|Home | About | Journals | Submit | Contact Us | Français|
A number of studies have found an increased risk of lower extremity injuries in obese patients. Most studies, however, are unable to provide stable population-based estimates based on the degree of obesity and few assess the risk pertaining to more detailed fracture location in the lower extremities.
We therefore investigated the relationship between obesity and lower extremity fractures in different age and fracture locations in a stable population.
This is a population-based, cross-sectional study from the electronic medical records of 913,178 patients aged 2 to 19 years. The body mass index (BMI) for each patient in the cohort was used to stratify patients into five weight classes (underweight, normal weight, overweight, moderate obesity, and extreme obesity) based on BMI for age. Records were assessed for the occurrence of lower extremity fractures for each cohort member. The associations among the five weight classes and specific lower extremity fractures were estimated using multiple logistic regression models and expressed with odds ratios (ORs) and 95% confidence intervals (CIs) using multivariate analysis to adjust for patient demographic variables.
Overweight, moderately obese, and extremely obese patients all had an increased OR of fractures of the foot (OR, 1.14, 1.23, and 1.42, respectively, with 95% CI, 1.04–1.24, 1.12–1.35, and 1.26–1.61, respectively) along with the ankle, knee, and leg (OR, 1.27, 1.28, and 1.51, respectively, with 95% CI, 1.16–1.39, 1.15–1.42, and 1.33–1.72, respectively). The association was strongest in the 6- to 11-year-old age group. We found no association between increasing BMI and increased risk of fractures of the femur and hip.
Increasing BMI is associated with increased odds of foot, ankle, leg, and knee fractures in children.
Level III, prognostic study. See Guidelines for Authors for a complete description of levels of evidence.
Obesity is a growing problem in industrialized nations. Ogden et al. [28, 29] have documented the increasing number of overweight children in the United States. In addition to putting children at risk for long-term medical issues such as diabetes and hypertension, numerous studies [2, 3, 14, 26, 27, 30, 40] have documented the association between obesity and an increased risk of musculoskeletal injuries. Most of these studies, however, are not population-based estimates based on the degree of obesity.
Recently, we assessed a large population-based cohort of children and adolescents and documented the association between increasing body mass index (BMI) and lower extremity injuries of all types, including fractures, sprains, strains, and dislocations . We found the risk of having a lower extremity fracture increased monotonically with increasing weight class. That study did not, however, assess the association between BMI and the specific fracture locations in the lower extremities. Lower extremity fractures account for between 15% and 22% of long bone fractures in large-scale studies of pediatric fractures [5, 21, 22, 41].
The purpose of the present study is to more clearly elucidate the associations between BMI and a specific fracture region in the lower extremities.
As described by Adams et al. , we used a subset of patients enrolled in a large population-based cohort study, the Kaiser Permanente Southern California (KPSC) Children’s Health Study 2007–2009. During the study period, KPSC provided coverage for 1,295,971 children aged 2 to 19 years. After exclusion of 265,241 members who did not have any medical encounters in the study period, 1,030,730 patients were eligible for participation. For this study we included only subjects with at least one healthcare encounter, a minimum of one valid height and weight measurement, and patients who were not pregnant. Of note, Smith et al.  has previously demonstrated the low error rates in height and weight recorded in the electronic medical record. Of these 1,030,730 patients, 920,034 (89.2%) had at least one valid weight and height. After excluding 6856 pregnant patients, we were left with 913,178 patients for the final cohort. Institutional review board approval was obtained for the study.
BMI was calculated as weight in kilograms divided by the square of height in meters. The median BMI for age of all encounters in the year of study enrollment for each patient was used for analysis. Children were categorized as underweight (BMI-for-age < 5th percentile), normal weight (BMI-for-age ≥ 5th and < 85th percentile), overweight (BMI-for-age ≥ 85th and < 95th percentile or BMI ≥ 25 kg/m2 and < 30 kg/m2), moderately obese (BMI-for age ≥ 95th and < 1.2 × 95th percentile or BMI ≥ 30 kg/m2 and < 35 kg/m2), and extremely obese (BMI-for age ≥ 1.2 × 95th percentile or BMI ≥ 35 kg/m2) based on a combination of sex-specific BMI-for-age growth charts developed by the Centers for Disease Control and Prevention and the World Health Organization for overweight and obesity in adults [1, 8, 20].
