- Email: [email protected]
Four score years and ten
Injury, Int. J. Care Injured 40 (2009) 1111–1114
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Four score years and ten An analysis of the epidemiology of fractures in the very elderly C.M. Court-Brown a,*, N. Clement b a b
University of Edinburgh, United Kingdom Royal Infirmary of Edinburgh, United Kingdom
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 11 June 2009
Fragility fractures are becoming more common and with increasing longevity it seems likely that orthopaedic surgeons will see more such fractures in progressively older patients. We have examined the epidemiology of fractures in patients aged 90 years or more. Introduction: The purpose of this study was to examine the epidemiology of fractures in the very elderly to see if it differed from the overall fracture epidemiology. The requirement for hospital admission and surgical treatment was assessed as was the patients’ domicile, length of hospital stay and discharge location. Materials and methods: All fractures presenting to the Royal Infirmary of Edinburgh and the Royal Hospital for Sick Children in 2000 were examined prospectively and recorded on databases. A retrospective examination of the clinical and social details of the 90+ age group was undertaken. Results: The 90+ age group comprises 0.58% of the population but accounts for 3.02% of the fractures in the community, 8.7% of the in-patient admissions and 7.6% of the acute orthopaedic trauma surgery. About 56% of fractures in this age group are proximal femoral fractures but high prevalences were noted for all femoral fractures. About 65% of the fractures were in the lower limb. Over 86% of fractures occurred in falls and over 81% of patients were treated as in-patients. The average hospital stay was 9 days but only 23.9% were discharged to their domicile. Discussion: It is forecast that this age group will increase and become less fit. This has significant implications for hospital treatment and costs. ß 2009 Elsevier Ltd. All rights reserved.
Keywords: Epidemiology Fractures Nonagenarians Osteoporosis Fragility fractures
In recent years fragility fractures have assumed considerable importance in orthopaedic surgery. With increasing longevity they are becoming more common and a recent study has suggested that about 30% of male fractures and 66% of female fractures should now be regarded as fragility fractures.3 It seems likely that the prevalence of fragility fractures will correlate with age but there is very little known about the epidemiology of fractures in the very elderly. There have been a number of studies of proximal femoral fractures in nonagenarians6,7,15 but there have been no previous studies of the other fractures that might occur in this age group. As it seems likely that these fractures will become more common in the next few decades we have undertaken an epidemiological analysis of all fractures in patients aged 90 years or more.
* Corresponding author at: Department of Orthopaedic Surgery, Royal Infirmary of Edinburgh, Old Dalkeith Road, Edinburgh EH16 4SU, United Kingdom. Tel.: +44 0131 242 3516; fax: +44 0131 660 4227. E-mail address: [email protected] (C.M. Court-Brown). 0020–1383/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2009.06.011
Materials and methods All fractures presenting to the Royal Infirmary of Edinburgh and The Royal Hospital for Sick Children in 2000 were analysed. All in-patients and out-patients were included. These hospitals are the only hospitals treating orthopaedic trauma in the City of Edinburgh, Midlothian and East Lothian. All patients resident outside these areas were excluded from analysis but all patients who were injured elsewhere but resident in these areas were included. Both hospitals maintain prospective in-patient and outpatient databases of all fractures. The databases were studied to allow us to detail the fracture type, mode of injury, patient age and gender and whether the patients were treated on an in-patient or out-patient basis. A retrospective analysis of the relevant hospital databases was undertaken to analyse the domicile at the time of injury, the length of hospital stay and the location after discharge of all patients aged 90 years or more. Statistical analysis of any changes in the prevalence of fractures in this age group during the decade of the 1990s was undertaken using Spearman rank correlations.
C.M. Court-Brown, N. Clement / Injury, Int. J. Care Injured 40 (2009) 1111–1114
1112
Fig. 1. The changes in prevalences of fractures of the lower limb, upper limb and the axial skeleton with increasing decades.
Results In the year 2000 the Royal Infirmary of Edinburgh and the Royal Hospital for Sick Children treated 7907 fractures. Two hundred and thirty-six (3.02%) fractures occurred in patients of at least 90 years of age (90+ group). The 236 fractures occurred in 208 patients with 12 (5.8%) of the patients presenting with more than one fracture and 10 patients presenting with different fractures on more than one occasion. Of the 208 patients 184 (88.5%) were female and 24 (11.5%) were male. The age range was 90–102 years. Analysis of the different fractures showed that 154 (65.3%) were lower limb fractures, 72 (30.5%) were upper limb fractures and 10 (4.2%) were fractures of the axial skeleton of which one was a spinal fracture and nine were pelvic fractures (Fig. 1). Table 1 gives a breakdown of the fractures according to their location. The overall number of fractures treated in 2000 is shown together with the numbers, prevalence and incidence of the fractures in the 90+ group. The percentage of each fracture that presented in patients of at least 90 years of age is also shown. An analysis of the fractures shows that 192 (81.4%) were treated on an in-patient basis (Fig. 2) with the remaining 44 (18.6%) Table 1 The numbers of fractures seen in 2000 together with the numbers and prevalences of fractures in the 90+ group. The incidence of fractures in the 90+ group is also shown.
