REVIEW ARTICLE

Blood Is Thicker Than Water

The Management of Hyperviscosity in Adults With Cyanotic Heart Disease
Andrew Paul DeFilippis, MD,* Karen Law, MD, Sarah Curtin, MD, and James R. Eckman, MD

Abstract: Complications of chronic hypoxia, including erythrocytosis, hyperviscosity, abnormalities of hemostasis, cerebral abscesses, stroke, and endocarditis, are among the most common consequences of cyanotic heart disease in adults. The compensatory erythrocytosis of cyanotic heart disease can become pathologic by causing an increase in blood viscosity, thereby decreasing perfusion and resulting in decreased total oxygen delivery and increased risk of venoocclusive/hyperviscosity syndrome. Treatment of hyperviscosity secondary to erythrocytosis in cyanotic heart disease is controversial. Data is limited but suggest that phlebotomy has the potential to increase exercise capacity, reduce the symptoms of hyperviscosity, and reduce the potential risk of vasoocclusive disease in selected patients with polycythemia secondary to cyanotic heart disease. Unfortunately, repeated phlebotomy can quickly lead to iron deciency, resulting in microcytic erythrocytes that induce higher viscosity than normocytic erythrocytes, which may increase the risk for venoocclusive events. There are limited data on the use of hydroxyurea to suppress erythrocytosis in this patient population. The authors conclude that until newer approaches to decreasing hematocrit without inducing iron deciency are shown to be safe and efcacious, phlebotomy should only be used for the acute resolution of hyperviscosity symptoms. In addition, the use of hydroxyurea should be limited to patients with recurrent symptoms. Key Words: cyanotic heart disease, congenital heart disease, hyperviscosity, secondary polycythemia (Cardiology in Review 2007;15: 3134)

ith the survival of patients with cyanotic heart disease into adulthood, new management issues arise. Tetralogy of Fallot (TOF) is the most common cyanotic congenital heart defect after infancy, characterized by ventricular septal defect, right ventricular outow tract obstruction, overriding aorta, and right ventricular hypertrophy. The classic childhood symptoms include hypoxic spells of tachypnea, hyperpnea, cyanosis, loss of consciousness, seizure, cerebral vascular accidents, and death. The denitive treatment of TOF is complete surgical correction in infancy, including ventricularseptal defect closure and relief of right ventricular outow tract obstruction. However, not all patients undergo surgery as a result of an undetected defect, inadequate access to care, or poor surgical candidacy. Survival to age 20 years without surgical intervention is 11%; survival declines to 6% at age 30 and 3% at age 40.1 Other less common cyanotic congenital heart diseases include Ebstein malformation, transposition of the great arteries, and Eisenmenger syndrome. In these patients with uncorrected cyanotic heart disease who survive into adulthood, a new series of symptoms arises. Complications of chronic hypoxia predominate, including erythrocytosis, hyperviscosity, abnormalities of hemostasis, cerebral abscesses, stroke, and endocarditis.1 This article reviews the pathophysiology and treatment strategies for erythrocytosis and hyperviscosity in adults with cyanotic heart disease.

PATHOPHYSIOLOGY OF ERYTHROCYTOSIS AND ITS SEQUELAE

Erythrocytosis in adults with cyanotic heart disease is the result of a compensatory mechanism designed to deliver more oxygen to chronically hypoxic tissue. Red blood cell mass is increased through renal release of erythropoietin, whereas overall blood volume is maintained. The compensatory erythrocytosis in cyanotic heart disease ultimately becomes harmful by inducing a pathologic increase in blood viscosity. Animal studies conrm these ndings. In dogs, normovolemic increases in hematocrit results in hyperviscosity responsible for decreasing oxygen transport at hematocrits greater than 40%.2 Mouse studies show increased oxygen delivery and cardiac output only with increases in both hematocrit and blood volume.2 Only when a patient can increase blood volume or cardiac output to overcome the

