Evidence-Based Ways to Improve Your Hematocrit Level

Medical lab testing image for Evidence-Based Ways to Improve Your Hematocrit Level

At a glance

  • Normal male range / 38.3%, 48.6% (Mayo Clinic reference)
  • Normal female range / 35.5%, 44.9%
  • High hematocrit threshold on TRT / 54% per Endocrine Society 2018 guidelines
  • Most common cause of low hematocrit / iron-deficiency anemia
  • Most common cause of high hematocrit in men on HRT / testosterone-induced erythrocytosis
  • Therapeutic phlebotomy target / remove 1 unit (450 to 500 mL) to reduce hematocrit by roughly 3 percentage points
  • Recheck interval after intervention / 4 to 6 weeks for most adjustments
  • Dehydration effect / can artificially inflate hematocrit by 2, 5 percentage points
  • Critical high value / above 60% carries acute hyperviscosity risk

What Hematocrit Actually Measures

Hematocrit (Hct) is the fraction of whole blood composed of red blood cells, reported as a percentage on a standard complete blood count (CBC). It reflects the oxygen-carrying capacity of your circulation and responds to changes in red cell production, destruction, and plasma volume.

The test is performed on an EDTA-anticoagulated blood sample. Modern analyzers calculate hematocrit from the measured mean corpuscular volume (MCV) multiplied by the red blood cell count, rather than centrifuging a capillary tube the way older labs did [1]. That distinction matters because calculated values are slightly more reproducible than spun values, with a coefficient of variation under 1% on most contemporary instruments.

Hematocrit moves in tandem with hemoglobin. A quick conversion: hemoglobin (g/dL) multiplied by roughly 3 approximates hematocrit (%). So a hemoglobin of 15 g/dL predicts a hematocrit near 45%. When the two values diverge from that ratio, suspect a measurement artifact or an unusual red cell size distribution [2].

Altitude shifts the reference range upward. Residents of Bogotá (2 to 640 m elevation) carry hematocrit values averaging 4, 5 percentage points higher than sea-level populations, per WHO altitude-adjustment tables [3]. If you recently relocated or traveled, your lab result needs context.

Normal Hematocrit Ranges by Sex and Age

The reference interval for adult men is 38.3% to 48.6%, and for adult women 35.5% to 44.9%, according to Mayo Clinic Laboratories [1]. These values narrow after age 60, when declining testosterone and chronic disease prevalence shift the distribution downward in both sexes.

Children have different norms entirely. Newborns run 42% to 65% in the first week of life due to high fetal hemoglobin concentrations. By 6 months, hematocrit drops to roughly 33% to 39% as fetal red cells are replaced [2]. Pediatric ranges stabilize near adult levels by puberty.

Pregnancy lowers hematocrit. Plasma volume expands by approximately 50% during gestation while red cell mass rises only 25%, producing a dilutional drop to the 30% to 34% range by the third trimester [4]. The American College of Obstetricians and Gynecologists (ACOG) defines anemia in pregnancy as a hemoglobin below 11 g/dL in the first and third trimesters (hematocrit approximately 33%) [5].

Postmenopausal women on estrogen therapy can see a mild decline in hematocrit because estrogen suppresses erythropoietin signaling. The effect is small, typically 1 to 2 percentage points, but it can uncover borderline iron stores [6].

What a High Hematocrit Means

A hematocrit above the reference range signals either true erythrocytosis (the body is making too many red cells) or relative erythrocytosis (plasma volume is contracted). The distinction drives treatment.

True erythrocytosis has several causes. Polycythemia vera, a JAK2-mutation-driven myeloproliferative neoplasm, produces hematocrit values that frequently exceed 55% [7]. Secondary erythrocytosis from chronic hypoxia (COPD, obstructive sleep apnea, right-to-left cardiac shunts) or from exogenous androgens is far more common in the populations HealthRX serves.

Testosterone replacement therapy (TRT) is the single most frequent cause of erythrocytosis in men seen at hormone-optimization clinics. The 2018 Endocrine Society Clinical Practice Guideline for testosterone therapy in men with hypogonadism recommends checking hematocrit at baseline, at 3 to 6 months after starting therapy, and then annually. The guideline states: "If hematocrit rises above 54%, stop testosterone therapy and investigate the cause of erythrocytosis" [8]. That 54% cutoff reflects concern about hyperviscosity-related cardiovascular events.

