Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
Contents lists available at ScienceDirect
Best Practice & Research Clinical Rheumatology journal homepage: www.elsevierhealth.com/berh
3
Other non-vertebral fractures Hiroshi Hagino School of Health Science and Rehabilitation Division, Tottori University, Nishicho 86, Yonago, Tottori 683-8503, Japan
a b s t r a c t Keywords: Distal radius fracture Proximal humerus fracture Pelvic fracture Epidemiology Osteoporosis Quality of life Surgical treatment
Non-vertebral non-hip (NVNH) fractures account for 90% of all fractures in patients up to 80 years of age and for 59% thereafter. There is a significant relationship between reductions in peripheral bone mineral density and the risk of fractures at various NVNH sites except for the face. Fractures of the clavicle, upper arm, forearm, spine, ribs, hip, pelvis, upper leg and lower leg elevate the risk of future fractures. Among NVNH fractures in women aged 80 years or over, forearm fractures have the highest incidence, and proximal humerus fractures have the second highest incidence. There is a large variation in incidence across geographical regions, with incidence higher in Northern Europe and lower in Asia and Africa. NVNH fractures are associated with higher mortality and significantly higher health-care costs than controls with osteoporosis. Reductions in health-related quality of life (HRQOL) for women with major NVNH fractures are of a similar magnitude as reductions for women with incident hip fractures; however, forearm fractures do not significantly affect HRQOL. Therapeutic options for NVNH fractures differ by fracture location. The recent development of implants for internal fixation made it a more popular choice for treating distal radius and proximal humerus fractures; however, treatment decisions should take into account patient age, activity levels, co-morbidities and injury characteristics. The recent increase in the number of patients with osteoporotic pelvic fractures is drastic, although they can generally be treated non-surgically with pain management and mobilisation. Ó 2014 Elsevier Ltd. All rights reserved.
E-mail address: [email protected]. 1521-6942/$ – see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.berh.2014.01.003
732
H. Hagino / Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
Introduction Osteoporosis has been called ‘the silent disease’ because there are no symptoms until a fragility fracture occurs. Fragility fractures increase societal burdens in terms of quality of life and mortality, as well as economic costs. As vertebral and hip fractures have the greatest impact on the ambulatory status of the elderly, most attention is focussed on these types of fractures. Consequently, there are insufficient data on the epidemiology and outcomes of other types of fragility fractures. Non-vertebral non-hip (NVNH) fractures account for 90% of all fractures until 80 years of age and for 59% thereafter [1]. The most common NVNH fractures in children and young adults involve the carpal bones, bones of the feet, tibia and fibula, while fractures of the spine, distal radius, hip, proximal humerus, ribs, clavicle and pelvis occur more frequently in older patients. In women aged 80 years or over, forearm fractures have the highest incidence among NVNH fractures, followed by proximal humerus fractures [1]. For individuals aged 50, the lifetime risks for forearm and proximal humeral fractures are 22.9% and 12.9% in women and 10.7% and 4.1% in men, respectively (Table 1) [2]. What are NVNH fragility fractures? Fragility fractures are associated with osteoporosis, as opposed to trauma. However, there is no conclusive definition of fragility fracture. Based on recent epidemiological studies, fragility fractures are those due to mild injuries, such as a fall from standing height, and not the result of significant trauma, such as a motor vehicle collision. Fragility fractures could be regarded as fractures that occur with a reduction in bone mass, an increased risk of subsequent fragility fractures or both. Fractures related to low bone density Although there is no accurate marker of overall bone strength, bone mineral density (BMD), which accounts for approximately 70% of bone strength, is frequently used as a proxy [3]. In a large cohort of US women aged 65 or older, Stone et al. found relationships between BMD measured at several sites and subsequent fracture risk at multiple sites over more than 8 years of follow-up [4]. They found a significant relationship between a reduction of peripheral BMD and all fractures observed, except for fractures involving the face. These results indicate that NVNH fractures at various sites are related to osteoporosis. Fractures associated with a high risk of subsequent fractures A new fragility fracture not only reduces daily activity levels, but can also increase the risk of subsequent fractures, either refracture at the same site or new fractures elsewhere. In patients with a pre-existing vertebral fracture, the relative risk of a new vertebral fracture is approximately 3–4 times greater than in individuals without pre-existing fractures after correcting for BMD, while the relative risk of a new fracture in the proximal femur is 3–5 times greater [5,6]. In patients with a pre-existing hip fracture, the relative risk of a new fracture is 2.5–6.8 times [7] greater than in those without preexisting fractures. Forearm fractures also increase the risk of subsequent forearm, vertebral and hip fractures by as much as 3.3-, 1.7- and 1.9-fold, respectively, in perimenopausal or postmenopausal women [5]. However, there is a paucity of data available on the role of other fracture sites in conferring risk. Recently, results from the prospective Global Longitudinal Study of Osteoporosis in Women Table 1 Remaining lifetime risk of fracture (%) in men and women at 50 and 80 years of age. At 50 years
Forearm Proximal humerus Cited from Ref. [2].
