You are here

Article Text

Article menu
Original research
Hypothalamic-pituitary-ovarian axis suppression is common among women during US Army Basic Combat Training
Free
  1. Kristin L Popp1,2,3,
  2. Brittany N Bozzini1,
  3. Marinaliz Reynoso1,
  4. Jennifer Coulombe1,4,
  5. Katelyn I Guerriere1,
  6. Susan P Proctor1,
  7. Colleen M Castellani1,
  8. Leila A Walker1,
  9. Nicholas Zurinaga1,
  10. Katherine Kuhn5,
  11. Stephen A Foulis1,
  12. Mary L Bouxsein4,6,
  13. Julie M Hughes1,
  14. Nanette Santoro5
  1. 1 Military Performance Division, US Army Research Institute of Environmental Medicine, Natick, Massachusetts, USA
  2. 2 TRIA Orthopaedic Center, HealthPartners Institute, Bloomington, Minnesota, USA
  3. 3 Wu Tsai Female Athlete Program, Division of Sports Medicine, Boston Children's Hospital, Boston, Massachusetts, USA
  4. 4 Department of Orthopedic Surgery, Harvard Medical School and Center for Advanced Orthopedic Studies, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA
  5. 5 Department of Obstetrics and Gynecology, University of Colorado Anschutz Medical Campus, Aurora, Colorado, USA
  6. 6 Endcrine Unit, Massachusetts General Hospital, Boston, Massachusetts, USA
  1. Correspondence to Dr Kristin L Popp; kpopp{at}mgh.harvard.edu

Abstract

Objective Less than half of servicewomen report loss of menses during initial military training. However, self-reported menstrual status may not accurately reflect hypothalamic-pituitary-ovarian (HPO) axis suppression and may underestimate reproductive health consequences of military training. Our aim was to characterise HPO axis function during US Army Basic Combat Training (BCT) in non-hormonal contraceptive-using women and explore potential contributors to HPO axis suppression.

Methods In this 10-week prospective observational study, we enrolled multi-ethnic women entering BCT. Trainees provided daily first-morning voided urine, and weekly blood samples during BCT. Urinary luteinising hormone, follicle stimulating hormone, and metabolites of estradiol and progesterone were measured by chemiluminescent assays (Siemens Centaur XP) to determine hormone patterns and luteal activity. We measured body composition, via dual-energy X-ray absorptiometry, at the beginning and end of BCT.

Results Trainees (n=55) were young (mean (95% CI): 22 (22, 23) years) with average body mass index (23.9 (23.1, 24.7) kg/m2). Most trainees (78%) reported regular menstrual cycles before BCT. During BCT, 23 (42%) trainees reported regular menses. However, only seven trainees (12.5%) had menstrual cycles with evidence of luteal activity (ELA) (ie, presumed ovulation), all with shortened luteal phases. 41 trainees (75%) showed no ELA (NELA), and 7 (12.5%) were categorised as indeterminant. Overall, women gained body mass and lean mass, but lost fat mass during BCT. Changes in body mass and composition appear unrelated to luteal activity.

Conclusions Our findings reveal profound HPO axis suppression with NELA in the majority of women during BCT. This HPO axis suppression occurs among women who report normal menstrual cycles.

  • Exercise
  • Stress, Physiological
  • Sleep
  • Endocrine System
  • Female

Data availability statement

Data are available upon reasonable request.

https://doi.org/10.1136/bjsports-2023-107716

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THE TOPIC

    • Previous literature assessing reproductive health in US military trainees is limited to self-report and most studies include both hormonal contraceptive users and non-users, which limits the ability to interpret findings.

    WHAT THIS STUDY ADDS

    • This is the first study to collect daily reproductive hormone measures that show the reproductive dysfunction that occurs during US military training, even in women reporting normal menstrual cycles. It further indicates that this dysfunction is due to a combination of stressors experienced in a military training environment.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • This research highlights the discrepancy between self-reported menses and actual reproductive health. It further indicates that treating factors beyond low energy availability should be considered for young, active individuals experiencing or at risk of reproductive and menstrual dysfunction.

    Introduction

    US Initial Military Training is a rigorous 6 to 13-week period at the beginning of a servicemember’s career that varies according to the military branch. Female military personnel frequently report menstrual disturbances during Initial Military Training with 53% of US Military Academy cadets reporting menstrual cycle interruptions within their first 3 months of training.1 2 Moreover, 86% of women report changes to menstrual cycle length, symptoms and bleeding during US Army Basic Combat Training (BCT).3 The hypothalamic-pituitary-ovarian (HPO) axis regulates reproductive function, including the orchestration of ovulation and menstrual cyclicity. HPO axis suppression leads to altered hormonal patterns and consequently short luteal phases, anovulation and amenorrhoea, some of which will be detectable by routine menstrual cycle monitoring and some of which may not.4 HPO axis suppression can have long-term health consequences, including infertility, impaired bone health and an increased risk of heart disease.5

    HPO axis suppression can stem from diverse aetiologies, including insufficient sleep, psychosocial stress and nutritional deficiencies.4 The most commonly recognised cause of reproductive dysfunction in premenopausal female athletes is insufficient caloric intake relative to exercise energy expenditure, leading to low energy availability.6 7 Low energy availability is often associated with maintenance or loss of body mass.4 6 7 In well-controlled laboratory settings, energy intake and expenditure can be accurately measured. However, field-based methods have poor accuracy,8 and are particularly challenging in environments such as BCT where the range of physical activities is highly variable.9 Several laboratory-based studies in women have demonstrated disruption in the hormonal milieu and metabolism over 3–6 days of low energy availability, indicating that blood markers, such as leptin, thyroid hormones and insulin-like growth factor (IGF-1), can reflect short-term low energy availability.10–12 Taken together, changes in blood markers, body mass and body composition can help estimate energy status in the field. The only study to rigorously assess HPO axis function in a multi-stressor field environment reported that 14 of 24 British Military trainees of European ancestry did not ovulate during the first 30 days of military training.13 However, the relationships among estimates of energy availability and HPO axis suppression in a larger, multi-ethnic cohort throughout US Army BCT are incompletely understood.

