Approximately 66% of the US population is overweight (body mass index [BMI], 25 to 30) or obese (BMI ≥30.0).1,2 The annual costs of obesity-related medical expenses and loss of worker productivity range from $147 billion to $200 billion.3 Workers who are obese are more likely than those who are not obese to have work limitations, sustain unintentional injuries, and file workers’ compensation claims.4 From the trauma caregiver’s standpoint, patients with a BMI ≥30.0 and blunt trauma have an increased risk for renal, pulmonary, and cardiovascular complications5 as well as longer hospital stays.6 In addition, the rate of mortality7-12 and the likelihood of multiple organ failure11 is increased in patients who are obese and have injuries associated with motor vehicle collisions (MVCs) or other blunt trauma.
BMI and body morphology are associated with unique injury predispositions. Persons who are obese have a greater risk of knee dislocation13,14 secondary to low velocity falls and sustain more severe ankle fractures than do persons who are not obese.15 In the setting of MVCs, patients who are obese also have a higher risk of injuries to the chest, particularly injuries to the heart, ribs, and lungs.16 This predisposition persists even when controlling for sex, seat belt use, and air-bag deployment.16 Other studies of MVCs have reported more injuries to the upper and lower extremities17,18 and more severe fractures of the distal femur19 and pelvis20 in persons who are obese. Obesity may have a protective effect against certain injuries caused by MVC, with a “cushion effect” resulting in a lower Maximum Abbreviated Injury Score (mAIS) for pelvic and abdominal injuries.21,22
Limited information exists on the influence of body morphology (ie, BMI, height, weight) on spinal injuries sustained in a MVC. In one study, obesity significantly (P <0.01) correlated with a higher frequency of cervical spine injuries with an Abbreviated Injury Score (AIS) 2+ and death.23,24 Data from the National Automotive Sampling System Crashworthiness Data System show a positive correlation between AIS 3+ spine injury and BMI.18 Rao et al25 reported associations between patient-, vehicle-, and crash-related factors and thoracolumbar (T/L) fractures sustained in MVCs. The authors found an association between T/L extension injuries and a BMI >36.
Our study analyzes the associations between spinal fracture patterns and body morphology. Clinical and imaging data from the Crash Injury Research and Engineering Network (CIREN) database were analyzed, and fracture subtypes were categorized according to a modified Denis classification system.26 Data were then stratified across BMI groups and analyzed in relation to several patient characteristics.
The CIREN database is maintained by the National Highway Traffic Safety Administration and contains data on MVCs. Six trauma centers in the US collect information pertaining to vehicular crashes and de-identified clinical information on injuries from hospital medical records of patients in involved vehicles. Data on persons of all ages with injuries to the thoracic and lumbar spine that occurred between 1996 and 2011 were extracted from the database. Data on motorcycle, bicycle, and pedestrian accidents as well as crashes involving vehicles manufactured >5 years before the year of the accident were excluded. Only data from persons who sustained systemic injuries with an AIS ≥3 (regardless of region of injury) or an AIS of 2 in at least two different regions of the body were included.23 Data and engineering specialists logged impact and engineering information pertaining to the collision after inspection of the vehicle(s), road, and collision scene.27
From 1996 to 2011, data from 4,572 persons involved in MVCs were registered in the CIREN database. Of these, 631 persons had 2,626 thoracic and lumbar spine injuries. CIREN files for all 631 persons were reviewed for demographics, injuries, and crash data. Patient demographics included age, sex, body weight, height, and BMI. The data were stratified into underweight (BMI, ≤18.49), normal weight (BMI, 18.5 to 24.9), overweight (BMI, 24.9 to 29.9), and obese (BMI, ≥30.0) groups.28 Information on thoracic and lumbar injuries included the type and level of spinal column injury and neurologic injury. Associated systemic injuries recorded in the database include those to the head, face, cervical spine, thorax, abdomen, pelvis, spinal cord, and lower extremities. Information on systemic injuries included the AIS in each body region23 and the injury severity score.29 Only injuries with an AIS ≥2 (ie, moderate to severe injuries) were used to analyze whether there was an association between those injuries and other system injuries. The injury severity score, maximum AIS, and fatality (if applicable) were recorded for each patient.