At 80 years
Men
Women
Risk ratio
Men
Women
Risk ratio
10.7 4.1
22.9 12.9
2.1 3.1
9.1 2.5
19.3 7.7
2.1 3.1
H. Hagino / Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
733
(GLOW) indicate that there is a relationship between pre-existing fractures at 10 skeletal locations, such as the clavicle, upper arm, forearm (wrist), spine (clinical vertebrae), rib, hip, pelvis, upper leg, lower leg, ankle and subsequent fractures [8,9]. Among the 12,009 women with prior fractures, an estimated 11.6% will experience another fracture by 2 years after baseline. The incidence of subsequent fracture was significantly associated with all prior fracture locations except for the clavicle (Fig. 1) [8]. In another report from GLOW, patients with pre-existing clavicle fractures had a significantly higher hazard ratio for subsequent fractures [9]. Thus, a broader range of NVNH fractures than is commonly thought is potentially associated with an increased risk of subsequent fractures. Thus, evaluation of osteoporosis and appropriate prevention strategies are essential in the management of patients with NVNH fractures. What is the burden? Incidence Forearm fractures The incidence of forearm (distal radius) fractures increases just after menopause and is the highest among non-vertebral fractures in women under 75 years of age [1,10]. This suggests that like vertebral fractures, forearm fractures are the first sign of osteoporosis in women who will benefit from the initiation of prevention strategies for subsequent fragility fractures. The incidence of forearm (distal radius) fractures highly varies across geographical regions. The incidence is higher in Northern Europe and lower in Asia and Africa, similar to trends for hip (Fig. 2) [2,10–18]. As the bone mass of Asians is known to be similar to or lower than that of Caucasians, differences in bone mass do not explain differences in the incidence of forearm fractures. One possible explanation is a lower risk of falls in the Asian population [19], presumably based on differences in lifestyles. The prevalence of forearm fractures increases with increasing degrees of urbanisation in both genders [20]. After adjustment for age and other explanatory factors, the odds ratios for having sustained a forearm fracture living in densely populated areas and cities were 1.12 and 1.38 in men, and 1.12 and 1.37 in women, respectively. Different lifestyles, or more falls in urban populations compared to rural populations, might account for this difference. The age-standardised incidence of forearm (distal radius) fractures among Icelandic women almost doubled between 1989–1992 and 1997–2000. After peaking in 1997–2000, the incidence declined by
Fig. 1. Risk of incident fractures by location of prior fracture. Data are presented as hazard ratios (95% confidence intervals). (Data are adopted from Ref. [8]).
734
H. Hagino / Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
Fig. 2. Age-specific incidence of forearm (distal radius) fractures in women from different geographic regions. There is a wide variation in incidence across geographical regions, with incidence higher in Northern Europe and lower in Asia and Africa. (Data are adopted from previous surveys: Rochester (13), Freideriskberg (14), Bergen (15), Uppsala (16), Malmo (2), Idaban (17), and Tottori (10)).