    Therefore, this study aimed to characterise HPO axis function in incoming US Army BCT trainees using daily hormone measures throughout BCT. Additionally, we investigated the relationships between HPO axis suppression and estimates of energy availability, sleep and stress. We hypothesised that HPO axis suppression, reflected by a short luteal phase or presumed anovulation, would occur in approximately half of the trainees. We further hypothesised that trainees who experience HPO axis suppression during BCT will have lower estimates of energy availability and sleep but higher estimates of stress than those with healthy HPO axis function.

    Methods

    Study design

    The research reported here is a substudy of a prospective observational study of US Army trainees enrolled in BCT.14 A subset of female trainees participating in that study were recruited from an incoming 10-week BCT class at an Army base in August of 2021. BCT is a standardised and documented training programme with an emphasis on physical readiness.15 16 Training includes callisthenics, running, marching, combatives, cadence marching and tactical marching. Trainees average approximately 16 000 steps per day, covering an average of 12 km per day with a peak of 22 000 steps per day or 16 km/day occurring near the end of BCT.16 Field training exercises were completed prior to week 6 and week 8 measurements in our cohort. Time spent performing moderate to intense physical activity increases progressively for the first 6 weeks of BCT.15

    Participants

    We enrolled 104 female trainees. The number of participants was limited to consenting women beginning BCT at one Army base in August of 2021 and, thus, represents a convenience sample. Additional details on power analyses performed for this project can be found in online supplemental materials. Inclusion and exclusion criteria were previously described.14 For the current study, trainees could not: have restricted physical activity, have an endocrine disorder (eg, diabetes or hypoparathyroidism), be pregnant, be using hormonal contraceptives or have been diagnosed with polycystic ovarian syndrome. Trainees were briefed on the voluntary nature of the study without military command staff present before providing informed consent.

    Supplemental material

    Questionnaires

    Trainees completed questionnaires at baseline and each week throughout BCT. The initial questionnaire captured demographics, lifestyle, sleep, and menstrual history and status. Baseline menstrual status was determined by self-reported number of menstrual cycles in the year prior to BCT. Each week during BCT, trainees reported whether their period started or ended. During the final week of BCT, trainees reported if their menstruation differed (duration, frequency, flow and menstrual cramp severity) compared with the year prior to BCT. At baseline and during the final week of BCT trainees were asked about their typical sleep habits during the prior month, including what time they had gone to bed and gotten up in the morning, the number of minutes it took them to fall asleep and the hours of actual sleep obtained each night.

    HPO axis function

    In order to determine HPO axis function, participants collected daily first-morning voided urine throughout BCT. Urine was aliquoted into polypropylene tubes prefilled with glycerol to a final concentration of 7%, frozen and transported to the University of Colorado for batched analysis. We normalised samples for the amount of creatinine.17 We assessed urinary luteinising hormone (LH) and follicle stimulating hormone (FSH) by competitive immunoassay using direct chemiluminescent technology (Centaur XP; Siemens Healthcare Diagnostics) as previously reported.18 19 Interassay and intra-assay coefficients of variation (CV) were 4.8% and 3.4% for LH, respectively, and 6.6% and 5.0% for FSH, respectively. Using immunoassays adapted for the Siemens Centaur XP, we assessed urinary reproductive hormone metabolites, estrone conjugates (E1c; estradiol) and pregnanediol glucuronide (Pdg; progesterone). Interassay and intra-assay CV were 11.5% and 8.1% for E1c, respectively, and 17.8% and 7.7% for Pdg, respectively.

    49 women of the subset of 104 women either had <60% of urine samples collected or more than four consecutive missing samples within 35 days so their urine samples were not analysed (online supplemental figure 1). Among the 55 women with sufficient urine compliance, we examined hormone and hormone metabolite profiles for ovulation, estimated by evidence of luteal activity (ELA). To be categorised as ELA, profiles had to meet the following criteria at any point during BCT20 21: (1) A threefold increase in Pdg concentrations above the nadir for ≥3 consecutive days using the Kassam et al validated algorithm.22 If 1 of 3 days was missing, we used simple linear interpolation; (2) corresponding rise in FSH, LH and E1c (detailed in online supplemental material). Individuals who did not meet these criteria were classified as having no ELA (NELA). If it was impossible to determine whether an individual achieved the above criteria due to missing sample(s) during a potential Pdg rise, she was designated ‘indeterminant’.