All electronic health records were assessed for the first occurrence of an International Classification of Disease, 9th Revision (ICD-9) code for lower extremity fractures for each cohort member during the period of study enrollment. Only the first fracture for any particular specific location and side was included in the analysis. For lower extremity fractures, we included all ICD-9 codes for all fractures from the hip to the tips of the toes. Patients diagnosed with multiple concurrent fractures were not included as a result of lack of clarity as to the specific fracture location from the coding alone. For the same reason, patients diagnosed with unspecified fractures were not included in the analysis. Fractures were grouped into three groups based on fracture region. Group A included all fractures of the foot; Group B included all fractures of the ankle, leg, and knee; and Group C included all fractures of the femur from the most proximal aspect of the femur down to the supracondylar area of the femur.
The age at the time of the fracture diagnosis was recorded and patients were grouped into three age groups: 2 to 5, 6 to 11, and 12 to 19 years of age. Patient race and ethnicity information was obtained from the health plan medical records and birth certificates and categorized as non-Hispanic white, Hispanic white, black (regardless of ethnicity), Asian or Pacific Islander, other or multiple race/ethnicity, and unknown as a result of missing information. For unknown race and ethnicity information, administrative records were supplemented using a validated imputation algorithm based on surname lists and address information derived from the US Census Bureau [8, 17, 34]. Wang and Beydoun  and Wang and Zhang  have shown lower socioeconomic status is associated with a higher obesity risk. To minimize socioeconomic biases we included two measures of socioeconomic status: neighborhood education and participation in Medi-Cal, the California state-subsidized program providing healthcare coverage for more than six million low-income children and families as well as elderly, blind, or disabled individuals. Because socioeconomic status is not available in the electronic medical record, neighborhood education and Medi-Cal participation were used as a proxy to adjust for socioeconomic inequalities. Neighborhood education (likelihood of neighborhood education below high school versus high school graduate or higher) was estimated based on the linkage of the health plan members’ addresses through geocoding with US Census block data as per Chen et al.  and Krieger et al. .
The associations among the five different weight classes and fractures were assessed using multiple logistic regression models to estimate odds ratios (ORs) and 95% confidence intervals (CIs). All models were adjusted for sex, race, age, neighborhood education, and Medi-Cal benefit use (yes versus no). Odds ratios for the different sexes were also assessed using multiple logistic regression models. Analyses of lower extremity fractures were also stratified by age and by fracture location (A, B, C), and differences in fracture risk between the different ethnicities were also assessed. All analyses were conducted using SAS 9.2 (SAS Inc, Cary, NC, USA).
The prevalence of underweight, normal weight, overweight, obese, and extremely obese patients in our cohort was 3.0%, 60.4%, 17.4%, 13.0%, and 6.2%, respectively. Thus, a total of 19.2% of the patients in the cohort were classified as obese (BMI ≥ 95%). Patients who were overweight, moderately, or extremely obese were more likely to be older (p < 0.001), male (p < 0.001), and Hispanic or black (p < 0.001) than those of normal weight (Table 1). Medi-Cal enrollees were more likely to be in the moderately obese group (15.1%) or extremely obese group (14.5%).