Proximal femur Femoral diaphysis Distal femur Pelvis Proximal humerus Tibial diaphysis Spine Distal radius Distal humerus Proximal forearm Distal tibia Proximal tibia Forearm Clavicle Finger phalanges Metacarpals Ankle Metatarsal Patella Humeral diaphysis Scapula Midfoot Toe phalanges Carpus Talus Calcaneus Sesamoid
Overall, n
90+, n
Overall (%)
90+ (%)
n/105
694 71 29 95 375 159 45 1720 191 356 74 75 181 331 870 812 599 497 61 75 17 29 277 179 19 75 1
132 8 3 9 19 4 1 32 3 5 1 1 2 3 4 4 3 2 0 0 0 0 0 0 0 0 0
19.0 11.3 10.3 9.5 5.1 2.5 2.2 1.9 1.6 1.4 1.3 1.3 1.1 0.9 0.5 0.5 0.5 0.4 0 0 0 0 0 0 0 0 0
56.2 3.4 1.3 3.8 8.1 1.7 0.4 13.6 1.3 2.1 0.4 0.4 0.9 1.3 1.7 1.7 1.3 0.9 0 0 0 0 0 0 0 0 0
3314.1 203.1 75.3 226.0 477.1 100.4 25.1 803.4 75.3 125.5 25.1 25.1 50.2 75.3 100.4 100.4 75.3 50.2 0 0 0 0 0 0 0 0 0
Fig. 2. The prevalences of in-patient admissions in the different decades of life.
fractures being treated in the out-patient department. Thirty-nine (88.6%) of the out-patient fractures involved the upper limb the remaining five (11.4%) fractures being fractures of the distal tibia, ankle and metatarsals. Thirty-three (17.2%) of the 192 in-patient fractures were in the upper limb, 149 (77.2%) were in the lower limb and 10 (5.2%) were in the axial skeleton. Of the 180 patients admitted with fractures 88 (48.9%) of the patients came from home, 82 (45.6%) came from residential or nursing homes and 10 (5.6%) were admitted from other hospitals. The average length of stay was 9 days (range 1–54 days). Table 2 shows the discharge location and length of stay for all in-patient admissions and it can be seen that only 11 (6.1%) of all patients admitted with a fracture went directly home, in an average of 8.2 days. However a further 43 (23.9%) patients who were admitted from a residential or nursing home returned to their domicile in an average of 8.8 days. Thus a total of 54 (30%) of in-patients in the 90+ group returned to their domicile after admission with one or more fractures. Table 3 shows the discharge location and the average duration of hospitalisation for in-patients depending on their site of fracture. Where patients were admitted with both upper and lower limb fractures they have been included in the lower limb group as it was the lower limb fracture that usually dictated their discharge location. Three patients presented with two lower limb fractures. The lower limb fractures have been divided into proximal femoral and other fractures to allow comparison. Table 2 The numbers of in-patients living at home, in a residential or nursing home (RH/NH) or in another hospital prior to admission. Their outcome and average stay is given. Domicile
Home
RH/NH
Hospital
Patient number Died Home RH/NH Hospital Average stay (days)
88 5 (5.7%) 11 (12.5%) 0 72 (81.8%) 11.6
82 8 (9.8%) 0 43 (52.4%) 31 (37.8%) 11.2
10 0 0 0 10 (100%) 8.6
Table 3 The outcomes of in-patients depending on the location of the fracture. Where the patient has been admitted with upper and lower limb fractures the result has been included in the lower limb group. (RH/NH = residential or nursing home).
Patient number Died Home RH/NH Hospital Average stay (days)
Upper limb
Proximal femur
Other lower limb
Axial skeleton
24 1 (4.2%) 7 (29.2%) 5 (20.8%) 11 (45.8%) 6.0
129 10 (7.8%) 4 (3.1%) 32 (24.8%) 83 (64.3%) 12.2
17 2 (11.8%) 0 6 (35.3%) 9 (52.9%) 17.8
10 0 0 1 (10%) 9 (90%) 6.3
C.M. Court-Brown, N. Clement / Injury, Int. J. Care Injured 40 (2009) 1111–1114
Fig. 3. The prevalences of falls that caused fractures in the different decades of life.
