have a type E distribution curve and should be regarded as fragility fractures. It is educational to note that Buhr and Cooke,13 in 1959, drew attention to the number of bimalleolar and trimalleolar fractures in the elderly over 50 years ago. 
It is likely that ankle fractures are increasing in incidence. In 2000 there was an incidence of 100.8/105/year,24 whereas it is now 137.7/105/year. However Kannus et al.61 have suggested that in Finland the incidence of ankle fractures is decreasing in the elderly, having reached a peak of 169/105/year in 1997. They recorded an incidence of 137/105/year in females and 100/105/year in males in patients aged ≥60 years. Our figures for the incidence of ankle fractures in the ≥60-year group are 225.1/105/year and 122.5/105/year in females and males respectively, indicating a much higher incidence of ankle fractures in the elderly in Scotland than in Finland. The reason for these differences is probably methodologic as the Finnish study only looked at inpatients. 
Table 3-42 shows the ankle fractures divided according to the OTA classification.43 It shows that type B transsyndesmotic fractures accounted for about two-thirds of all ankle fractures and that about 80% of all fractures occur as a result of a standing fall, although many of these fractures may well have occurred as a result of the twisting injury that preceded the fall. Overall 77.3% of Type A, 79.6% of Type B, and 68.1% of Type C fractures were caused by falls or twisting injuries. Examination of the Type C fractures shows that 20.8% were caused by sporting injuries. The average age of this group was 32.5 years and all were male. Tables 3-4 and 3-5 show that open ankle fractures and multiple fractures are relatively unusual. Table 3-42 shows that when multiple fractures occur they tend to be in the foot or the spine. 
 
Table 3-42
The Basic Epidemiologic Characteristics of Ankle Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Infrasyndesmotic 24.1 50.7 38/62
Transsyndesmotic 65.8 49.7 48/52
Suprasyndesmotic 10.1 40.2 51/49
Common Modes of Injury
Fall (standing height) 79.8 51.9 40/60
Sport 11.2 32 78/22
Low fall (stairs) 4.3 45.5 42/58
Associated Fractures
Metatarsus 23.8
Calcaneus 14.2
Spine 14.2
X

Talus

Talar fractures were the least common fractures seen in the study year, accounting for only 0.2% of the fractures (Table 3.3). This is less than in the 2007 to 2008 study year when 32 fractures were treated. There is no explanation for the difference which may simply be coincidental. Talar fractures are seen in young males and females and have a type C distribution. In Table 3-43 the talar fractures have been divided according to the OTA classification.43 The avulsion fractures, process fractures, and head fractures have been combined together and the neck and body fractures have been recorded separately. Talar neck fractures are the most common and all fractures are seen in younger adults. 
 
Table 3-43
The Basic Epidemiologic Characteristics of Talar Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Avulsion, process, head 25 31 33/67
Neck 41.7 32.6 100/0
Body 33.3 24.5 100/0
Fall (standing height) 33.3 29 100/0
Sport 33.3 29 75/25
Direct blow/assault 16.6 17.5 100/0
Associated Fractures
Midfoot 40
Ankle 40
Distal tibia 20
X
Talar neck and body fractures tend to be high-energy injuries with 44.4% occurring as a result of a fall from a height, although 33.3% were also caused by sports injuries. Given the severity of these fractures, it is not surprising that 8.3% were open fractures and that 41.6% of the patients presenting with talar fractures had multiple fractures. Table 3-43 shows that the multiple fractures were in the foot and distal tibia. 

Calcaneus

Calcaneal fractures are relatively uncommon accounting for 0.9% of the fractures in the study year. Tables 3-4 and 3-5 show that they are more commonly seen in males, although there has been a recent increase of calcaneal fractures in older females and overall calcaneal fractures have a type G distribution. Buhr and Cooke,13 in 1959, indicated that hindfoot fractures were “wage earners fractures” and mainly affected males. Kannus et al.60 drew attention to the rising incidence of low-trauma fractures of the calcaneus and foot in Finnish patients aged ≥50 years, so it is clear that in the last 50 to 60 years more older patients have been getting calcaneal fractures. 
In Table 3-44 calcaneal fractures have been divided into extra-articular tuberosity fractures, extra-articular body fractures, and intra-articular fractures, the latter being a fracture which has received considerable attention in the orthopedic literature in recent years. Because of this recent interest, the implication is that most calcaneal fractures are intra-articular, but Table 3-44 shows that extra-articular fractures are relatively common. Intra-articular and extra-articular calcaneal fractures have different distribution curves. Intra-articular fractures tend to present in younger males and they have a type B distribution. Extra-articular fractures have a type G distribution with older males and females also being affected. 
Table 3-44
The Basic Epidemiologic Characteristics of Calcaneal Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Tuberosities 20 31.6 73/27
Extra-articular body 24.6 38.5 81/19
Intra-articular 55.4 45.6 71/29
Common Modes of Injury
Fall height 56.9 33.1 86/14
Fall (standing height) 18.5 57.7 33/67
Low fall (stairs) 16.9 47 82/18
Associated Injuries
Bilateral calcaneus 31.3
Spine 31.3
Ankle 12.5
X
The commonest mode of injury is a fall from a height. Tables 3-4 and 3-5 show that open fractures tend to occur in males and this study all open fractures were intra-articular fractures resulting from a fall from a height. As with other high-energy fractures, multiple fractures are not uncommon, particularly in males. Table 3-43 shows that patients who presented with multiple fractures usually had bilateral calcaneal fractures or spinal fractures. Five patients (8.3%) presented with bilateral calcaneal fractures. Their average age was 32.8 years, all were male and all had occurred as a result from a fall from a height. 

Midfoot

Midfoot fractures are also comparatively rare and accounted for only 0.4% of fractures in the study year. Tables 3-63-11 show that they occur in younger patients and have a type C distribution. Table 3-45 shows that the cuboid bone had the highest prevalence of fractures and females were more commonly affected, with males presenting with more fractures of the cuneiforms and navicular. The overall modes of injury are shown in Table 3-45, but if one examines the bones separately, cuboid fractures were mainly caused by falls (27.3%) and sports injuries (27.3%), whereas navicular fractures were mainly caused by falls (50%) and cuneiform fractures by falls from a height (44.4%). Open fractures are rare, but Tables 3-4 and 3-5 show that patients with midfoot fractures have a high prevalence of multiple fractures. Table 3-45 shows that these are usually fractures of the other midfoot bones, metatarsus, or talus. 
Table 3-45
The Basic Epidemiologic Characteristics of Midfoot Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Cuboid 39.3 43 36/64
Cuneiforms 32.1 30.4 78/22
Navicular 28.6 42.6 75/25
Fall (standing height) 32.1 39.4 67/33
Fall height 25 32.3 100/0
Sport 17.9 32.3 60/40
Associated Fractures
Metatarsus 57.1
Multiple midfoot bones 42.9
Talus 42.9
X