Lower extremity fractures were identified in 7158 patients with 3720 foot fractures (Group A); 3145 fractures of the ankle, leg, and knee (Group B); and 416 fractures from the hip to the distal femur (Group C). Fractures occurred in 675 patients (9% of all fractures) in the youngest age group, 1945 patients (27% of all fracture) in the 6- to 11-year-old group, and 4538 patients (63% of all fractures) in the 12- to 19-year-old group. Sixty-two percent of all fractures occurred in males, and the risk for having a fracture was higher in males as compared with females in all age groups with the OR for any fracture occurring in a male increasing from 1.30 and 1.10 in the youngest and middle age groups (95% CI, 1.11–1.52 and 0.97–1.16, respectively) to 2.00 (95% CI, 1.88–2.12) in the oldest age group. Participation in Medi-Cal and neighborhood education did not have an effect on fracture risk, whereas patients of all races had a lower fracture risk when compared with non-Hispanic whites. Asian/Pacific Islanders had the lowest OR of a fracture when compared with non-Hispanic whites (OR, 0.57; 95% CI, 0.51–0.64) (Table 2).
For the cohort as a whole, the risk of having a lower extremity fracture of any kind increased in a monotonic fashion with increasing weight even after adjustment for all patient demographics (Table 3). Extremely obese patients were 1.45 times as likely to have had a lower extremity fracture as normal-weight patients (OR, 1.45; 95% CI, 1.33–1.58), and moderately obese and overweight patients had a 23% and 17% increased fracture risk (95% CI, 1.15–1.32 and 1.10–1.25, respectively). Underweight patients, on the other hand, had a 23% decreased risk of having a lower extremity fracture (OR, 0.77; 95% CI, 0.66–0.92). This increasing lower extremity fracture risk with increasing weight class was consistent in both males and females.
We repeated the multivariate logistic regression analysis stratified by age group (Table 3) and then further stratifying by both age group and fracture region (Table 4). This demonstrated that fracture risk in the extremely obese patients was increased compared with their normal-weight counterparts in all three age groups. This increase was the highest in 6 to 11 year olds (OR, 2.23; 95% CI, 1.92–2.59) and the lowest in the 12 to 19 year olds at 1.11 times that of normal (OR, 1.11; 95% CI, 1.03–1.24) (the effect of obesity for the injury risk is modified [p = 0.008] by age). This fracture risk increased monotonically with increasing weight in only the 6 to 11 year olds. Assessment of fracture risk by region for all ages combined indicated extremely obese patients had a 42% increased fracture risk, 51% increased risk, and 12% increased risk for Groups A, B, and C fractures, respectively (Table 4). Although the underweight patients had a decreased risk for Group A and B fractures, the risk of a Group C fracture in underweight patients was 66% higher than in normal-weight patients (OR, 1.66; 95% CI, 1.06–2.59). For Group C fractures as a whole, fracture risk bore little correlation to increasing weight class above normal-weight patients. Similarly to the cohort as a whole, the vast majority of Type C fractures (52%) occurred in the adolescent group. In the assessment of fracture risk stratified by both age and fracture location, the 6- to 11-year-old extremely obese patients had a 2.21 to 2.33 increased fracture risk for all lower extremity fractures, whereas the 2- to 5-year-old extremely obese patients had greater than double the fracture risk for all but Type C fractures.
Ogden et al. [28, 29] and Strauss and Pollack  demonstrated an increasing incidence of childhood obesity in our country with Ogden et al. noting that 14% to 17% of all US children between 2 and 19 years of age were obese from 1999 to 2004  and 16.9% from 2009 to 2010 . Our cross-section of patients in southern California had an even higher prevalence of obesity of 19.2%. Bazelmans et al.  showed in a cross-sectional study that childhood obesity increases the occurrence of injuries in general, and multiple authors have noted the increased risk of lower extremity injuries in obese patients [2, 3, 14, 26, 27, 30–32, 39, 41]. In regard to fracture risk, Chan and Chen  stated obese patients have an increased fracture risk and both Haricharan et al.  and Pollack et al.  demonstrated an increased lower extremity fracture risk in obese patients in motor vehicle collisions. More recently, we demonstrated the substantially increased risk of all lower extremity sprains/strains, fractures, dislocations, and pain with increasing BMI in a large population-based cohort of children and adolescents . The present study further clarifies from this same cohort the increased lower extremity fracture risk by fracture location and age with increasing body weight.