Analysis of the 12 patients who presented with more than one fracture showed that 8 (66.6%) were admitted from home but that only 1 (8.3%) was discharged home. Their average length of hospital stay was 9.1 days. Analysis of the causes of fracture showed that 204 (86.4%) of fractures were caused by a simple fall (Fig. 3). This figure is in fact likely to be higher as 16 patients could not give a history, usually because of dementia. Only one fracture, a proximal femoral fracture in a 91-year-old female, was caused by a road traffic accident. Overall 151 (64%) fractures were treated surgically but there were significant differences depending on the location of the fracture. Only 12 (36.4%) of the upper limb fractures were treated surgically, seven of these being distal radial fractures. This compares with 97% of proximal femoral fractures and 50% of the other lower limb fractures. None of the spinal or pelvic fractures were treated surgically. A review of the fractures in the 90+ group treated in the decade between 1991 and 2000 showed a significant increase in numbers for certain fractures. There were significant upward trends in proximal femoral fractures (p = 0.002), distal femoral fractures (p = 0.008), proximal humeral fractures (p = 0.007), distal humeral fractures (p = 0.035), forearm fractures (p = 0.039) and distal radius fractures (p = 006). However an analysis of the overall population numbers for each year showed that the apparent increase in fracture numbers was caused by an overall progressive increase in the numbers of 90+ patients in the population. There was no statistical evidence of a relative increase in any fracture during the decade. Discussion Nonagenarians and centenarians are becoming an increasingly important part of society. They are increasing in number and the National Office of Statistics has shown that they are, in fact, the age group which is increasing most quickly.12 In 2001 at the time of the last UK Census there were 29,385 people aged 90+ years in Scotland this being 0.58% of the population.5 This figure matches that of the US Census website which estimated that there were 389,717 people age 90+ in the United Kingdom population in 2000, this being 0.65% of the population.17 It is estimated that by 2025 there will be 764,472 people aged 90+ in the United Kingdom population and that this will represent 1.2% of the population.17 Unfortunately this increase in the population is not being matched by better health. In 2008 the National Office of Statistics stated that while the population of the United Kingdom had been living longer over the previous 23 years, the time that both sexes could expect to be in poor health or have a limiting illness or disability had risen between 1981 and 2004.11 Despite there being some minor improvements after 2004 it is obvious that increasing longevity is going to be paired with increasing morbidity and hospitalisation for a range of
1113
different conditions. The fact that this is already happening in orthopaedic trauma is highlighted by extrapolating the Edinburgh figures which show that the 0.58% section of the population, which the 90+ patients make up, accounted for 3.02% of the fractures in the community, 8.7% of the in-patient admissions and 7.6% of all acute orthopaedic trauma surgery. There has been no previous epidemiological analysis of fractures in the 90+ age group. Several papers have been written about proximal femoral fractures in nonagenarians6,7,15 with a recent study showing that the extreme elderly were more likely to be in institutional care at the time of the fracture.7 They were also less likely to return home or to previous levels of mobility. We have no doubt that this is true but we believe that it would be inappropriate to restrict interest in fractures in the 90+ group to proximal femoral fractures. Table 1 shows that while proximal femoral fractures are the commonest fracture in the 90+ group, and have the highest prevalence, 44.1% of the fractures in this group are not proximal femoral fractures. Many surgeons may be surprised at the high prevalences of femoral diaphyseal and distal femoral fractures shown in Table 1. It is interesting to note that 18.6% of all femoral fractures occurred in the 90+ group. There were only two other fractures with a prevalence of more than 5% these being pelvic fractures and proximal humeral fractures. There was only one spinal fracture seen during the year. However it is likely that there were many more spinal fractures that did not present to hospital. It has been estimated that the overall incidence of spinal fractures is 117/105/year2 there is no available figure for the 90+ group. Distal radial fractures were the second most common fracture but because of their frequency in childhood and in middle aged patients their overall prevalence in the 90+ group was low. In a previous study of the epidemiology of fractures in the adult population Court-Brown and Caesar3 drew attention to the fact that fractures of the hand and foot were rarely found in older patients. Table 1 confirms that these fractures are extremely uncommon in the 90+ population. We were however surprised at the scarcity of ankle fractures. Analysis of the 599 ankle fractures in the overall population showed that 38.7% were caused by falls and since at least 86% of the fractures in the 90+ age group were caused by falls it is surprising that there were only 3 ankle fractures in the year. It is also surprising that there were no humeral diaphyseal fractures in the 90+ group given that 53.3% of the 75 humeral diaphyseal fractures in the overall fracture group were caused by falls. Fig. 3 shows the prevalence of fractures caused by falls in the overall population. There is a bimodal distribution affecting children and older adults but there is an almost linear increase in prevalence from about 30 years of age. In the older population it has been suggested that about 30% of falls result in injury and 10% in fractures. A review of the literature shows that a number of factors have been associated with falls and the injuries caused by falls. Examples are age1,2,4, general health1,2 medications2,4, shoe wear1,2,8 time of day,14 handedness,10 poverty,13 location,4,9,14 and type of fall.16 We assume that that the scarcity of ankle and humeral diaphyseal fractures relates to the type of fall in this very elderly group. There is no doubt that the overall distribution of fractures between the upper and lower limbs and the axial skeleton changes with age. Fig. 1 shows the distribution of fractures in the overall population. In younger patients upper limb fractures are much more common with the prevalence of lower limb fractures only exceeding that of upper limb fractures when patients reach 80–89 years of age. Axial skeleton fractures slowly increase with age with the maximum prevalence also being seen between 80 and 89 years. It is the increased prevalence of lower limb fractures in the 90+ group that accounts for the fact that this