19% as of 2005–2008 [21]. In the Asian population, a significant increase in the incidence over time was observed in women [10]. Proximal humerus fractures The age-specific and gender-specific incidence of proximal humeral fractures increases with age (Fig. 3) [2,10,22]. There are substantial differences in age-adjusted incidence across geographic regions. It is higher in Northern Europe and lower in Asia, similar to the incidence of hip and distal radius fractures [10,18]. The age-adjusted incidence in Finland increased between 1970 and 2002 in both genders [23]. In Iceland, the incidence increased until 2001, when it started to decline for women over the last decade, but not for men [21]. Among Asian men and women, significant increases were observed from 1986 to 1995 [10]. Other non-vertebral fractures Only scant data are available for the incidence of NVNH fractures in sites other than the distal radius and proximal humerus. In a large 3-year population-based study in England, the annual incidence of fracture was 6.5 per 10,000 for the ankle and 3.2 per 10,000 for the clavicle (Table 2) [24]. Similar data were observed in a 10-year trial in England and Wales [1]. The most frequent fracture sites were the distal radius in women (30.2 per 10,000 person-years) and the carpal bones in men (26.2 per 10,000 person-years). Pelvic fracture is an important clinical entity as the excess treatment costs over 1 year for pelvic fractures in patients with osteoporosis are the second highest among non-vertebral fractures [25]. The incidence of pelvic fractures increases with age and 94% of pelvic fractures in patients aged 60 years or older are classified as osteoporotic fractures [26,27]. With a rapid increase in the proportion of the very elderly, the number of osteoporotic pelvic fractures increased 4.6-fold from 1970 to 1997. Mortality It has been well documented that mortality in patients after hip and vertebral fractures is much higher than that in the general population [28]. The relative risk of dying is sixfold higher following a
H. Hagino / Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
735
Fig. 3. Age-specific incidence of proximal humerus fractures in women from different geographic regions. There are substantial differences in age-adjusted incidence between geographic regions, with incidence higher in Northern Europe and lower in Asia. (Data are adopted from previous surveys: Malmo (2), Edinburgh (22), Tottori (10)).
hip fracture and ninefold greater following a vertebral fracture. By contrast, the risk of dying is unchanged after a forearm fracture [28]. A case–control study from the UK showed that mortality was twofold higher after a humerus fracture compared to controls, whereas there was no difference after a forearm fracture compared to controls [29]. The Dubbo Osteoporosis Epidemiology Study, a recent prospective cohort study, demonstrated that mortality rates associated with all major fractures (pelvis, distal femur, proximal tibia, three or more simultaneous ribs and proximal humerus) at all ages and even with minor fractures (all remaining NVNH fractures) in older age groups were increased for the first 5 years following the fracture [30]. In this study, the age-specific and gender-specific standardised mortality ratio was 1.65 for major fractures and 1.42 for minor fractures. Economic burden Estimated health-care costs for patients in the USA with NVNH fractures (e.g., femur, pelvis and forearm) range from $3000 to $29,600 in the year following a fracture, depending on the fracture site. [25] In Europe, NVNH fracture cost estimates range from V565 to V3651 [31]. Fracture-specific costs in women aged 50 years or older were evaluated recently using data from the Health Improvement Network (THIN) in the UK [29]. One-year costs in patients with closed fractures were calculated from the date of diagnosis and compared to patients without fractures. The mean 1-year incremental cost for fracture versus no fracture was £690 for NVNH fractures (£1152 for vertebral fractures). The 1-year incremental cost per woman was £525 (total cost, £1390) among Table 2 Average annual incidence (per 10,000 population) of fractures at different sites. Fracture site
Men
Women
Both genders
Radius and ulna (lower end) Ankle Humerus (other than upper end) Metacarpals Tarsals and metatarsals Radius and ulna (upper end and shaft) Clavicle Tibia and fibula (shaft) Tibia and fibula (upper end) Pelvis Humerus (upper end)
18.2 8.7 3.6 11.7 7.3 6.1 6.7 5.8 3.4 1.6 2.2
24.3 6.5 3.6 1.9 5.2 5.5 3.2 2.4 1.9 1.8 4.0
21.3 7.6 3.6 6.7 6.2 5.8 4.9 4.1 2.6 1.7 3.1
Cited from Ref. [24].