Metatarsus

Metatarsal fractures are common and accounted for 6.6% of the fractures in the study year. Tables 3-4 and 3-5 show that they tend to occur in younger males and older females and therefore have a type A distribution. They are mostly commonly caused by a fall from a standing height with 83.1% of fractures in females being caused by a twist or a fall. This compares to 51.8% of male metatarsal fractures. 
Table 3-46 shows that fractures of the hallux metatarsal were least commonly seen and only 17.6% of them were caused by a standing fall. A further 29.4% resulted from sport with 29.4% also being caused by direct blows or assaults. Metatarsal fractures become more common as one moves toward the lateral border of the foot and 69% of all metatarsal fractures affect the fifth metatarsal. The average age is higher and the majority occur in females. Further analysis shows that 78.8% of fifth metatarsal fractures follow a fall or a twisting injury. The average age of this group is 48.1 years and the gender ratio was 26/74. 
Table 3-46
The Basic Epidemiologic Characteristics of Metatarsal Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Hallux 3.7 40.5 59/41
Second 8.2 41 39/61
Third 9.5 36.5 39/61
Fourth 9.7 43.8 36/64
Fifth 69 45.7 35/65
Fall (standing height) 71.6 47.4 26/74
Sport 8.8 26.9 73/27
Direct blow 8.6 37.7 50/50
Associated Injuries
Other metatarsals 47.8
Midfoot 8.7
Ankle 8.7
X
The commonest site of a fifth metatarsal fracture is at the base of the metatarsal. There were 274 of these fractures, meaning that 58.9% of all metatarsal fractures and 87.7% of all fifth metatarsal fractures were basal fractures of the fifth metatarsal. This makes the fifth metatarsal base fracture one of the ten fractures mostly commonly seen by surgeons and they have a very similar incidence to distal fractures of the fifth metacarpal. Overall 81.2% are caused by twisting injuries or falls, the average age is 46 years and the gender ratio is 34/66. 
Given the frequency with which metatarsal fractures occur, it is perhaps surprising that there were no open fractures. Table 3-46 shows that the patients who presented with multiple fractures usually had multiple metatarsal fractures. Of these 31 patients, 11 (35.5%) had two fractures, 19 (61.3%) had three fractures, and one (3.2%) had four metatarsal fractures. The average age of the patients who presented with multiple metatarsal fractures was 43.6 years, the gender ratio was 24/76 and 61.3% had sustained a fracture following a twisting injury or a standing fall. 

Toes

Toe fractures are relatively common accounting for about 3.5% of fractures in the study year. They have a type C distribution affecting younger males and females. Unsurprisingly, 61.8% were caused by direct blows. 

Pelvis and Acetabulum

Fractures of the pelvis and acetabulum accounted for 1.7% of the fractures in the study year. There has been considerable interest in their management over the last 25 to 30 years and, as with some other fractures, the implication is that they occur as a result of high-energy trauma. Clearly some do, but Tables 3-6 and 3-9 show that the incidence of pelvic fractures in the 16- to 35-year age group is very low, although most were caused by high-energy injuries. The incidence rises with age such that pelvic fractures are the seventh most common fracture in males aged ≥65 years (Table 3-8) and the fifth most common fracture in females ≥65 years (Table 3-11). Overall pelvic fractures have a type E distribution, but if acetabular fractures are considered separately they have a type G distribution affecting younger males and females and older females. 
Table 3-47 shows that in the study year about 86% of pelvic and acetabular fractures involved the pelvis and 14% the acetabulum. The average age of patients with acetabular fractures was slightly lower and the gender ratio of the two fracture types was markedly different. The modes of injury were as expected with the majority of fractures following standing falls, although 8.4% were caused by motor vehicle accidents. Tables 3-4 and 3-5 show that there is a high prevalence of patients with multiple fractures and these obviously usually occur in high-energy injuries. Table 3-47 shows that the associated fractures are usually lower limb fractures. 
Table 3-47
The Basic Epidemiologic Characteristics of Pelvic and Acetabular Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Pelvis 85.7 77.5 21/79
Acetabulum 14.3 64.5 88/12
Common Modes of Injury
Fall (standing height) 82.4 81.3 22/78
MVA 8.4 51.4 70/30
Sport 3.4 37.5 50/50
Associated Injuries
Femoral diaphysis 16.6
Distal tibia 16.6
Calcaneus 11.1
X
Tables 3-4 and 3-5 show that open pelvic fractures are rare with none being seen in females during the study year. They are high-energy injuries and are associated with a significant mortality. Data from a Level 1 Trauma Center in the United States7 indicates that even in specialized trauma centers the prevalence of open pelvic fractures is low. In a 10-year study period they admitted 3,053 pelvic fractures, of which 52 (1.7%) were open. Of these 43 (82.7%) were in males. They commented that motorcycle injuries were the commonest cause of open pelvic fractures. 

Spinal Fractures

The incidence of vertebral fractures during the study year was not analyzed because of the difficulty in retrieving them and the impossibility of producing accurate figures. It seems likely that spinal fractures are by far the commonest fracture that occurs because osteoporotic fractures of the spine are extremely common and the majority are never seen by a doctor as many elderly ladies merely accept a bit more back pain! 
Table 3-48 shows an analysis of the spinal fractures that were admitted during the study year. Most were thoracolumbar and were caused by a fall from a height or a standing fall. The majority were seen in males, but as has already been stated many elderly females sustain thoracolumbar fragility fractures, but do not seek medical help. In Level 1 Trauma Centers thoracolumbar fractures have a Type A distribution but overall Table 3-13 shows a Type F distribution as there are so many elderly patients who have this fracture. Cervical spine fractures have a type H distribution. Open spinal fractures are extremely unusual and one must assume that open spinal fractures are often fatal. Overall 57.5% of the patients admitted with spinal fractures had multiple fractures and Table 3-48 shows that these were usually multiple spinal fractures or fractures of the distal tibia and hindfoot. 
Table 3-48
The Basic Epidemiologic Characteristics of Spinal Fractures
Prevalence (%) Average Age (yrs) Male/Female (%)
Cervical 6.7 68.4 57/43
Thoracic 42.3 57.1 61/39
Lumbar 51 47 70/30
Common Modes of Injury
Fall height 26.9 36 86/14
Fall (standing height) 23.1 69.7 71/29
Low fall 22.1 67 83/17
Associated Injuries
Two spinal fractures 20.6
Calcaneus 10.3
Distal tibia 8.8
X
Cooper et al.21 estimated the age and gender-adjusted incidence of clinically diagnosed vertebral fractures, between 1985 and 1989, in the United States as 117/105/year. Grados et al.45 analyzed the prevalence of vertebral fractures in elderly French women. They found that 22.8% of women with an average age of 80.1 years had a vertebral fracture. The prevalence and the number of fractures increased with age such that 41.4% of women aged ≥85 years had vertebral fractures. Recently attempts have been made to assess the frequency of vertebral fractures in postmenopausal females using radiologic techniques. El Moghraoui et al.34 studied 228 postmenopausal women and showed that 25.6% had vertebral fractures. Ferrar et al.40 undertook a similar study in premenopausal and postmenopausal women. They found vertebral fractures in 1.4% of the premenopausal women and in 6.8% of the postmenopausal women. A further 3% of the postmenopausal women developed vertebral fractures within 6 years. A Dutch study100 showed that 30.7% of women ≥50 years had a previously undiagnosed vertebral fracture. Obviously these studies had different results, but if one assumes a 25% incidence of vertebral fractures in women ≥50 years, the overall incidence of these fractures is about 18 times the incidence of all other fractures in this segment of the population. Clearly further work is required! 