1114
C.M. Court-Brown, N. Clement / Injury, Int. J. Care Injured 40 (2009) 1111–1114
0.58% section of the population requires 7.6% of the acute orthopaedic trauma surgical procedures. The requirement for surgery and the presence of medical comorbidities in this very elderly, frail group of patients means that the number that can be treated as out-patients is comparatively small. Fig. 2 shows that the percentage of in-patients correlates with increasing age and that fewer than 40% of patients are treated as in-patients until about 70 years of age. This highlights the problems associated with treatment of fractures in the very elderly and also indicates the significant expense in treating fractures in this group of patients. Table 2 shows that 48.9% of the in-patients lived at home at the time of fracture. However very few were discharged straight home. Of the patients who were discharged home those with upper limb fractures spent 3.7 days in hospital compared with 15.5 days for the very few patients with lower limb fractures who went straight home. This fact, combined with the data shown in Table 3 that about 30% of patients with upper limb fractures went straight home suggests that the patients with upper limb fractures were the fittest patients that were treated. Table 3 also shows that those patients with non-hip lower limb fractures had a higher hospital mortality and a longer average stay suggesting that these patients are probably less fit than the proximal femoral fracture population. It is obvious that the patients who present with fractures of the spine or pelvis are the least fit patients as all were admitted from other hospitals. It is a paradox that this group of patients had the same length of hospital stay as the fitter upper limb fracture population but this group is the only group with a guaranteed downstream hospital bed. Very little attention has been paid to the elderly non-hip fracture population. Our data suggests that patients with upper limb fractures are fitter and that less than half need secondary hospital care. However patients who present with other lower limb fractures would seem to be less fit and should be included in targets for early surgery that are currently reserved for patients with proximal femoral fractures. We believe that this group of patients will assume greater importance in the future. There will be more nonagenarians and centenarians in the population and it is likely that they will have more disabilities and medical comorbidities. Currently over 80% of this patient group needs in-patient treatment for their fractures
but fewer that 25% can be discharged directly to their domicile, which is usually a nursing or residential home. The difficulties and expense associated with their management is likely to increase considerably over the next few decades. Conflict of interest Neither author has any conflict of interest. References 1. Berry SD, Miller RR. Falls: epidemiology, pathophysiology, and relationship to fracture. Curr Osteoporos Rep 2008;6:149–54. 2. Cooper C, Atkinson EJ, O’Fallon WM, Melton LJ. Incidence of vertebral fractures: a population-based studying Rochester, Minnesota. J Bone Miner Res 1992;7: 221–7. 3. Court-Brown CM, Caesar B. Epidemiology of adult fractures: a review. Injury 2006;37:691–7. 4. Formiga F, Lopez-Soto A, Duaso E, Chivite D, Ruiz D, Perez-Castejon JM, et al. Characteristics of falls producing hip fractures in nonagenarians. J Nutr Health Aging 2008;12:664–7. 5. General Register Office for Scotland, 2001. http://www.gro-scotland.gov.uk/ census/censushm/index/html. 6. Holt G, Macdonald D, Fraser M, Reece AT. Outcome after surgery for fractures of the hip in patients aged over 95 years. J Bone Joint Surg Br 2006;88:1060–4. 7. Holt G, Smith R, Duncan K, Hutchison JD, Gregori A. Outcome after surgery for the treatment of hip fracture in the extremely elderly. J Bone Joint Surg Am 2008;90:1899–905. 8. Keegan THM, Kelsey JL, King AC, Quesenberry CP, Sidney S. Characteristics of fallers who fracture at the foot, distal forearm, proximal humerus, pelvis, and shaft of the tibia/fibula compared with fallers who do not fracture. Am J Epidemiol 2004;159:192–203. 9. Li W, Keegan THM, Sternfeld B, Sidney S, Quesenberry CP, Kelsey JL. Outdoor falls among middle-aged and older adults: a neglected public health problem. Am J Public Health 2006;96:1192–8. 10. Luetters CM, Kelsey JL, Keegan TH, Quesenberry CP, Sidney S. Left-handedness as a risk factor fro fractures. Osteoporos Int 2003;14:918–22. 11. National Statistics. http://www.statistics.gov.uk/cci/nugget_print.asp?ID=934. 12. National Statistics. http://www.statistics.gov.uk/cci/nugget_print.asp?ID=1875. 13. Navarro MC, Sosa M, Saavedra P, Lainez P, Marrero M, Torres M, et al. Poverty is a risk factor for osteoporotic fractures. Osteoporos Int 2009;20:393–8. 14. Niino N, Kozakai R, Eto M. Epidemiology of falls among community-dwelling elderly people. Nippon Ronen Igakkai Zasshi 2003;40:484–6. 15. Oliver CW, Burke C. Hip fractures in centenarians. Injury 2004;35:1025–30. 16. Palvanen M, Kannus P, Parkkari J, Pita¨ja¨rvi T, Pasanen M, Vuori I, et al. The injury mechanisms of osteoporotic upper extremity fractures among older adults: a controlled study of 287 consecutive patients and their 108 controls. Osteoporos Int 2000;11:822–31. 17. US Census Bureau. http://www.census.gov/ipc/www/idb.