736
H. Hagino / Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
those with forearm (wrist) fractures and £602 (total cost, £1604) among those with humerus fractures. Pike et al. investigated direct costs of treating patients with osteoporosis who experience nonvertebral fractures during the 2 years following the incident fracture using a private health insurance claims database in the US from 1999 to 2006 [25]. Mean direct excess costs for non-vertebral fracture patients fell from $5267 in the first year to $2072 in the second year after a fracture. They found that patients with fractures of the pelvis, hip, femur and forearm had excess costs of $7,083, $13,334, $4211 and $5009 in the first year, and $5,121, $3,930, $3828 and $1562 in the second year, respectively. NVNH fracture patients made up a larger proportion of the sample and had significantly higher costs than controls with osteoporosis but without a fracture history, although hip fractures had the highest excess costs over both years. NVNH patients accounted for over 6 times the total costs to payers as hip fracture patients over the entire 2-year period ($21.7 billion versus $3.4 billion). Thus, NVNH fracture patients represent an important subpopulation from an economic point of view, even though they are often overlooked. What are the outcomes? In addition to pain and limitations in physical function, fragility fractures decrease mobility, social interaction and emotional well-being. In the Canadian Multicentre Osteoporosis Study, pelvic, lower limb and rib fractures were associated with very low health-related quality of life (HRQOL) scores [32]. Baseline data, as well as prospective data from GLOW, suggest that a history of fracture at various sites is associated with significant reductions in HRQOL [33,34]. Roux et al. reported changes in HRQOL when classifying NVNH fractures as major fractures (pelvis/leg and shoulder/arm) and minor fractures (wrist/hand, ankle/foot, rib/clavicle) [34]. They found that both minor and major incident NVNH fractures have a strong impact on HRQOL in women aged 55 years or more, as measured by EuroQOL five dimensions (EQ-5D) over 1 year, independent of baseline HRQOL. Reductions in EQ-5D and Short Form (36) (SF-36) physical functioning scores in women with major NVNH fractures were of a magnitude similar to reductions in women with incident hip fractures, but the reduction was smaller than for women who experienced incident spine fractures. By contrast, forearm fractures entailed fewer consequences. In patients with Colles’ fractures, Dolan et al. observed considerable reductions in HRQOL in the first 3 months, but recovery was fast and the impairment after recovery was small [35]. In a 1-year prospective study of patients with new distal radius fractures, EQ-5D values at 6 months showed recovery to pre-fracture levels [36]. Forearm fractures cause pain and loss of function, but the prognosis for fracture healing and regaining of function is usually favourable, especially in the elderly who do not have jobs. What is the optimal early management for ensuring the best outcomes in the short term and the long term? Therapeutic options for NVNH fractures vary across fracture sites. The recent development of implants for internal fixation made open reduction internal fixation (ORIF) more common for distal radius fractures and proximal humerus fractures. Decisions on treatment strategy should take into account the patient’s age, activity levels, co-morbidities and injury characteristics. Distal radius fractures The goal of distal radius fracture treatment is to restore acceptable mobility and durability to an upper extremity. There are four treatment options for distal radius fractures: closed reduction and immobilisation, closed reduction and pinning, external fixation and ORIF using plates. Closed reduction, external fixation and ORIF each has advantages and disadvantages. There has been a shift towards the use of volar locking plates for the treatment of displaced distal radius fractures (Fig. 4) [37]. However, there is no definitive consensus on an appropriate treatment for unstable distal radius fractures in elderly patients, because only minor differences exist between functional outcomes and activities of daily living after surgical and non-surgical treatment 1 year after injury [38,39]. The American Academy of Orthopaedic Surgeons (AAOS) clinical practice guidelines state that the Academy is unable to recommend for or against operative treatment in patients with distal radius fractures over
H. Hagino / Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
737
Fig. 4. Radiograph of a distal radius fracture. A. First visit at 2 months after the injury and B. after fixation with bone grafting using a locking plate. A 72-year-old woman was injured when she fell in her home. Two weeks after the injury she visited a clinic due to persistent pain. Manual reduction was attempted and cast immobilisation was performed. Two months after the injury, she visited our hospital due to pain around the ulnar end. Open reduction with osteotomy and autologous bone grafting was performed using a locking plate.