From the *Department of Medicine, Division of General Medicine, and the Department of Internal Medicine, Emory University School of Medicine, Atlanta, Georgia; Family Medicine, St. Marys Family Practice, Grand Junction, Colorado; and the Departments of Hematology and Oncology and Medicine, Georgia Comprehensive, Sickle Cell Center, Clinical Hematology at Grady Health System, The Robert W. Woodruff Health Science Center, Emory University, Atlanta, Georgia. Correspondence: Andrew Paul DeFilippis, MD, Assistant Professor of Medicine, Division of General Medicine, Emory University School of Medicine, 49 Jesse Hill Jr. Drive, Atlanta, GA 30303. E-mail: APDeF@ yahoo.com. Copyright 2006 by Lippincott Williams & Wilkins ISSN: 1061-5377/07/1501-0031 DOI: 10.1097/01.crd.0000214959.82120.97

Cardiology in Review Volume 15, Number 1, January/February 2007

31

DeFilippis et al

Cardiology in Review Volume 15, Number 1, January/February 2007

increased resistance of polycythemia would one expect an increase in oxygen delivery. In addition, patients with cyanotic heart disease are at additional risk of hyperviscosity because many patients with chronic erythrocytosis often become iron-decient. Iron deciency results from total body depletion of iron stores to support erythropoiesis as well as from therapeutic phlebotomy, which is discussed later. The subsequent microcytic erythrocytes are less deformable than their normocytic counterparts, thereby elevating blood viscosity at any hematocrit level.35 A study by Kaemmerer et al of 52 adults with cyanotic heart disease found the incidence of iron deciency among their population to be 37%.6 The major consequence of hyperviscosity is thrombosis, because the increased red blood cell mass and abnormal platelet function induce sludging in the microvasculature. Clinically, this hyperviscosity may manifest as a cerebrovascular accident, myocardial infarction, or other overt thrombotic event. Although it is widely accepted that pediatric patients with cyanotic heart disease are at increased risk of cerebral vascular accidents, the literature presents conicting reports of the prevalence of these events among adults with cyanotic heart disease.7,8 Engelfriet et al report a 10% prevalence of stroke or transient ischemic attack among their cohort of adult patients with cyanosis over a median follow up of 5.1 years.9 Ammash et al report a 13.6% prevalence of a history of stroke among a similar population of adult patients.10 In contrast, Perloff et al report following a similar group of 124 adults in which zero strokes were observed during an average 6.1-year follow-up period.11 Differences in the incidence of iron deciency and arrhythmias may explain the observed differences in the increased risk of stroke in children or different study populations of adults with cyanotic heart disease. Evidence of hyperviscosity producing cerebral perfusion abnormalities not resulting in an overt thrombotic event may also manifest itself with more subtle clinical ndings, including headache, slow mentation, dizziness, blurry vision, muscle weakness, or paresthesias caused by reduced tissue perfusion.3 A prospective study in healthy asymptomatic adults reported an increased risk of vascular occlusive disease with hematocrits greater than 46% as compared with those with a hematocrit less than 46%.12 This provides indirect evidence that erythrocytosis in adults with cyanotic heart disease may have increased risk for cerebral vascular accidents. A small number of case reports suggest the risk of myocardial infarction is increased in adults with cyanotic heart disease and elevated hematocrits.13,14

In 22 patients with cyanotic heart disease ages 3 to 25 years, Rosenthal et al report that decreasing the hematocrit from an average of 73.5% to 62% signicantly decreased blood viscosity and systemic vascular resistance, resulting in an increased cardiac index and a 10% to 40% increase in oxygen delivery.15 There were no signicant changes in heart rate, respiratory rate, or oxygen consumption. These results were observed despite a decrease in arterial saturation in 17 of the 22 patients with severe right ventricular outow obstruction from pulmonary hypertension. Oldershaw et al report increased exercise capacity and cardiac index after removing 3 units of blood in 6 adults with cyanotic heart disease. Three and 14 days after phlebotomy, maximal exercise capacity increased, as did the cardiac index at rest and at 50% maximal exercise, without an increase in heart rate. Oxygen uptake increased signicantly during exercise but not at rest.16 These objective ndings suggest that tissue oxygen delivery may increase after phlebotomy in some individuals with cyanotic heart disease despite a reduction in blood oxygen-carrying capacity. Phlebotomy among select patients with erythrocytosis has also been shown to improve cerebral blood ow and neurologic symptoms of hyperviscosity.12 In patients with polycythemia secondary to hypoxemia, York et al demonstrate an increase in cerebral blood ow by 78% after phlebotomy.17 Although many studies report subjective improvement of symptoms thought to be secondary to poor cerebral blood ow such as headache, confusion, and lethargy,18 1 study was able to show increased neuropsychiatric performance at lower hematocrits in patients with secondary polycythemia.19