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Four score years and ten An analysis of the epidemiology of fractures in the very elderly C.M. Court-Brown a,*, N. Clement b a b
University of Edinburgh, United Kingdom Royal Infirmary of Edinburgh, United Kingdom
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 11 June 2009
Fragility fractures are becoming more common and with increasing longevity it seems likely that orthopaedic surgeons will see more such fractures in progressively older patients. We have examined the epidemiology of fractures in patients aged 90 years or more. Introduction: The purpose of this study was to examine the epidemiology of fractures in the very elderly to see if it differed from the overall fracture epidemiology. The requirement for hospital admission and surgical treatment was assessed as was the patients’ domicile, length of hospital stay and discharge location. Materials and methods: All fractures presenting to the Royal Infirmary of Edinburgh and the Royal Hospital for Sick Children in 2000 were examined prospectively and recorded on databases. A retrospective examination of the clinical and social details of the 90+ age group was undertaken. Results: The 90+ age group comprises 0.58% of the population but accounts for 3.02% of the fractures in the community, 8.7% of the in-patient admissions and 7.6% of the acute orthopaedic trauma surgery. About 56% of fractures in this age group are proximal femoral fractures but high prevalences were noted for all femoral fractures. About 65% of the fractures were in the lower limb. Over 86% of fractures occurred in falls and over 81% of patients were treated as in-patients. The average hospital stay was 9 days but only 23.9% were discharged to their domicile. Discussion: It is forecast that this age group will increase and become less fit. This has significant implications for hospital treatment and costs. ß 2009 Elsevier Ltd. All rights reserved.
Keywords: Epidemiology Fractures Nonagenarians Osteoporosis Fragility fractures
In recent years fragility fractures have assumed considerable importance in orthopaedic surgery. With increasing longevity they are becoming more common and a recent study has suggested that about 30% of male fractures and 66% of female fractures should now be regarded as fragility fractures.3 It seems likely that the prevalence of fragility fractures will correlate with age but there is very little known about the epidemiology of fractures in the very elderly. There have been a number of studies of proximal femoral fractures in nonagenarians6,7,15 but there have been no previous studies of the other fractures that might occur in this age group. As it seems likely that these fractures will become more common in the next few decades we have undertaken an epidemiological analysis of all fractures in patients aged 90 years or more.
* Corresponding author at: Department of Orthopaedic Surgery, Royal Infirmary of Edinburgh, Old Dalkeith Road, Edinburgh EH16 4SU, United Kingdom. Tel.: +44 0131 242 3516; fax: +44 0131 660 4227. E-mail address: [email protected] (C.M. Court-Brown). 0020–1383/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2009.06.011
Materials and methods All fractures presenting to the Royal Infirmary of Edinburgh and The Royal Hospital for Sick Children in 2000 were analysed. All in-patients and out-patients were included. These hospitals are the only hospitals treating orthopaedic trauma in the City of Edinburgh, Midlothian and East Lothian. All patients resident outside these areas were excluded from analysis but all patients who were injured elsewhere but resident in these areas were included. Both hospitals maintain prospective in-patient and outpatient databases of all fractures. The databases were studied to allow us to detail the fracture type, mode of injury, patient age and gender and whether the patients were treated on an in-patient or out-patient basis. A retrospective analysis of the relevant hospital databases was undertaken to analyse the domicile at the time of injury, the length of hospital stay and the location after discharge of all patients aged 90 years or more. Statistical analysis of any changes in the prevalence of fractures in this age group during the decade of the 1990s was undertaken using Spearman rank correlations.
C.M. Court-Brown, N. Clement / Injury, Int. J. Care Injured 40 (2009) 1111–1114
1112
Fig. 1. The changes in prevalences of fractures of the lower limb, upper limb and the axial skeleton with increasing decades.
Results In the year 2000 the Royal Infirmary of Edinburgh and the Royal Hospital for Sick Children treated 7907 fractures. Two hundred and thirty-six (3.02%) fractures occurred in patients of at least 90 years of age (90+ group). The 236 fractures occurred in 208 patients with 12 (5.8%) of the patients presenting with more than one fracture and 10 patients presenting with different fractures on more than one occasion. Of the 208 patients 184 (88.5%) were female and 24 (11.5%) were male. The age range was 90–102 years. Analysis of the different fractures showed that 154 (65.3%) were lower limb fractures, 72 (30.5%) were upper limb fractures and 10 (4.2%) were fractures of the axial skeleton of which one was a spinal fracture and nine were pelvic fractures (Fig. 1). Table 1 gives a breakdown of the fractures according to their location. The overall number of fractures treated in 2000 is shown together with the numbers, prevalence and incidence of the fractures in the 90+ group. The percentage of each fracture that presented in patients of at least 90 years of age is also shown. An analysis of the fractures shows that 192 (81.4%) were treated on an in-patient basis (Fig. 2) with the remaining 44 (18.6%) Table 1 The numbers of fractures seen in 2000 together with the numbers and prevalences of fractures in the 90+ group. The incidence of fractures in the 90+ group is also shown.
