RightRep EN · PT Open the app
BLOG

Can You Train With Sickle Cell Disease or Trait? The Hemoglobin Audit: The 7.74-Million File (Up 41.4% Since 2000, 515,000 Born in 2021, 376,000 Excess Deaths a Year) and the Audit That Inverts Twice — for the ~300 Million Trait Carriers: Kark's 2-Million-Recruit File With Sudden Unexplained Death 32.2 vs 1.2 per 100,000 at RR 27.6, Boden's Gridiron File With Sickling Behind 12.1% of Nontraumatic Football Fatalities Until the 2010 NCAA Bylaw Cut It to 0.12×, Nelson's 47,944-Soldier Cohort With Exertional Rhabdomyolysis HR 1.54 While Mortality Sits Flat at 0.99, and the ASH-Convened 2025 Verdict That the Trait Does Not Cause Crises; for the 7.74 Million Patients: Liem's CPET at VO₂ 26.9 vs 37.0, Ogunsile's Adults at 16.1 vs 42.7, Melo's Brazilian Kids Already Moving Less at 19.2 vs 27.1 min/day — and EXDRE, the RCT That Flipped the Default: 8 Weeks of Moderate Work +7.2 W With ZERO Training-Group Hospitalizations Against 5 in the Sedentary Controls — Plus the Brazil File That Makes This Structural (1:650 Bahia Births, Trait 1:17, 60,017 Living, Median Age at Death 32 vs 69) and the PCDT 2024 Dose: Breaks Every 20 Minutes, 50–90% HRmax, Stop at the First Sign of Fatigue

By RightRep AI Assistant · · 21 min read

There are two fences at the top of this post, because this is one subject wearing two very different programs. If you have sickle cell disease, training is a hematologist partnership: every program decision — starting load, progression, when to pause — is made with your care team, and Brazil's own clinical protocol (the PCDT 2024) says it outright: physical activity can benefit patients, but under "criterious evaluation" by the team [24]. A pain crisis in the last 24–48 hours, a fever, or new pain means the session is off and the team hears about it (when to train sick has the general file). If you carry the trait — one gene copy, essentially healthy blood at rest — your program is an ordinary one with four preparations: status known, progression built, hydration engineered, heat respected. And for both populations, one fence shares the same words: an athlete who collapses mid-effort with cramping pain, weakness, or struggling posture is a stop-and-treat emergency — stop, cool, hydrate, emergency care — never "walk it off" [11]. With those fences standing, the file underneath is the strangest dose-response curve this library has audited: it bends in two places, in opposite directions, for two populations carrying the same mutation.

The short answer: yes — twice, in two different doses

Strip the file to its spine and five findings carry it. For trait carriers, the risk is real, small, and preparation-shaped: among 2 million army recruits, sudden unexplained death ran 32.2 vs 1.2 per 100,000 with vs without the trait (RR 27.6) in the era of day-one maximal effort [7]; across 20 years of football, exertional sickling stood behind 12.1% of nontraumatic fatalities — until the 2010 NCAA screening-and-preparation bylaw cut that risk to 0.12× [8]; and in the modern precision cohort of 47,944 soldiers, the trait raised exertional rhabdomyolysis HR 1.54 while overall mortality sat indistinguishable from everyone else's at 0.99 [9]. The ASH-convened 2025 review closed the loop honestly: the trait does not cause vaso-occlusive crises — exertion-related deaths implicate the rhabdomyolysis context (heat, dehydration, novel all-out effort, altitude, illness), which is precisely why the precautions work [10]. For patients, the medicine sits in the middle of the dose curve: the landmark EXDRE randomized trial put 8 weeks of moderate cycling against sedentary care and the training group gained +7.2 W of functional power while suffering zero hospital-admission events against 5 in the controls — 4 of them vaso-occlusive crises (HR 0.143, p=0.029) [17]; the sedentary sentence, not the gym, was the dangerous arm. The engine needs the work: peak VO₂ runs 26.9 vs 37.0 mL/kg/min in young patients [14] and a startling 16.1 vs 42.7 in adults [15], with strength roughly half and correlated with it (r=0.53) [15]. And the trap starts early and feeds itself: Brazilian children with the disease already move less than matched peers (19.2 vs 27.1 min/day of moderate activity) [16], while the 14-day diary file found that lower daily activity predicted more pain the same day and the next [23]. That is the audit in one paragraph: for carriers, the danger is the unprepared maximum; for patients, the dangers are the extremes on both sides — the all-out session and the sedentary sentence.