Epidemiology of Adolescent Fractures

There is very little information available about adolescent fractures.75 This is because the epidemiologic studies tend to concentrate on adults or pediatric fractures with a dividing age of 14, 16, or 18 years. Unfortunately, adolescent fractures are lost in the division. They are an important group because fractures in adolescent males in particular are common and the curves shown in Figure 3-2 do not emphasize this. To study adolescent fracture epidemiology, the epidemiologic data from the year 2000, which was presented in the sixth edition of Rockwood and Green,24 was combined with the pediatric data from the same year,85 Adolescent fractures were defined as being between 10 and 19 years of age. Table 3-49 shows the incidences of different fractures in the adolescent population. It can be seen that there is a significant rise in the incidence of adolescent fractures from 10 to 19 years of age compared with the incidences of fractures in children and adults. Male adolescents in 2000 had a fracture incidence of 3,830/105/year. There was a progressive decrease in fracture incidence in boys after 13 years and in girls after 11 years and at 19 years of age the fracture incidence in males was 3.6 times that in females. The overall incidence in adolescents in 2000 was 2,430/105/year and the gender ratio was 72/28. 
Table 3-49
The Incidence of Fractures in Adolescents, Children, and Adults in 2000
Adolescents (10–19 yrs) Children (0–13 yrs) Adults (≥14 yrs)
Distal radius 659 689.7 195.2
Finger phalanges 439.9 294.7 107.3
Metacarpus 405.3 111.8 130.3
Clavicle 139.8 137.9 36.5
Metatarsus 132.7 99.3 75.4
Ankle 118.6 60.6 100.8
Toe phalanges 110.1 63.7 39.6
Carpus 69.2 19.9 29.7
Forearm diaphysis 63.5 111.8 13.8
Proximal forearm 55.1 59.6 55.5
Tibial diaphysis 52.2 44.9 21.5
Distal tibia 35.3 33.4 7.9
Distal humerus 32.5 166.2 5.8
Proximal humerus 29.7 38.7 63
Spine 12.7 5.2 7.5
Proximal tibia 11.3 4.2 13.3
Humeral diaphysis 11.3 5.2 12.9
Patella 9.9 4.2 10.7
Pelvis 9.9 4.2 17
Femoral diaphysis 8.5 16.7 10.3
Calcaneus 7.1 2.1 13.7
Midfoot 5.7 4.2 5
Talus 5.7 1 3.2
Proximal femur 5.7 1 129.4
Distal femur 2.8 5.2 4.5
Scapula 2.8 0 3.2
2,430.2 1,986.5 1,113.3
 

Data used in this Table is from Court-Brown and Caesar24 and Rennie et al.85 Incidence is n/105/year.

X
Table 3-49 shows the incidence of the different fractures seen in children, adolescents, and adults in 2000. What is striking is the very high incidence of fractures of the distal radius, finger phalanges, metacarpus, clavicle, metatarsus, and ankle in adolescents. Some fractures have a lower incidence in adolescents. These tend to be fragility fractures although calcaneal fractures are also rare in the adolescent period. In other fractures such as distal humeral fractures the adolescent group is clearly midway between the high incidence seen in childhood and the lower incidence seen in adulthood. Menon et al.75 divided the adolescents into male and female junior and senior adolescents of 10 to 14 years and 15 to 19 years. They examined the influence of social deprivation in these groups and showed a correlation between social deprivation and fracture incidence in senior male and female adolescents and junior male adolescents. They also found that social deprivation was an independent predictor of fractures of the hand in senior adolescent males, fractures of the upper limb in junior adolescent males and in fractures of the upper limb and distal radius in senior adolescent females. 

The Epidemiology of Dislocations

Paul Hindle and Eleanor K. Davidson 
This is the first edition of Rockwood and Green to discuss the overall epidemiology of dislocations. Obviously the assessment of their epidemiology is subject to the same methodologic considerations as were discussed in the section on the epidemiology of fractures. As with fractures there are relatively few hospitals with a captive, well-defined population that allows for accurate epidemiologic assessment. Thus surgeons have often employed an overview of emergency department admissions102 or they have analyzed insurance company records11 or, in the case of shoulder and knee dislocations, the records of the United States military.53,81 Clearly this will define epidemiology in a particular, usually younger, subset of the population but not the whole population. To our knowledge no one has, as yet, attempted to assess the epidemiology of dislocations by postal questionnaire! 
There are some other complications when assessing the epidemiology of dislocations that one is not faced with when estimating fracture epidemiology. It can be subjectively difficult to differentiate between a subluxation and a dislocation and one simply has to accept the surgeon’s view. This is a particular problem in situations such as fracture dislocations of the ankle or finger joints where, because of the fracture, there may be a significant subluxation. The other issue which can prove impossible to resolve is whether the joint was dislocated before a well-meaning doctor, physical therapist, or bystander reduced it. Clearly some of the joints will have been dislocated but others will not have been. 
In this assessment of the epidemiology of dislocations we have examined the incidence of dislocations in the Edinburgh catchment area in a 1-year period between November 2008 and October 2009. We have included both children and adults treated during the year. Unlike the fracture epidemiology assessment this was a retrospective study with the data being obtained from the three hospitals in the Edinburgh area that deal with adult and childhood trauma or provide an emergency minor injury service. The question of whether a dislocation was present prior to a prehospital reduction was resolved as best as it could be by careful analysis of the clinical records. However this is a persistent problem with the assessment of dislocations and we accept that there might be some errors. 
It should be emphasized that there are very few studies of the overall epidemiology of dislocations. Yang et al.104 studied the incidence of dislocations in Taiwan between 2000 and 2005 using data from their National Health Insurance Program. They analyzed a randomly selected 1 million people from the National database stating that the demographics of the selected population was similar to the overall Taiwanese population. They estimated the incidence of dislocations of the shoulder, elbow, wrist, fingers, hip, knee, ankle, and foot during each year between 2000 and 2005 and noted that the incidence rose annually. They also analyzed the dislocations to see if they were simple dislocations or fracture dislocations and they monitored the prevalence of recurrent dislocations. Their overall average incidence of dislocations was 42.1/105/year which is relatively low and their reported incidence of dislocations of different joints is lower than that of many other studies. Thus, as with the incidence of fractures, one does not know if there is a different incidence of dislocations around the world or if the different methodologies used to assess epidemiology give different results. 
Brinker and O’Connor,11 in their insurance-based study in the United States looked at the common dislocations referred to orthopedic surgeons. They found that the commonest dislocation was the patellofemoral joint which accounted for 55% of their dislocations and that 78% of their dislocations involved the patellofemoral, shoulder, and acromioclavicular joints. In the section on the epidemiology of fractures it was pointed out that the average age of their capitated population was low and it seems unlikely that it represents the complete population. 
The basic epidemiology of the dislocations treated in the 1-year study period is shown in Table 3-50. It shows that the overall incidence was 157.4/105/year with an incidence of 188/105/year in males and 128/105/year in females. There were 50 dislocations in children and adolescents <15 years in the study giving an overall incidence of 48.9/105/year for this group and an incidence of 53.6/105/year in males and 44.1/105/year in females. The overall incidence of dislocations in adults was 178.5/105/year with 215.9/105/year being recorded in males and 145.2/105/year in females. Table 3-50 shows the numbers, prevalence, incidence, average age, and gender ratio for all dislocations in the population. The distribution curve for all dislocations is also shown as is the percentage of patients ≥65 years and ≥80 years. It should be noted that because dislocations of prosthetic hips are so common they have been included in the analysis. However if only dislocations of native joints are considered the overall incidence is 138.4/105/year and the average age is 39.3 years. The percentage of patients ≥65 years is 15.5% with 5.1% of patients ≥80 years. The gender ratio changes to 62/38. 
Table 3-50
The Prevalence, Incidence, and Epidemiologic Characteristics of Different Dislocation.
Dislocation No. % n/105/yr Average Age (yrs) ≥65 yrs (%) ≥80 yrs (%) M/F Distribution Curve
Shoulder 317 32.5 51.2 43 23.6 9.4 59/41 H
Hand (MCPJ, PIPJ, and DIPJ) 185 19 29.9 40.7 13.5 5.9 79/21 G
Patellofemoral 134 13.8 21.6 24.8 2.2 0 51/49 C
Prosthetic hip 114 11.7 19 75.9 87.7 35.1 30/70 F
Ankle 71 7.3 11.5 49.8 31 4.2 30/70 H
Acromioclavicular 55 5.6 8.9 37.1 5.4 0 87/13 B
Elbow 37 3.8 5.5 33.4 2.7 0 49/51 C
Toes (MTPJ, PIPJ, and DIPJ) 33 3.4 5.3 35.5 9.1 0 64/36 H
Carpometacarpal 9 0.9 1.5 27.2 11.1 0 67/33 C
Native Hip 4 0.4 0.6 22.5 0 0 75/25 C
Tarsometatarsal 4 0.4 0.6 25.5 0 0 75/25 C
Knee 3 0.3 0.5 43 0 0 67/33 C
Perilunate 3 0.3 0.5 25.7 0 0 100/0 B
Distal radioulnar 2 0.2 0.3 44 0 0 50/50 ?
Sternoclavicular 2 0.2 0.3 15.5 0 0 50/50 ?
Subtalar 1 0.1 0.2 47 0 0 0/100 ?
974 100 157.4 43 23.9 8.6 57/43 H
 