Proximal femur Femoral diaphysis Distal femur Pelvis Proximal humerus Tibial diaphysis Spine Distal radius Distal humerus Proximal forearm Distal tibia Proximal tibia Forearm Clavicle Finger phalanges Metacarpals Ankle Metatarsal Patella Humeral diaphysis Scapula Midfoot Toe phalanges Carpus Talus Calcaneus Sesamoid
Overall, n
90+, n
Overall (%)
90+ (%)
n/105
694 71 29 95 375 159 45 1720 191 356 74 75 181 331 870 812 599 497 61 75 17 29 277 179 19 75 1
132 8 3 9 19 4 1 32 3 5 1 1 2 3 4 4 3 2 0 0 0 0 0 0 0 0 0
19.0 11.3 10.3 9.5 5.1 2.5 2.2 1.9 1.6 1.4 1.3 1.3 1.1 0.9 0.5 0.5 0.5 0.4 0 0 0 0 0 0 0 0 0
56.2 3.4 1.3 3.8 8.1 1.7 0.4 13.6 1.3 2.1 0.4 0.4 0.9 1.3 1.7 1.7 1.3 0.9 0 0 0 0 0 0 0 0 0
3314.1 203.1 75.3 226.0 477.1 100.4 25.1 803.4 75.3 125.5 25.1 25.1 50.2 75.3 100.4 100.4 75.3 50.2 0 0 0 0 0 0 0 0 0
Fig. 2. The prevalences of in-patient admissions in the different decades of life.
fractures being treated in the out-patient department. Thirty-nine (88.6%) of the out-patient fractures involved the upper limb the remaining five (11.4%) fractures being fractures of the distal tibia, ankle and metatarsals. Thirty-three (17.2%) of the 192 in-patient fractures were in the upper limb, 149 (77.2%) were in the lower limb and 10 (5.2%) were in the axial skeleton. Of the 180 patients admitted with fractures 88 (48.9%) of the patients came from home, 82 (45.6%) came from residential or nursing homes and 10 (5.6%) were admitted from other hospitals. The average length of stay was 9 days (range 1–54 days). Table 2 shows the discharge location and length of stay for all in-patient admissions and it can be seen that only 11 (6.1%) of all patients admitted with a fracture went directly home, in an average of 8.2 days. However a further 43 (23.9%) patients who were admitted from a residential or nursing home returned to their domicile in an average of 8.8 days. Thus a total of 54 (30%) of in-patients in the 90+ group returned to their domicile after admission with one or more fractures. Table 3 shows the discharge location and the average duration of hospitalisation for in-patients depending on their site of fracture. Where patients were admitted with both upper and lower limb fractures they have been included in the lower limb group as it was the lower limb fracture that usually dictated their discharge location. Three patients presented with two lower limb fractures. The lower limb fractures have been divided into proximal femoral and other fractures to allow comparison. Table 2 The numbers of in-patients living at home, in a residential or nursing home (RH/NH) or in another hospital prior to admission. Their outcome and average stay is given. Domicile
Home
RH/NH
Hospital
Patient number Died Home RH/NH Hospital Average stay (days)
88 5 (5.7%) 11 (12.5%) 0 72 (81.8%) 11.6
82 8 (9.8%) 0 43 (52.4%) 31 (37.8%) 11.2
10 0 0 0 10 (100%) 8.6
Table 3 The outcomes of in-patients depending on the location of the fracture. Where the patient has been admitted with upper and lower limb fractures the result has been included in the lower limb group. (RH/NH = residential or nursing home).
Patient number Died Home RH/NH Hospital Average stay (days)
Upper limb
Proximal femur
Other lower limb
Axial skeleton
24 1 (4.2%) 7 (29.2%) 5 (20.8%) 11 (45.8%) 6.0
129 10 (7.8%) 4 (3.1%) 32 (24.8%) 83 (64.3%) 12.2
17 2 (11.8%) 0 6 (35.3%) 9 (52.9%) 17.8
10 0 0 1 (10%) 9 (90%) 6.3
C.M. Court-Brown, N. Clement / Injury, Int. J. Care Injured 40 (2009) 1111–1114
Fig. 3. The prevalences of falls that caused fractures in the different decades of life.
