Sickle cell trait (SCT) is a hemoglobinopathy that results from inheriting one copy of the normal HbA gene and one copy of the HbS variant. Although a benign carrier state in most cases (1 2–4 5–7 8 2
The survival advantage conferred by hemoglobin S (Hgb S) during infection with Plasmodium falciparum malaria is well established (9 10,11 falciparum malaria are less common (12 13,14 10
Aerobic and anaerobic exercise discrepancies between those with and without SCT have been reported in several small trials, but their results are contradictory (15 16,17 anaerobic fitness .
METHODS
All recruits who entered US Air Force Basic Military Training (BMT) (Joint Base San Antonio–Lackland, Texas) between January 2009 and December 2014 were included in this retrospective cohort study. Per Air Force policy, recruits were universally screened for SCT with a sickle solubility test within 5 d of arrival; aliquots screening positive were automatically referred for hemoglobin electrophoresis testing, which provides a complete hemoglobin percentage profile. During the first and last week of training—usually separated by 6 wk—recruits completed the US Air Force Fitness Assessment, which includes a 1.5-mile timed run, 1 min of push-ups, 1 min of sit-ups, and body composition measurements (height, weight, and abdominal circumference). To pass the initial or final assessment, recruits were required to pass each component and achieve a minimum total score, based on age- and sex-specific cutoffs (18
We queried the Trainee Health Squadron’s SCT database for hemoglobinopathy laboratory results and merged these with data recorded in the US Air Force Basic Training Management System (BTMS): age, sex, fitness assessment dates and results, and dates of training entry and exit (i.e., by graduation or separation). Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared. Run time was recorded in seconds, and push-ups and sit-ups were recorded as counts. Because of manual data entry into BTMS, fitness assessment data contain occasional errors; implausible results (run time <420 or >1800 s, push-ups and sit-ups >100, and abdominal circumference <19 or >42 inches) were excluded. Time-sensitive variables of age and BMI were based on the start date of training. Rarely, recruits are removed from training during the first week and have separation dates but missing entry dates. In these cases we calculated the entry date by subtracting 14 d from the separation date (to account for the usual processing time between removal from training and separation) and rounding to the previous Monday. Self-reported race was obtained from the US Air Force Personnel Center database. Individuals reporting more than one race were categorized as “multiracial,” and those declining to report were grouped with those who had missing race information.
Because weather can affect physical performance (19 am from March through October and at 4:00 pm from November through February, we selected the minimum daily temperature for the former and maximum daily temperature for the latter. These fitness assessment times correspond closely to the historical minimum and maximum daily temperatures in the San Antonio metropolitan area. Given the lack of date information for the final fitness assessment, the results could not be adjusted for ambient temperature during the final assessment.
We used summary statistics to describe the population and two-sided t tests to compare fitness parameters between those with and without SCT, both overall and stratified by sex. We conducted multivariate linear regression—adjusting for age, sex, race, BMI, and ambient temperature—to assess the association between SCT (as a binary variable) and Hgb S percentage (as a categorical variable) on each component of the initial and final fitness assessment and change from initial to final. To elucidate the relationship between race, SCT, and fitness, we also conducted multivariate linear regression after stratification for race and SCT status. Robust variance–covariance estimations were used to determine the standard error (SE) of the β coefficients, and the coefficient of determination (R 2 ) was calculated to evaluate the relationship between Hgb S percentage and fitness results. We used multivariate logistic regression, adjusting for the same variables, to calculate adjusted odds ratios (aOR) and 95% confidence intervals (CI) of failing the initial and final fitness assessment, and of changing from a failing to passing score, based on SCT status and Hgb S percentage. After testing various quantile cutpoints for the Hgb S percentage for linear and logistic regressions, we chose to establish natural cutpoints, resulting in the categories of 20% to 29.99%, 30% to 39.99%, and ≥40%. To aid interpretation, run times were converted from seconds to minutes:seconds.
All analyses were performed with Stata/IC v14.2 (StataCorp, College Station, Texas). This study was approved by the 59th Medical Wing Institutional Review Board at Joint Base San Antonio-Lackland, TX.