In addition to the seating design and orientation and the seat-belt design (lap or lap and shoulder), the make, model, and year of vehicle manufacture are recorded in the database. By reconciling information from police and emergency medical services as well as other forensic methods, researchers determined how the seat belts were worn. Bruises and skin contusions from the seat belt were noted by the clinical team, and distortion of belt webbing and examination of the seat belt retractors and pretensioners were noted by the crash investigation team. Seat belt use was determined by the investigator in the field by evaluating restraint components, such as the D-ring, belt webbing, belt attachment at both ends, and if available, information from the electronic data recorder. This data was cross-referenced with clinical data, such as the seat belt sign observed on CT and/or photographs taken by the medical team, to ensure consistency of the pattern of injury with restraint use. The deployment and location of air bags in the vehicle and collision characteristics, including the direction of impact and the change in velocity at impact, were recorded, as well.
Radiographs, CT scans, and MRIs (when available) of the thoracic and lumbar spine in each patient were reviewed independently by two spine surgeons. Thoracic and lumbar fractures were classified according to a modified Denis classification system.26 A group of patients displayed a spinal injury pattern that did not conform to any of those originally described in the Denis classification. These injuries were characterized by distractive failure of the anterior column through the disk or vertebral body with or without additional distraction of the middle column or translation at the fracture site.25 Distractive extension injuries were added as a distinct “major” injury group. Of the 631 occupants with thoracic and lumbar injuries, 299 were found to have major T/L fractures. The vertebral level of each injury was noted, and major injuries were categorized by level into three regions: thoracic (T1 to T10-T11), T/L junction (T11 to L2), and lumbar (L2-L3 to L5-S1).
All subjects in the CIREN database, including those with major and minor injuries to the T/L spine, were stratified by their BMI (ie, underweight, normal weight, overweight, obese). Mean BMI, height, and weight were recorded for each region of T/L injury, each major fracture type, minor fractures, and each associated systemic injury. Mean BMI was compared for spinal injuries associated with fatality and survival. Associations between major and minor T/L injuries and individual BMI, height, weight, and BMI subgroups were studied.
Of 4,572 patients in the CIREN database, 631 (13.8%) sustained T/L spine injuries. Of these, 299 patients sustained major injuries at 330 vertebral levels and the remaining 332 patients had minor injuries to the facets and transverse, pars, and/or spinous processes. The mean age of all 631 patients was 45.8 years at the time of injury (range, 21 months to 96 years), and the mean age of those with major injuries was 48.0 years. The mean age of patients with minor injuries was 44.0 years. Major injuries to the T/L spine occurred with similar frequency in both sexes (148 female, 151 male), whereas minor injuries occurred in a greater number of women than men (189 female, 143 male). Among the 299 patients with major T/L fractures, 238 had adequate data to permit fracture classification. Of those patients, adequate height and weight data for BMI analysis was available for 234 who sustained 326 classifiable major T/L fractures. Height and weight data used for BMI calculation were present in 329 of 332 patients with 702 minor injuries.
Data on the body morphology of 4,401 patients in the CIREN database indicated that the average BMI, weight, and height were 26.62, 74 kg, and 165 cm, respectively. In patients with major T/L injuries, body morphology was generally similar (average BMI, 26.04; average weight, 74 kg; and average height 168 cm; Table 1). In contrast, patients with minor T/L injuries were generally larger, with an average BMI, weight, and height of 28.3, 81 kg, and 170 cm, respectively. Compared with the overall major T/L cohort, men and women displayed similar trends in fracture types and BMI (Tables 2 and 3).
Body Morphology and Location of Spinal Fracture
Higher BMI values were associated with spinal injury at more cranial levels. The mean BMI associated with spine injuries was highest in patients with thoracic injuries followed by patients with injuries to the T/L junction and those with injuries to the lumbar spine (Figure 1). The mean body weight was also highest among patients with thoracic injury (81 kg; range, 10 to 158 kg) compared with those who had injury to the lumbar spine (63 kg; range, 19 to 108 kg) and T/L junction (73 kg; range, 26 to 139 kg). The mean height of patients with injuries to the thoracic spine, T/L junction, or lumbar spine was 171 cm, 168 cm, and 164 cm, respectively.