Iron Deficiency and Iron Supplementation

Repeated phlebotomy can quickly lead to iron deciency. As previously discussed, microcytic erythrocyte shows reduced deformability and induces higher viscosity than normocytic erythrocytes at comparable hematocrits.35 Erythrocyte count can also be increased, adding to viscosity, with phlebotomyinduced iron deciency. Thus, aggressive phlebotomy that causes severe iron deciency may paradoxically induce persistent hyperviscosity. In practice, this nding may explain the additional risk for cerebral vascular accidents among patients with both iron deciency anemia and cyanotic heart disease.7,11,20,21 In addition, iron deciency may have detrimental effects beyond the reduced oxygen-carrying capacity of the red blood cell. Iron-decient animals have shown reduced exercise performance independent of hemoglobin level restored by exchange transfusion.2224 This effect is thought to be mediated by a reduction in iron-dependent enzymes, impairing tissue oxygen utilization.24 Evaluation of iron deciency among patients with cyanotic heart disease using calculated erythrocyte indices is insufcient. In a study of patients with cyanotic heart disease, Kaemmerer et al reported 19 of 52 patients with iron deciency; however, only 3 of these 19 patients had the expected ndings of microcytosis and hypochromia.6 Most patients showed a trend toward hyperchromia (31%), macrocytosis (31%), or both (27%), suggesting concomitant iron, folate, and B-complex vitamin deciency. The authors recommend full
2006 Lippincott Williams & Wilkins

TREATMENT OPTIONS FOR HYPERVISCOSITY IN CYANOTIC HEART DISEASE Phlebotomy

Treatment of hyperviscosity secondary to erythrocytosis in cyanotic heart disease is controversial. Data suggest that phlebotomy has the potential to increase exercise capacity, reduce the symptoms of hyperviscosity, and reduce the potential increased risk of vasoocclusive disease in selected patients with polycythemia secondary to cyanotic heart disease.

32

Cardiology in Review Volume 15, Number 1, January/February 2007

Management of Hyperviscosity

evaluation of iron metabolism in patients with cyanosis, including serum ferritin and transferrin in addition to erythrocyte indices, to determine the need for iron supplementation.6 Iron replacement therapy may be indicated in selected patients with iron deciency, microcytosis, or signicant erythrocytosis. However, patients with cyanotic heart disease receiving iron replacement therapy must be closely monitored, because marked polycythemia and hyperviscosity may rapidly occur as iron stores are repleted.3,25

this patient population, although approaches using dextran or fresh-frozen plasma confer additional considerations such as higher cost and risk for anaphylaxis or exposure to bloodborne pathogens.