the age of 55 [37]. This recommendation is based on previous prospective, randomised trials in patients with displaced distal radius fractures who were all above the age of 55. It is important to take patient expectations and activity levels into account over age when considering operative versus non-operative treatment. Less aggressive treatment for older patients is advised [37]. When counselling, patients with distal radius fractures, especially low-energy fragility fractures, the health-care provider should discuss both operative and non-operative options. The surgeon should also discuss important health concerns, such as osteoporosis, and make appropriate referrals to either a primary care physician or an osteoporosis specialist for follow-up as needed [40]. Proximal humerus fractures Most patients with proximal humerus fractures regain a functional shoulder without surgery. Surgery should only be considered in approximately 20% of patients, either because they require better shoulder function or because their fracture is more complex. Murray et al. proposed a flowchart depicting the treatment algorithm (Fig. 5) [41]. For elderly patients over 85 years of age, surgical treatment is rarely indicated. Surgery is absolutely indicated in
Contents lists available at ScienceDirect
Best Practice & Research Clinical Rheumatology journal homepage: www.elsevierhealth.com/berh
3
Other non-vertebral fractures Hiroshi Hagino School of Health Science and Rehabilitation Division, Tottori University, Nishicho 86, Yonago, Tottori 683-8503, Japan
a b s t r a c t Keywords: Distal radius fracture Proximal humerus fracture Pelvic fracture Epidemiology Osteoporosis Quality of life Surgical treatment
Non-vertebral non-hip (NVNH) fractures account for 90% of all fractures in patients up to 80 years of age and for 59% thereafter. There is a significant relationship between reductions in peripheral bone mineral density and the risk of fractures at various NVNH sites except for the face. Fractures of the clavicle, upper arm, forearm, spine, ribs, hip, pelvis, upper leg and lower leg elevate the risk of future fractures. Among NVNH fractures in women aged 80 years or over, forearm fractures have the highest incidence, and proximal humerus fractures have the second highest incidence. There is a large variation in incidence across geographical regions, with incidence higher in Northern Europe and lower in Asia and Africa. NVNH fractures are associated with higher mortality and significantly higher health-care costs than controls with osteoporosis. Reductions in health-related quality of life (HRQOL) for women with major NVNH fractures are of a similar magnitude as reductions for women with incident hip fractures; however, forearm fractures do not significantly affect HRQOL. Therapeutic options for NVNH fractures differ by fracture location. The recent development of implants for internal fixation made it a more popular choice for treating distal radius and proximal humerus fractures; however, treatment decisions should take into account patient age, activity levels, co-morbidities and injury characteristics. The recent increase in the number of patients with osteoporotic pelvic fractures is drastic, although they can generally be treated non-surgically with pain management and mobilisation. Ó 2014 Elsevier Ltd. All rights reserved.
E-mail address: [email protected]. 1521-6942/$ – see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.berh.2014.01.003
732
H. Hagino / Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
Introduction Osteoporosis has been called ‘the silent disease’ because there are no symptoms until a fragility fracture occurs. Fragility fractures increase societal burdens in terms of quality of life and mortality, as well as economic costs. As vertebral and hip fractures have the greatest impact on the ambulatory status of the elderly, most attention is focussed on these types of fractures. Consequently, there are insufficient data on the epidemiology and outcomes of other types of fragility fractures. Non-vertebral non-hip (NVNH) fractures account for 90% of all fractures until 80 years of age and for 59% thereafter [1]. The most common NVNH fractures in children and young adults involve the carpal bones, bones of the feet, tibia and fibula, while fractures of the spine, distal radius, hip, proximal humerus, ribs, clavicle and pelvis occur more frequently in older patients. In women aged 80 years or over, forearm fractures have the highest incidence among NVNH fractures, followed by proximal humerus fractures [1]. For individuals aged 50, the lifetime risks for forearm and proximal humeral fractures are 22.9% and 12.9% in women and 10.7% and 4.1% in men, respectively (Table 1) [2]. What are NVNH fragility fractures? Fragility fractures are associated with osteoporosis, as opposed to trauma. However, there is no conclusive definition of fragility fracture. Based on recent epidemiological studies, fragility fractures are those due to mild injuries, such as a fall from standing height, and not the result of significant trauma, such as a motor vehicle collision. Fragility fractures could be regarded as fractures that occur with a reduction in bone mass, an increased risk of subsequent fragility fractures or both. Fractures related to low bone density Although there is no accurate marker of overall bone strength, bone mineral density (BMD), which accounts for approximately 70% of bone strength, is frequently used as a proxy [3]. In a large cohort of US women aged 65 or older, Stone et al. found relationships between BMD measured at several sites and subsequent fracture risk at multiple sites over more than 8 years of follow-up [4]. They found a significant relationship between a reduction of peripheral BMD and all fractures observed, except for fractures involving the face. These results indicate that NVNH fractures at various sites are related to osteoporosis. Fractures associated with a high risk of subsequent fractures A new fragility fracture not only reduces daily activity levels, but can also increase the risk of subsequent fractures, either refracture at the same site or new fractures elsewhere. In patients with a pre-existing vertebral fracture, the relative risk of a new vertebral fracture is approximately 3–4 times greater than in individuals without pre-existing fractures after correcting for BMD, while the relative risk of a new fracture in the proximal femur is 3–5 times greater [5,6]. In patients with a pre-existing hip fracture, the relative risk of a new fracture is 2.5–6.8 times [7] greater than in those without preexisting fractures. Forearm fractures also increase the risk of subsequent forearm, vertebral and hip fractures by as much as 3.3-, 1.7- and 1.9-fold, respectively, in perimenopausal or postmenopausal women [5]. However, there is a paucity of data available on the role of other fracture sites in conferring risk. Recently, results from the prospective Global Longitudinal Study of Osteoporosis in Women Table 1 Remaining lifetime risk of fracture (%) in men and women at 50 and 80 years of age. At 50 years
Forearm Proximal humerus Cited from Ref. [2].
