yrhsbaactA

dtdlS

Cr

0h

Sickle Cell Trait Associated With Sudden Death inCompetitive Athletes

Kevin M. Harris, MDa, Tammy S. Haas, RNa, E. Randy Eichner, MDb, and Barry J. Maron, MDa,*

Sickle cell trait (SCT; hemoglobin AS) occurs in 8% of African Americans and althoughtypically benign has been associated with sudden death in military recruits during intensephysical activity. However, the role of SCT in the deaths of trained athletes is less welldocumented or acknowledged. The 31-year United States Sudden Death in AthletesRegistry was interrogated to determine the frequency, epidemiology, and clinical profile ofSCT-related death in a large population of competitive athletes. Of 2,462 athlete deaths, 23(0.9% overall, 3.3% of African Americans) occurred in association with SCT: ages 12 to 22years, 21 male (91%), and all African Americans. SCT diagnosis was made by solubilitytesting (n � 13) and/or hemoglobin electrophoresis (n � 16). Most victims competed incollege (n � 17) and in football (n � 19). Of 271 African American football deaths in theregistry, 7% (1 in 14) were known to be associated with SCT. Each athlete experienceddistinctive noninstantaneous collapse with gradual deterioration over several minutesassociated with vigorous or exhaustive physical exertion, usually during conditioning drills(n � 22) and typically early in the training season. Ambient temperatures were >80°F in20 patients (87%), with most events in southern or border states during the summer andautumn (n � 17 [74%]). In conclusion, SCT can be associated with largely unpredictablesudden collapse and death and apparent predilection for African American college footballplayers during conditioning. Understanding the risks, mechanisms, and event triggers ofSCT may allow lifesaving alterations in training methods to be implemented. © 2012

Published by Elsevier Inc. (Am J Cardiol 2012;110:1185–1188)

ci

M

nsos

pec

Sickle cell trait (SCT), in which a normal hemoglobingene and an abnormal mutated �-globin sickle gene areinherited, occurs in 8% of African Americans in the UnitedStates.1 Although typically a benign condition, SCT has nev-ertheless been occasionally associated with sudden collapseand death or other life-threatening events.2–5 One study 25ears ago associated SCT with deaths in basic military trainingecruits during vigorous exercise.2 Recently, anecdotal reportsave raised concern that SCT could be responsible for theudden deaths of young athletes during training,5 underscoredy the National Collegiate Athletic Association’s decision todopt mandatory SCT screening for Division I athletes.6 Inddition, preventive training modification guidelines and pre-autions for SCT-affected athletes are now recommended byhe National Collegiate Athletic Association and the Nationalthletic Trainers’ Association.7,8

Although there is increasing interest in potential risks posedby SCT in trained athletes,9,10 most available data are anec-otal and derived from case reports. Systematic reporting ofhe epidemiology, prevalence, and clinical presentation ofeaths associated with SCT in large athlete populations isacking. We interrogated the unique, long-standing Unitedtates Sudden Death in Athletes Registry11–14 for SCT-asso-

aThe Hypertrophic Cardiomyopathy Center, Minneapolis Heart Insti-tute Foundation at Abbott-Northwestern Hospital, Minneapolis, Minne-sota; and bUniversity of Oklahoma Health Sciences Center, Oklahoma

ity, Oklahoma. Manuscript received March 30, 2012; revised manuscripteceived and accepted June 2, 2012.

*Corresponding author: Tel: 612-863-3996; fax: 612-863-3875.

E-mail address: hcm.maron@mhif.org (B.J. Maron).

002-9149/12/$ – see front matter © 2012 Published by Elsevier Inc.ttp://dx.doi.org/10.1016/j.amjcard.2012.06.004

iated events, to investigate this rapidly emerging medicalssue.

ethods

The registry is a forensic database instituted at the Min-eapolis Heart Institute Foundation for the purpose of pro-pectively and retrospectively assembling data on the deathsf young athletes participating in organized competitiveports.11–14 Over a 31-year period (1980 to 2010), 2,462

such sudden deaths (and survivors of cardiac arrest) havebeen tabulated.