RESULTS
Of all recruits (N = 210,461) who entered training during the 6-yr surveillance period, the mean (standard deviation [SD]) age was 20.6 yr (3.0 yr) and BMI was 23.7 kg·m−2 (2.7 kg·m−2 ). The majority (78.9%) were male. By self-reported race, most were white (72.6%), followed by black (15.9%), multiracial (4.4%), and Asian (3.0%). Of the 2161 (1.0%) recruits positive for SCT, mean (SD) Hgb S percentage was 38.0% (3.1), with an interquartile range of 35.6% to 40.1% and an overall range of 20.1% to 44.7%. The majority of recruits with SCT were black (84.2%), followed by white (7.6%) and multiracial (5.6%) (Table 1 ). On their initial fitness assessment, the total recruit population completed the 1.5-mile run in a mean (SD) of 13:20 (2:19) and completed 31.7 (14.4) push-ups and 35.1 (11.4) sit-ups. Results improved to 11:09 (1:22), 48.7 (13.9) push-ups, and 53.1 (8.8) sit-ups on the final assessment.
TABLE 1: US Air Force recruit population characteristics, 2009–2014 (N = 210,461).
In the unadjusted model, recruits with SCT had slower initial and final run times, but made greater improvements from initial to final than their peers without SCT—both overall and sex-stratified (P < 0.05 for all). Males with SCT completed more push-ups and sit-ups than their counterparts without SCT on both the initial and final assessments, whereas females with SCT completed fewer push-ups than their peers on the initial assessment (P < 0.05 for all) (Table 1 ).
After adjusting for age, sex, race, BMI, and approximate ambient temperature while conducting the fitness assessment, recruits with SCT were slower on their initial run than their peers without SCT by a mean (SE) of 9.7 s (2.6 s) (P < 0.001) and completed 0.6 (0.2) fewer push-ups (P < 0.05); baseline sit-up completion was similar between those with and without SCT. Recruits with SCT improved their run times by a margin of 4.2 s (2.1 s) over their peers (P < 0.05), whereas change in push-ups and sit-ups between those with and without SCT was statistically equivalent (Table 2 ).
TABLE 2: Physical fitness performance by sickle cell trait status and Hgb S percentage among US Air Force recruits, 2009–2014 (N = 210,461).
On baseline testing, increased percentages of Hgb S were modestly correlated with faster run times (R 2 = 0.375) and the completion of fewer push-ups (R 2 = 0.338) and sit-ups (R 2 = 0.141), after adjusting for age, sex, race, BMI, and ambient temperature. Correlations between Hgb S percentage and change in run time, push-ups, or sit-ups over the course of training were weak, with an R 2 < 0.10 for all (Table 2 ).
Recruits with SCT were 13% more likely to fail their initial fitness assessment (aOR, 1.13; 95% CI, 1.01–1.27]) but nonsignificantly less likely to fail their final assessment (aOR, 0.92; 95% CI, 0.66–1.27). Progression from a failing score to a passing score was nonsignificantly higher among those with SCT (aOR, 1.05; 95% CI, 0.86–1.28). No Hgb S percentage stratum was statistically associated with failing the initial or final fitness assessment, or with the likelihood of improving from a failing score to a passing score (Table 3 ).
TABLE 3: Risk of fitness assessment failure by sickle cell trait status and Hgb S percentage among US Air Force recruits, 2009–2014 (N = 210,461).
With race removed from the model (i.e., adjusting only for age, sex, BMI, and ambient temperature), most associations reverted toward the unadjusted results. Notably, in the non–race-adjusted model, recruits with SCT had slower run times (by 23.7 [2.5] s) and completed more push-ups (0.7 [0.2]) and sit-ups (1.2 [0.2]) than their peers on the initial assessment (P < 0.01 for all). White recruits without SCT had faster initial run times than white recruits with SCT (by 29.8 [8.1] s; P < 0.001), black recruits without SCT (by 22.1 s [0.7 s]; P < 0.001) and with SCT (by 28.3 s [2.7 s]; P < 0.001), and recruits of other races without SCT (by 12.2 s [0.8 s]; P < 0.001) and with SCT (by 30.1 s [9.9 s]; P < 0.01). White recruits without SCT completed a statistically equivalent amount of push-ups and sit-ups as white recruits with SCT on initial testing, but they completed fewer push-ups and sit-ups as black recruits without SCT (1.9 [0.1] push-ups and 2.2 [0.1] sit-ups; P < 0.001 for both exercises) and black recruits with SCT (1.3 [0.3] push-ups and 1.7 [0.2] sit-ups; P < 0.001 for both exercises) (Table 4 ).