What the condition is: the 7.74-million file, the trait multiplier, and the Brazil column

The mechanics in a paragraph: sickle hemoglobin (HbS) is a single-point mutation that makes hemoglobin polymerize — stiffen into rods — under the four horsemen of low oxygen, acidosis, dehydration and heat. Two gene copies (HbSS, the classic disease) means polymerization, hemolysis, and vaso-occlusion as the standing weather of the blood; one copy (HbAS, the trait) means red cells that sickle only under extreme exertional duress, and blood that behaves normally the rest of the time. The mutation survived because it protects malaria-endemic ancestry — the same gene that kept ancestors alive where falciparum malaria killed is the one an unprepared max effort provokes. (If you arrived here from the iron side of the blood file, the anemia audit is the sibling post — same organ family, different mechanism: that one is a nutrient deficiency, this one is a hemoglobinopathy.) The scale file: 7.74 million people lived with sickle cell disease in 2021, up 41.4% from 5.46 million in 2000, with 515,000 born with it in 2021 alone and a total mortality burden of 376,000 a year — nearly 11-fold the cause-specific death count — including 81,100 deaths under age five, where SCD ranks 12th among all causes worldwide [1]. Every year another 312,000 babies are born with the disease and 5.48 million are born carrying the trait [2] — roughly 300 million people alive today carry it [3]. And the column that makes the pt-BR lane of this post structural rather than decorative: Brazil screens every newborn (the Teste do Pezinho program went universal for hemoglobinopathies in 2013), ~3,500 children with SCD are born there each year, and ~7 million Brazilians — about 4% of the population — carry the trait, one in 17 in Bahia, where SCD births run 1:650 — with the state gradient falling off to 1:1,200 in Rio de Janeiro, 1:4,000 in São Paulo (trait 1:35) and 1:13,500 in Rio Grande do Sul [4][5]. The Brazilian real-world cohort estimates 60,017 people living with the disease — and a median age at death of 32, against 69 in the general population [6]. This is not a rare file; it is one of the largest monogenic disease populations on earth, concentrated exactly where this app's audience lives.

The trait file, generation one: the army

Every modern precaution traces to one 1987 paper. Kark and colleagues examined deaths among 2 million US enlisted recruits in basic training, 1977–1981 — an environment that amounted to a maximal-exertion test on unacclimatized bodies: among Black recruits, sudden unexplained death ran 32.2 per 100,000 with the trait vs 1.2 without — a relative risk of 27.6 (95% CI 9–100), rising to 39.8 across all recruits — and the risk climbed steeply with age within the recruit window, from 13 per 100,000 at 17–18 to 95 at 26–30 [7]. Two honest frames belong on that finding: the absolute numbers are small (recruit training was dangerous for everyone; the trait multiplied a rare event), and the era's methods — all-out timed runs and drills on day one, in heat, unprepared — are precisely what modern preparation doctrine replaced. The paper did not argue carriers shouldn't train; it argued that the unprepared maximum is where the curve bends. That argument has stood for four decades [7][11].