    Among women with ELA, we determined the day of luteal transition using the Waller et al method23 or due to missing samples, as the first day of the sustained Pdg rise. Luteal phase lengths <12 days were considered short based on an idealised 28-day cycle with a 14-day luteal phase.24

    Body composition

    We assessed height by stadiometer and body mass via a calibrated electronic scale and calculated body mass index (BMI=body mass (kg)/height (m)2). We measured body composition (lean body mass (kg), fat mass (kg), body fat percentage (%)) using whole body dual-energy X-ray absorptiometry (Lunar Prodigy, GE Healthcare, Madison, WI, USA) at baseline and post-BCT. Trainees wore standardised physical training shorts and T-shirts.

    Blood collection and fasting markers

    Study staff trained in phlebotomy acquired fasted, rested morning blood samples weekly by venipuncture throughout BCT to assess estimates of energy availability and stress. Blood was separated for serum, frozen at −30°C, shipped to USARIEM and Pennington Biomedical Research Center (PBRC, Baton Rouge, LA) and frozen at −80°C until batched analyses. We used Immulite analyzers (Siemens Healthineers) to assess cortisol and IGF-1 each week during BCT, and free triiodothyronine (FT3), triiodothyronine (T3), thyroxine (T4) and leptin at baseline, weeks 2, 4, 6, 8 and post-BCT. Baseline testosterone was measured by liquid chromatography-tandem mass spectrometry (Brigham Research Assay Core, Boston, MA, USA). Interassay and intra-assay CVs were <9.5% and <7.5% for all assays, respectively.

    Statistical analysis

    Baseline characteristics were reported in trainees classified as having ELA, NELA, or indeterminant HPO axis function. These same characteristics were reported for the cohort with adequate urine collection compliance compared with those who were excluded due to poor compliance. Changes in body mass, body composition, BMI, sleep and testosterone were compared within luteal activity groups by paired t-tests and between luteal activity groups by one-way analysis of variance (ANOVA). Differences in the proportion of trainees by age of menarche, race/ethnicity and smoking status between the luteal activity groups were determined via χ2 test.

    Secondary analyses were conducted to compare serum biomarkers between the three luteal activity groups as well as between the included versus excluded cohorts, using one-way ANOVA for unadjusted differences and multivariate linear regression with repeated measures to adjust for covariates: age, BMI, race/ethnicity, smoking history and week of BCT. These covariates were identified through an a priori causal directed acyclic graph. The interaction of group and week of BCT was also assessed with repeated measures. Model selection was based on Akaike information criteria (AIC). Among the 104 women enrolled in this study, one trainee was missing BMI and one missing smoking history. We had complete covariate data for 98% of our cohort.

    Urinary hormone and hormone metabolite comparisons between ELA, NELA and indeterminant groups were normalised to 28 days. Data were organised to the peak E1c during the follicular phase of the first cycle to achieve all indicators of luteal activity (ELA) or during the 35 days of highest compliance (NELA, indeterminant). For each trainee, integrated hormone levels were calculated over the 28 days surrounding the E1c peak. Data interpolation was used if missing values were ≤2 subsequent days, where the average of the surrounding days with data was used to replace the missing values. If missing values spanned ≥3 days, values were replaced with the luteal activity group average for that day. Integrated and peak hormone levels between luteal activity group were compared using one-way ANOVA and Tukey’s HSD pairwise comparison for unadjusted differences and multivariate linear regression to adjust for covariates: age, BMI, race/ethnicity and smoking history. Model selection was based on AIC.

    Statistical test and model assumptions and fit were assessed including linear associations of outcomes with continuous predictors, and normally distributed residuals with constant variance. Analyses were performed in RStudio V.2023.06.1+524 ‘Mountain Hydrangea’ release.

    Equity, diversity and inclusion

    Our study included a multi-ethnic cohort of women of all socioeconomic levels from all regions of the US race/ethnicity was a covariate in our analyses. We excluded men, because the goal of the study was to characterise reproductive health in women. Our research team consisted of ten women and two men with ranks from professor to first-time researcher. The author’s disciplines include exercise physiology, reproductive endocrinology, bone physiology, biology and epidemiology.

    Patient and public involvement

    Patients or the public were not involved in any aspect of this research.

    Results

    Baseline characteristics

    Trainees with adequate daily urine compliance were 22 (95% CI 22, 23) years old and had normal BMI (mean (95% CI): 23.9 (23.1, 24.7) kg/m2). They reported 7.5 (95% CI 7.0, 8.0) hours of sleep per night prior to BCT. All but one individual had testosterone levels (mean (95% CI): 36.0 (34.7, 37.3) ng/dL) within the normal reference range (15–70 ng/dL) for age (table 1). The individual outside the reference range had a testosterone level of 71.1 ng/dL. Age of menarche was younger in the ELA group compared with the NELA and indeterminant groups. Other baseline characteristics appear to be similar among luteal activity groups. Trainees with adequate versus inadequate daily urine compliance also had similar baseline characteristics (online supplemental table 1).

    View this table:
    Table 1

    Characteristics of trainees in the overall cohort, and those with evidence of luteal activity (ELA), with no evidence of luteal activity (NELA) and with indeterminant luteal activity

    HPO axis function

    During BCT, 42% of women reported regular (every 21–35 days apart) menstrual periods (figure 1). However, we did not observe a single menstrual cycle that indicated both ELA and normal luteal phase length among our cohort. Seven trainees (12.5%) achieved one or more menstrual cycles that met ELA criteria. However, all seven of these trainees displayed short luteal phases, ranging from 7 to 11 days. 41 trainees (75%) showed NELA, and 7 (12.5%) were classified as indeterminant (figure 1).