The registry population was identified by targeted searchesusing a variety of sources at the time each of these strategiesbecame available during the duration of the study, includingLexisNexis archival database. Athletes were included in theregistry if 2 criteria were met: (1) participation in organizedteam or individual sports requiring regular competition, andplacing a premium on excellence and achievement, and (2)sudden death at �39 years of age.11 A systematic trackingrocess was established to assemble detailed information onach case, including the complete autopsy report and pertinentlinical and demographic information.

Data are expressed as mean � SD. Proportions werecompared using chi-square or Fisher’s exact tests. Contin-uous variables were compared using unpaired Student’s ttests or Mann-Whitney rank-sum tests.

Results

There were 2,462 athlete deaths recorded in the registry

as of December 2010. Ages ranged from 8 to 39 years

www.ajconline.org

cdpn

Sorwc

bp

ntotvd(rea

p

dASi

tywpaomN

vaiti(rtaftu

D

b

globin

1186 The American Journal of Cardiology (www.ajconline.org)

(mean 18.7 � 6); 1,459 patients were white (59%), 34(1.4%) Asian, and 699 (28%) African American. The mostcommon causes of death were blunt trauma (n � 476),ommotio cordis (n � 98), and a variety of cardiovasculariseases (n � 1,396). The latter were most frequently hy-ertrophic cardiomyopathy (n � 375) and congenital coro-ary artery anomalies of wrong sinus origin (n � 131).

Notably, 23 of the deaths (0.9%) were in athletes withCT, documented by testing during life or at autopsy; eachf these athletes was African American (Table 1). Agesanged from 12 to 22 years (mean 18.6 � 2.6); 21 (91%)ere male. At the time of their deaths, these athletes were

ompeting in organized college (n � 17), high school (n �5), or junior high school (n � 1) sports.

The 23 athletes with SCT-related death represent 3.3% of699 African American athletes in the registry. SCT diagno-sis was made by solubility testing alone (n � 7, including 3y newborn or university screening), hemoglobin electro-horesis alone (n � 10, 5 antemortem), or both (n � 6).

All athletes died under similar distinctive circumstances:oninstantaneous collapse with gradual but rapid deteriora-ion (i.e., dyspnea, fatigue, weakness, and muscle cramping)ver an estimated 10 to 45 minutes, associated with exer-ional rhabdomyolysis in 11. Each event occurred duringigorous or exhaustive maximal physical exertion, usuallyuring training and conditioning (n � 22), or competitionn � 1), and often after brief sustained effort such asepetitive short-distance sprints or endurance running. SCTvents were most common early in the conditioning seasonnd in 6 cases were documented on the first day of practice.

Notably, SCT-deaths occurred predominately in football

Table 1Deaths associated with sickle cell trait in 23 competitive athletes

Patient Age (years)/Gender Sport Level State

1 12/M FB JHS Ohio2 14/F BB HS Texas3 15/M FB HS Florida4 15/M FB HS Texas5 15/M TR HS Florida6 18/M FB C Tennessee7 18/M FB C Florida8 18/M FB HS California9 18/M FB C Ohio

10 19/M FB C Texas11 19/M FB C Arkansas12 19/M FB C Missouri13 19/M FB C Florida14 20/F BB C Texas15 20/M FB C Mississippi16 20/M FB C Georgia17 20/M FB C North Carolina18 20/M FB C Indiana19 21/M BB C Florida20 21/M FB C New Mexico21 21/M FB C Mississippi22 21/M FB C Arizona23 22/M FB C North Carolina

BB � basketball; C � college; F � female; FB � football; HS � highTR � track (sprinter); � � present; 0 � absent.