Patients who were obese (BMI ≥30.0) frequently had injuries at the thoracic level (23 of 54 patients, 42.59%), whereas those who were underweight (BMI ≤18.49) frequently sustained injury to the lumbar region (15 of 23 patients, 65.21%). In patients with T/L fractures, there was a trend toward a higher level of injury as the BMI category increased from underweight to obese (Figure 2).
Body Morphology Associated with Fracture Type
Compared with other fracture types, patients with extension injuries had the highest BMI and body weight (Figures 3 and 4). Fracture-dislocation, compression fractures, and burst fractures were associated with a lower BMI. The lowest average BMI was associated with flexion-distraction injuries (Table 1). Compared with the mean weight of the entire cohort (74 kg), patients with flexion-distraction injuries had a lower mean weight of 48 kg. Height was similar in patients with different fracture patterns, except that 28 patients with flexion-distraction injury had an average height of 147 cm, much lower than the mean of 168 cm for the entire cohort.
Fatalities and Systemic Injury
BMI and fatality were positively correlated for all cohorts. In the entire T/L fracture cohort, the mean BMI was higher in patients who had a fatal outcome (67 patients) than in those who survived (557 patients; Table 1). This trend in mean BMI was consistent when minor and major T/L injuries were analyzed separately and when the mean BMI of fatalities and survivors was compared within each major fracture subtype. Patients who sustained a T/L fracture and another system injury with an AIS ≥2 had a greater tendency to be overweight and obese compared with the overall CIREN cohort (Table 4). The difference was most pronounced in subjects with injuries to the face and spinal cord followed by those with head injuries. Patients with T/L fracture and cervical spine injuries were the only ones to deviate from the trend; this group displayed lower overweight frequencies compared with groups from CIREN that had T/L fractures and other system injuries. A relationship between spinal cord injury and underweight status was also found. Compared with all CIREN subjects, those who sustained a T/L fracture with an associated spinal cord injury were twice as likely to be underweight (Table 4).
The role obesity plays in contributing to mortality in patients who sustain trauma has been reported. In a meta-analysis of 18 studies involving blunt and penetrating trauma, Liu et al30 found that increasing BMI was a positive predictor of mortality. In the setting of MVCs, Mock et al10 found that patients with a BMI of 30 to 34 or 35 to 49 were 1.68 and 3.18 times more likely, respectively, to have a fatal outcome than those with a BMI of <20. Each unit increase in BMI was associated with a 1.04 increase in the odds ratio associated with mortality. Similarly, in a study of 22,107 patients involved in a MVC, Zhu et al9 found an increased risk of mortality among drivers with a BMI ≥35.0 after controlling for collision type, seat belt use, and drug or alcohol use. The results of our study, which includes data from a large, multicenter prospectively maintained database, confirm an increased risk of fatality associated with increasing BMI in patients with major or minor T/L fractures sustained in MVCs.
The correlation between BMI and specific patterns of T/L injury has not been addressed extensively in the literature. Our study includes 21 subjects with an extension T/L fracture mechanism—the largest such cohort, to our knowledge (Figures 3 and 4). Extension injury of the T/L spine occurred almost exclusively in the thoracic region (20 of 21 patients; 95%), and patients with this injury pattern had a much higher mean BMI (36.0) than did those with all other fracture patterns. Thirteen of 21 patients (62%) with extension injuries were obese (BMI ≥30.0); this is much greater than the 57 of 295 subjects (19.3%) with all other T/L fracture patterns who were obese (Figure 5). In a study of 965 patients with injuries to the T/L spine, Matejka31 reported that all four patients with extension pattern injuries were obese (mean BMI, 32.75; range, 30.9 to 36.4) and hypothesized that higher BMI resulted in increased hyperextension forces.