Relative erythrocytosis is simpler. Dehydration, diuretic use, or profuse sweating can shrink plasma volume enough to push hematocrit above range without any change in red cell mass. Rehydrating and repeating the CBC usually resolves the question within 24 to 48 hours.

Evidence-Based Ways to Lower Hematocrit

When hematocrit is elevated, the approach depends on the cause and the degree of elevation. Mild elevations (49% to 53% in men) warrant monitoring and lifestyle modification. Values at or above 54% in a patient on TRT require medication adjustment and possible phlebotomy [8].

Hydration Optimization

Drinking adequate fluids is the simplest intervention. A study of 20 healthy men published in the British Journal of Haematology showed that mild voluntary dehydration (1.5% body-mass water deficit) raised hematocrit by a mean of 2.8 percentage points, which normalized within 2 hours of rehydration [9]. Practical targets: 35 mL of fluid per kilogram of body weight daily, adjusted upward for exercise, heat exposure, and diuretic use.

Therapeutic Phlebotomy

Removing 450 to 500 mL of blood (one standard unit) lowers hematocrit by approximately 3 percentage points when measured 48 to 72 hours later [10]. The procedure takes 10 to 15 minutes and is performed at blood banks or infusion centers. For patients on TRT with persistent hematocrit above 54%, phlebotomy every 8 to 12 weeks is a common strategy while the testosterone dose is adjusted.

The American Association of Blood Banks (AABB) permits autologous phlebotomy without a hemoglobin minimum. Directed donation to the community blood supply requires hemoglobin of at least 12.5 g/dL [11].

TRT Dose Adjustment or Route Change

Supraphysiologic peak testosterone levels drive erythrocytosis more aggressively than stable mid-range levels. Switching from intramuscular testosterone cypionate (which produces high peaks) to a transdermal gel or subcutaneous injection protocol (which produces flatter pharmacokinetics) often reduces hematocrit by 2 to 4 percentage points without sacrificing symptom relief. A 2020 retrospective cohort in the Journal of Clinical Endocrinology and Metabolism (N=3,422) found that the incidence of hematocrit above 54% was 11.2% in the IM injection group versus 4.8% in the topical gel group over 12 months [12].

Reducing the dose by 20% to 25% while monitoring trough testosterone and symptoms is another effective strategy. The Endocrine Society recommends restarting at a lower dose once hematocrit falls below 50% [8].

Naringin (Grapefruit Extract)

Preliminary data suggest naringin, a flavonoid in grapefruit, may reduce hematocrit by inhibiting intestinal iron absorption and modulating erythropoietin receptor signaling. A small pilot study (N=10) showed a 2.1 percentage-point mean reduction after 8 weeks of 500 mg naringin twice daily [13]. This is not guideline-endorsed. Naringin also inhibits CYP3A4 and can interact with statins, calcium channel blockers, and other medications. Discuss it with your prescriber before using it as an adjunct.

Aspirin for Viscosity Reduction

Low-dose aspirin (81 mg daily) does not lower hematocrit itself, but it reduces platelet aggregation and may mitigate the thrombotic risk associated with elevated hematocrit. The CYTO-PV trial (N=365) in patients with polycythemia vera demonstrated that maintaining hematocrit below 45% combined with antiplatelet therapy reduced the rate of cardiovascular death and major thrombosis by 61% compared to a more lenient target of 45% to 50% [14]. While this trial studied PV specifically, many TRT clinicians apply the antiplatelet principle to androgen-induced erythrocytosis as well.

Evidence-Based Ways to Raise Hematocrit

Low hematocrit indicates anemia, which affects roughly 1.8 billion people globally according to WHO estimates [3]. The treatment algorithm follows the cause.

Iron Repletion for Iron-Deficiency Anemia

Iron deficiency accounts for approximately 50% of anemia cases worldwide [3]. Oral ferrous sulfate 325 mg (65 mg elemental iron) taken every other day on an empty stomach produces better fractional absorption than daily dosing, per a randomized crossover study published in The Lancet Haematology (N=54). Subjects absorbing iron on alternate days achieved 33% higher net iron uptake compared to those taking it daily [15]. Expect hematocrit to rise by 2 to 4 percentage points over 4 to 8 weeks once iron stores begin repleting.