Analysis of the 12 patients who presented with more than one fracture showed that 8 (66.6%) were admitted from home but that only 1 (8.3%) was discharged home. Their average length of hospital stay was 9.1 days. Analysis of the causes of fracture showed that 204 (86.4%) of fractures were caused by a simple fall (Fig. 3). This figure is in fact likely to be higher as 16 patients could not give a history, usually because of dementia. Only one fracture, a proximal femoral fracture in a 91-year-old female, was caused by a road traffic accident. Overall 151 (64%) fractures were treated surgically but there were significant differences depending on the location of the fracture. Only 12 (36.4%) of the upper limb fractures were treated surgically, seven of these being distal radial fractures. This compares with 97% of proximal femoral fractures and 50% of the other lower limb fractures. None of the spinal or pelvic fractures were treated surgically. A review of the fractures in the 90+ group treated in the decade between 1991 and 2000 showed a significant increase in numbers for certain fractures. There were significant upward trends in proximal femoral fractures (p = 0.002), distal femoral fractures (p = 0.008), proximal humeral fractures (p = 0.007), distal humeral fractures (p = 0.035), forearm fractures (p = 0.039) and distal radius fractures (p = 006). However an analysis of the overall population numbers for each year showed that the apparent increase in fracture numbers was caused by an overall progressive increase in the numbers of 90+ patients in the population. There was no statistical evidence of a relative increase in any fracture during the decade. Discussion Nonagenarians and centenarians are becoming an increasingly important part of society. They are increasing in number and the National Office of Statistics has shown that they are, in fact, the age group which is increasing most quickly.12 In 2001 at the time of the last UK Census there were 29,385 people aged 90+ years in Scotland this being 0.58% of the population.5 This figure matches that of the US Census website which estimated that there were 389,717 people age 90+ in the United Kingdom population in 2000, this being 0.65% of the population.17 It is estimated that by 2025 there will be 764,472 people aged 90+ in the United Kingdom population and that this will represent 1.2% of the population.17 Unfortunately this increase in the population is not being matched by better health. In 2008 the National Office of Statistics stated that while the population of the United Kingdom had been living longer over the previous 23 years, the time that both sexes could expect to be in poor health or have a limiting illness or disability had risen between 1981 and 2004.11 Despite there being some minor improvements after 2004 it is obvious that increasing longevity is going to be paired with increasing morbidity and hospitalisation for a range of
1113
different conditions. The fact that this is already happening in orthopaedic trauma is highlighted by extrapolating the Edinburgh figures which show that the 0.58% section of the population, which the 90+ patients make up, accounted for 3.02% of the fractures in the community, 8.7% of the in-patient admissions and 7.6% of all acute orthopaedic trauma surgery. There has been no previous epidemiological analysis of fractures in the 90+ age group. Several papers have been written about proximal femoral fractures in nonagenarians6,7,15 with a recent study showing that the extreme elderly were more likely to be in institutional care at the time of the fracture.7 They were also less likely to return home or to previous levels of mobility. We have no doubt that this is true but we believe that it would be inappropriate to restrict interest in fractures in the 90+ group to proximal femoral fractures. Table 1 shows that while proximal femoral fractures are the commonest fracture in the 90+ group, and have the highest prevalence, 44.1% of the fractures in this group are not proximal femoral fractures. Many surgeons may be surprised at the high prevalences of femoral diaphyseal and distal femoral fractures shown in Table 1. It is interesting to note that 18.6% of all femoral fractures occurred in the 90+ group. There were only two other fractures with a prevalence of more than 5% these being pelvic fractures and proximal humeral fractures. There was only one spinal fracture seen during the year. However it is likely that there were many more spinal fractures that did not present to hospital. It has been estimated that the overall incidence of spinal fractures is 117/105/year2 there is no available figure for the 90+ group. Distal radial fractures were the second most common fracture but because of their frequency in childhood and in middle aged patients their overall prevalence in the 90+ group was low. In a previous study of the epidemiology of fractures in the adult population Court-Brown and Caesar3 drew attention to the fact that fractures of the hand and foot were rarely found in older patients. Table 1 confirms that these fractures are extremely uncommon in the 90+ population. We were however surprised at the scarcity of ankle fractures. Analysis of the 599 ankle fractures in the overall population showed that 38.7% were caused by falls and since at least 86% of the fractures in the 90+ age group were caused by falls it is surprising that there were only 3 ankle fractures in the year. It is also surprising that there were no humeral diaphyseal fractures in the 90+ group given that 53.3% of the 75 humeral diaphyseal fractures in the overall fracture group were caused by falls. Fig. 3 shows the prevalence of fractures caused by falls in the overall population. There is a bimodal distribution affecting children and older adults but there is an almost linear increase in prevalence from about 30 years of age. In the older population it has been suggested that about 30% of falls result in injury and 10% in fractures. A review of the literature shows that a number of factors have been associated with falls and the injuries caused by falls. Examples are age1,2,4, general health1,2 medications2,4, shoe wear1,2,8 time of day,14 handedness,10 poverty,13 location,4,9,14 and type of fall.16 We assume that that the scarcity of ankle and humeral diaphyseal fractures relates to the type of fall in this very elderly group. There is no doubt that the overall distribution of fractures between the upper and lower limbs and the axial skeleton changes with age. Fig. 1 shows the distribution of fractures in the overall population. In younger patients upper limb fractures are much more common with the prevalence of lower limb fractures only exceeding that of upper limb fractures when patients reach 80–89 years of age. Axial skeleton fractures slowly increase with age with the maximum prevalence also being seen between 80 and 89 years. It is the increased prevalence of lower limb fractures in the 90+ group that accounts for the fact that this