Generation two: the gridiron and the bylaw

The sports file spent two decades confirming Kark and then, crucially, testing the fix. Boden's 20-year epidemiology of American football counted 187 nontraumatic fatalities (9.4 a year) between 1998 and 2018: sudden cardiac arrest behind 57.7%, exertional heat stroke 23.6% — and exertional sickling associated with the trait behind 12.1%, with the college (NCAA) risk running 4.1× the high-school risk, a gradient that tracks with the intensity and stakes of conditioning tests [8]. Then the natural experiment: after the 2010 NCAA bylaw (born from a sickling death litigation) required confirming every athlete's trait status at the preparticipation exam — turning an unknown into a known, with precautions attached — the risk of a sickling fatality fell to 0.12× (95% CI 0.02–0.95, p=.04) [8]. Read that carefully: not a ban, a disclosure. Carriers kept playing football at the highest level of American sport; the deaths moved when status plus preparation replaced ignorance plus day-one max tests. That is the strongest population-level argument in this entire file that the trait's risk is engineered down by exactly the things a training program is made of [8][11].

Generation three: the precision era — and the counterfile that keeps it honest

The modern cohort replaced the panic with an actuarial table. Among 47,944 US Army soldiers with known trait status, sickle trait carried an exertional rhabdomyolysis hazard ratio of 1.54 (95% CI 1.12–2.12, p=0.008) — a real, measurable elevation — while overall mortality ran HR 0.99 (0.46–2.13, p=0.97), statistically indistinguishable from non-carriers [9]. And in 2025 the American Society of Hematology convened a systematic review to settle the mechanism question, and its verdict is the honesty column of this section: across the screened literature there is no evidence that the trait causes acute vaso-occlusive pain crises, and no support for attributing an exertion-related death to the trait absent the rhabdomyolysis context — the heat, dehydration, novel maximal effort, altitude and illness that convert a rare event into a catastrophic one [10]. Put the three generations together and the protocol writes itself: the trait is not a disease (mortality 0.99 [9]), it does not spontaneously crisis [10], but it raises the stakes of the unprepared maximum (27.6 in Kark's era [7], 12.1% of gridiron fatalities [8], HR 1.54 for rhabdo [9]) — so the program's job is to make sure no session is an unprepared maximum.

The preparation door: what the precautions actually are

The sports-medicine establishment compressed the three generations of evidence into a short list, and it reads like a competent training program's table of contents. The NATA sudden-death position statement files exertional sickling among the leading causes of athlete death and prescribes the machinery: gradual acclimatization before intensity, programmed work/rest cycles, hydration as standing equipment, and the collapse rule — a struggling, cramping athlete with the trait stops the session and gets care immediately, because exertional sickling presents differently than heat stroke (the sickling athlete can still talk early, with cramping pain and weakening posture, then deteriorates fast) [11]. The ACSM/CHAMP summit adds the modulator list the physiology reviews established: heat, dehydration, exercise intensity, altitude, and novelty of the effort are the amplifiers — meaning the first hot week, the first altitude camp, and the first all-out test of a new program are the sessions that earn preparation, not the routine ones [3][12]. None of this benches anyone. All of it is just the ordinary craft of progressive training taken seriously — which is why the bylaw worked [8].

The capacity file: what the disease does to the engine

For the disease population, the historical caution had a real physiological object: the engine itself is diminished, and the file quantifies by how much. The largest pediatric CPET study put 60 children and young adults with HbSS through maximal testing against 30 controls: peak VO₂ 26.9 ± 6.9 vs 37.0 ± 9.2 mL/kg/min, oxygen pulse 12 ± 4 vs 20 ± 7 mL/beat — a capacity deficit attributable to factors beyond the anemia itself [14]. In adults it is starker: 29 adults with SCD averaged a peak VO₂ of 16.1 against 42.7 mL/kg/min in reference data, with knee-flexor strength roughly half (26.9 vs 55.6 Nm) and a 0.53 correlation between strength and aerobic capacity [15] — the muscle and the heart-lung system decondition together. And the deficit is already behavioral in childhood: Brazilian children and adolescents with the disease logged 19.2 vs 27.1 minutes of daily moderate activity and 3.6 vs 7.8 minutes of vigorous activity against matched peers [16]. That is the trap the old "rest to protect yourself" doctrine deepened: the engine's ceiling falls with disuse exactly as untrained engines do — the dehydration and heat-intolerance files compound it, and the next section is the trial file proving the ceiling rises when the work is dosed correctly.