At 80 years
Men
Women
Risk ratio
Men
Women
Risk ratio
10.7 4.1
22.9 12.9
2.1 3.1
9.1 2.5
19.3 7.7
2.1 3.1
H. Hagino / Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
733
(GLOW) indicate that there is a relationship between pre-existing fractures at 10 skeletal locations, such as the clavicle, upper arm, forearm (wrist), spine (clinical vertebrae), rib, hip, pelvis, upper leg, lower leg, ankle and subsequent fractures [8,9]. Among the 12,009 women with prior fractures, an estimated 11.6% will experience another fracture by 2 years after baseline. The incidence of subsequent fracture was significantly associated with all prior fracture locations except for the clavicle (Fig. 1) [8]. In another report from GLOW, patients with pre-existing clavicle fractures had a significantly higher hazard ratio for subsequent fractures [9]. Thus, a broader range of NVNH fractures than is commonly thought is potentially associated with an increased risk of subsequent fractures. Thus, evaluation of osteoporosis and appropriate prevention strategies are essential in the management of patients with NVNH fractures. What is the burden? Incidence Forearm fractures The incidence of forearm (distal radius) fractures increases just after menopause and is the highest among non-vertebral fractures in women under 75 years of age [1,10]. This suggests that like vertebral fractures, forearm fractures are the first sign of osteoporosis in women who will benefit from the initiation of prevention strategies for subsequent fragility fractures. The incidence of forearm (distal radius) fractures highly varies across geographical regions. The incidence is higher in Northern Europe and lower in Asia and Africa, similar to trends for hip (Fig. 2) [2,10–18]. As the bone mass of Asians is known to be similar to or lower than that of Caucasians, differences in bone mass do not explain differences in the incidence of forearm fractures. One possible explanation is a lower risk of falls in the Asian population [19], presumably based on differences in lifestyles. The prevalence of forearm fractures increases with increasing degrees of urbanisation in both genders [20]. After adjustment for age and other explanatory factors, the odds ratios for having sustained a forearm fracture living in densely populated areas and cities were 1.12 and 1.38 in men, and 1.12 and 1.37 in women, respectively. Different lifestyles, or more falls in urban populations compared to rural populations, might account for this difference. The age-standardised incidence of forearm (distal radius) fractures among Icelandic women almost doubled between 1989–1992 and 1997–2000. After peaking in 1997–2000, the incidence declined by
Fig. 1. Risk of incident fractures by location of prior fracture. Data are presented as hazard ratios (95% confidence intervals). (Data are adopted from Ref. [8]).
734
H. Hagino / Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
Fig. 2. Age-specific incidence of forearm (distal radius) fractures in women from different geographic regions. There is a wide variation in incidence across geographical regions, with incidence higher in Northern Europe and lower in Asia and Africa. (Data are adopted from previous surveys: Rochester (13), Freideriskberg (14), Bergen (15), Uppsala (16), Malmo (2), Idaban (17), and Tottori (10)).