    Figure 1
    Figure 1

    (A) Per cent of trainees self-reporting regular versus irregular menstrual cycles during the 1 year prior to (n=43; n=12, respectively), and during Basic Combat Training (BCT) (n=23; n=32, respectively). (B) Per cent of trainees with evidence of luteal activity (n=7), without evidence of luteal activity (n=41) and with indeterminant luteal activity (n=7), assessed by analyses of daily urine samples.

    Comparisons of integrated hormones between groups revealed higher E1c and Pdg in women with ELA than those with NELA or indeterminant cycles (figure 2, table 2). Peak LH, E1c and Pdg were higher among women with ELA than those with NELA. Peak Pdg was also higher among trainees with ELA than those with indeterminant cycles (table 2). Daily hormone values per group are presented in figure 2 and online supplemental tables 2–5.

    Figure 2
    Figure 2

    Daily levels of urinary hormones measured over 28 days surrounding the peak in metabolites of estradiol (E1c). (A) Luteinising hormone (LH), (B) follicle-stimulating hormone (FSH), (C) metabolites of E1c and (D) progesterone (Pdg) were measured in each sample. Data are presented as mean (SEM) plotted against cycle day, where day 0 is the day of E1c peak in trainees with evidence of luteal activity (ELA; n=7), with no evidence of luteal activity (NELA; n=41) and with indeterminant luteal activity (n=7).

    View this table:
    Table 2

    Peak and integrated hormone values among those with evidence of luteal activity (ELA), with no evidence of luteal activity (NELA) and with indeterminant luteal activity

    Changes in body mass, body composition and sleep

    From baseline to post-BCT, on average, trainees gained body mass (p=0.012) and lean mass (p<0.001), while losing fat mass (p<0.001) (table 3). Reported hours of sleep per night decreased from before to during BCT (p=0.0055). Changes in body mass, body composition and sleep appear similar by luteal activity group and between trainees with adequate versus inadequate daily urine compliance (online supplemental table 6).

    View this table:
    Table 3

    Changes in body mass and composition and self-reported sleep from baseline to post-BCT among those with evidence of luteal activity (ELA), with no evidence of luteal activity (NELA) and with indeterminant luteal activity

    Fasting blood markers

    Serum analyses revealed variable responses in estimates of energy availability. Specifically, leptin increased from baseline to weeks 6 (+16%, p=0.041), 8 (+9%, p=0.023) and post-BCT (+38%, p<0.001) among the cohort (figure 3, online supplemental table 7). FT3 and T3 were lower at week 2 (FT3: −7%, p=0.0067; T3: −6%, p=0.0078) compared with baseline but similar to baseline at all other timepoints (figure 3, online supplemental figure 2 and table 7). T4 and FT4 were lower at all time points compared with baseline (T4: −23% to −17 %, p<0.001 for all; FT4: −22% to −11%, p<0.001 for all; figure 3, online supplemental figure 2 and table 7). We did not detect differences between groups in changes in leptin, FT3, T3 or T4. FT4 was lower among the ELA group compared with NELA at weeks 4, 6, 8 and post-BCT (p<0.021 for all) and lower than the indeterminant group than NELA at week 4 (p=0.027, online supplemental table 7). IGF-1 values varied during BCT and were lower at weeks 2 (−17%, p<0.001) and 6 (−5%, p=0.041) but higher at weeks 4, 5 and post-BCT (+8 to +18%, p<0.0059 for all) compared with baseline with no differences detected by group (figure 3, online supplemental table 7). Cortisol, an estimate of stress, was lower than baseline at weeks 1, 3, 4, 5, 6 and post-BCT (−14% to −5%, p<0.048 for all) and was lower in the NELA compared with the ELA group at six of the 10 timepoints (−14% to −4%%; p<0.0098 for all; figure 3, online supplemental table 7).

    Figure 3
    Figure 3

    Serum (A) leptin, (B) free triiodothyronine (FT3), (C) insulin-like growth factor (IGF-1) and (D) cortisol in trainees with evidence of luteal activity (ELA; n=7), with no evidence of luteal activity (NELA; n=41) and with indeterminant luteal activity (n=7). Data are presented as mean (SEM). BCT, Basic Combat Training.

    Discussion

    Hormone profiles in young female US Army trainees demonstrate profound suppression of the HPO axis throughout 10 weeks of BCT. Though many trainees reported menstrual bleeding, we did not observe a single hormone profile that indicated both ELA and adequate luteal phase length. Indeed, nearly all trainees in our cohort failed to achieve a progesterone surge indicative of ovulation and normal reproductive function. This observation indicates that self-reported menstrual status drastically underestimates the number of trainees who experience decreased ovarian function during military training. Thyroid and pituitary hormone profiles coupled with gains in lean mass and body mass suggest adequate energy intake during BCT. Trainees display moderately elevated cortisol throughout BCT and report fewer hours of sleep during BCT compared with prior to BCT. It is likely that HPO axis suppression results from a compounding effect of stressors during BCT.