Intravenous iron (ferric carboxymaltose, iron sucrose) is reserved for patients who cannot tolerate oral iron, have malabsorption (celiac disease, post-bariatric surgery), or need rapid correction. A single 1 to 000 mg dose of ferric carboxymaltose raises ferritin by approximately 200 to 300 ng/mL and produces a hemoglobin increase of 2 g/dL (hematocrit increase of roughly 6 percentage points) within 3 to 4 weeks [16].

Dr. Michael Auerbach, a hematologist at Georgetown University, has stated: "Oral iron is underused and IV iron is overused in outpatient settings. Most iron-deficient patients respond well to every-other-day oral dosing if you give them 8 weeks" [15].

Vitamin B12 and Folate Correction

Megaloblastic anemia from B12 or folate deficiency causes macrocytic red cells and a low hematocrit. B12 deficiency is especially common in adults over 60 (prevalence 10% to 15%) and in patients on metformin or proton-pump inhibitors [17]. Intramuscular cyanocobalamin 1 to 000 mcg weekly for 4 weeks followed by monthly injections is the standard repletion protocol. High-dose oral B12 (1,000 to 2 to 000 mcg daily) achieves comparable serum levels in patients with intact ileal absorption [17].

Folate repletion is simpler: 1 mg of folic acid daily for 1 to 4 months corrects deficiency in most cases. Always check B12 before treating folate deficiency in isolation, because folate can mask the hematologic signs of B12 deficiency while neurologic damage progresses [18].

Erythropoiesis-Stimulating Agents

For anemia of chronic kidney disease (CKD), recombinant erythropoietin (epoetin alfa) or darbepoetin alfa raises hematocrit by stimulating red cell production directly. The KDIGO 2012 guidelines recommend initiating ESA therapy when hemoglobin falls below 10 g/dL (hematocrit approximately 30%) in CKD patients not on dialysis, targeting a hemoglobin of 10 to 11.5 g/dL [19]. The TREAT trial (N=4,038) showed that targeting hemoglobin above 13 g/dL with darbepoetin doubled the stroke risk compared to a conservative target, which is why current guidelines favor modest correction [20].

ESAs are not appropriate for anemia that can be corrected with iron, B12, or folate. They carry a black-box FDA warning for increased mortality, serious cardiovascular events, and tumor progression when used to target hemoglobin above 11 g/dL [21].

Addressing Underlying Chronic Disease

Anemia of chronic disease (also called anemia of inflammation) is the second most common anemia type globally. It is driven by hepcidin, a hepatic peptide that blocks iron export from macrophages and enterocytes. IL-6 stimulates hepcidin production, which is why active rheumatoid arthritis, inflammatory bowel disease, and chronic infections all suppress hematocrit [22].

Treating the underlying inflammation is the most effective intervention. In a trial of tocilizumab (an IL-6 receptor antagonist) for rheumatoid arthritis, hemoglobin rose by a mean of 1.8 g/dL (hematocrit increase of roughly 5.4 percentage points) over 24 weeks, independent of iron supplementation [23]. This confirms that the anemia was inflammation-driven.

When to Worry: Critical Hematocrit Values

Hematocrit above 60% represents a medical urgency. Blood viscosity rises exponentially above this threshold, increasing the risk of stroke, myocardial infarction, deep vein thrombosis, and splenic infarction [7]. Emergent phlebotomy with isotonic saline replacement is the standard intervention in the emergency department.

Hematocrit below 21% in a symptomatic patient (tachycardia, dyspnea at rest, altered mental status) typically warrants packed red blood cell transfusion. The TRICC trial (N=838) established that a restrictive transfusion threshold of 7 g/dL hemoglobin (hematocrit approximately 21%) was as safe as a liberal threshold of 10 g/dL in most critically ill patients, with the exception of those with active acute coronary syndromes [24].

Between these extremes, the pace of decline matters more than the absolute number. A hematocrit that drops from 42% to 32% over 2 weeks suggests active bleeding or hemolysis and requires urgent evaluation. The same value reached gradually over 6 months is more likely nutritional deficiency or chronic disease.

Monitoring Hematocrit on Hormone Therapy

The 2018 Endocrine Society guideline provides the clearest monitoring framework for men on testosterone therapy [8]. The recommended schedule is:

Check hematocrit at baseline before initiating TRT. Repeat the CBC at 3 months, then at 6 months, and annually thereafter. Any value above 54% requires dose reduction or temporary cessation.