19% as of 2005–2008 [21]. In the Asian population, a significant increase in the incidence over time was observed in women [10]. Proximal humerus fractures The age-specific and gender-specific incidence of proximal humeral fractures increases with age (Fig. 3) [2,10,22]. There are substantial differences in age-adjusted incidence across geographic regions. It is higher in Northern Europe and lower in Asia, similar to the incidence of hip and distal radius fractures [10,18]. The age-adjusted incidence in Finland increased between 1970 and 2002 in both genders [23]. In Iceland, the incidence increased until 2001, when it started to decline for women over the last decade, but not for men [21]. Among Asian men and women, significant increases were observed from 1986 to 1995 [10]. Other non-vertebral fractures Only scant data are available for the incidence of NVNH fractures in sites other than the distal radius and proximal humerus. In a large 3-year population-based study in England, the annual incidence of fracture was 6.5 per 10,000 for the ankle and 3.2 per 10,000 for the clavicle (Table 2) [24]. Similar data were observed in a 10-year trial in England and Wales [1]. The most frequent fracture sites were the distal radius in women (30.2 per 10,000 person-years) and the carpal bones in men (26.2 per 10,000 person-years). Pelvic fracture is an important clinical entity as the excess treatment costs over 1 year for pelvic fractures in patients with osteoporosis are the second highest among non-vertebral fractures [25]. The incidence of pelvic fractures increases with age and 94% of pelvic fractures in patients aged 60 years or older are classified as osteoporotic fractures [26,27]. With a rapid increase in the proportion of the very elderly, the number of osteoporotic pelvic fractures increased 4.6-fold from 1970 to 1997. Mortality It has been well documented that mortality in patients after hip and vertebral fractures is much higher than that in the general population [28]. The relative risk of dying is sixfold higher following a
H. Hagino / Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
735
Fig. 3. Age-specific incidence of proximal humerus fractures in women from different geographic regions. There are substantial differences in age-adjusted incidence between geographic regions, with incidence higher in Northern Europe and lower in Asia. (Data are adopted from previous surveys: Malmo (2), Edinburgh (22), Tottori (10)).
hip fracture and ninefold greater following a vertebral fracture. By contrast, the risk of dying is unchanged after a forearm fracture [28]. A case–control study from the UK showed that mortality was twofold higher after a humerus fracture compared to controls, whereas there was no difference after a forearm fracture compared to controls [29]. The Dubbo Osteoporosis Epidemiology Study, a recent prospective cohort study, demonstrated that mortality rates associated with all major fractures (pelvis, distal femur, proximal tibia, three or more simultaneous ribs and proximal humerus) at all ages and even with minor fractures (all remaining NVNH fractures) in older age groups were increased for the first 5 years following the fracture [30]. In this study, the age-specific and gender-specific standardised mortality ratio was 1.65 for major fractures and 1.42 for minor fractures. Economic burden Estimated health-care costs for patients in the USA with NVNH fractures (e.g., femur, pelvis and forearm) range from $3000 to $29,600 in the year following a fracture, depending on the fracture site. [25] In Europe, NVNH fracture cost estimates range from V565 to V3651 [31]. Fracture-specific costs in women aged 50 years or older were evaluated recently using data from the Health Improvement Network (THIN) in the UK [29]. One-year costs in patients with closed fractures were calculated from the date of diagnosis and compared to patients without fractures. The mean 1-year incremental cost for fracture versus no fracture was £690 for NVNH fractures (£1152 for vertebral fractures). The 1-year incremental cost per woman was £525 (total cost, £1390) among Table 2 Average annual incidence (per 10,000 population) of fractures at different sites. Fracture site
Men
Women
Both genders
Radius and ulna (lower end) Ankle Humerus (other than upper end) Metacarpals Tarsals and metatarsals Radius and ulna (upper end and shaft) Clavicle Tibia and fibula (shaft) Tibia and fibula (upper end) Pelvis Humerus (upper end)
18.2 8.7 3.6 11.7 7.3 6.1 6.7 5.8 3.4 1.6 2.2
24.3 6.5 3.6 1.9 5.2 5.5 3.2 2.4 1.9 1.8 4.0
21.3 7.6 3.6 6.7 6.2 5.8 4.9 4.1 2.6 1.7 3.1
Cited from Ref. [24].