Biomechanical studies show how obese persons may be predisposed to extension mechanisms in MVCs. In a study of crash injury mechanisms, Kent et al32 compared obese and nonobese cadavers and found greater forward translation of the hips in relation to the shoulders in belted, obese, postmortem human surrogates (PMHSs) in frontal collisions. The authors postulated that the lap belt loads the soft tissue of the abdomen first, delaying the loading onset of the bony prominences of the pelvis. Another similar study of frontal collisions used PMHSs and found that “forward pelvis motion was great enough to cause a person who was obese to fall off of the front edge of the seat.”33 Furthermore, Reed et al34 showed that persons who are obese wear the lap belt higher on the abdomen, rendering them more susceptible to extension injuries than persons with normal body morphology.
An increased incidence of chest cage injury has also been reported in persons with higher BMI. Cormier16 found that patients who were overweight or obese had an increased risk of severe thorax injuries caused by frontal MVCs. Boulanger et al35 reported a 30% increased risk of rib fractures in patients35 who are obese and sustain blunt trauma. The intact chest cage contributes to the three-dimensional stability of the thoracic spinal column,36,37 and the reduced ability of the injured chest cage to dissipate impact energy may be a factor in the increased incidence of thoracic spine fractures we found in persons with a high BMI. Our findings show that subjects who were obese were disproportionately more likely to sustain injuries to the thoracic spine (23 of 54 patients, 42.6%) and the T/L junction (21 of 54 patients, 38.9%) than injuries to the lumbar spine (10 of 54 patients, 18.5%). Conversely, patients who were underweight suffered a disproportionately greater number of injuries to the lumbar spine (15 of 23 patients, 65.2%) compared with injuries to the thoracic spine (3 of 23 patients, 13.0%) and T/L junction (7 of 23 patients, 30.4%). Although the data in this study does not permit a definitive conclusion, patients who are obese are predisposed to chest cage injury and extension mechanisms; these predispositions likely contribute to the higher incidence of thoracic spine injury.
Flexion-distraction injuries, referred to as seat belt injuries, often result from an inappropriately worn three-point seat belt or the use of a two-point restraint. In the absence of appropriate shoulder restraint, the lap belt acts as a fulcrum around which the spine abruptly flexes, leading to posterior column distraction injuries. In this study, 11 of 21 patients with flexion-distraction injuries appropriately used a three-point seat belt at the time of collision. However, these patients were generally smaller in terms of height (147 cm) and weight (48 kg), with an average BMI of 21.71 compared with the height, weight, and average BMI of the overall T/L cohort (168 cm, 74 kg, 27.2, respectively). Seat belts are optimized for median male anthropometry (total body mass, 75 kg; BMI, 28), and deviations from median body mass and/or size likely result in varied restraint and susceptibility to injury, even for the appropriately belted person. Reed et al34 analyzed three-point seat belt fit and reported greater belt webbing length in persons with a higher BMI. In a study of the kinematics of three-point belted PMHSs in impact sled tests, Forman et al33 noted that, unlike normal subjects, subjects who were obese did not have any torso rotation around the shoulder portion of the restraint. In our study, 23 of 28 (82%) subjects with a flexion-distraction injury were underweight or normal weight, lending support to the argument that the snug fit of the three-point restraint resulted in reduced torso rotation and greater limitation in spinal flexion in obese persons, providing a protective effect against flexion-distraction injuries (Figure 5). Our data did not show a correlation between BMI and an axial loading mechanism in compression or burst-type injuries.