The distribution curves are shown in Figure 3-3

X

Shoulder Dislocations

Shoulder dislocations are the commonest dislocations that present to orthopedic surgeons. In the study year we had an overall incidence of 51.2/105/year but this covered both primary and recurrent dislocations. The incidence in males was 63.1/105/year and in females it was 40.2/105/year. Table 3-50 shows that the distribution curve is Type H (Fig. 3-3) with bimodal peaks in both males and females. Table 3-50 shows that with the exception of dislocations of prosthetic hips shoulder dislocations have the highest prevalence of patients ≥80 years. A review of the fracture dislocations that occurred during the study year showed that they have a different distribution curve. These tend to occur in older patients and they have a Type F curve affecting older males and females (Fig. 3-3). It is well established that posterior dislocations are much less common than anterior dislocation and in the study year the incidence of posterior dislocations was 2.4/105/year. 
It is important to differentiate between primary and secondary or recurrent dislocations when considering the epidemiology. In the study year 58% of the shoulder dislocations were primary and the remaining 42% were recurrent. Thus the incidence of primary dislocations was 29.7/105/year and the incidence of recurrent dislocations was 21.5/105/year. These figures are very similar to those published by Liavaag et al.68 who analyzed the incidence of shoulder dislocations in Oslo, Norway in 2009. They had an overall dislocation rate of 56.3/105/year with a primary dislocation rate of 26.2/105/year. Their male and female dislocation rates were a little different from ours at 82.2/105/year and 30.9/105/year, respectively. 
Liavaag et al.68 drew attention to the differences in the published rates of shoulder dislocation over the last 40 years. Some researchers have investigated specific groups in the population. An example is Owen et al.81 who published an overall incidence of 435/105/year in the United States military. However other authors have looked at their whole population. Simont et al.92 published an incidence of 11.2/105/year in males and 5/105/year in females in Olmsted County, Minnesota in 1970 to 1979. Kroner et al.65 published an incidence of 12.3/105/year in 1989 and Zacchilli and Owens105 reported an incidence of 23.9/105/year in 2002 to 2009 using a randomized sample of US hospitals with emergency departments. They had fewer elderly patients with shoulder dislocations than seen in Oslo or Edinburgh. They were not able to document recurrence and they stated that only 2.1% of their dislocations were recurrent. This would seem to be an underestimate of the prevalence of recurrent dislocations but it may well be that they mainly recorded primary dislocation which would explain the lower incidence of dislocations. 
It may well be that dislocations, like fractures, are increasing in incidence. There are an increasing number of older active people in most communities and it therefore seems logical to assume that the rate of dislocations, such as shoulder dislocations, is rising. Simont et al.92 in 1970 to 1979 noted that 15.2% of his patients were ≥40 years whereas Liavaag et al.68 in 2009 found that 39.5% of his patients were ≥40 years of age. Presumably this trend will continue. 

Sternoclavicular and Acromioclavicular Dislocations

There were only two sternoclavicular dislocations during the study year giving an incidence of 0.3/105/year. However acromioclavicular dislocations are considerably more common and accounted for 5.6% of the dislocations. Type I dislocations were excluded from analysis as they are simply joint sprains. However we have included Type II subluxations as they are usually referred to as dislocations. The overall incidence was 8.9/105/year and these dislocations have a Type B distribution being commoner in young males. Very few occur in females. 
Analysis of the severity of the acromioclavicular dislocations during the study year shows that there were no Type VI dislocations but 28% of the dislocations were Type II, 37% were Type II, 2% were Type IV, and 33% were Type V. 

Elbow Dislocations

Table 3-50 shows that elbow dislocations are relatively uncommon with an incidence of 5.5/105/year. Overall we believe that simple dislocations of the elbow, which do not have an associated fracture, have a Type C distribution affecting young males and females. However in the study year we collected both simple fractures and fracture dislocations and the fracture dislocations had a Type G distribution with a bimodal distribution in males and a unimodal distribution in older females. Thus the fractures tend to occur in older patients. Of the 35 patients who had confirmatory radiographs 28 (80%) had a dislocation of both the humeroulnar and radiocapitellar joints with the remaining 7 (20%) simply having a dislocation of the radiocapitellar joint. 
A review of the complete elbow dislocations involving both joints showed that in 2 (7.1%) there was anterior displacement with posterior displacement being seen in the other 26 (92.3%) dislocations. Fifteen (53.6%) of the elbow dislocations were associated with a fracture. Four (57.1%) of the radiocapitellar dislocations were associated with a fracture. Two (5.4%) of the elbow dislocations were open. 
In a previous analysis of simple elbow dislocations over a period of 10 years in Edinburgh the overall incidence was noted to be 2.9/105/year.1 The average patient age was 38.8 years and the gender ratio was 54/46. The main causes of simple dislocations in males were a standing fall (46%) or sport (24%) with 71% of dislocations in females being caused by a standing fall. Stoneback et al.96 analyzed the incidence of simple elbow dislocations in the United States. They had a slightly higher overall incidence at 5.21/105/year and noted that the incidence was similar in males and females at 5.26/105/year in males and 5.16/105/year in females. Most of the dislocations occurred as a result of sports or gymnastic activity. Yang et al.104 reported an incidence of 7.7/105/year in Taiwan but it is likely that these included fracture dislocations. 