There are several limitations of the present study. First, we assessed patients only for the first fracture during the study period and we therefore did not take into account refractures. It is unclear how this may have affected the results but it did decrease the number of overall fractures for the group as a whole. Second, the exclusion of multiple fractures occurring simultaneously clearly excluded the highest energy injuries such as high-speed motor vehicle accidents involving multilevel trauma. It is again unclear how, if at all, this affected the study results, but it did decrease the number of overall fractures for the group as a whole. Third, we did not assess activity level, mode of injury, differences in treatment settings, fracture severity, or differences in course of care; all of these are factors important in the frequency, location, and outcome of fractures of the lower extremities [3, 27]. This did not affect the study numbers but inclusion of these variables would have provided more information pertaining to fracture risk in this cohort and potential confounding effects on the association between weight class and lower extremity fracture. These variables could have been particularly helpful in understanding the different results seen in the Group C fractures and those of the oldest age group.
Our findings suggest obese patients have an increased risk of lower extremity fractures with the risk increasing monotonically with increasing weight while adjusting for sex, race, age, neighborhood education, and Medi-Cal benefit use. These findings were consistent among both sexes. Stratification by different regions of the lower extremity revealed a similarly monotonically increased fracture risk with increasing weight in fractures of the foot, ankle, leg, and knee region with a 42% increased risk of foot fractures; 51% increased risk of ankle, leg, and knee fractures in extremely obese patients; and a 31% and 23% decreased fracture risk in these same locations for underweight patients. Overall, overweight, moderately obese, and extremely obese patients all had an increased risk of fractures of the foot, ankle, leg, and knee as compared with normal-weight children. In terms of lower extremity fracture risk based on both age and fracture location, severely obese patients in the 6- to 11-year-old group all had a dramatically increased risk of Type A, B, and C fractures of 2.21, 2.20, and 2.33 times that of normal patients, respectively, whereas the youngest severely obese patients also had more than double the risk of Type A and B lower extremity fractures. The markedly increased fracture risk seen in the obese 6- to 11-year-old group was consistent with the study of Haricharan et al. , in which the risk of lower extremity injury in a motor vehicle accident was most markedly increased in the obese 10 to 13 year olds and 6 to 9 year olds.
The only slightly increased fractured risk seen in the obese patients of the oldest group was an interesting finding. There are two possible explanations for this. The literature has consistently demonstrated that in older children and adolescents, the number of fractures resulting from sports-related injuries increases dramatically, especially beyond the age of 10 years [3, 6, 21, 22]. Landin , in fact, demonstrated that between the ages of 10 and 15 years, the incidence of sports-related injuries doubles in girls and more than triples in boys. At the same time, multiple articles have demonstrated that obese children participate in sports much less and have substantially more difficulty with physical activities than their nonobese counterparts [3, 13, 33, 36]. Thus, the effective exposure of obese adolescents to one of the most common causes of injuries—sports, and especially competitive sports—is much lower, thus possibly lowering their risk of fracture. In addition, it may be that older obese children have a bone density that either is the same or higher than that of normal or underweight children. Several studies have shown that obese adults have increased bone density [16, 23]. Given the somewhat conflicting results in previous studies showing that obese children either have increased bone density [6, 7, 15, 24, 25] or decreased bone density/mass or bone mineral content relative to body size [11, 12], it may be that some older obese children could have an increased bone density, similar to the pattern seen in obese adults.
The inconsistent association between obesity and fractures of the proximal femur and femoral shaft was the other interesting finding in this study that was different than most of the rest of the data indicating a relationship between increasing obesity and increasing fracture risk. In particular, the 1.66 times increased risk of Type C fractures in underweight patients along with the decreased Type C fracture risk in the overweight teenagers was at first surprising. However, this may also explained by multiple factors. First, like with the rest of the cohort, more than half of these fractures occurred in the adolescent group. It is well known that in this group, as opposed to younger patients, femur fractures in adolescents tend to be higher energy injuries, many of which are likely related to competitive sports [21, 22]; as mentioned previously, obese children participate substantially less in such activities. In addition, it is distinctly possible that the underweight patients have decreased bone density while at the same time being much more active in sports than the obese children.