TABLE 4: Physical fitness performance by sickle cell trait status and race among US Air Force recruits, 2009–2014 (N = 210,461).
DISCUSSION
In this population of over 210,000 recruits who entered US Air Force BMT between 2009 and 2014, those with SCT had slower 1.5-mile run times and completed more push-ups and sit-ups on baseline fitness testing. However, these associations changed considerably after adjustment for age, sex, race, BMI, and ambient temperature. The difference in run time was reduced by half, and sit-up completion was inverted. All post-adjustment differences were small: recruits with SCT were 9.4 s slower on the run and completed 0.5 and 0.3 fewer push-ups and sit-ups, respectively. The run time discrepancy was further reduced by the end of the training course, approximately 6 wk thereafter. Although recruits with SCT were 13% more likely to fail their initial fitness assessment, failure on the final fitness assessment was statistically equivalent between those with and without SCT. Hgb S percentage only modestly correlated to baseline fitness performance and negligibly correlated to change in performance over time.
The small but statistically significant difference in baseline exercise capacity between recruits with and without SCT in this large observational study may help clarify the contradictory findings from multiple small trials, mostly involving controlled laboratory exercise. Several of these studies have suggested improved anaerobic performance in SCT carriers: Bilé and colleagues (20 21 22 23 24,25
In the present study, SCT carriers initially completed 0.7 more push-ups and 1.2 more sit-ups—anaerobic activities consisting of maximal effort exercise for 60 s—as compared with their colleagues with normal hemoglobin, after adjusting for age, sex, BMI, and ambient temperature. However, these findings inverted after the additional adjustment for race, suggesting that race is a qualitative confounder in the relationship between SCT and anaerobic fitness performance. Prior military studies have found that black recruits (26 27 Table 2 ), black recruits with and without SCT and recruits of other races without SCT completed more push-ups and sit-ups than white recruits without SCT (Table 4 ). This suggests that the race-independent presence of SCT is not responsible for the anaerobic performance discrepancies observed in this large population. Although an evolutionary hypothesis linking the HbS gene to fast-twitch muscle predominance is not supported by these findings, the interaction between race, hemoglobin concentration, and exercise metabolism should be studied further.
The slower baseline 1.5-mile run time among those with SCT in this population, which shifted toward the null but persisted after adjustment for race, may reflect true physiologic differences associated with SCT, such as reduced cardiopulmonary or ventilatory capability (e.g., heart rate and maximal oxygen uptake). This finding is consistent with some (28–30 21,23 31
Stratifying SCT recruits by Hgb S percentage, as opposed to keeping them in one group, only slightly changed β coefficients and OR, and all R 2 values were small or negligible. This corresponds to studies in which the concomitant presence of alpha-thalassemia, which generally reduces Hgb S percentage (17 29,30
This study is not without limitations. First, because the date of the final fitness assessment is not recorded in BTMS, analyses could not be adjusted for ambient temperature during the final assessment. Because full acclimatization should have occurred by the time of the final assessment; however, it is less important to account for this variable. Second, because approximately 6% of recruits are discharged over the course of the training program, not all recruits completed a final fitness assessment. These two limitations only affect results associated with the change from initial to final. Third, 1.7% of the population had no race information, either due to a missing value in the Air Force demographic database or declining to report. Because the SCT prevalence in those without race information (0.8%) was similar to the total population (1.0%); however, it is unlikely that this substantively biased the results. Fourth, the study examines a relatively young and healthy cohort of Americans who passed a baseline medical examination at a Military Entrance Processing Station; its findings may not be generalizable to the broader population or to the student-athlete subpopulation.