    Our findings of widespread ovarian suppression among trainees from the onset of BCT were unexpected. Most trainees in our cohort reported normal menses prior to BCT and at least one menstrual bleed during BCT, which is in agreement with our previous report3 and prior literature.2 13 25 Specifically, we previously reported that although 86% of trainees self-report changes in menstrual cycle symptoms and length, 78% of trainees self-report having a period during BCT.3 The only study to report ovulatory status using frequent hormone measures in a military training environment found that 14 of 24 women showed no evidence of ovulation in the first 30 days of an 11-month UK Royal Military Academy Commissioning Course despite all 24 reporting a menstrual bleed during the first 14 weeks of training.13 Similarly, in a study of recreational and competitive athletes, 25% of those reporting regular menses were anovulatory by daily hormone measurements.26 Thus, self-reported menstrual status underestimates HPO axis suppression among young, active populations. Our findings highlight the severity of HPO axis suppression experienced by trainees undergoing US Army BCT, and the near-immediate onset of that suppression during BCT.

    Metabolic and pituitary hormones suggest that energy availability was largely adequate during BCT. Leptin, a typically sensitive marker of energy status, increased during BCT, suggesting adequate energy. T3 and FT3, IGF-1 were lower during one to two timepoints during BCT but similar to or higher than baseline at eight to nine timepoints. When exposed to prolonged low energy availability, we would expect these markers to decrease in as few as 5 days and remain low.10 12 27 28 In specialised military training environments with a known energy deficit of ~4000 kcal/day, IGF-1, T3 and T4 decreased after 7 days.29 Similarly, smaller (~1000 kcal/day) energy deficits lead to decreased T3 and IGF-1 over 8 weeks of US Army Ranger Training.30 Still, it is possible that the loss of fat mass observed in our cohort could be indicative of low energy availability. Prior studies demonstrated retention of or increased lean mass coupled with decreased fat mass during energy deficit.31 32 At an estimated 30% reduction in energy intake over 4 weeks combined with resistance and interval training, men who consumed high protein diets were able to gain lean body mass while losing fat mass.31 Notably, total body mass decreased over 4 weeks in the male cohort, compared with our cohort, where trainees gained body mass. Given the observed increase in body mass and stable or increased serum surrogates of low energy availability in our cohort, it is unlikely that low energy availability is the cause of the widespread HPO axis suppression we observed.

    Mechanisms underlying the HPO axis suppression in our study are unclear. Metabolic and psychosocial stress, including, lack of social support, nutritional deficiencies and disturbed sleep, can have compounding effects on reproductive function.4 11 33 34 Preclinical evidence in primates supports this concept and demonstrates that psychosocial and metabolic stressors are additive in their contribution to reproductive dysfunction.35 It is well-established that military personnel experience insufficient sleep quantity and quality during training courses, including US Army BCT,36–39 so it is not surprising that self-reported sleep decreased in our cohort during BCT. Similarly, psychosocial stress is an acknowledged aspect of military training.13 40 Stress-induced alterations to HPO axis function include reduced LH pulse frequency and insufficient LH and FSH levels to maintain ovulatory ovarian function.4 34 Our findings of moderately elevated cortisol levels,41 42 particularly in the NELA group, align with prior work demonstrating that women with functional hypothalamic amenorrhoea have higher cortisol43 and a greater increase in serum cortisol levels in response to psychological stress exposure44 than their eumenorrheic counterparts. Taken together, our observed changes in sleep and cortisol provide evidence of multiple stressors experienced by trainees in the NELA group. These in addition to the changes in exercise, living quarters and social support on military entrance, may have contributed to the HPO axis suppression observed in this cohort and should be considered in future work.

    Clinical implications

    Clinicians treating women living and performing in stressful environments should consider mechanisms beyond low energy availability as a potential cause of HPO axis suppression. Based on these data, relying on self-reported menses is not a sufficient estimate of HPO axis function. Hormonal abnormalities, especially those of the luteal phase, may be present in the absence of obvious menstrual irregularities and if they are maintained over time, could contribute to health concerns related to HPO axis function, including bone and cardiovascular health, and infertility.

    Strengths/limitations

    Limitations of our study include the small sample size and unequal distribution of subjects between groups, which limited the statistical power of our analysis. Though compliance and subject retention were challenging, this is the most extensive study detailing female reproductive health during military training and the only study in US military BCT. The lack of longer-term follow-up is a limitation as we were unable to determine if and when HPO axis function returns to normal. Similarly, we did not have the ability to assess HPO axis function in the months leading up to BCT and therefore, do not know whether HPO axis suppression was present prior to BCT. We underestimated the percentage of women who would have normal reproductive function and did not have a control group, thus our ability to determine contributing factors was limited. We did not capture exercise energy expenditure or energy intake to assess energy availability which have been documented to be variable throughout military training.45 46 Sleep outcomes were based on self-report and may not accurately reflect sleep duration or quality. Though we did not identify confounders in our directed acyclic graph and we adjusted for key covariates known to impact HPO axis function, there is potential for residual confounding due to potential unmeasured confounders that were not considered or accounted for in our study design.

    Strengths of our study include the comprehensive assessment of reproductive function through daily hormone measurements in a racially diverse cohort of military trainees. Further, due to the logistical challenges of accessing military populations during training, our longitudinal serum measurements are unique to military populations. Our ability to acquire frequent biological specimens allowed the discovery of the near-universal HPO axis suppression that occurs during US Army BCT.

    Conclusion

    Our findings reveal that HPO axis suppression with NELA is widespread during US Army BCT and occurs among women reporting regular menstruation.

    Data availability statement

    Data are available upon reasonable request.