1114
C.M. Court-Brown, N. Clement / Injury, Int. J. Care Injured 40 (2009) 1111–1114
0.58% section of the population requires 7.6% of the acute orthopaedic trauma surgical procedures. The requirement for surgery and the presence of medical comorbidities in this very elderly, frail group of patients means that the number that can be treated as out-patients is comparatively small. Fig. 2 shows that the percentage of in-patients correlates with increasing age and that fewer than 40% of patients are treated as in-patients until about 70 years of age. This highlights the problems associated with treatment of fractures in the very elderly and also indicates the significant expense in treating fractures in this group of patients. Table 2 shows that 48.9% of the in-patients lived at home at the time of fracture. However very few were discharged straight home. Of the patients who were discharged home those with upper limb fractures spent 3.7 days in hospital compared with 15.5 days for the very few patients with lower limb fractures who went straight home. This fact, combined with the data shown in Table 3 that about 30% of patients with upper limb fractures went straight home suggests that the patients with upper limb fractures were the fittest patients that were treated. Table 3 also shows that those patients with non-hip lower limb fractures had a higher hospital mortality and a longer average stay suggesting that these patients are probably less fit than the proximal femoral fracture population. It is obvious that the patients who present with fractures of the spine or pelvis are the least fit patients as all were admitted from other hospitals. It is a paradox that this group of patients had the same length of hospital stay as the fitter upper limb fracture population but this group is the only group with a guaranteed downstream hospital bed. Very little attention has been paid to the elderly non-hip fracture population. Our data suggests that patients with upper limb fractures are fitter and that less than half need secondary hospital care. However patients who present with other lower limb fractures would seem to be less fit and should be included in targets for early surgery that are currently reserved for patients with proximal femoral fractures. We believe that this group of patients will assume greater importance in the future. There will be more nonagenarians and centenarians in the population and it is likely that they will have more disabilities and medical comorbidities. Currently over 80% of this patient group needs in-patient treatment for their fractures
but fewer that 25% can be discharged directly to their domicile, which is usually a nursing or residential home. The difficulties and expense associated with their management is likely to increase considerably over the next few decades. Conflict of interest Neither author has any conflict of interest. References 1. Berry SD, Miller RR. Falls: epidemiology, pathophysiology, and relationship to fracture. Curr Osteoporos Rep 2008;6:149–54. 2. Cooper C, Atkinson EJ, O’Fallon WM, Melton LJ. Incidence of vertebral fractures: a population-based studying Rochester, Minnesota. J Bone Miner Res 1992;7: 221–7. 3. Court-Brown CM, Caesar B. Epidemiology of adult fractures: a review. Injury 2006;37:691–7. 4. Formiga F, Lopez-Soto A, Duaso E, Chivite D, Ruiz D, Perez-Castejon JM, et al. Characteristics of falls producing hip fractures in nonagenarians. J Nutr Health Aging 2008;12:664–7. 5. General Register Office for Scotland, 2001. http://www.gro-scotland.gov.uk/ census/censushm/index/html. 6. Holt G, Macdonald D, Fraser M, Reece AT. Outcome after surgery for fractures of the hip in patients aged over 95 years. J Bone Joint Surg Br 2006;88:1060–4. 7. Holt G, Smith R, Duncan K, Hutchison JD, Gregori A. Outcome after surgery for the treatment of hip fracture in the extremely elderly. J Bone Joint Surg Am 2008;90:1899–905. 8. Keegan THM, Kelsey JL, King AC, Quesenberry CP, Sidney S. Characteristics of fallers who fracture at the foot, distal forearm, proximal humerus, pelvis, and shaft of the tibia/fibula compared with fallers who do not fracture. Am J Epidemiol 2004;159:192–203. 9. Li W, Keegan THM, Sternfeld B, Sidney S, Quesenberry CP, Kelsey JL. Outdoor falls among middle-aged and older adults: a neglected public health problem. Am J Public Health 2006;96:1192–8. 10. Luetters CM, Kelsey JL, Keegan TH, Quesenberry CP, Sidney S. Left-handedness as a risk factor fro fractures. Osteoporos Int 2003;14:918–22. 11. National Statistics. http://www.statistics.gov.uk/cci/nugget_print.asp?ID=934. 12. National Statistics. http://www.statistics.gov.uk/cci/nugget_print.asp?ID=1875. 13. Navarro MC, Sosa M, Saavedra P, Lainez P, Marrero M, Torres M, et al. Poverty is a risk factor for osteoporotic fractures. Osteoporos Int 2009;20:393–8. 14. Niino N, Kozakai R, Eto M. Epidemiology of falls among community-dwelling elderly people. Nippon Ronen Igakkai Zasshi 2003;40:484–6. 15. Oliver CW, Burke C. Hip fractures in centenarians. Injury 2004;35:1025–30. 16. Palvanen M, Kannus P, Parkkari J, Pita¨ja¨rvi T, Pasanen M, Vuori I, et al. The injury mechanisms of osteoporotic upper extremity fractures among older adults: a controlled study of 287 consecutive patients and their 108 controls. Osteoporos Int 2000;11:822–31. 17. US Census Bureau. http://www.census.gov/ipc/www/idb.