A final consideration is the use of maximal push-up and sit-up completion in 1 min to estimate anaerobic capacity. Although these exercises are primarily anaerobic endeavors, the proportion of energy system used (i.e., phosphagen, anaerobic, and aerobic) varies between persons based on exercise capacity, including upper-extremity muscular strength, lumbopelvic core muscular strength, muscular endurance, body weight and composition, and V˙O2max (32 26
The presence of SCT, after adjustment for known confounding variables, was associated with slightly inferior aerobic fitness and largely similar anaerobic fitness upon entry to BMT. These gaps, which were small at baseline, further narrowed over 6 wk of training. Stratifying those with SCT by their electrophoresis-derived Hgb S percentage did not noticeably change the magnitude of association on initial testing. Likewise, there was no clear relationship between Hgb S percentage and degree of aerobic or anaerobic improvement over 6 wk of training. Recruits with SCT—irrespective of Hgb S percentage—were more likely to fail their initial fitness assessment but not their final assessment, as compared with their counterparts without SCT. Policymakers should be aware of the small but increased risk of exertional rhabdomyolysis and death with SCT and the ethical considerations associated with screening for SCT (33
The authors thank Sandy Kawano, technical editor, Aeromedical Research Department, US Air Force School of Aerospace Medicine, for her assistance editing this article.
The authors declare no financial conflicts of interest. The results of the present study do not constitute endorsement by ACSM. The results of the present study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. The views expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Air Force, the Department of Defense, or the US Government.
REFERENCES
1. Castro O, Rana SR, Bang KM, Scott RB. Age and prevalence of sickle-cell trait in a large ambulatory population.
Genet Epidemiol . 1987;4(4):307–11.
2. Webber BJ, Witkop CT. Screening for sickle-cell trait at accession to the United States military.
Mil Med . 2014;179(11):1184–9.
3. Tsaras G, Owusu-Ansah A, Boateng FO, Amoateng-Adjepong Y. Complications associated with sickle cell trait: a brief narrative review.
Am J Med . 2009;122(6):507–12.
4. Nelson DA, Deuster PA, Carter R 3rd, Hill OT, Wolcott VL, Kurina LM. Sickle cell trait, rhabdomyolysis, and mortality among U.S. Army soldiers.
N Engl J Med . 2016;375(5):435–42.
5. Kark JA, Posey DM, Schumacher HR, Ruehle CJ. Sickle-cell trait as a risk factor for sudden death in physical training.
N Engl J Med . 1987;317(13):781–7.
6. Harmon KG, Drezner JA, Klossner D, Asif IM. Sickle cell trait associated with a RR of death of 37 times in National Collegiate Athletic Association football athletes: a database with 2 million athlete-years as the denominator.
Br J Sports Med . 2012;46(5):325–30.
7. Loosemore M, Walsh SB, Morris E, Stewart G, Porter JB, Montgomery H. Sudden exertional death in sickle cell trait.
Br J Sports Med . 2012;46(5):312–4.
8. O’Connor FG, Bergeron MF, Cantrell J, et al. ACSM and CHAMP summit on sickle cell trait: mitigating risks for warfighters and athletes.
Med Sci Sports Exerc . 2012;44(11):2045–56.
9. Taylor SM, Parobek CM, Fairhurst RM. Haemoglobinopathies and the clinical epidemiology of malaria: a systematic review and meta-analysis.
Lancet Infect Dis . 2012;12(6):457–68.
10. Irving R, Charlton V, Morrison E, Facey A, Buchanan O. Demographic characteristics of world class Jamaican sprinters.
Scientific World Journal . 2013;670217.
11. Scott RA, Irving R, Irwin L, et al. ACTN3 and ACE genotypes in elite Jamaican and US sprinters.
Med Sci Sports Exerc . 2010;42(1):107–12.
12. Piel FB, Patil AP, Howes RE, et al. Global distribution of the sickle cell gene and geographic confirmation of the malaria hypothesis.
Nat Commun . 2010;1:104.
13. Pitsiladis Y, Wang G, Wolfarth B, et al. Genomics of elite sporting performance: what little we know and necessary advances.
Br J Sports Med . 2013;47(9):550–5.
14. Tucker R, Santos-Concejero J, Collins M. The genetic basis for elite running performance.
Br J Sports Med . 2013;47(9):545–9.
15. Connes P, Reid H, Hardy-Dessources MD, Morrison E, Hue O. Physiological responses of sickle cell trait carriers during exercise.
Sports Med . 2008;38(11):931–46.