Wrist and Hand Dislocations

Table 3-50 shows that dislocations of the wrist and hand are relatively common with an overall incidence of 32.2/105/year. Dislocations of the fingers are by far the most common and it is surprising that there is comparatively little written about these injuries. In this study year there were only two dislocations of the distal radioulnar joint and little useful information could be gained from them. In the study year documented in the section on the epidemiology of fractures there were three Galeazzi fractures associated with distal radioulnar dislocation. All occurred in males with an average age of 29 years. The overall incidence of Galeazzi fractures is therefore 0.6/105/year. 
There were three perilunate dislocations giving an overall incidence of 0.5/105/year although all occurred in adults and therefore the adult incidence was 0.6/105/year. All were associated with carpal or distal radial fractures but it is well recognized that carpal dislocations can occur without an associated fracture. All occurred in young males and the distribution curve is therefore type B. There were nine complete carpometacarpal dislocations. Carpometacarpal subluxations are more commonly seen but these were not included. As with most hand injuries these occurred in young adults and they had a type C distribution. Of the nine dislocations five (55.6%) were single and the remaining four involved two or three joint. Five (55.6%) had an associated fracture. 
The second most common dislocation, after the shoulder dislocation is that of the joints of the fingers. The overall incidence of dislocations in all the metacarpophalangeal and interphalangeal joints is 29.9/105/year with the incidence in males and females being 59.9/105/year and 12.1/105/year. There is a bimodal distribution in males and a unimodal distribution affecting older females. Hand dislocations therefore have a type G curve. Overall 6.4% of the dislocations occurred in patients <15 years and the overall incidence in this group was 11.8/105/year with incidences of 13.4/105/year and 10/105/year being recorded in males and females. In patients ≥15 years the overall incidence was 35.7/105/year with the incidences in males and females being 59.9/105/year and 14.3/105/year. There were 22 (11.9%) open injuries and 60 (32.4%) fracture dislocations giving fracture dislocations of the hand an overall incidence of 11.6/105/year. 
Figure 3-6 shows the prevalence of dislocations in the different joints of the hand. It can be seen that 59.4% of all the dislocations involved the joints of the thumb or little finger with the highest dislocation rates being in the proximal interphalangeal joint of the little finger, the metacarpophalangeal joint of the thumb and the proximal interphalangeal joint of the ring finger. Figure 3-6 shows that 58.4% of all hand dislocations affect the proximal interphalangeal joints of the fingers or the interphalangeal joint of the thumb, 27.6% occur in the metacarpophalangeal joints and 14% at the distal interphalangeal joints. There was no significant difference in the average ages or the gender ratios of the patients relative to which digit was affected. 
Figure 3-6
The prevalences in each finger are also shown.
View Original | Slide (.ppt)
Figure 3-6
The prevalence of metacarpophalangeal and interphalangeal dislocations.
The prevalences in each finger are also shown.
The prevalences in each finger are also shown.
View Original | Slide (.ppt)
X
The incidence of dislocations in this study is considerable higher than that recorded by Yang et al.104 in Taiwan between 2000 and 2005 who found in incidence of finger dislocations of 4.6/105/year. Mall et al.71 studied the incidence of dislocations in American football players and showed that subluxations or dislocations comprised 49% of all finger injuries. As in our series the commonest site of a finger dislocation was the proximal interphalangeal joint. 

Hip Dislocations

Hip dislocations tend to be high-energy injuries with the majority being caused by motor vehicle accidents although in the study year one dislocation occurred in a rugby match. There were four in the study year all of which occurred in adults giving an overall incidence in patients ≥15 years of 0.8/105/year with an incidence in males and females of 1.2/105/year and 0.4/105/year, respectively. The condition has a Type C distribution curve. Three (75%) of the hip dislocations were associated with fractures with one both column fracture, two posterior lip fractures, and one femoral head fracture being recorded. 
Dislocations of prosthetic hips are much commoner and presumably are increasing in incidence. Obviously all occurred in adults and the overall incidence in patients aged ≥15 years was 22/105/year with the incidence in adult males and females being 13.9/105/year and 29.3/105/year. They have a Type F distribution. 

Knee Dislocations

Dislocation of the knee joint is very rare and is usually caused by high-energy injuries although, as with hip dislocations during the study year, one knee dislocation occurred as a result of playing rugby with the other two following motor vehicle accidents. All the dislocations occurred in adults and if the overall incidence is calculated in the population aged ≥15 years it is 0.6/105/year with the incidences in males and females being 0.8/105/year and 0.4/105/year. As with hip dislocations they have a Type C distribution curve. It is possible that the incidence of knee dislocations will rise in developed countries in the future. In a study from Finland, Peltola et al.82 drew attention to the fact that knee dislocation usually followed a high-energy injury but they found a significant incidence of knee dislocation in obese patients following a fall. They estimated the incidence as 0.1/105/year. 
Dislocations of the patellofemoral joint are extremely difficult to evaluate accurately. A significant number of dislocations will have been reduced prior to attendance at an emergency department and patients will also have been told that they have had a knee dislocation without any good evidence that this is the case. It is also very difficult to separate an actual or perceived subluxation from a dislocation. Thus the best that can be done is to calculate the incidence from those patients who are believed to have had a dislocation but we accept that there may be inaccuracies. 
Table 3-50 shows that the overall incidence for the whole population was recorded at 21.6/105/year. The overall incidence for primary dislocators was 11.9/105/year with 9.7/105/year being recorded for secondary or recurrent dislocators. The overall incidence of dislocations in males and females was 22.9/105/year and 20.4/105/year, respectively. However this is a condition that is common in children and adolescents and the male and female incidences in the <15-year groups were 21.1/105/year and 16/105/year, respectively. The equivalent incidences in the male and female ≥15-year groups were 23.8/105/year and 20.8/105/year. The distribution curve is Type C. 
A review of the literature shows widespread variation in the results that have been published probably because of the problems that have already been stated. Nielsen and Yde77 examined the incidence of patellar dislocation in Denmark in 1986 and stated that the overall incidence was 30/105/year with incidences of 20/105/year in males and 50/105/year in females. Nietosvaara et al.78 investigated the incidence of patellar dislocation in children and adolescents <16 years of age in the early 1990s in Finland and documented an incidence of 43/105/year. Fithian et al.42 studied the condition between 1992 and 1997 in the United States in members of a health plan. The overall incidence of primary dislocators was 5.8/105/year and for secondary dislocators it was 3.8/105/year. The incidence was age dependent, the highest incidence being reported in primary female dislocators aged 10 to 17 years. Hsiao et al.53 investigated patellar dislocation in the United States military between 1996 and 2007. They reported an overall incidence of 69/105/year with a higher incidence in females and in patients <20 years of age. A similar incidence of 77/105/year was reported in males in the Finnish military90 whereas Yang et al.104 reported an incidence of 1.4/105/year in Taiwan although they were not specific is to whether they were documenting both knee and patellofemoral dislocations. These results demonstrate the difficulty in assessing the incidence of patellar dislocations accurately. 