* Usually defined as bands present for hemoglobin A and S, with hemo

layers (n � 19 of 23 [83%]), constituting 2.6% of all 723 i

eaths occurring in that sport and 7.0% of deaths in the 271frican American football players (i.e., 1 of 14). OtherCT-deaths occurred in basketball (n � 3) and track sprint-

ng (n � 1).SCT deaths were similar to other cardiovascular deaths in

he registry with respect to age (mean 18.6 � 2.6 vs 17.6 � 5.0ears, p � 0.092) and gender (91% vs 88% male, p � 1.00) butere more common in African Americans (100% vs 37%,�0.001). Deaths were most frequent in the summer and early

utumn months of June to October (17 of 23 [74%]). Twentyf the 23 deaths (87%) occurred in southern or border states,ost commonly Florida (n � 5), Texas (n � 4), Mississippi,orth Carolina, and Ohio (n � 2 each).Potential associated triggers for SCT collapse were en-

ironmental temperature (�80°F) in 20 (87%) and highltitude in 2 (Albuquerque, New Mexico, and Tucson, Ar-zona), as well as maximum intensity of exercise. Four ofhe 23 athletes had associated structural cardiac abnormal-ties identified at autopsy: 2 with tunneled coronary arteries1 left anterior descending and 1 right coronary), 1 withedundant mitral valve and features consistent with hyper-rophic cardiomyopathy, and 1 with 80% right coronaryrtery narrowing by atherosclerosis. Heart weights rangedrom 225 to 640 g (mean 423); 5 were �500 g. Largely onhe basis of the circumstances of collapse, these features arenlikely to be the primary or a major cause of death.

iscussion

From our forensic-based registry study, we have assem-led the first sizable series of competitive athletes (n � 23)

T Solubility Test Hemoglobin Electrophoresis* Rhabdomyolysis

� NA �NA � 0� � 0� NA 0� NA 0

NA � NA� NA 0� NA 0

NA � �� � �

NA � �NA � 0� � �

NA � 0� � 0� NA �� � 0

NA � �NA � 0� � �

NA � �� NA �

NA � �

; JHS � junior high school; M � male; NA � no information available;

S 35% to 45%.

SC

school

n whom SCT was associated with otherwise unexplained

n2lwphsocin

ecedhadpcaca

iS

eepspstgtccmwttpp

dt

1187Miscellaneous/Sickle Cell Trait and Sudden Death

sudden and unexpected collapse and death. These athletesconstituted 0.9% of our large registry of almost 2,500 sud-den deaths, with a predominance in college football players.

The outcomes in these athletes can most reasonably belinked to the presence of SCT and its unpredictable conse-quences,15 particularly with no reasonable alternative expla-ation or disease to explain demise. Clinical profiles in the3 SCT athletes were similar and distinctive. Cardiovascu-ar collapse followed several minutes of gradual clinicalorsening with dyspnea, fatigue, and weakness (sometimesermitting verbalization) during and after vigorous or ex-austive physical activity, often early in the workout sea-on.5,16 Such scenarios differ distinctly from sudden deathccurring from ventricular tachyarrhythmias due to structuralardiovascular diseases (e.g., hypertrophic cardiomyopathy,on channelopathies) in which collapse is virtually instanta-eous.

Although the pathophysiology and determinants of SCTvents are not fully understood, it is likely that cardiovas-ular collapse occurs under conditions that increase risk forxertional sickling of red blood cells, including heat stress,ehydration, illness, and altitude.1–3,17,18 It is reasonable toypothesize that a cascade of events ensues with intramuscularnd microvascular sickling leading to vascular occlusion, en-othelial damage, and impaired blood flow to muscles, whichromotes ischemic rhabdomyolysis and disseminated intravas-ular coagulation. Rhabdomyolysis leads to hyperkalemia andcidosis, which in the setting of hypoxia adversely affectsardiac function and lowers the threshold for a lethal cardiacrrhythmia.1,3,17–21 Indeed, although rhabdomyolysis was

present in almost 50% of our SCT athletes, we found only5 other athletes in the registry in whom rhabdomyolysis wasthe sole cause of death, all of whom were unassociated withSCT. Each was white, 17 to 28 years old, and died duringrunning sports (exclusive of football).

However, the finding of widespread sickling in the heartand other organs at autopsy does not represent unequivocalevidence for this mechanism of death because of the obsta-cle in distinguishing histologic postmortem sickling (whichcan be caused by the diminished oxygen environment afterdeath) from antemortem sickling.1,4 This unavoidable lim-tation contributes to the debate over the pathogenesis ofCT and its relation to sudden death.