CONCLUSIONS
The available data suggest that a normovolemic reduction in hematocrit in patients with cyanotic heart disease may increase systemic oxygen delivery and relieve symptoms of hyperviscosity, thereby decreasing the potential risk of central nervous system or cardiac ischemia and thrombosis. In the acute setting, phlebotomy, while maintaining intravascular volume, is appropriate for patients experiencing symptomatic hyperviscosity. Unfortunately, phlebotomy for the long-term management of erythrocytosis in otherwise asymptomatic patients is less feasible. Chronic phlebotomy in an attempt to optimize oxygen delivery and prevent hyperviscosity may lead to iron deciency anemia and a paradoxically increased risk of hyperviscosity at a lower hematocrit. Furthermore, iron replacement in these patients can induce increased erythropoiesis, ultimately trading one problem for another. Although it appears that hydroxyurea has been used in this patient population without harm, there are no controlled clinical trials demonstrating efcacy. For the chronic management of these patients while asymptomatic, Castle and Jandl in a 1966 review of polycythemia state In achieving the best balance between increased cardiac work and decreased tissue hypoxia, the wisdom of the body probably exceeds that of the physician.28 Until well-designed, controlled trials of decreasing the hematocrit are shown to be safe and efcacious, phlebotomy should only be used for the acute resolution of hyperviscosity symptoms, severe iron deciency should be avoided, and the use of hydroxyurea should be limited to recurrently symptomatic patients. REFERENCES
1. Brickner ME, Hillis LD, Lange RA. Congenital heart disease in adults. Second of two parts. N Engl J Med. 2000;342:334 342. 2. Erslev AJ, Caro J. Secondary polycythemia: a boon or a burden? Blood Cells. 1984;10:177191. 3. Baum VC. The adult patient with congenital heart disease. J Cardiothorac Vasc Anesth. 1996;10:261282. 4. Hutton RD. The effect of iron deciency on whole blood viscosity in polycythaemic patients. Br J Haematol. 1979;43:191199. 5. Linderkamp O, Klose HJ, Betke K, et al. Increased blood viscosity in patients with cyanotic congenital heart disease and iron deciency. J Pediatr. 1979;95:567569. 6. Kaemmerer H, Fratz S, Braun SL, et al. Erythrocyte indexes, iron metabolism, and hyperhomocysteinemia in adults with cyanotic congenital cardiac disease. Am J Cardiol. 2004;94:825 828. 7. Phornphutkul C, Rosenthal A, Nadas AS, et al. Cerebrovascular accidents in infants and children with cyanotic congenital heart disease. Am J Cardiol. 1973;32:329 334. 8. Wintrobe MM, Lee GR. Wintrobes Clinical Hematology, 11th ed. Philadelphia: Lippincott Williams & Wilkins; 2004. 9. Engelfriet P, Boersma E, Oechslin E, et al. The spectrum of adult congenital heart disease in Europe: morbidity and mortality in a 5 year follow-up period: the Euro Heart Survey on adult congenital heart disease. Eur Heart J. 2005;26:23252333. 10. Ammash N, Warnes CA. Cerebrovascular events in adult patients with cyanotic congenital heart disease. J Am Coll Cardiol. 1996;28:768 772. 11. Perloff JK, Rosove MH, Child JS, et al. Adults with cyanotic congenital

Suppression of Erythrocyte Production

The addition of erythrocyte production suppression agents like hydroxyurea to a regimen of phlebotomy has successfully been used in patients with polycythemia vera. In contrast to phlebotomy, this regimen reduces erythrocytosis without inducing iron deciency. Triadou et al report treating 64 patients with cyanotic heart disease aged 8 to 47 years with hydroxyurea for 2 to 15 years, with dosing adjusted to keep white blood cells (WBC) 3000 and platelets 50,000.26 Compared with a control group, the hydroxyurea group showed a signicant increase in the mean corpuscular volume (MCV) from 92.8 to 109.6 and mean fetal hemoglobin from 0.8% to 2.67%, whereas both groups were maintained at a mean hemoglobin of 21 with phlebotomy.26 Although not supported with data, the authors report clinical improvement was superior with hydroxyurea as compared with phlebotomy only and persisted over a long period of time.

Practical Considerations in Phlebotomy: Target Hematocrit and Blood Volume Replacement