Sex has been shown to influence injury severity following MVC. In restrained subjects, women had a 47% increase in the risk of mAIS 3+ injury and a 71% increase in the risk of mAIS 2+ injury compared with men.38 In another study, men had fewer AIS 3+ injuries to the upper and lower extremities, head, and thorax than did women.39 The combined impact of sex and BMI on injury caused by MVC is less clear. Studies have reported an increased risk of serious injury8 and mortality40 in female drivers who are obese compared with these risks in male drivers who are obese. Conversely, Zhu et al41 reported an increased risk of mortality associated with increasing BMI in male drivers, but not in female drivers. Rupp et al18 reported that both sexes had a higher risk of AIS 3+ spine injuries with increasing BMI, although men had a higher risk than did women. Our data showed no clear influence of sex and BMI on the incidence of major T/L injuries, with the exception of 9 of 14 men (64.3%) with extension injuries who were obese compared with 4 of 7 women (57.1%). This small difference could be due to sex variations in body fat deposition and body morphology.42,43 Men tend to deposit fat around the belly, leading to larger waist-hip ratios and resulting in delayed pelvic loading by the lap belt.32 Furthermore, increased stature in men results in the lap belt resting at a greater distance from the pelvis,34 further increasing lap-belt loading time.
Global automotive safety standards exist for body regions in front-impact (eg, the head, neck, chest, femur); side-impact (eg, head, chest, pelvis); and rear-impact (eg, head-neck) collisions. However, similar injury assessment criteria based on mechanisms are not available for the T/L spine. The results of our study underscore a need to determine the biomechanical tolerance criteria of the T/L spine based on factors such as injury patterns and severity to derive and specify mitigation criteria. Future studies should quantitatively examine the effect that the rib cage anatomy, chest compliance, adipose tissue thickness surrounding the anterior and lateral chest walls, and restraint characteristics have on the observed increased incidence of thoracic spine fractures associated with increasing BMI. Computer models and PMHS crash tests should to be used to reproduce T/L injuries in controlled laboratory settings to derive injury criteria to protect passengers with varying body morphology.
The limitations of this study relate to the CIREN database, which gathers data only on subjects who have been injured in MVCs, making it impossible to compare an injured group with an uninjured group. Additionally, although the availability of BMI data makes it a commonly used metric for body morphology,44-47 more specific measurements of waist and chest circumference along with skin-fold tests would lead to increased accuracy with regard to patient shape and tissue composition, permitting better correlation of body morphology with injury. Although the CIREN database provides information on body morphology and its associations with T/L trauma sustained in MVCs, outcomes data were not investigated. Future steps would include analysis of the short- and long-term outcomes of these injuries. Notwithstanding these limitations, the large sample size from a multicenter database and the availability of detailed subject and crash data provides valuable field data to characterize the relationship between body morphology and T/L injury patterns resulting from MVC. Subjects with minor T/L injury and those who had T/L injury and an injury to another organ system tended to have a higher BMI. Among the various T/L fracture patterns, a higher BMI was associated with extension injury, whereas a lower BMI was associated with flexion-distraction injury. In patients who sustained a T/L injury, higher BMI correlated with fatality. These findings establish a rationale for specifically including T/L spine parameters in crash safety standards and serve as a fundamental dataset for advancing MVC outcomes and research in automotive safety mechanisms.
References printed in bold type are those published within the past 5 years.
1. Ogden CL, Carroll MD, Kit BK, Flegal KM: Prevalence of childhood and adult obesity
in the United States, 2011-2012. JAMA 2014;311(8):806–814.
2. Fakhouri TH, Ogden CL, Carroll MD, Kit BK, Flegal KM: Prevalence of obesity
among older adults in the United States, 2007-2010. NCHS Data Brief 2012;106:1–8.
3. Russell GV, Pierce CW, Nunley L: Financial implications of obesity
. Orthop Clin North Am 2011;42(1):123–127, vii.
4. Sabharwal S, Root MZ: Impact of obesity
on orthopaedics. J Bone Joint Surg Am 2012;94(11):1045–1052.
5. Byrnes MC, McDaniel MD, Moore MB, Helmer SD, Smith RS: The effect of obesity
on outcomes among injured patients. J Trauma 2005;58(2):232–237.
6. Duane TM, Dechert T, Aboutanos MB, Malhotra AK, Ivatury RR: Obesity
and outcomes after blunt trauma. J Trauma 2006;61(5):1218–1221.
7. Brown CV, Neville AL, Rhee P, Salim A, Velmahos GC, Demetriades D: The impact of obesity
on the outcomes of 1,153 critically injured blunt trauma patients. J Trauma 2005;59(5):1048–1051.
