Relation of Vitamin D Level, BMI, and Location of Lower Extremity Stress Fractures in Military Trainees

ABSTRACT

Introduction

Military trainees are at an increased risk of stress fractures. Vitamin D availability is known to play an important role in both fracture prevention and healing. The purpose of this investigation was to assess 25-hydroxy vitamin D (25(OH)D) levels in soldiers with confirmed lower extremity stress fractures and assess the predictors of fracture location.

Materials and Methods

Following Institutional Review Board approval, military trainees at a large training base presenting to the orthopedic clinic with a radiographically verified stress fracture were identified. Demographic data and 25(OH)D levels were collected. A descriptive analysis was performed in regard to patient age, body mass index (BMI), and 25(OH)D level. Interactions between variables were assessed using one-way analysis of variance for four fracture location groups (femoral neck, femoral shaft, tibial shaft, and foot and ankle). Bivariate correlations were examined between age, BMI, and vitamin D level.

Results

A total of 155 lower extremity stress fractures were identified in 144 males and 11 females over 30 months. The mean age was 22.7 ± 4.85 years. The majority (60.7%) of fractures were located in the femoral neck. The average 25(OH)D level was 26.8 ± 8.37 ng/mL. Overall, 26% (N = 41) of enrolled patients had normal 25(OH)D levels, 48% (N = 74) had insufficient 25(OH)D levels, and 26% (N = 40) had deficient 25(OH)D levels. Patients with femoral neck fractures and tibial shaft fractures had significantly lower BMI than patients with foot and ankle fractures (23.3 vs. 27.7, P < .001 and 24.2 vs. 27.7, P = .003, respectively). Patients with foot and ankle fractures had significantly lower 25(OH)D levels than patients with femoral shaft fractures (21.1 vs. 30.1, P = .02). There were no significant findings regarding age and fracture location. Age correlated positively (but weakly) with BMI (0.338, P < .001). There was no correlation between age and vitamin D level or BMI and vitamin D level.

Conclusion

Overall, 74% of patients in military training with lower extremity stress fractures had insufficient or deficient levels of 25(OH)D, highlighting a persistent area of concern in this population. Patients with femoral neck and tibial shaft stress fractures had significantly lower BMI than patients with foot and ankle stress fractures. This suggests that in stress fracture–prone patients, BMI may play a role in predicting fracture location.

INTRODUCTION

Military training puts new soldiers at increased risk for stress fracture. Possible etiologies include sudden onset of training, with strenuous and regular exercise often involving prolonged endurance activities and load bearing, combined with acute weight loss.¹ The incidence of stress fracture in the military population ranges from ∼3% in males to 9.2% in females.^1,2 The implications of stress fracture can be significant, with 40% of men and 60% of women in the U.S. Military not completing basic training as a result.³ It has also been said that stress fractures are the single most costly injury in military training.⁴

Decreased circulating vitamin D levels, in the form of 25-hydroxyvitamin D (25(OH)D), have been associated with increased risk for stress fracture.^5–8 The vitamin D receptor has also been investigated at the gene–environment level in military recruits, with findings further supporting the role of low vitamin D levels in stress fracture pathogenesis.⁹ Furthermore, supplementation has been shown to play an important role in decreasing the incidence of stress fracture in military recruits and athletes.^3,10 In female Navy recruits, it has been shown that supplementation with vitamin D and calcium led to a 20% decreased incidence of stress fracture in the supplemented group.³ In collegiate athletes, patients who remained low in vitamin D had a 12% higher rate of stress fracture compared to those who were initially low but improved their vitamin D level to ≥40 ng/mL with supplementation.¹⁰ Low serum vitamin D also appears to play a role in the duration of recovery. In British Army recruits, it has been shown that the mean time to recover from stress fractures in individuals with sufficient levels of vitamin D was less compared to deficient or insufficient groups.¹¹ Other risk factors for stress fracture in military recruits include older age, lower body weight, and lower body mass index (BMI).¹² In both military and civilian populations, female sex^2,13–15 and lower body fat percentage have also been shown to increase the risk of stress fracture.¹⁶

Stress fracture location can have important clinical consequences. For instance, femoral neck stress fractures can have devastating long-term sequelae.¹⁷ In young military patients who sustained a displaced femoral neck stress fracture, medical discharge rates may be as high as 100%.^17,18 Therefore, predicting fracture location, utilizing known risk factors for stress fracture, may be an important tool for military clinicians. The purpose of this study was to examine 25(OH)D levels in military trainees who presented to an orthopedic clinic with symptomatic, radiographically confirmed stress fractures and then assess whether age, vitamin D level, or BMI plays a role in predicting fracture location.