Notably, the epidemiology and pathophysiology of SCTvents have vital implications and provide clues that cannhance prevention with targeted measures and tailoredrecautions during training. For example, SCT eventshowed a clear predilection for male athletes during footballractice and were often associated with brief bursts ofustained maximal exertion. These considerations have ledo specific recommendations for athletes with known SCT:radual conditioning at the beginning of the season, atten-ion to modifying pace and providing periods of rest duringonditioning drills, adequate hydration, and a high index ofaution for immediate cessation of physical activity withuscle cramping, fatigue, and excessive dyspnea. Collapseith SCT is a medical emergency requiring the administra-

ion of oxygen, intravenous hydration, and cooling to pro-ect against rhabdomyolysis. Although cardiac arrest is aossible sequela, and an external defibrillator should be in

lace,1,5,7 its effectiveness in this setting is unpredictable.22

There has been some skepticism and substantial emotionsurrounding the legitimacy of SCT as a causative factor forcardiovascular sudden death. However, our data support theprinciple that SCT may well be primarily linked to suchuncommon events with vigorous exercise in young physi-cally active individuals, often associated with elevated en-vironmental temperatures as a trigger.

SCT should be included among the myriad of nontrau-matic risks of sports participation.9–13 However, in thisanalysis we were unable to document the overall associationof SCT and sudden death in the registry or to show that suchevents are more (or less) common compared to athleteswithout SCT. This is an unavoidable limitation, becauselaboratory-based SCT diagnoses were not available to theregistry for all athletes, given confidentiality restrictions.However, it is particularly notable that 7% of 271 AfricanAmerican football players in the registry died of SCT,which approximates the 8% of African American footballplayers who would be estimated to carry the trait in thispopulation.15

Controversy surrounding SCT screening in athlete pop-ulations is a complex and sometimes polarizing ethical andlegal debate.4,6,23 Screening proponents emphasize the po-tential for prevention of death by prospective recognition ofSCT-affected individuals. Others regard SCT testing as po-tentially discriminatory practice used for selective targeting,and an infringement on individual privacy and liberty. Nev-ertheless, in 2010, the National Collegiate Athletic Associ-ation mandated SCT screening for all Division I studentathletes as part of a legal settlement concerning an athletedeath due to SCT.6 Recently, hematologists rejected man-atory SCT screening, instead recommending universalraining interventions.24 We take neither side of this con-

troversy and have confined ourselves to reporting a substan-tial series of athlete deaths associated with (and apparentlydue to) SCT.

Our data also underscore the unpredictability of SCTevents. With an 8% prevalence in African Americans, wewould expect that thousands of individuals with SCT havecompeted safely during vigorous and competitive sports, butonly a small fraction appear susceptible to catastrophe fromthis genetic defect, which has been generally regarded as be-nign. This observation raises the possibility of a SCT subset atparticular risk because of currently undefined and unpredict-able genetic, physiological, or environmental factors.

Acknowledgment: We appreciate the statistical assistanceof Ross F. Garberich, MSc, of the Minneapolis Heart Insti-tute Foundation.

1. Tsaras G, Owusu-Ansah A, Boateng FO, Amaoteng-Adjepong Y.Complications associated with sickle-cell trait: a brief narrative re-view. Am J Med 2009;122:507–512.

2. Kark JA, Posey DM, Schumacher HR, Ruehle CJ. Sickle-cell trait asa risk factor for sudden death in physical training. N Engl J Med1987;317:781–787.

3. Jones SR, Binder RA, Donowho EM. Sudden death in sickle-cell trait.N Engl J Med 1970;282:323–325.

4. Mitchell BL. Sickle cell trait and sudden death—bringing it home.J Natl Med Assoc 2007;99:300–305.

5. Eichner ER. Sickle cell trait in sports. Curr Sports Med Rep 2010;9:

347–351.