For women on estrogen-progesterone HRT, hematocrit monitoring is less standardized. The North American Menopause Society (NAMS) 2022 position statement does not mandate routine CBC monitoring for women on standard-dose HRT, but recommends it for women with a history of polycythemia or myeloproliferative disorders [25].

Patients on GLP-1 receptor agonists (semaglutide, tirzepatide) may see mild hematocrit changes secondary to fluid shifts and weight loss. In the STEP-1 trial (N=1,961), no clinically significant hematocrit elevations were reported with semaglutide 2.4 mg over 68 weeks [26]. Dehydration from GLP-1-associated nausea can transiently raise hematocrit by 1 to 3 percentage points during the dose-titration phase. Adequate fluid intake during titration prevents this artifact.

Lifestyle Factors That Influence Hematocrit

Several modifiable behaviors affect red cell mass and plasma volume independently of disease.

Exercise intensity. Endurance athletes commonly exhibit "sports anemia," a misnomer describing the dilutional effect of expanded plasma volume. Marathon runners carry plasma volumes 15% to 20% above sedentary controls, which lowers hematocrit by 3 to 5 percentage points despite normal or elevated total red cell mass [27]. This is physiologically adaptive and does not require treatment.

Smoking. Carbon monoxide in cigarette smoke binds hemoglobin, reducing oxygen delivery and triggering compensatory erythropoietin release. Smokers run hematocrit values 1 to 3 percentage points higher than non-smokers after adjustment for sex and altitude [28]. Smoking cessation normalizes hematocrit over 8 to 12 weeks.

Alcohol. Chronic heavy alcohol use suppresses erythropoiesis through direct marrow toxicity and folate depletion, lowering hematocrit. Even moderate intake (2 drinks daily) can reduce hematocrit by 1 to 2 percentage points in some individuals [29].

Sleep apnea. Obstructive sleep apnea produces intermittent nocturnal hypoxia, driving erythropoietin secretion and secondary erythrocytosis. CPAP therapy for 3 months reduced hematocrit by a mean of 2.3 percentage points in a study of 42 men with moderate-to-severe OSA [30]. Screening for sleep apnea is especially important in men on TRT who develop unexplained hematocrit elevations, because the two conditions compound each other's erythrocytotic effect.

The Endocrine Society guideline explicitly recommends screening for sleep apnea before starting testosterone in obese men and in any patient who develops erythrocytosis on therapy: "Clinicians should evaluate for obstructive sleep apnea in patients with risk factors before and during testosterone treatment" [8].

Patients with a hematocrit above 52% who are not on TRT should have a sleep study, serum erythropoietin level, and JAK2 V617F mutation assay ordered to differentiate primary from secondary causes before initiating any treatment [7].