Ankle and Foot Dislocations

An analysis of the literature indicates that most dislocations of the ankle are fracture dislocations and a review of the ankle dislocations during the study year confirmed this as only four (5.6%) dislocations were not associated with a fracture. Table 3-50 shows that the overall incidence of ankle fractures is 11.5/105/year but the incidence of pure dislocations in adults is only 0.8/105/year. The distribution curve for fracture dislocations is Type H, with bimodal distributions in both males and females, but for pure dislocations it is Type C with younger patients being affected. Table 3-50 shows that ankle fracture dislocations are more common in females and this is probably due to the osteoporotic nature of bimalleolar and trimalleolar fractures (Table 3-13). The overall incidence of all ankle dislocations in males is 7.2/105/year compared with 15.5/105/year in females. A review of the fractures associated with ankle dislocations showed that 54% were trimalleolar, 23% were bimalleolar, and 14% were lateral malleolar fractures. The remaining fractures were talar and distal tibial fractures. Six (8.5%) of the ankle dislocations were open. 
Hindfoot dislocations are extremely rare and there is little information available about their epidemiology. There was only one subtalar dislocation in the study year which was associated with a talar neck fracture. There were no dislocations affecting Chopart’s joint. Lisfranc injuries to the tarsometatarsal joint are more commonly seen although the true epidemiology is difficult to determine as a substantial number remain undiagnosed. There were four Lisfranc injuries all of which occurred in adults giving an overall incidence in the ≥15-year population of 0.8/105/year. Three (75%) of the dislocations were associated with fractures Lisfranc dislocations have a Type C distribution curve. 
Toe dislocations had an overall incidence of 5.3/105/year. They have a Type H distribution curve with bimodal distributions in both males and females. The overall incidence in males and females is 7.1/105/year and 3.7/105/year, respectively. As with finger dislocations the first and fifth toes are most affected with 27% of the dislocations being in each toe. These were followed by the second (21%), third (15%), and the fourth (10%) toes. Again as with finger dislocations the proximal interphalangeal joints and the interphalangeal joint of the hallux were most affected (61%). A further 27% of the dislocations occurred in the metatarsophalangeal joints and the remaining 12% occurred in the distal interphalangeal joints. There were 8 (24.2%) fracture dislocation and one (3%) dislocation was open. 