METHODS

After obtaining Institutional Review Board approval, 25(OH)D levels and demographic data in all military patients presenting to the orthopedic surgery clinic at Martin Army Community Hospital, Fort Benning, GA, with verified stress fractures were collected. Inclusion criteria for the study included active duty patients and verified stress fractures based on history, physical exam, and diagnostic imaging (plain radiographs, magnetic resonance imaging, or bone scan) determined by a radiologist and the treating orthopedic surgeon. All patients received standard treatment for their stress fracture to include internal fixation, casting or bracing, and activity modification as deemed appropriate by the treating orthopedic surgeon depending on fracture type and location. At the time of diagnosis, all patients underwent standard serum 25(OH)D testing (Beckman Coulter, Brea, CA). For the purpose of this study (and the laboratory test reference range), 25(OH)D levels were defined as normal (30-100 ng/mL), insufficient (20-30 ng/mL), or deficient (<20 ng/mL).

A descriptive analysis was performed in regard to patient age, BMI, and vitamin D level. Interactions between variables were assessed using one-way analysis of variance for four fracture location groups (femoral neck, femoral shaft, tibial shaft, and foot and ankle). Bivariate correlations were also examined between age, BMI, and vitamin D level.

RESULTS

Demographics

Over the course of 30 months (2014-2017), 155 lower extremity stress fractures were identified in 144 males and 11 females (Table 1). Patients’ age ranged from 17 to 42 years (mean = 22.7 ± 4.85 years). Overall, 94 fractures were located in the femoral neck (60.7%), 13 in the femoral shaft (8.4%), 34 in the tibial shaft (21.9%), and 14 in the foot and ankle (9.0%).

Vitamin D Levels

The average 25(OH)D level was 26.8 ± 8.37 ng/mL. Overall, 26% (N = 41) of enrolled patients had normal vitamin D levels, 48% (N = 74) were insufficient, and 26% (N = 40) were deficient (Table 1).

Predictors of Fracture Location

Patients with femoral neck fractures and tibial shaft fractures had significantly lower BMIs than patients with foot and ankle fractures (23.3 vs. 27.7, P < .001 and 24.2 vs. 27.7, P = .003, respectively) (Table 2). Patients with foot and ankle fractures had significantly lower vitamin D levels than patients with femoral shaft fractures (21.1 vs. 30.1, P = .02).

There were no significant findings regarding age and fracture location. Age correlated positively (but weakly) with BMI (0.338, P < .001). There was no correlation between age and vitamin D level or BMI and vitamin D level.

DISCUSSION

The purpose of this study was to examine 25(OH)D levels in military trainees who presented with a stress fracture. A secondary aim of this study was to assess whether age, vitamin D level, or BMI plays a role in predicting fracture location. In our study population of military trainees, we found that 74% of patients with lower extremity stress fractures had insufficient or deficient levels of 25(OH)D (<30 ng/mL). We also found a high rate of stress fracture in this population, with 155 fractures over just 30 months, averaging approximately five fractures per month. This underscores the importance and relevance of this topic in the military population.

Our findings regarding vitamin D deficiency are in keeping with the existing literature. Davey et al. examined 25(OH)D concentrations in Royal Marine recruits and found that those who stress fractured within the first 10 weeks of training had baseline serum 25(OH)D levels of 49.5 ± 18.7 nmol L⁻¹ (∼14.3 ng/mL).⁵ Those who fractured later in training had a baseline serum level of 66.4 ± 29.2 nmol L⁻¹ (∼19.2 ng/mL). Furthermore, soldiers with a 25(OH)D level of <50 nmol L⁻¹ (∼14.4 ng/mL) at the start of training had an odds ratio for stress fracture of 1.6 (95% confidence interval [CI], 1.0-2.6) compared to recruits with a baseline serum level >50 nmol L⁻¹.

Dao et al. reported on 25(OH)D levels in a meta-analysis of eight observational studies including 2,634 military personnel (age, 18-30 years; 44% male) with 761 cases (16% male) and 1,873 controls (61% male) of both military trainees and active duty military.⁸ Three of the studies included in the meta-analysis measured serum 25(OH)D levels at the time of diagnosis.^6,19,20 Dao et al. calculated that the pooled mean serum 25(OH)D level was significantly lower in stress fracture cases compared with controls (mean difference, −2.26 ng/mL; 95% CI, −3.89 to −0.63; P = .007). In the three studies measuring serum 25(OH)D levels at the time of diagnosis, 25(OH)D ranged from 25.3 ± 10 to 28.2 ± 12.1 ng/mL. Our findings fall within the range of the previously reported data, with an average 25(OH)D level of 26.8 ± 8.37 ng/mL measured at the time of diagnosis.