    Ethics statements

    Patient consent for publication

    Ethics approval

    This study involves human participants and was approved by the Institutional Review Board of the US Army Research Institute of Environmental Medicine (Protocol # 17-18HC) and the US Army Medical Research and Development Command, Fort Detrick, Maryland (Protocol # M-10678). Participants gave informed consent to participate in the study before taking part.

    Acknowledgments

    The authors thank all participants and the researchers on the ARIEM Reduction in Musculoskeletal Injury team for their support. We thank the Command staff and cadre at Fort Jackson, South Carolina for access to the trainees and logistics support. We thank the US Army Center for Initial Military Training for their support of this study.

    References

    1. ↵
      2. Gifford RM ,
      3. Reynolds RM ,
      4. Greeves J , et al
      . Reproductive dysfunction and associated pathology in women undergoing military training. J R Army Med Corps 2017;163:301–10. doi:10.1136/jramc-2016-000727
      OpenUrlAbstract/FREE Full Text
    2. ↵
      2. Cosman F ,
      3. Ruffing J ,
      4. Zion M , et al
      . Determinants of stress fracture risk in united states military academy cadets. Bone 2013;55:359–66. doi:10.1016/j.bone.2013.04.011
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      2. Bozzini BN ,
      3. Nguyen VT ,
      4. Reynoso MC , et al
      . The risk of menstrual dysfunction increases for women during U.S. army basic combat training. Med Sci Sports Exerc 2023;55:1533–9. doi:10.1249/MSS.0000000000003183
      OpenUrl
    4. ↵
      2. Gordon CM ,
      3. Ackerman KE ,
      4. Berga SL , et al
      . Functional hypothalamic amenorrhea: an endocrine society clinical practice guideline. J Clin Endocrinol Metab 2017;102:1413–39. doi:10.1210/jc.2017-00131
      OpenUrlCrossRefPubMed
    5. ↵
      2. Shufelt CL ,
      3. Torbati T ,
      4. Dutra E
      . Hypothalamic amenorrhea and the long-term health consequences. Semin Reprod Med 2017;35:256–62. doi:10.1055/s-0037-1603581
      OpenUrl
    6. ↵
      2. Joy E ,
      3. De Souza MJ ,
      4. Nattiv A , et al
      . 2014 female athlete triad coalition consensus statement on treatment and return to play of the female athlete triad. Curr Sports Med Rep 2014;13:219–32. doi:10.1249/JSR.0000000000000077
      OpenUrlCrossRefPubMed
    7. ↵
      2. Mountjoy M ,
      3. Ackerman KE ,
      4. Bailey DM , et al
      . 2023 international olympic committee’s (IOC) consensus statement on relative energy deficiency in sport (reds). Br J Sports Med 2023;57:1073–97. doi:10.1136/bjsports-2023-106994
      OpenUrlAbstract/FREE Full Text
    8. ↵
      2. Burke LM ,
      3. Lundy B ,
      4. Fahrenholtz IL , et al
      . Pitfalls of conducting and interpreting estimates of energy availability in free-living athletes. Int J Sport Nutr Exerc Metab 2018;28:350–63. doi:10.1123/ijsnem.2018-0142
      OpenUrlPubMed
    9. ↵
      2. Gifford RM ,
      3. Greeves JP ,
      4. Wardle SL , et al
      . Measuring the exercise component of energy availability during arduous training in women. Med Sci Sports Exerc 2021;53:860–8. doi:10.1249/MSS.0000000000002527
      OpenUrl
    10. ↵
      2. Loucks AB ,
      3. Thuma JR
      . Luteinizing hormone pulsatility is disrupted at a threshold of energy availability in regularly menstruating women. J Clin Endocrinol Metab 2003;88:297–311. doi:10.1210/jc.2002-020369
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      2. Hilton LK ,
      3. Loucks AB
      . Low energy availability, not exercise stress, suppresses the diurnal rhythm of leptin in healthy young women. Am J Physiol Endocrinol Metab 2000;278:E43–9. doi:10.1152/ajpendo.2000.278.1.E43
      OpenUrlPubMedWeb of Science
    12. ↵
      2. Loucks AB ,
      3. Heath EM
      . Induction of low-T3 syndrome in exercising women occurs at a threshold of energy availability. Am J Physiol 1994;266:R817–23. doi:10.1152/ajpregu.1994.266.3.R817
      OpenUrlWeb of Science
    13. ↵
      2. Gifford RM ,
      3. O’Leary TJ ,
      4. Wardle SL , et al
      . Reproductive and metabolic adaptation to multistressor training in women. Am J Physiol Endocrinol Metab 2021;321:E281–91. doi:10.1152/ajpendo.00019.2021
      OpenUrl
    14. ↵
      2. Hughes JM ,
      3. Foulis SA ,
      4. Taylor KM , et al
      . A prospective field study of U.S. army trainees to identify the physiological bases and key factors influencing musculoskeletal injuries: a study protocol. BMC Musculoskelet Disord 2019;20:282. doi:10.1186/s12891-019-2634-9
    15. ↵
      2. Knapik JJ ,
      3. Darakjy S ,
      4. Hauret KG , et al
      . Ambulatory physical activity during united states army basic combat training. Int J Sports Med 2007;28:106–15. doi:10.1055/s-2006-924147
      OpenUrlCrossRefPubMed
    16. ↵
      2. Simpson K ,
      3. Redmond JE ,