16. Le Gallais D, Prefaut C, Mercier J, Bile A, Bogui P, Lonsdorfer J. Sickle cell trait as a limiting factor for high-level performance in a semi-marathon.
Int J Sports Med . 1994;15(7):399–402.
17. Steinberg MH, Embury SH. Alpha-thalassemia in blacks: genetic and clinical aspects and interactions with the sickle hemoglobin gene.
Blood . 1986;68(5):985–90.
18. U.S. Air Force.
Fitness Program . Washington (DC): Department of the Air Force; 2013. Air Force Instruction 36–2905; [cited 2017 Aug 23]. Available from:
http://static.e-publishing.af.mil/production/1/af_a1/publication/afi36-2905/afi36-2905.pdf .
19. Racinais S, Alonso JM, Coutts AJ, et al. Consensus recommendations on training and competing in the heat.
Sports Med . 2015;45(7):925–38.
20. Bilé A, Le Gallais D, Mercier B, et al. Blood lactate percentages during incremental exercise in subjects with sickle cell trait.
Med Sci Sports Exerc . 1998;30(5):649–54.
21. Bilé A, Le Gallais D, Mercier J, Boqui P, Préfaut C. Sickle cell trait in Ivory Coast athletic throw and jump champions, 1956–1995.
Int J Sports Med . 1998;19(3):215–9.
22. Sara F, Hardy-Dessources MD, Voltaire B, Etienne-Julan M, Hue O. Lactic response in sickle cell trait carriers in comparison with subjects with normal hemoglobin.
Clin J Sport Med . 2003;13(2):96–101.
23. Marlin L, Sara F, Antoine-Jonville S, Connes P, Etienne-Julan M, Hue O. Ventilatory and lactic thresholds in subjects with sickle cell trait.
Int J Sports Med . 2007;28(11):916–20.
24. Samb A, Kane MO, Ba A, et al. Physical performance and thermoregulatory study of subjects with sickle cell trait during a sub-maximal exercise [Article in French].
Dakar Med . 2005;50(2):46–51.
25. Bilé A, Le Gallais D, Mercier B, Martinez P, Ahmaidi S, Préfaut C. Anaerobic exercise components during the force-velocity test in sickle cell trait.
Int J Sports Med . 1996;17(4):254–8.
26. Knapik JJ, Sharp MA, Jones BH, et al. Secular trends in the physical fitness of American youth, young adults, and Army recruits. Aberdeen Proving Ground, MD: U.S. Army Center for Health Promotion and Preventive Medicine, 2004. Report No.: 12-HF-01Q9B-04.
27. Fitzgerald PI, Vogel JA, Daniels WL, et al. The body composition project: a summary report and descriptive data. Natick, MA: U.S. Army Research Institute of Environmental Medicine, 1986. Report No.: USARIEM-T 5/87.
28. Connes P, Rachinais S, Sra F, et al. Does the pattern of repeated sprint ability differ between sickle cell trait carriers and healthy subjects.
Int J Sports Med . 2006;27(12):937–42.
29. Connes P, Monchanin G, Perrey S, et al. Oxygen uptake kinetics during submaximal exercise: effect of sickle cell trait with or without alpha-thalassemia.
Int J Sports Med . 2006;27(7):517–25.
30. Vincent L, Féasson L, Oyono-Enquéllé S, et al. Skeletal muscle structural and energetic characteristics in subjects with sickle cell trait, alpha-thalassemia, or dual hemoglobinopathy.
J Appl Physiol (1985) . 2010;109(3):728–34.
31. Connes P, Hue O, Tripette J. Hardy-Dessources. Blood rheology abnormalities and vascular cell adhesion mechanisms in sickle cell trait carriers during exercise.
Clin Hemorheol Microcirc . 2008;39(1–4):179–84.
32. Mann T, Lamberts RP, Lambert MI. Methods of prescribing relative exercise intensity: physiological and practical considerations.
Sports Med . 2013;43(7):613–25.
33. Webber BJ, Casa DJ, Beutler AI, Nye NS, Trueblood WE, O’Connor FG. Preventing exertional death in military trainees: recommendations and treatment algorithms from a multidisciplinary working group.
Mil Med . 2016;181(4):311–8.