736
H. Hagino / Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
those with forearm (wrist) fractures and £602 (total cost, £1604) among those with humerus fractures. Pike et al. investigated direct costs of treating patients with osteoporosis who experience nonvertebral fractures during the 2 years following the incident fracture using a private health insurance claims database in the US from 1999 to 2006 [25]. Mean direct excess costs for non-vertebral fracture patients fell from $5267 in the first year to $2072 in the second year after a fracture. They found that patients with fractures of the pelvis, hip, femur and forearm had excess costs of $7,083, $13,334, $4211 and $5009 in the first year, and $5,121, $3,930, $3828 and $1562 in the second year, respectively. NVNH fracture patients made up a larger proportion of the sample and had significantly higher costs than controls with osteoporosis but without a fracture history, although hip fractures had the highest excess costs over both years. NVNH patients accounted for over 6 times the total costs to payers as hip fracture patients over the entire 2-year period ($21.7 billion versus $3.4 billion). Thus, NVNH fracture patients represent an important subpopulation from an economic point of view, even though they are often overlooked. What are the outcomes? In addition to pain and limitations in physical function, fragility fractures decrease mobility, social interaction and emotional well-being. In the Canadian Multicentre Osteoporosis Study, pelvic, lower limb and rib fractures were associated with very low health-related quality of life (HRQOL) scores [32]. Baseline data, as well as prospective data from GLOW, suggest that a history of fracture at various sites is associated with significant reductions in HRQOL [33,34]. Roux et al. reported changes in HRQOL when classifying NVNH fractures as major fractures (pelvis/leg and shoulder/arm) and minor fractures (wrist/hand, ankle/foot, rib/clavicle) [34]. They found that both minor and major incident NVNH fractures have a strong impact on HRQOL in women aged 55 years or more, as measured by EuroQOL five dimensions (EQ-5D) over 1 year, independent of baseline HRQOL. Reductions in EQ-5D and Short Form (36) (SF-36) physical functioning scores in women with major NVNH fractures were of a magnitude similar to reductions in women with incident hip fractures, but the reduction was smaller than for women who experienced incident spine fractures. By contrast, forearm fractures entailed fewer consequences. In patients with Colles’ fractures, Dolan et al. observed considerable reductions in HRQOL in the first 3 months, but recovery was fast and the impairment after recovery was small [35]. In a 1-year prospective study of patients with new distal radius fractures, EQ-5D values at 6 months showed recovery to pre-fracture levels [36]. Forearm fractures cause pain and loss of function, but the prognosis for fracture healing and regaining of function is usually favourable, especially in the elderly who do not have jobs. What is the optimal early management for ensuring the best outcomes in the short term and the long term? Therapeutic options for NVNH fractures vary across fracture sites. The recent development of implants for internal fixation made open reduction internal fixation (ORIF) more common for distal radius fractures and proximal humerus fractures. Decisions on treatment strategy should take into account the patient’s age, activity levels, co-morbidities and injury characteristics. Distal radius fractures The goal of distal radius fracture treatment is to restore acceptable mobility and durability to an upper extremity. There are four treatment options for distal radius fractures: closed reduction and immobilisation, closed reduction and pinning, external fixation and ORIF using plates. Closed reduction, external fixation and ORIF each has advantages and disadvantages. There has been a shift towards the use of volar locking plates for the treatment of displaced distal radius fractures (Fig. 4) [37]. However, there is no definitive consensus on an appropriate treatment for unstable distal radius fractures in elderly patients, because only minor differences exist between functional outcomes and activities of daily living after surgical and non-surgical treatment 1 year after injury [38,39]. The American Academy of Orthopaedic Surgeons (AAOS) clinical practice guidelines state that the Academy is unable to recommend for or against operative treatment in patients with distal radius fractures over
H. Hagino / Best Practice & Research Clinical Rheumatology 27 (2013) 731–741
737
Fig. 4. Radiograph of a distal radius fracture. A. First visit at 2 months after the injury and B. after fixation with bone grafting using a locking plate. A 72-year-old woman was injured when she fell in her home. Two weeks after the injury she visited a clinic due to persistent pain. Manual reduction was attempted and cast immobilisation was performed. Two months after the injury, she visited our hospital due to pain around the ulnar end. Open reduction with osteotomy and autologous bone grafting was performed using a locking plate.
the age of 55 [37]. This recommendation is based on previous prospective, randomised trials in patients with displaced distal radius fractures who were all above the age of 55. It is important to take patient expectations and activity levels into account over age when considering operative versus non-operative treatment. Less aggressive treatment for older patients is advised [37]. When counselling, patients with distal radius fractures, especially low-energy fragility fractures, the health-care provider should discuss both operative and non-operative options. The surgeon should also discuss important health concerns, such as osteoporosis, and make appropriate referrals to either a primary care physician or an osteoporosis specialist for follow-up as needed [40]. Proximal humerus fractures Most patients with proximal humerus fractures regain a functional shoulder without surgery. Surgery should only be considered in approximately 20% of patients, either because they require better shoulder function or because their fracture is more complex. Murray et al. proposed a flowchart depicting the treatment algorithm (Fig. 5) [41]. For elderly patients over 85 years of age, surgical treatment is rarely indicated. Surgery is absolutely indicated in