8. Viano DC, Parenteau CS, Edwards ML: Crash injury risks for obese occupants using a matched-pair analysis. Traffic Inj Prev 2008;9(1):59–64.
9. Zhu S, Layde PM, Guse CE, et al.: Obesity
and risk for death due to motor vehicle crashes. Am J Public Health 2006;96(4):734–739.
10. Mock CN, Grossman DC, Kaufman RP, Mack CD, Rivara FP: The relationship between body weight and risk of death and serious injury in motor vehicle crashes. Accid Anal Prev 2002;34(2):221–228.
11. Neville AL, Brown CV, Weng J, Demetriades D, Velmahos GC: Obesity
is an independent risk factor of mortality in severely injured blunt trauma patients. Arch Surg 2004;139(9):983–987.
12. Donnelly JP, Griffin RL, Sathiakumar N, McGwin G Jr: Obesity
and vehicle type as risk factors for injury caused by motor vehicle collision. J Trauma Acute Care Surg 2014;76(4):1116–1121.
13. Edwards GA, Sarasin SM, Davies AP: Dislocation of the knee: An epidemic in waiting? J Emerg Med 2013;44(1):68–71.
14. Shetty RR, Mostofi SB, Housden PL: Knee dislocation of a morbidly obese patient: A case report. J Orthop Surg (Hong Kong) 2005;13(1):76–78.
15. Spaine LA, Bollen SR: ‘The bigger they come ...’: The relationship between body mass index and severity of ankle fractures. Injury 1996;27(10):687–689.
16. Cormier JM: The influence of body mass index on thoracic injuries in frontal impacts. Accid Anal Prev 2008;40(2):610–615.
17. Chong M, Sochor M, Ipaktchi K, Brede C, Poster C, Wang S: The interaction of ‘occupant factors’ on the lower extremity fractures in frontal collision of motor vehicle crashes based on a level I trauma center. J Trauma 2007;62(3):720–729.
18. Rupp JD, Flannagan CA, Leslie AJ, Hoff CN, Reed MP, Cunningham RM: Effects of BMI on the risk and frequency of AIS 3+ injuries in motor-vehicle crashes. Obesity
(Silver Spring) 2013;21(1):E88–E97.
19. Maheshwari R, Mack CD, Kaufman RP, et al.: Severity of injury and outcomes among obese trauma patients with fractures of the femur and tibia: A crash injury research and engineering network study. J Orthop Trauma 2009;23(9):634–639.
20. Bansal V, Conroy C, Lee J, Schwartz A, Tominaga G, Coimbra R: Is bigger better? The effect of obesity
on pelvic fractures after side impact motor vehicle crashes. J Trauma 2009;67(4):709–714.
21. Zarzaur BL, Marshall SW: Motor vehicle crashes obesity
and seat belt use: A deadly combination? J Trauma 2008;64(2):412–419.
22. Arbabi S, Wahl WL, Hemmila MR, Kohoyda-Inglis C, Taheri PA, Wang SC: The cushion effect. J Trauma 2003;54(6):1090–1093.
23. Gennarelli TA, Wodzin E: The Abbreviated Injury Scale 2005: Update 2008. Des Plaines, IL, Association for the Advancement of Automotive Medicine, 2008.
24. Funk JR, Cormier JM, Manoogian SJ: Comparison of risk factors for cervical spine, head, serious, and fatal injury in rollover crashes. Accid Anal Prev 2012;45:67–74.
25. Rao RD, Berry CA, Yoganandan N, Agarwal A: Occupant and crash characteristics in thoracic and lumbar spine injuries resulting from motor vehicle collisions. Spine J 2014;14(10):2355–2365.
26. Denis F: The three column spine and its significance in the classification of acute thoracolumbar spinal injuries. Spine (Phila Pa 1976) 1983;8(8):817–831.
27. National Highway Traffic Safety Administration: Crash injury research. http://www.nhtsa.gov/CIREN
. Accessed September 25, 2015.
28. Centers for Disease Control and Prevention: Assessing your weight. http://www.cdc.gov/healthyweight/assessing/index.html
. Accessed September 28, 2015.