References

Anakwe RE, Middleton SE, Jenkins PJ, et al. Patient-reported outcomes after simple dislocation of the elbow. J Bone Joint Surg Am. 2011;93:1220–1226.
Arakaki H, Owan I, Kudoh H, et al. Epidemiology of hip fractures in Okinawa, Japan. J Bone Miner Metab. 2011;29:309–314.
Barakat K, Stevenson S, Wilkinson P, et al. Socioeconomic differentials in recurrent ischaemia and mortality after acute myocardial infarction. Heart. 2000;85:390–394.
Baron JA, Farahmand BY, Weiderpass E, et al. Cigarette smoking, alcohol consumption and risk of hip fracture in women. Arch Intern Med. 2001;161:983–988.
Bengnér U, Johnell O, Redlund-Johnell I. Changes in the incidence of fracture of the upper end of the humerus during a 30-year period. Clin Orthop. 1988;231:179–182.
Bergström U, Jonsson H, Gustafson Y, et al. The hip fracture incidence curve is shifting to the right. A forecast of the age-quake. Acta Orthopaedica. 2009;80:520–524.
Black EA, Lawson M, Smith S, et al. Open pelvic fractures: The University of Tennessee medical center at Knoxville experience over ten years. Iowa Orthop J. 2011;31:193–198.
Bradley C, Harrison J. Descriptive epidemiology of traumatic fractures in Australia. Injury Research and Statistics Series Number 17. Adelaide AIHW(AIHW cat no INJCAT 57), 2004.
Brennan SL, Henry MJ, Wluka AE, et al. Socioeconomic status and bone mineral density in a population-based sample of men. Bone. 2010;46:993–999.
Bridgeman S, Wilson R. Epidemiology of femoral fractures in children in the West Midlands region of England 1991 to 2001. J Bone Joint Surg Br. 2004;86B:1152–1157.
Brinker MR, O’Connor DP. The incidence of fractures and dislocations referred for orthopaedic services in a capitated population. J Bone Joint Surg Am. 2004;86-A:290–297.
Brogren E, Petranek M, Atroshi I. Incidence and characteristics of distal radius fractures in a southern Swedish region. BMC Musculoskelet Disord. 2007;8:48.
Buhr AJ, Cooke AM. Fracture patterns. Lancet 1959;1(7072):531–536.
Byrd GS, Edwards CL, Kelkar VA, et al. Recruiting intergenerational African American males for biomedical research studies: A major research challenge. J Natl Med Assoc. 2011;103:480–487.
Carstairs V, Morris R. Deprivation and health in Scotland. Health Bull. 1990;48:162–175.
Cauley JA, Wampler NS, Barnhart JN. Incidence of fractures compared to cardiovascular disease and breast cancer: The Women’s Health Observational study. J Bone Min Res. 2004;19(4):532–536.
Cauley JA. Defining ethnic and racial differences in osteoporosis and fragility fractures. Clin Orthop. 2011;469:1891–1899.
Chang KP, Center JR, Nguyen TV, et al. Incidence of hip and other osteoporotic fractures in elderly men and women: Dubbo osteoporosis epidemiology study. J Bone Miner Res. 2004;19(4):532–536.
Chevally T, Guilley E, Herrmann FR, et al. Incidence of hip fracture over a 10-year period (1991-2000): Reversal of a secular trend. Bone. 2007;40(5):1284–1289.
Cole PA, Gauger EM, Schroder LK. Management of scapular fractures. J Am Acad Orthop Surg. 2012;20:130–141.
Cooper C, Atkinson EJ, O’Fallon WM, et al. Incidence of clinically diagnosed vertebral fractures: A population-based study in Rochester, Minnesota, 1985–1989.
Court-Brown CM, McBirnie J. The epidemiology of tibial fractures. J Bone Joint Surg Br. 1995;77B:417–421.
Court-Brown CM, Caesar B. Epidemiology of adult fractures; A review. Injury. 2006;30(11):691–697.
Court-Brown CM, Caesar B. The epidemiology of fractures. In: Heckman JD, Buchholz RW, Court-Brown CM, eds. Rockwood and Green’s Fractures in Adults. 6th ed. Philadelphia, PA: Lippincott, Williams and Wilkins; 2006:95–113.
Court-Brown CM, Brydon A. Social deprivation and adult tibial diaphyseal fractures. Injury. 2007;38:750–754.
Court-Brown CM, Aitken SA, Forward D, et al. The epidemiology of fractures. In: Bucholz RW, Court-Brown CM, Heckman JD, Tornetta P, eds. Rockwood and Green’s Fractures in Adults. 7th ed. Philadelphia, PA: Lippincott, Williams and Wilkins; 2010:95–113.
Court-Brown CM, Aitken SA, Ralston SH, et al. The relationship of fall-related fractures to social deprivation. Osteoporos Int. 2011;22:1211–1218.
Court-Brown CM, Bugler KE, Clement ND, et al. The epidemiology of open fractures in adults. A 15-year review. Injury. 2012;43:891–897.
Court-Brown CM, Aitken SA, Duckworth AD, et al. The relationship between social deprivation and the incidence of adult fractures. J Bone Joint Surg Am. 2013;95(6):e321–e327.
Dhanwal DK, Dennison EM, Harvey NC, et al. Epidemiology of hip fracture: Worldwide geographic variation. Indian J Orthop. 2011;45:15–22.
Donaldson LJ, Cook A, Thomson RG. Incidence of fractures in a geographically defined population. J Epidemiol Community Health. 1990;44(3):241–245.
Donaldson LJ, Reckless IP, Scholes S, et al. The epidemiology of fractures in England. J Epidemiol Community Health. 2008;62(2):174–180.
Dunn L, Henry J, Beard D. Social deprivation and adult head injury: A national study. J Neurol Neurosurg Psychiatry. 2003;74:1060–1064.
El Moghraoui A, Morjane F, Nouijai A, et al. Vertebral fracture assessment in Moroccan women: Prevalence and risk factors. Maturitas. 2009;62:171–175.
Elliot JR, Gilchrist NL, Wells JE. The effect of socioeconomic status on bone density in a male Caucasian population. Bone. 1996;18:371–373.
Emami A, Mjöberg B, Ragnarsson B, et al. Changing epidemiology of tibial shaft fractures. 513 cases compared between 1971-1975 and 1986-1990. Acta Orthop Scand. 1996;67:557–561.
Emmett JE, Breck LW. A review and analysis of 11,000 fractures seen in a private practice of orthopaedic surgery 1937–1956. J Bone Joint Surg Am. 1958;40A:1169–1175.
Evans JM, Newton RW, Ruta DA, et al. Socio-economic status, obesity and prevalence of type 1 and type 2 diabetes mellitus. Diabet Med. 2000;17:478–480.
Farahmand BY, Persson PG, Michaëlsson K, et al. Socioeconomic status, marital status and hip fracture risk: A population-based case-control study. Osteoporos Int. 2000;11:803–808.
Ferrar L, Roux C, Felsenberg D, et al. Association between incident and baseline vertebral fractures in European women: Vertebral fracture assessment in the Osteoporosis and Ultrasound Study (OPUS). Osteoporos Int. 2012;23:59–65.
Fife D, Barancik J. Northeastern Ohio Trauma Study III: Incidence of fractures. Ann Emerg Med. 1985;14(3);244–248.
Fithian DC, Paxton EW, Stone ML, et al. Epidemiology and natural history of acute patellar dislocation. Am J Sports Med. 2004;32:1114–1121.
Fracture and dislocation compendium (Orthopaedic Trauma Association Committee for coding and Classification). J Orthop Trauma. 1996;10(suppl 1):v–ix,1–154.
General Register Office for Scotland 2001 Census. Available at http://www.gro-scotland.gov.uk/census/censushm/index.html. Accessed May 18, 2012.
Grados F, Marcelli C, Dargent-Molina P, et al. Prevalence of vertebral fractures in French women older than 75 years from the EPIDOS study. Bone. 2004;34(2):362–367.
Guilley E, Chevally F, Herrmann D, et al. Reversal of the hip fracture secular trend is related to a decrease in the incidence of institution-dwelling elderly women. Osteoporos Int. 2008;19(12);1741–1748.
Gustilo RB, Mendoza RM, Williams DM. Problems in the management of type III (severe) open fractures. A new classification of type III open fractures. J Trauma. 1984;24(8):742–746.
Hagino H, Yamamoto K, Ohshiro H, et al. Changing incidence of hip, distal radius and proximal humerus fractures in Tottori Prefecture, Japan. Bone. 1999;24:265–270.
Hakanson R, Nussman D, Gorman RA, et al. Gunshot injuries: A medical, social and economic analysis. Orthopaedics. 1994;17:519–523.
Ho SC, Chen YM, Woo JLF. Educational level and osteoporosis risk in postmenopausal Chinese women. Am J Epidemiol. 2005;161:680–690.
Hole DJ, McArdle CS. Impact of socioeconomic deprivation on outcome after surgery for colorectal cancer. Br J Surg. 2002;89:586–589.
Horton TC, Dias JJ, Burke FD. Social deprivation and hand injury. J Hand Surg Eur. 2007;26:29–35.