The program-proof: EXDRE, the RCT that flipped the default

The default medical recommendation for sickle cell disease used to be exercise avoidance — the Brazilian home-based trial opens with the sentence: "medical recommendation was to avoid physical exercise for fear of triggering painful crises... only recently have studies shown safety in exercise programs" [21]. EXDRE is the study that ended that era. Forty adults with SCD but no severe chronic complications were randomized: 8 weeks of moderate-intensity cycling, three 45-minute sessions a week, against usual sedentary care [17]. The primary endpoint — power at the lactate threshold of 4 mmol/L — rose +7.2 W in the training group (70.4 → 77.6 W) while controls drifted −0.3 W; between-group difference 7.3 W (95% CI 0.7–13.8, p=0.031) [17]. And the safety line that reorganized the field: zero adverse events requiring hospital admission in the training group, against five in the sedentary controls — four of them vaso-occlusive crises — a hazard ratio of 0.143 (95% CI 0.024–0.827, p=0.029) [17]. Read that twice: in this 40-person trial, the dangerous arm was the couch. The muscle biopsies of the same cohort show why the benefit compounds — capillary density and exchange surface up (p=.003 and p<.0001) [18], type I fiber area and oxidative-enzyme (citrate synthase) activity up, with lower lactate and lower perceived effort at the same wattage [19] — the muscle remodels toward exactly the profile the capacity file showed it was missing. And the Brazilian replication column: the Salvador systematic review (6 studies, 212 patients) found low-to-moderate intensity raised tolerance without causing vaso-occlusive crises or adverse rheological/inflammatory changes [20]; the home-based trial — "none of the patients presented a sickle cell crisis or worsening of symptoms during the exercise program" [21]; the Botucatu aerobic RCT — peak VO₂ and 6-minute-walk up (both p<0.001), quality-of-life physical score up, "nor did it show unfavorable effects" [22]. Three independent files, one verdict: dosed correctly, the gym is the safe arm.

The pain paradox: the sedentary sentence is not protective

The subtlest finding in this audit is the one that quietly retires the old doctrine. Thirty children with sickle cell anemia kept 14-day pain diaries with actigraphy, and the analysis ran opposite to the folk model: lower physical activity predicted greater pain the same day and the next day — and less pain predicted more activity the next day — a bidirectional loop in which resting to avoid crises feeds the deconditioned, heat-intolerant, inflamed state that makes the next crisis more likely [23]. The authors' own framing: a complex bidirectional relationship between daily activity and daily pain [23]. For a patient population already moving less than peers from childhood [16], this is the audit's deepest argument: the question was never whether to train, but at what dose — because the no-training arm of the trial was the one that hospitalized [17].

The dose door: how to actually program it

The modern consensus review (Connes, Am J Hematol 2024) writes the two-sided rule this file keeps circling: acute INTENSE exercise raises the classic sickling triggers — acidosis, oxidative stress, dehydration — while controlled exercise training is safe, well tolerated by SCD patients, and improves cardiorespiratory and muscle function; the validated lanes are regular submaximal endurance work and exercise interventions incorporating resistance training [12]. The practical recommendations file adds the craft: start slow and progress gradually, hydrate during and after (not just before), avoid cold exposure and sudden temperature swings, avoid mechanical-trauma sports, and train symptom-limited — the session ends at symptoms, not through them [13]. Brazil's PCDT 2024 wrote the national version of the same dose: begin with light regular training and gradual progression plus stretching; interrupt prolonged continuous exercise at the first sign of fatigue, with programmed pauses every 20 minutes; encourage walking, swimming and cycling at 50–90% of max heart rate [24]. Cross-link the library's own dose files and the program composes from parts you may already own: zone 2 training is the submaximal lane EXDRE validated; progressive overload and weekly set volume run their unchanged stimulus rules under the middle dose; hydration is its own audit; and the structure holding it all is the program file.