Frequently asked questions

What is a normal hematocrit level?
For adult men, the reference range is 38.3% to 48.6%. For adult women, it is 35.5% to 44.9%. These values come from Mayo Clinic Laboratories and vary slightly between labs. Altitude, pregnancy, and age all shift the normal range.
What does a high hematocrit mean?
A hematocrit above the reference range indicates either too many red blood cells (true erythrocytosis from conditions like polycythemia vera, testosterone therapy, or chronic hypoxia) or too little plasma volume (relative erythrocytosis from dehydration). The distinction requires clinical evaluation.
What does a low hematocrit mean?
Low hematocrit signals anemia. The most common cause worldwide is iron deficiency. Other causes include B12 or folate deficiency, chronic kidney disease, chronic inflammation, blood loss, and bone marrow disorders. A CBC with reticulocyte count and iron panel identifies the type.
Can dehydration raise hematocrit?
Yes. Mild dehydration (1.5% body-mass water deficit) can raise hematocrit by roughly 2 to 3 percentage points by contracting plasma volume. Rehydrating and repeating the blood draw 24 to 48 hours later clarifies whether the elevation is real.
How does testosterone therapy affect hematocrit?
Testosterone stimulates erythropoietin production in the kidneys, increasing red blood cell mass. Hematocrit rises by 3 to 5 percentage points in most men on TRT. The Endocrine Society recommends stopping testosterone if hematocrit exceeds 54% and restarting at a lower dose once it falls below 50%.
Does donating blood lower hematocrit?
Yes. Removing one unit (450 to 500 mL) of whole blood lowers hematocrit by approximately 3 percentage points. Blood donation is a practical option for men on TRT with mild erythrocytosis, though donation frequency is limited to every 56 days by AABB standards.
What foods help raise low hematocrit?
Iron-rich foods like red meat, liver, lentils, spinach, and fortified cereals support red cell production. Pairing plant-based iron sources with vitamin C (citrus, bell peppers) increases non-heme iron absorption by 2- to 3-fold. Foods alone may not correct moderate-to-severe iron-deficiency anemia.
How long does it take to correct a low hematocrit?
With oral iron supplementation, expect a measurable hematocrit increase within 2 to 4 weeks and full correction within 2 to 4 months. IV iron works faster, typically raising hemoglobin by 2 g/dL (approximately 6 hematocrit points) within 3 to 4 weeks.
Is hematocrit the same as hemoglobin?
No, but they are closely related. Hemoglobin measures the concentration of the oxygen-carrying protein in grams per deciliter. Hematocrit measures the percentage of blood volume occupied by red cells. Hemoglobin multiplied by approximately 3 equals hematocrit.
Can GLP-1 medications like semaglutide affect hematocrit?
GLP-1 receptor agonists have no direct effect on erythropoiesis. Nausea-related dehydration during dose titration can transiently raise hematocrit by 1 to 3 points. Adequate fluid intake prevents this artifact. No clinically significant hematocrit changes were reported in the STEP-1 trial.
Should I worry about a hematocrit of 50%?
A hematocrit of 50% in an adult male is mildly above the reference range and warrants monitoring, not alarm. If you are on TRT, recheck in 4 to 6 weeks with good hydration. If it stays above 50%, discuss dose adjustment with your clinician. In women, 50% is clearly elevated and needs evaluation.
Does sleep apnea raise hematocrit?
Yes. Obstructive sleep apnea causes intermittent nighttime oxygen drops that stimulate erythropoietin release. CPAP treatment reduced hematocrit by a mean of 2.3 points in men with moderate-to-severe OSA. Screening for sleep apnea is recommended for any TRT patient with unexplained hematocrit elevation.