Hsiao M, Owens BD, Burks R, et al. Incidence of acute traumatic patellar dislocation among active-duty United States military service members. Am J Sports Med. 2010;38:1997–2004.
Jacobsen SJ, Goldberg J, Miles TP, et al. Seasonal variation in the incidence of hip fracture among white persons aged 65 years or older in the United States, 1984-1987. Am J Epidemiol. 1991;133:996–1004.
Jenkins PJ, Perry PW, Ng CW, et al. Deprivation influences the functional outcome from hip arthroplasty. Surgeon. 2009;7:351–356.
Johansen A, Evans RJ, Stone MD, et al. Fracture incidence in England and Wales: A study based on the population of Cardiff. Injury. 1997;28:655–660.
Johnell O, Kanis J. Epidemiology of osteoporotic fractures. Osteoporos Int. 2005;16(suppl 2):S3–S7.
Kanis J, Oden A, Johnell O, et al. The burden of osteoporotic fractures: A method for setting intervention thresholds. Osteoporsis Int. 2001;12:417–427.
Kanis JA, Odén A, McCloskey EV, et al. A systematic review of hip fracture incidence and probability of fracture worldwide. Osteoporos Int. 2012;23(9):2239–2256.
Kannus P, Niemi S, Palvanen M, et al. Rising incidence of low-trauma fractures of the calcaneus and the foot among Finnish older adults. J Gerontol A Biol A Sci Med Sci. 2008;63:642–645.
Kannus P, Palvanen M, Niemi S, et al. Stabilizing influence of low-trauma ankle fractures in elderly people. Finnish statistics for 1970 to 2006 and prediction for the future. Bone. 2008;43:340–342.
Kannus P, Palvanen M, Niemi S, et al. Rate of proximal humeral fractures in older Finnish women between 1970 and 2007. Bone. 2009;44:656–659.
Kaye JA, Jick H. Epidemiology of lower limb fractures in general practice in the United Kingdom. Inj Prev. 2004;10:368–374.
Knowelden J, Buhr AJ, Dunbar O. Incidence of fractures in persons over 35 years of age. A report to the MRC working party on fractures in the elderly. Brit J Prev Soc Med. 1964;18:130–141.
Kroner K, Lind T, Jensen J. The epidemiology of shoulder dislocations. Arch Orthop Trauma Surg. 1989;108:288–290.
Lau EMC, Lee JK, Suriwongpaisal P, et al. The incidence of hip fracture in four Asian countries: The Asian osteoporosis study (AOS). Osteoporos Int. 2001;12:239–243.
Lawson DH, Sherman V, Hollowell J. The general practice research database. Q J Med. 1998;91:445–452.
Liavaag S, Svenningsen S, Reikerås O, et al. The epidemiology of shoulder dislocations in Oslo. Scand Med Sci Sports. 2011;21:3334–3340.
MacKenzie EJ, Bosse MI, Kellam JF, et al. Characterization of patients with high-energy lower extremity trauma. J Orthop Trauma. 2000;7:455–466.
Malgaigne JF. A Treatise on Fractures. Philadelphia, PA: JP Lippincott; 1859.
Mall NA, Carlisle JC, Matava MJ, et al. Upper extremity injuries in the national football league. Am J Sports Med. 2008;36:1938–1944.
Mattila VM, Jormanainen V, Sahi T, et al. An association between socioeconomic, health and health behavioural indicators and fractures in young adult males. Ostoporos Int. 2007;18:1609–1615.
Melton LJ, Therneau TM, Larson DR. Long-term trends in hip fracture prevalence: The influence of hip fracture incidence and survival. Osteoporos Int. 1998;8:68–74.
Melton LJ, Crowson CS, O’Fallon WM. Fracture incidence in Olmsted County, Minnesota: Comparison of urban with rural rates and changes in urban rates over time. Ostoporos Int. 1999;9:29–37.
Menon MRG, Walker JL, Court-Brown CM. The epidemiology of fractures in adolescents with reference to social deprivation. J Bone Joint Surg Br. 2008;90-B:1482–1486.
Navarro MC, Sosa M, Saavedra P, et al. Poverty is a risk factor for osteoporotic fractures. Osteoporos Int. 2009;20:393–398.
Nielsen AB, Yde J. Epidemiology of acute knee injuries: A prospective hospital investigation. J Trauma. 1991;31:1644–1648.
Nietosvaara Y, Aalto K, Kallio PE. Acute patellar dislocation in children: Incidence and associated ostechondral fractures. J Pediatr Orthop. 1994;14:513–515.
O’Neill TW, Cooper C, Finn JD, et al. Incidence of distal forearm fracture in British men and women. Osteoporos Int. 2001;12(7):555–558
Orces CH, Martinez FJ. Epidemiology of fall related forearm and wrist fractures among adults treated in US hospital emergency department. Inj Prev. 2011;17:33–36.
Owens BD, Dawson L, Burks R, et al. Incidence of shoulder dislocation in the united States military: Demographic considerations from a high-risk population. J Bone Joint Surg Am. 2009b:91:791–796.
Peltola EK, Lindahl J, Hietaranta H, et al. Knee dislocation in overweight patients. AJR. 2009;192:101–106.
Pillai A, Ativa S, Costigan PS. The incidence of Perthes’ disease in Southwest Scotland. J Bone Joint Surg (Br). 2008;90B:1482–1486.
Pressley JC, Kendig TD, Frencher SK, et al. Epidemiology of bone fracture across the age span in blacks and whites. J Trauma. 2011;71:S541–S548.
Rennie L, Court-Brown CM, Mok JY, et al. The epidemiology of fractures in children. Injury. 2007;38:913–922.
Rosengren BE, Ahlborg HG, Gärdsell P, et al. Bone mineral density and incidence of hip fracture in Sweden urban and rural women 1987-2002. Acta Orthopaedica. 2010;81:453–459.
Sahlin Y. Occurrence of fractures in a defined population: A 1-year study. Injury. 1990;21:158–160.
Sanders KM, Nicholson GC, Ugoni AM, et al. Fracture rates lower in rural than urban communities: The Geelong osteoporosis study. J Epidemiol Community Health. 2002;56:466–470.
Sakuma M, Endo N, Oinuma T, et al. Incidence and outcome of osteoporotic fractures in 2004 in Sado City, Niigata Prefecture, Japan. J Bone Miner Metab. 2008;26:373–378.
Sillanpää P, Mattila VM, Livonen T, et al. Incidence and risk factors of acute traumatic primary patellar dislocation. Med Sci Sports Exerc. 2008;40:606–611.
Silversides JA, Gibson A, Glasgow JF, et al. Social deprivation and childhood injuries in North and West Belfast. Ulster Med J. 2005;74:22–28.
Simont WT, Melton LJ, Cofield RH, et al. Incidence of anterior shoulder dislocation in Olmsted County, Minnesota. Clin Orthop. 1984;186:186–191.
Stark AD, Bennet GC, Stone DH, et al. Association between childhood fractures and poverty: Population based study. BMJ. 2002;324:457.
Sterling RS. Gender and race/ethnicity differences in hip fracture incidence, morbidity, mortality, and function. Clin Orthop Relat Res. 2011;469;1913–1918.
Stimson LA. A Practical Treatise on Fractures and Dislocations. 4th ed. New York, NY, Philadelphia, PA: Lea Brothers & Co; 1905.
Stoneback JW, Owens BD, Sykes J, et al. Incidence of elbow dislocations in the United States population. J Bone Joint Surg Am. 2012;94:240–245.
Tracy JK, Meyer WA, Flores RH, et al. Racial differences in rate of decline in bone mass in older men: The Baltimore men’s osteoporosis study. J Bone Mine Res. 2005;20:1228–1234.
Trauer T, Eagar K, Mellsop G. Ethnicity, deprivation and mental health outcomes. Aust Health Rev. 2006;30:310–321.
Urwin M, Symmons D, Allison T, et al. Estimating the burden of musculoskeletal disorders in the community: The comparative prevalence of symptoms at different anatomical sites, and the relation to social deprivation. Ann Rheum Dis. 1998;57:649–655.
van den Berg M, Verdiik NA, van den Bergh JP, et al. Vertebral fractures in women aged 50 years and older with clinical risk factors for fractures in primary care. Maturitas. 2011;70:74–79.
Wang X-F, Seeman E. Epidemiology and structural basis of racial differences in fragility fractures in Chinese and Caucasians. Ostoporos Int. 2012;23:411–422.
Waterman BR, Belmont PJ, Owens BD. Patellar dislocation in the United States: Role of sex, age, race, and athletic participation. J Knee Surg. 2012;25:51–57.
Weiss RJ, Montgomery SM, Ehlin A, et al. Decreasing incidence of tibial shaft fractures between 1998 and 2004: Information based on 10,627 Swedish inpatients. Acta Orthop. 2008;79:526–533.
Yang N-P, Chen H-C, Phan D-V, et al. Epidemiological survey of orthopedic joint dislocations based on nationwide insurance data in Taiwan, 2000-2005. BMC Musculoskelet Disord. 2011;12:253.
Zacchilli MA, Owens BD. Epidemiology of shoulder dislocations presenting to emergency departments in the United Stataes. J Bone Joint Surg Am. 2010;92A:542–549.