      4. Cohen BS , et al
      . Quantification of physical activity performed during US army basic combat training. US Army Med Dep J 2013;55–65.
    17. ↵
      2. Saketos M ,
      3. Sharma N ,
      4. Adel T , et al
      . Time-resolved immunofluorometric assay and specimen storage conditions for measuring urinary gonadotropins. Clin Chem 1994;40:749–53.
      OpenUrlAbstract/FREE Full Text
    18. ↵
      2. Rovner P ,
      3. Keltz J ,
      4. Allshouse A , et al
      . Induction of the LH surge in premenarchal girls confirms early maturation of the hypothalamic-pituitary-ovarian axis. Reprod Sci 2018;25:33–8. doi:10.1177/1933719117741377
      OpenUrl
    19. ↵
      2. Ross LA ,
      3. Polotsky AJ ,
      4. Kucherov A , et al
      . Profound reduction of ovarian estrogen by aromatase inhibition in obese women. Obesity (Silver Spring) 2014;22:1464–9. doi:10.1002/oby.20713
      OpenUrl
    20. ↵
      2. O’Connor KA ,
      3. Brindle E ,
      4. Miller RC , et al
      . Ovulation detection methods for urinary hormones: precision, daily and intermittent sampling and a combined hierarchical method. Hum Reprod 2006;21:1442–52. doi:10.1093/humrep/dei497
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      2. Weiss G ,
      3. Skurnick JH ,
      4. Goldsmith LT , et al
      . Menopause and hypothalamic-pituitary sensitivity to estrogen. JAMA 2004;292:2991–6. doi:10.1001/jama.292.24.2991
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      2. Kassam A ,
      3. Overstreet JW ,
      4. Snow-Harter C , et al
      . Identification of anovulation and transient luteal function using a urinary pregnanediol-3-glucuronide ratio algorithm. Environ Health Perspect 1996;104:408–13. doi:10.1289/ehp.96104408
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      2. Waller K ,
      3. Swan SH ,
      4. Windham GC , et al
      . Use of urine biomarkers to evaluate menstrual function in healthy premenopausal women. Am J Epidemiol 1998;147:1071–80. doi:10.1093/oxfordjournals.aje.a009401
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      2. Vollman RF
      . The menstrual cycle. Major Probl Obstet Gynecol 1977;7:1–193.
      OpenUrlPubMed
    25. ↵
      2. Schneider MB ,
      3. Bijur PE ,
      4. Fisher M , et al
      . Menstrual irregularity in female military cadets: comparison of data utilizing short-term and long-term recall. J Pediatr Adolesc Gynecol 2003;16:89–93. doi:10.1016/s1083-3188(03)00008-1
      OpenUrlCrossRefPubMed
    26. ↵
      2. De Souza MJ ,
      3. Toombs RJ ,
      4. Scheid JL , et al
      . High prevalence of subtle and severe menstrual disturbances in exercising women: confirmation using daily hormone measures. Hum Reprod 2010;25:491–503. doi:10.1093/humrep/dep411
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      2. Loucks AB
      . The response of luteinizing hormone pulsatility to 5 days of low energy availability disappears by 14 years of gynecological age. J Clin Endocrinol Metab 2006;91:3158–64. doi:10.1210/jc.2006-0570
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      2. Loucks AB ,
      3. Callister R
      . Induction and prevention of low-T3 syndrome in exercising women. Am J Physiol 1993;264:R924–30. doi:10.1152/ajpregu.1993.264.5.R924
      OpenUrl
    29. ↵
      2. Hamarsland H ,
      3. Paulsen G ,
      4. Solberg PA , et al
      . Depressed physical performance outlasts hormonal disturbances after military training. Med Sci Sports Exerc 2018;50:2076–84. doi:10.1249/MSS.0000000000001681
      OpenUrlPubMed
    30. ↵
      2. Nindl BC ,
      3. Barnes BR ,
      4. Alemany JA , et al
      . Physiological consequences of U.S. army ranger training. Med Sci Sports Exerc 2007;39:1380–7. doi:10.1249/MSS.0b013e318067e2f7
      OpenUrlCrossRefPubMedWeb of Science
    31. ↵
      2. Pasiakos SM ,
      3. Cao JJ ,
      4. Margolis LM , et al
      . Effects of high-protein diets on fat-free mass and muscle protein synthesis following weight loss: a randomized controlled trial. FASEB J 2013;27:3837–47. doi:10.1096/fj.13-230227
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      2. Longland TM ,
      3. Oikawa SY ,
      4. Mitchell CJ , et al
      . Higher compared with lower dietary protein during an energy deficit combined with intense exercise promotes greater lean mass gain and fat mass loss: a randomized trial. Am J Clin Nutr 2016;103:738–46. doi:10.3945/ajcn.115.119339
      OpenUrlAbstract/FREE Full Text
    33. ↵
      2. Knutson KL ,
      3. Spiegel K ,
      4. Penev P , et al
      . The metabolic consequences of sleep deprivation. Sleep Med Rev 2007;11:163–78. doi:10.1016/j.smrv.2007.01.002
      OpenUrlCrossRefPubMedWeb of Science
    34. ↵
      2. Ackerman KE ,
      3. Patel KT ,
      4. Guereca G , et al
      . Cortisol secretory parameters in young exercisers in relation to LH secretion and bone parameters. Clin Endocrinol (Oxf) 2013;78:114–9. doi:10.1111/j.1365-2265.2012.04458.x