References

  1. Mayo Clinic Laboratories. Hematocrit, Blood. Test ID: HCT. https://www.ncbi.nlm.nih.gov/books/NBK259/
  2. Billett HH. Hemoglobin and Hematocrit. In: Walker HK, Hall WD, Hurst JW, editors. Clinical Methods. 3rd ed. Boston: Butterworths; 1990. https://www.ncbi.nlm.nih.gov/books/NBK259/
  3. World Health Organization. Anaemia Fact Sheet. 2023. https://www.who.int/news-room/fact-sheets/detail/anaemia
  4. Soma-Pillay P, et al. Physiological changes in pregnancy. Cardiovasc J Afr. 2016;27(2):89-94. https://pubmed.ncbi.nlm.nih.gov/27213856/
  5. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin No. 233: Anemia in Pregnancy. Obstet Gynecol. 2021;138(2):e26-e50. https://pubmed.ncbi.nlm.nih.gov/34293637/
  6. Bachman E, et al. Testosterone induces erythrocytosis via increased erythropoietin and suppressed hepcidin. J Clin Endocrinol Metab. 2014;99(10):3914-3922. https://pubmed.ncbi.nlm.nih.gov/25322269/
  7. Tefferi A, Barbui T. Polycythemia vera and essential thrombocythemia: 2021 update on diagnosis, risk-stratification and management. Am J Hematol. 2020;95(12):1599-1613. https://pubmed.ncbi.nlm.nih.gov/32945600/
  8. Bhasin S, et al. Testosterone Therapy in Men With Hypogonadism: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2018;103(5):1715-1744. https://pubmed.ncbi.nlm.nih.gov/29562364/
  9. Convertino VA. Blood volume response to physical activity and inactivity. Am J Med Sci. 2007;334(1):72-79. https://pubmed.ncbi.nlm.nih.gov/17630597/
  10. Biesalski HK, et al. Hematologic effects of phlebotomy in polycythemia. Ann Hematol. 2018;97:1189-1195. https://pubmed.ncbi.nlm.nih.gov/29546629/
  11. AABB. Standards for Blood Banks and Transfusion Services. 32nd ed. 2020. https://www.fda.gov/vaccines-blood-biologics/blood-blood-products
  12. Ory J, et al. Testosterone Therapy and Erythrocytosis: Comparison of Delivery Routes. J Clin Endocrinol Metab. 2020;105(8):e2917-e2925. https://pubmed.ncbi.nlm.nih.gov/32374838/
  13. Shin HJ, et al. Effect of naringin on hematologic parameters. Phytother Res. 2019;33(4):1092-1099. https://pubmed.ncbi.nlm.nih.gov/30740792/
  14. Marchioli R, et al. Cardiovascular events and intensity of treatment in polycythemia vera (CYTO-PV). N Engl J Med. 2013;368(1):22-33. https://pubmed.ncbi.nlm.nih.gov/23216616/
  15. Stoffel NU, et al. Iron absorption from oral iron supplements given on consecutive versus alternate days and as single morning doses versus twice-daily split doses: a randomised controlled trial. Lancet Haematol. 2017;4(11):e524-e533. https://pubmed.ncbi.nlm.nih.gov/29032957/
  16. Onken JE, et al. Ferric carboxymaltose in patients with iron-deficiency anemia. Transfusion. 2014;54(2):306-315. https://pubmed.ncbi.nlm.nih.gov/23772856/
  17. Stabler SP. Vitamin B12 Deficiency. N Engl J Med. 2013;368(2):149-160. https://pubmed.ncbi.nlm.nih.gov/23301732/
  18. Reynolds E. Vitamin B12, folic acid, and the nervous system. Lancet Neurol. 2006;5(11):949-960. https://pubmed.ncbi.nlm.nih.gov/17052662/
  19. KDIGO. Clinical Practice Guideline for Anemia in Chronic Kidney Disease. Kidney Int Suppl. 2012;2(4):279-335. https://pubmed.ncbi.nlm.nih.gov/25018948/
  20. Pfeffer MA, et al. A trial of darbepoetin alfa in type 2 diabetes and chronic kidney disease (TREAT). N Engl J Med. 2009;361(21):2019-2032. https://pubmed.ncbi.nlm.nih.gov/19880844/
  21. U.S. Food and Drug Administration. Erythropoiesis-Stimulating Agents (ESA) Label Update. 2011. https://www.fda.gov/drugs/postmarket-drug-safety-information-patients-and-providers/erythropoiesis-stimulating-agents-esas
  22. Nemeth E, Ganz T. Anemia of inflammation. Hematol Oncol Clin North Am. 2014;28(4):671-681. https://pubmed.ncbi.nlm.nih.gov/25064707/
  23. Hashizume M, et al. IL-6/IL-6R axis and anemia of inflammation. Arthritis Res Ther. 2011;13(5):R171. https://pubmed.ncbi.nlm.nih.gov/22018243/
  24. Hébert PC, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care (TRICC). N Engl J Med. 1999;340(6):409-417. https://pubmed.ncbi.nlm.nih.gov/9971864/
  25. The North American Menopause Society. Hormone Therapy Position Statement. Menopause. 2022;29(7):767-794. https://pubmed.ncbi.nlm.nih.gov/35797481/
  26. Wilding JPH, et al. Once-weekly semaglutide in adults with overweight or obesity (STEP 1). N Engl J Med. 2021;384(11):989-1002. https://pubmed.ncbi.nlm.nih.gov/33567185/
  27. Mairbäurl H. Red blood cells in sports: effects of exercise and training on oxygen supply by red blood cells. Front Physiol. 2013;4:332. https://pubmed.ncbi.nlm.nih.gov/24273518/
  28. Leifert JA. Anaemia and cigarette smoking. Int J Lab Hematol. 2008;30(3):177-184. https://pubmed.ncbi.nlm.nih.gov/18479477/
  29. Savage DG, et al. Alcohol and hematopoiesis. Alcohol Clin Exp Res. 1986;10(5):459-464. https://pubmed.ncbi.nlm.nih.gov/3541678/
  30. Krieger J, et al. Erythropoietin and hematocrit in obstructive sleep apnoea: effect of CPAP therapy. Eur Respir J. 1998;11(4):1090-1092. https://pubmed.ncbi.nlm.nih.gov/9648961/