29. Baker SP, O’Neill B, Haddon W Jr, Long WB: The injury severity score: A method for describing patients with multiple injuries and evaluating emergency care. J Trauma 1974;14(3):187–196.
30. Liu T, Chen JJ, Bai XJ, Zheng GS, Gao W: The effect of obesity
on outcomes in trauma patients: A meta-analysis. Injury 2013;44(9):1145–1152.
31. Matejka J: Hyperextension injuries of the thoracolumbar spine [German]. Zentralbl Chir 2006;131(1):75–79.
32. Kent RW, Forman JL, Bostrom O: Is there really a “cushion effect”?: A biomechanical investigation of crash injury mechanisms in the obese. Obesity
(Silver Spring) 2010;18(4):749–753.
33. Forman J, Lopez-Valdes FJ, Lessley D, Kindig M, Kent R, Bostrom O: The effect of obesity
on the restraint of automobile occupants. Ann Adv Automot Med 2009;53:25–40.
34. Reed MP, Ebert-Hamilton SM, Rupp JD: Effects of obesity
on seat belt fit. Traffic Inj Prev 2012;13(4):364–372.
35. Boulanger BR, Milzman D, Mitchell K, Rodriguez A: Body habitus as a predictor of injury pattern after blunt trauma. J Trauma 1992;33(2):228–232.
36. Watkins R IV, Watkins R III, Williams L, et al.: Stability provided by the sternum and rib cage in the thoracic spine. Spine (Phila Pa 1976) 2005;30(11):1283–1286.
37. Brasiliense LB, Lazaro BC, Reyes PM, Dogan S, Theodore N, Crawford NR: Biomechanical contribution of the rib cage to thoracic stability. Spine (Phila Pa 1976) 2011;36(26):E1686–E1693.
38. Bose D, Segui-Gomez M, Crandall JR: Vulnerability of female drivers involved in motor vehicle crashes: An analysis of US population at risk. Am J Public Health 2011;101(12):2368–2373.
39. Carter PM, Flannagan CA, Reed MP, Cunningham RM, Rupp JD: Comparing the effects of age, BMI and gender on severe injury (AIS 3+) in motor-vehicle crashes. Accid Anal Prev 2014;72:146–160.
40. Sivak M, Schoettle B, Rupp J: Survival in fatal road crashes: Body mass index, gender, and safety belt use. Traffic Inj Prev 2010;11(1):66–68.
41. Zhu S, Kim JE, Ma X, et al.: BMI and risk of serious upper body injury following motor vehicle crashes: Concordance of real-world and computer-simulated observations. PLoS Med 2010;7(3):e1000250.
42. Wells JC: Sexual dimorphism of body composition. Best Pract Res Clin Endocrinol Metab 2007;21(3):415–430.
43. Taylor RW, Grant AM, Williams SM, Goulding A: Sex differences in regional body fat distribution from pre- to postpuberty. Obesity
(Silver Spring) 2010;18(7):1410–1416.
44. Ranasinghe C, Gamage P, Katulanda P, Andraweera N, Thilakarathne S, Tharanga P: Relationship between body mass index (BMI) and body fat percentage, estimated by bioelectrical impedance, in a group of Sri Lankan adults: A cross sectional study. BMC Public Health 2013;13:797.
45. Deurenberg P, Andreoli A, Borg P, et al.: The validity of predicted body fat percentage from body mass index and from impedance in samples of five European populations. Eur J Clin Nutr 2001;55(11):973–979.
46. Larsson I, Henning B, Lindroos AK, Näslund I, Sjöström CD, Sjöström L: Optimized predictions of absolute and relative amounts of body fat from weight, height, other anthropometric predictors, and age 1. Am J Clin Nutr 2006;83(2):252–259.
47. Krachler B, Völgyi E, Savonen K, Tylavsky FA, Alén M, Cheng S: BMI and an anthropometry-based estimate of fat mass percentage are both valid discriminators of cardiometabolic risk: A comparison with DXA and bioimpedance. J Obes 2013;2013(862514):862514.