      OpenUrlPubMed
    35. ↵
      2. Williams NI ,
      3. Berga SL ,
      4. Cameron JL
      . Synergism between psychosocial and metabolic stressors: impact on reproductive function in cynomolgus monkeys. Am J Physiol Endocrinol Metab 2007;293:E270–6. doi:10.1152/ajpendo.00108.2007
      OpenUrlCrossRefPubMedWeb of Science
    36. ↵
      2. Ritland BM ,
      3. Hughes JM ,
      4. Taylor KM , et al
      . Sleep health of incoming army trainees and how it changes during basic combat training. Sleep Health 2021;7:37–42. doi:10.1016/j.sleh.2020.10.005
      OpenUrl
    37. ↵
      2. Crowley SK ,
      3. Wilkinson LL ,
      4. Burroughs EL , et al
      . Sleep during basic combat training: a qualitative study. Mil Med 2012;177:823–8. doi:10.7205/milmed-d-12-00022
      OpenUrlPubMed
    38. ↵
      2. Miller NL ,
      3. Shattuck LG
      . Sleep patterns of young men and women enrolled at the united states military academy: results from year 1 of a 4-year longitudinal study. Sleep 2005;28:837–41. doi:10.1093/sleep/28.7.837
      OpenUrlPubMed
    39. ↵
      2. Miller NL ,
      3. Shattuck LG ,
      4. Matsangas P
      . Longitudinal study of sleep patterns of united states military academy cadets. Sleep 2010;33:1623–31. doi:10.1093/sleep/33.12.1623
      OpenUrlPubMedWeb of Science
    40. ↵
      2. Jackson S ,
      3. Agius R ,
      4. Bridger R , et al
      . Occupational stress and the outcome of basic military training. Occup Med (Lond) 2011;61:253–8. doi:10.1093/occmed/kqr036
      OpenUrlCrossRefPubMed
    41. ↵
      2. Wilhelm I ,
      3. Born J ,
      4. Kudielka BM , et al
      . Is the cortisol awakening rise a response to awakening? Psychoneuroendocrinology 2007;32:358–66. doi:10.1016/j.psyneuen.2007.01.008
      OpenUrlCrossRefPubMedWeb of Science
    42. ↵
      2. Tanskanen MM ,
      3. Kyröläinen H ,
      4. Uusitalo AL , et al
      . Serum sex hormone-binding globulin and cortisol concentrations are associated with overreaching during strenuous military training. J Strength Cond Res 2011;25:787–97. doi:10.1519/JSC.0b013e3181c1fa5d
      OpenUrlPubMed
    43. ↵
      2. Berga SL ,
      3. Daniels TL ,
      4. Giles DE
      . Women with functional hypothalamic amenorrhea but not other forms of anovulation display amplified cortisol concentrations. Fertil Steril 1997;67:1024–30. doi:10.1016/s0015-0282(97)81434-3
      OpenUrlCrossRefPubMedWeb of Science
    44. ↵
      2. Gallinelli A ,
      3. Matteo ML ,
      4. Volpe A , et al
      . Autonomic and neuroendocrine responses to stress in patients with functional hypothalamic secondary amenorrhea. Fertil Steril 2000;73:812–6. doi:10.1016/s0015-0282(99)00601-9
      OpenUrlCrossRefPubMedWeb of Science
    45. ↵
      2. Edwards VC ,
      3. Myers SD ,
      4. Wardle SL , et al
      . Nutrition and physical activity in british army officer cadet training part 2-daily distribution of energy and macronutrient intake. Int J Sport Nutr Exerc Metab 2022;32:204–13. doi:10.1123/ijsnem.2021-0191
      OpenUrl
    46. ↵
      2. Edwards VC ,
      3. Myers SD ,
      4. Wardle SL , et al
      . Nutrition and physical activity during british army officer cadet training: part 1-energy balance and energy availability. Int J Sport Nutr Exerc Metab 2022;32:195–203. doi:10.1123/ijsnem.2021-0190
      OpenUrl

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • X @kristypopp

    • Contributors KLP, MLB and JMH conceived of the presented idea and initiating the funding acquisition. KK, NS, BNB, MR, SPP, MLB, JMH, SAF and KLP designed the study methodology. BNB, MR, KIG, LAW, NZ, SAF and KLP carried out the investigation. KLP, JMH, MR, BNB and SAF supervised aspects of data collection. JMH and SAF contributed resources to the investigation as PIs of a larger prospective study. KK, JC, CMC, NS and KLP completed formal analyses and tables and figures for visualisation. KLP and JC wrote the original draft of this manuscript and KLP, BNB, JC, KIG, LAW, SPP, MR, KK, NZ, SAF, MLB, JMH and NS reviewed and/or edited the manuscript. KLP is the guarantor.

    • Funding This study was funded by the Military Operational Medicine Research Programme, US Army Medical Research and Materiel Command. This research was also partly supported by appointments to the Department of Defense Research Participation Programme at the US Army Research Institute of Environmental Medicine administered by the Oak Ridge Institute for Science and Education. The funders played no role in study design, execution, data interpretation or dissemination of results. The authors have no relevant financial or non-financial interests to disclose. The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the US Army or the US Department of Defense.

    • Competing interests None declared.

    • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.