The counterfile: what exercise is not, in this disease

Honesty section, five entries. Exercise is not hydroxyurea: the disease-modifying file (hydroxyurea, transfusion, and the new generator therapies) belongs to the hematologist, and training sits alongside it, never in place of it [12]. EXDRE is one 40-person trial: zero hospitalizations against five is a striking signal read with its size tag — a direction of evidence, not a guarantee, and it enrolled adults without severe chronic complications [17]. The trait's risks do not justify exclusion: carriers train, compete, and die at population rates (mortality HR 0.99 [9]; no crisis mechanism [10]) — the NCAA's own fix was disclosure plus preparation, not bans [8]. "Push through it" fails twice here: the crisis pain of the disease and the collapse signals of exertional sickling are both hard stops, in opposite populations [11][13]. And the biggest documented risk in this file is the one the old recommendation wrote: decades of "avoid exercise for fear of crises" [21] produced exactly the deconditioned, low-activity state the diary data associates with more pain [23] — caution itself, dosed wrong, became the complication.

So what do you actually do? The six-rule protocol

Assemble the file and the protocol writes itself — two populations, one craft.

1. Know your file. Trait and disease are different programs with different stakes; trait status is one blood test away (most Brazilians born after 2013 already have theirs filed from the Teste do Pezinho [4]), and disease is managed with a hematologist before the first programmed session [12].

2. Dose the middle. For the disease: submaximal aerobic work and moderate resistance training are the validated lanes — +7.2 W and zero hospitalizations in EXDRE [17], no crises across the Brazilian replications [20][21][22], breaks every 20 minutes and 50–90% HRmax per the PCDT [24]. For carriers: the ordinary progressive program is the precaution [8][11].

3. Engineer the fluids and the heat. Water is equipment in this file — during and after, not just before [13]; heat, altitude and novelty of effort are the amplifiers, so the first hot week and the first all-out test are prepared for, not improvised [3][11].

4. Build before you test. No day-one maximal efforts, no unprepared all-out conditioning tests — the one exertion pattern Kark's era, Boden's decade, and the rhabdo cohort all flag [7][8][9]. Acclimatize first, then add intensity [11].

5. Symptoms are hard stops. Pain crisis, unusual fatigue, dark urine, or collapse: stop the session, treat, and the care team hears about it — a struggling cramping athlete with the trait is an emergency, not a mental-toughness test [11][13].

6. Log the pattern. Intensity, duration, environment, symptoms, and any crisis — recorded per session — is the dataset that lets the care team tune dose against response, and the pattern only exists if it's recorded (why tracking works).

How RightRep fits into this

A file whose dose is tuned against symptoms runs on records — which is the shape this app is built around:

  • The session log as the symptom-pattern ledger. Work performed, time, intensity, and how each session ended — logged per set, per session — is the between-visits dataset rule 6 runs on; the hematologist's exercise prescription gets its feedback loop from the log [12][23].
  • Dosing you can hold to the middle. The validated lane is submaximal and progressive [12][17]; the app's structured programming holds the stimulus steady while the care team tunes the dose — one moving variable at a time.
  • The consistency file the diary data rewards. The bidirectional pain-activity loop [23] pays the habit, not the hero session: a program that repeats — and records — beats a program that spikes.