These study results enhance prior literature on the association between obesity and fracture risk in a number of ways. By virtue of having access to an extremely large, self-contained electronic database that is truly a cross-section of our southern California population, we were able to provide stable population-based estimates of the associations between childhood obesity and lower extremity fractures that have not previously been demonstrated in such a way. In addition, rather than purely comparing obese (≥ 95th percentile) with nonobese patients or assess the association between obesity and all musculoskeletal complaints, like with all previously referenced studies [2, 3, 14, 26, 27, 30–32, 39, 41], we measured more specific weight classes and assessed only lower extremity fractures to understand how varying degrees of increasing (or decreasing) weight affect fracture risk in the lower extremities. Also, we further assessed fracture risk by both age and lower extremity region as opposed to that of all prior studies on the association of childhood or adult obesity and fracture/injury risk (Table (Table5).5). In doing so, we were able to unearth major differences between fracture risk in the proximal femur/femoral shaft and that of the rest of the lower extremity and to demonstrate the minimally increased fracture risk in obese middle schoolers and high schoolers as opposed to younger patients. Finally, all models were adjusted for sex, race, age, and multiple indicators of socioeconomic status to assess purely for the association between weight and fracture risk and eliminate the influence of these variables. Krieger et al.  has shown that failure to address socioeconomic status may bias almost all associations in the medical literature given the higher predilection toward almost all diseases in those of lower socioeconomic status, and others  have shown the complex relationship that exists between socioeconomic/racial factors and obesity.
In conclusion, the present study shows there is a stepwise increased risk of fractures in the foot, ankle, leg, and knee with increasing BMI in children of all ages and a decreased fracture risk in the same region with decreased body weight. The association between weight class and risk of Type C fractures is less consistent. The association between increasing BMI and increasing risk of fractures of the foot, ankle, leg, and knee is most pronounced in the 6- to 11-year-old age group and least important in the adolescent 12- to 19-year-old patients. Like many other large-scale fracture studies, the increased fracture risk with increasing age in childhood along with the increased fracture risk in boys as compared with girls was clearly illuminated, especially in the oldest age group [2, 5, 9, 21, 22]. In addition, this is one of the first studies to provide an in-depth assessment of the association between childhood weight and lower extremity fracture risk by region and age group in an extremely large, stable population. The implications of the study are manifold. First, this study highlights once again the increased musculoskeletal risks of obesity not just in middle age and beyond, but beginning in childhood, and demonstrates that even being overweight without being obese increases the risk of childhood lower extremity fractures. Thus, recognition of this fact emphasizes the need for uniform familial education on the deleterious musculoskeletal effects of obesity from early on to try to minimize the societal, financial, and medical burden imposed by this problem. Second, this study illustrates the efficacy of electronic medical records in assessing and studying important issues such as the association between childhood obesity and fractures and specifically demonstrates the need for developing large-scale fracture registries in electronic medical records to more clearly understand specific associations such as modes of injury and frequency of participation in sports and other activities among obese and normal-weight children (among other things). Lastly, this study highlights the need for a better understanding of bone density in children of all ages and sizes to understand how weight affects bone density in early, middle, and late childhood.
We thank Hooman Nikizad for preparation of references and Gez Bowman for help with table preparation.
Each author certifies that he or she, or a member of his or her immediate family, has no funding or commercial associations (eg, consultancies, stock ownership, equity interest, patent/licensing arrangements, etc) that might pose a conflict of interest in connection with the submitted article.
All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research editors and board members are on file with the publication and can be viewed on request.
Each author certifies that his or her institution approved the human protocol for this investigation, that all investigations were conducted in conformity with ethical principles of research, and that informed consent for participation in the study was obtained.