The distribution of fracture location in our study displayed a much higher representation of femoral neck fractures and lower representation of foot and ankle fractures, compared to the existing literature. We found 94 fractures located in the femoral neck (60.7%) and 14 fractures in the foot and ankle (9.0%). In a previous study of United States Military Academy (USMA) cadets, Cosman et al. reported the most common stress fracture location to be the metatarsals (58%), followed by the tibia (29%).¹⁵ The femur was relatively rare, with only 5% of cadets sustaining a stress fracture in this location. We demonstrated relatively similar results regarding stress fractures in the tibia (21.9%). In another study of 800 Finnish military recruits, Ruohola et al. reported a total of 30 stress fractures in 22 patients, with the most common location being the foot/ankle (50%) followed by the tibia (33%). They only reported one fracture found in the femur (3.3%).²¹

It is unclear why our results differ so greatly in regard to the prevalence of femoral neck fractures. In a previous study completed in 1988, also at Fort Benning, GA, Fullerton et al. reported 54 stress fractures in 49 patients over a period of 4 years.²² This is certainly less than our findings of 94 femoral neck stress fractures over the course of 30 months. It is possible that the diagnosis of femoral neck stress fractures has become more sensitive, as our study utilized both magnetic resonance imaging and bone scan compared to plain radiographs alone. It is also possible that not every patient with a diagnosis of stress fracture was referred to the orthopedic clinic, and this may favor an over-representation of higher-risk stress fractures, such as femoral neck stress fractures, compared to foot and ankle stress fractures that military primary care providers feel more comfortable managing. Finally, changing training demands in the recent era at this specific training center have also led to a heightened risk of femoral neck stress fractures. As femoral neck stress fractures can have devastating long-term sequelae if left untreated, this is an important area of further study.

Finally, we also analyzed the potential predictors of fracture location. We discovered that patients with femoral neck fractures and tibial shaft fractures had significantly lower BMI than patients with foot and ankle fractures (23.3 vs. 27.7, P < .001 and 24.2 vs. 27.7, P = .003, respectively). Lower BMI has been previously reported to be a risk for stress fracture, in general, in both male and female basic trainees.¹² It has also been shown that low body fat percentage is associated with higher risk of stress fracture.¹⁶ Knapik et al. reported that women and men with <8.9% body fat had a 20-27% higher relative risk of developing a stress fracture during basic training.¹⁶ The relationship between femoral neck fractures and tibial shaft fractures in patients with lower BMI and body fat percentage may be related to frame sizes. Previously, it has been demonstrated that lower body fat percentage has been associated with smaller bone width.²³ It has also been shown that military recruits with smaller tibias and femurs have a higher likelihood of stress fracture during basic training.²⁴ Therefore, it is possible that because of smaller frame sizes, the femoral neck and tibial shaft are at greater risk of stress fracture in military recruits with lower BMI compared to those with foot and ankle fractures. We also found that military recruits with foot and ankle fractures had significantly lower 25(OH)D levels than femoral shaft fractures (21.1 vs. 30.1, P = .02); however, clinic applicability of this finding remains unclear as there were no other significant findings revealed regarding 25(OH)D levels and specific fracture location.

Because of significant implications of vitamin D deficiency and stress fractures in military personnel, recent literature has focused on possible solutions in this patient population. Fogleman et al. published a commentary piece recommending lifestyle modifications, such as 20 minutes of sun exposure daily to the arms and legs, routine supplementation of vitamin D, standardized treatments of vitamin D deficiency, and screening of all recruits entering military service.⁴ They also recommend a military-wide campaign to increase education on bone health. Our study certainly supports the need for rapid intervention.

This study has several limitations based on its observational nature. This includes potential measurement errors in the determination of stress fracture at the time of diagnosis. We did not prospectively calculate the sample size; therefore, it is possible that the determination of demographic risk factors in regard to fracture location may be underpowered. Another limitation is that we did not analyze biologic sex in regard to fracture location, as we had a relatively small number of women in our study. This is likely due to the fact that Fort Benning is the training location and home for multiple combat arms units. As such, there are more active duty men than women on site at any given time, even compared to some other posts. Additionally, data collection began before the full integration of women into combat arms; therefore, this underrepresentation may be further exaggerated compared to more recent years. As sex has been shown to play a role in stress fracture, this may prevent the study from being applicable to the larger population. Finally, we did not analyze the role of race or ethnicity, which has also been shown to play a role in 25(OH)D levels and the risk of stress fracture.

CONCLUSION

25(OH)D insufficiency/deficiency remains an area of concern in U.S. military recruits. Our study demonstrated that 74% of basic trainees with lower extremity stress fractures had insufficient or deficient levels of 25(OH)D. Due to the role vitamin D plays in bone health, hypovitaminosis D certainly places basic trainees at increased risk of stress fracture. Our study demonstrated similar levels of 25(OH)D in military recruits at the time of diagnosis compared to the previous literature. We also found that patients with femoral neck and tibial shaft stress fractures had significantly lower BMI than patients with foot and ankle stress fractures. This is likely related to the relationship between BMI and frame size, and this finding has the potential to play a role in predicting fracture location in stress fracture–prone patients such as military recruits. Military-wide interventions are still needed to decrease the impact of vitamin D deficiency and stress fractures on basic trainees.

ACKNOWLEDGMENTS

Raywin R. Huang, Ph.D., Biostatistician.

FUNDING

None declared.

CONFLICT OF INTEREST STATEMENT

None declared.

REFERENCES

Author notes