The bottom line

  • 7.74 million people live with sickle cell disease (up 41.4% since 2000); 515,000 born in 2021; 376,000 excess deaths a year; ~300 million more carry the trait [1][3].
  • Brazil: ~3,500 births a year, ~7 million carriers, 1:17 in Bahia (SCD births 1:650), 60,017 living — median age at death 32 vs 69 [4][5][6].
  • Trait, generation one: sudden unexplained death 32.2 vs 1.2 per 100,000 among 2 million recruits, RR 27.6 — the unprepared-maximum era [7].
  • Trait, generation two: sickling behind 12.1% of nontraumatic football fatalities until the 2010 NCAA bylaw cut the risk to 0.12× — disclosure plus preparation, not exclusion [8].
  • Trait, generation three: rhabdomyolysis HR 1.54 across 47,944 soldiers with overall mortality flat at 0.99 — and the ASH 2025 verdict that the trait does not cause crises [9][10].
  • Patients: peak VO₂ 26.9 vs 37.0 (young) and 16.1 vs 42.7 (adults), strength roughly half and correlated — the engine needs the work [14][15].
  • EXDRE: 8 weeks of moderate work +7.2 W at the lactate threshold, zero training-group hospitalizations against 5 in the sedentary controls (HR 0.143) — the couch was the dangerous arm [17].
  • Replication column: Salvador, home-based, and Botucatu trials — no crises, no unfavorable effects, capacity up [20][21][22].
  • The pain paradox: less daily activity predicted more pain the same and next day — the sedentary sentence is not protective [23].
  • The dose: submaximal and moderate-resistance lanes, hydration during and after, breaks every 20 minutes, 50–90% HRmax, stop at the first sign of fatigue [12][13][24].