If the treatment course of phlebotomy is chosen, one must decide how much blood to remove and how to maintain appropriate blood volume in a population with compromised cardiac function. To date, there are no human studies dening optimal hematocrit in patients with cyanotic heart disease. Canine studies of experimentally induced normovolemic anemia and polycythemia showed optimal oxygen transport at an average hematocrit of 47.3%.27 Menon et al report a significant improvement in cerebral blood ow and neuropsychologic testing in patients with secondary polycythemia when hematocrit was reduced to less than 45% compared with a hematocrit of 46% to 51%.19 We believe that the available evidence demonstrates that a reduction of hematocrit in this patient population increases oxygen delivery. Unfortunately, a standard optimal hematocrit is unknown and likely to vary for each patient depending on the level of shunting and cardiac output. Phlebotomy based on symptomatology may presently be the only practical approach. Phlebotomy must be followed by infusion of equal volume replacement to maintain intravascular volume and blood ow as well as to provide a dilutional effect to reduce the hematocrit. Perloff et al recommend a 1:1 replacement with 0.9% normal saline,11 whereas Rosenthal suggests replacing the volume of blood removed 1:1 with fresh-frozen plasma infused over 30 to 60 minutes.15 Other authors suggest replacing the extracted volume with a similar volume of colloid such as dextran or hetastarch.21 No form of volume replacement has been shown to be superior to any other in
2006 Lippincott Williams & Wilkins

33

DeFilippis et al

Cardiology in Review Volume 15, Number 1, January/February 2007

12. 13. 14. 15. 16. 17. 18. 19.

heart disease: hematologic management. Ann Intern Med. 1988;109: 406 413. Humphrey PR. Changes in cerebral blood ow relating to haematocrit and viscosity. Scand J Clin Lab Invest Suppl. 1981;156:209 211. Grant P, Patel P, Singh S. Acute myocardial infarction secondary to polycythaemia in a case of cyanotic congenital heart disease. Int J Cardiol. 1985;9:108 110. Yeager SB, Freed MD. Myocardial infarction as a manifestation of polycythemia in cyanotic heart disease. Am J Cardiol. 1984;53:952953. Rosenthal A, Nathan DG, Marty AT, et al. Acute hemodynamic effects of red cell volume reduction in polycythemia of cyanotic congenital heart disease. Circulation. 1970;42:297308. Oldershaw PJ, Sutton MG. Haemodynamic effects of haematocrit reduction in patients with polycythaemia secondary to cyanotic congenital heart disease. Br Heart J. 1980;44:584 588. York EL, Jones RL, Menon D, et al. Effects of secondary polycythemia on cerebral blood ow in chronic obstructive pulmonary disease. Am Rev Respir Dis. 1980;121:813 818. Wade JP, Pearson TC, Russell RW, et al. Cerebral blood ow and blood viscosity in patients with polycythaemia secondary to hypoxic lung disease. BMJ. 1981;283:689 692. Menon D, York EL, Bornstein RA, et al. Optimal hematocrit and blood viscosity in secondary polycythemia as determined from cerebral blood ow. Clin Invest Med. 1981;4:117121.

20. Amitai Y, Blieden L, Shemtov A, et al. Cerebrovascular accidents in infants and children with congenital cyanotic heart disease. Isr J Med Sci. 1984;20:11431145. 21. Strong WB. Complications of polycythemia in patients who have cyanotic congenital heart disease. Pediatr Rev. 1992;13:379 380. 22. Cook JD, Lynch SR. The liabilities of iron deciency. Blood. 1986;68: 803 809. 23. Dallman PR. Manifestations of iron deciency. Semin Hematol. 1982; 19:19 30. 24. Finch CA, Miller LR, Inamdar AR, et al. Iron deciency in the rat. Physiological and biochemical studies of muscle dysfunction. J Clin Invest. 1976;58:447 453. 25. Vongpatanasin W, Brickner ME, Hillis LD, et al. The Eisenmenger syndrome in adults. Ann Intern Med. 1998;128:745755. 26. Triadou P, Maier-Redelsperger M, Krishnamoorty R, et al. Fetal haemoglobin variations following hydroxyurea treatment in patients with cyanotic congenital heart disease. Nouv Rev Fr Hematol. 1994;36:367 372. 27. Murray JF, Gold P, Johnson BL Jr. Systemic oxygen transport in induced normovolemic anemia and polycythemia. Am J Physiol. 1962; 203:720 724. 28. Castle WB, Jandl JH. Blood Viscosity and blood volume: opposing inuences upon oxygen transport in polycythemia. Semin Hematol. 1966;3:193198.

34

2006 Lippincott Williams & Wilkins