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

1188 The American Journal of Cardiology (www.ajconline.org)

6. Bonham VL, Dover GJ, Brody LC. Screening student athletes forsickle cell trait—a social and clinical experiment. N Engl J Med2010;363:997–999.

7. National Athletic Trainers’ Association. The National Athletic Train-ers’ Association (NATA) releases “Sickle Cell Trait and the Athlete”consensus statement. Available at: http://www.nata.org/NR062107.Accessed April 17, 2012.

8. Klossner D. 2009-10 NCAA sports medicine handbook. Available at:http://www.ncaapublications.com/productdownloads/MD10.pdf. Ac-cessed May 3, 2011.

9. Thompson PD. The cardiovascular complications of vigorous physicalactivity. Arch Intern Med 1996;156:2297–2302.

0. Van Camp SP, Bloor CM, Mueller FO, Cantu RC, Olson HG. Non-traumatic sports death in high school and college athletes. Med SciSports Exerc 1995;27:641–647.

1. Maron BJ, Doerer JJ, Haas TS, Tierney DM, Mueller FO. Suddendeaths in young competitive athletes: analysis of 1866 deaths in theU.S. 1980-2006. Circulation 2009;119:1085–1092.

2. Maron BJ, Shirani J, Poliac LC, Mathenge R, Roberts WC, MuellerFO. Sudden death in young competitive athletes: clinical, demographicand pathologic profiles. JAMA 1996;276:199–204.

3. Maron BJ. Sudden death in young athletes. N Engl J Med 2003;349:1065–1075.

4. Maron BJ, Carney KP, Lever HM, Lewis JF, Barac I, Casey SA, SherridMV. Relationship of race to sudden cardiac death in competitive athleteswith hypertrophic cardiomyopathy. J Am Coll Cardiol 2003;41:974–980.

5. Harmon KG, Drezner JA, Klossner D, Asif IM. Sickle cell traitassociated with a RR of death of 37 times in National Collegiate

Athletic Association football athletes: a database with 2 million ath-lete-years as the denominator. Br J Sports Med 2012;46:325–330.

6. Rosenthal MA, Parker DJ. Collapse of a young athlete. Ann EmergMed 1992;21:1493–1498.

7. Loosemore M, Walsh SB, Morris E, Stewart G, Porter JB, Montgom-ery H. Sudden exertional death in sickle cell trait. Br J Sports Med2012;46:312–314.

8. Rodgers GP. Sickle-cell trait and physical training: evidence for im-proved fitness. Arch Intern Med 1988;148:1019–1020.

9. Ramirez A, Hartley LH, Rhodes D, et al. Morphological feature of redblood cells in subjects with sickle cell trait: changes during exercise.Arch Intern Med 1976;136:1064–1066.

0. Martin TW, Weisman IM, Zeballos RJ, Stephenson SR. Exercise andhypoxia increase sickling in venous blood from an exercising limb inindividuals with sickle cell trait. Am J Med 1989;87:48–56.

1. Phillips M, Robinowitz M, Higgins JR, Boran KJ, Reed T, Virmani R.Sudden cardiac death in Air Force recruits. A 20-year review. JAMA1986;256:2696–2699.

2. Soar J, Perkins GD, Abbas G, Alfonzo A, Barelli A, Bierens JJ,Brugger H, Deakin CD, Dunning J, Georgiou M, Handley AJ, LockeyDJ, Paal P, Sandroni C, Thies KC, Zideman DA, Nolan JP. EuropeanResuscitation Council Guidelines for Resuscitation 2010 Section 8.Cardiac arrest in special circumstances: electrolyte abnormalities, poi-soning, drowning, accidental hypothermia, hyperthermia, asthma, ana-phylaxis, cardiac surgery, trauma, pregnancy, electrocution. Resusci-tation 2010;81:1400–1433.

3. Steinberg MH. In the clinic: sickle cell disease. Ann Intern Med2011;155:ITC31–ITC315.

4. American Society of Hematology. ASH policy opposes mandatorysickle cell trait screening for athletic participation, recommends uni-

versal training interventions & additional research. Available at: http://www.hematology.org/news/2012/7703.aspx. Accessed April 17, 2012.