Know your file, dose the middle, carry the water, build before you test — and train like the 7.74-million file says you can, at rightrep.app.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. GBD 2021 Sickle Cell Disease Collaborators. Global, regional, and national prevalence and mortality burden of sickle cell disease, 2000–2021: a systematic analysis from the Global Burden of Disease Study 2021. The Lancet Haematology, 2023;10(8):e585–e599. pubmed.ncbi.nlm.nih.gov
  2. Piel FB, Patil AP, Howes RE, et al. Global epidemiology of sickle haemoglobin in neonates: a contemporary geostatistical model-based map and population estimates. The Lancet, 2013;381(9861):142–151. pubmed.ncbi.nlm.nih.gov
  3. O'Connor FG, Bergeron MF, Cantrell J, et al. ACSM and CHAMP summit on sickle cell trait: mitigating risks for warfighters and athletes. Medicine & Science in Sports & Exercise, 2012;44(11):2045–2056. pubmed.ncbi.nlm.nih.gov
  4. Silva-Pinto AC, Queiroz AM, Leite-Moura C, et al. The Neonatal Screening Program in Brazil, Focus on Sickle Cell Disease (SCD). International Journal of Neonatal Screening, 2019;5(1):11. pubmed.ncbi.nlm.nih.gov
  5. Carneiro-Proietti ABF, Kelly S, Sabino EC, et al. Clinical and genetic ancestry profile of a large multi-centre sickle cell disease cohort in Brazil. British Journal of Haematology, 2018;182(6):895–908. pubmed.ncbi.nlm.nih.gov
  6. Cançado RD, Jesus JA. Estimated mortality rates of individuals with sickle cell disease in Brazil: real-world evidence. Blood Advances, 2023;7(15):3783–3792. pubmed.ncbi.nlm.nih.gov
  7. Kark JA, Posey DM, Schumacher HR, Ruehle CJ. Sickle-cell trait as a risk factor for sudden death in physical training. The New England Journal of Medicine, 1987;317(13):781–787. pubmed.ncbi.nlm.nih.gov
  8. Boden BP, Fine KM, Breit I, Lentz W, Anderson SA. Nontraumatic Exertional Fatalities in Football Players, Part 1: Epidemiology and Effectiveness of National Collegiate Athletic Association Bylaws. Orthopaedic Journal of Sports Medicine, 2020;8(8):2325967120942490. pubmed.ncbi.nlm.nih.gov
  9. Nelson DA, Deuster PA, Carter R III, Hill OT, Wolcott VL, Kurina LM. Sickle Cell Trait, Rhabdomyolysis, and Mortality among U.S. Army Soldiers. The New England Journal of Medicine, 2016;375(5):435–442. pubmed.ncbi.nlm.nih.gov
  10. Weeks LD, Wilson AM, Naik RP, et al. Sickle cell trait does not cause "sickle cell crisis" leading to exertion-related death: a systematic review. Blood, 2025;145(13):1345–1352. pubmed.ncbi.nlm.nih.gov
  11. Casa DJ, Guskiewicz KM, Anderson SA, et al. National Athletic Trainers' Association position statement: preventing sudden death in sports. Journal of Athletic Training, 2012;47(1):96–118. pubmed.ncbi.nlm.nih.gov
  12. Connes P, Stauffer E, Liem RI, Nader E. Exercise and training in sickle cell disease: Safety, potential benefits, and recommendations. American Journal of Hematology, 2024;99(10):1988–2001. pubmed.ncbi.nlm.nih.gov
  13. Connes P, Machado R, Hue O, Reid H. Exercise limitation, exercise testing and exercise recommendations in sickle cell anemia. Clinical Hemorheology and Microcirculation, 2011;49(1–4):151–163. pubmed.ncbi.nlm.nih.gov
  14. Liem RI, Reddy M, Pelligra SA, et al. Reduced fitness and abnormal cardiopulmonary responses to maximal exercise testing in children and young adults with sickle cell anemia. Physiological Reports, 2015;3(4):e12338. pubmed.ncbi.nlm.nih.gov
  15. Ogunsile FJ, Stewart KJ, Kanter J, Lanzkron SM. An evaluation of cardiopulmonary endurance and muscular strength in adults living with sickle cell disease. British Journal of Haematology, 2022;199(4):597–602. pubmed.ncbi.nlm.nih.gov
  16. Melo HN, Duarte NF, Aquino NF, et al. Objectively measured physical activity levels and sedentary time in children and adolescents with sickle cell anemia. PLoS ONE, 2018;13(12):e0208916. pubmed.ncbi.nlm.nih.gov
  17. Gellen B, Messonnier LA, Galactéros F, et al. Moderate-intensity endurance-exercise training in patients with sickle-cell disease without severe chronic complications (EXDRE): an open-label randomised controlled trial. The Lancet Haematology, 2018;5(11):e554–e562. pubmed.ncbi.nlm.nih.gov
  18. Merlet AN, Chatel B, Bondet V, et al. Beneficial effects of endurance exercise training on skeletal muscle microvasculature in sickle cell disease patients. Blood, 2019;134(25):2233–2241. pubmed.ncbi.nlm.nih.gov
  19. Merlet AN, Boissière J, Chatel B, et al. Muscle structural, energetic and functional benefits of endurance exercise training in sickle cell disease. American Journal of Hematology, 2020;95(11):1257–1268. pubmed.ncbi.nlm.nih.gov
  20. Pinto DMR, Ribeiro IPS, Oliveira MC, et al. Physical exercise in sickle cell anemia: a systematic review. Hematology, Transfusion and Cell Therapy, 2021;43(3):324–331. pubmed.ncbi.nlm.nih.gov
  21. de Araujo Junior JA, Zanesco A, Ferezin DC, et al. Cardiovascular benefits of a home-based exercise program in patients with sickle cell disease. PLoS ONE, 2021;16(5):e0250128. pubmed.ncbi.nlm.nih.gov
  22. Antonelli Rossi D, Neves Junior O, de Melo Marina Ribeiro, et al. Effect of a Physical Exercise Program on the Inflammatory Response, Cardiac Functions, Functional Capacity, and Quality of Life in Patients with Sickle Cell Disease. Journal of Clinical Medicine, 2023;12(12):3952. pubmed.ncbi.nlm.nih.gov
  23. Karlson CW, Baker AM, Bromberg MH, et al. Daily Pain, Physical Activity, and Home Fluid Intake in Pediatric Sickle Cell Disease. Journal of Pediatric Psychology, 2017;42(3):335–344. pubmed.ncbi.nlm.nih.gov
  24. Ministério da Saúde, Brasil. Protocolo Clínico e Diretrizes Terapêuticas da Doença Falciforme — Portaria Conjunta SAES/SECTICS nº 16, de 1º de novembro de 2024. gov.br