Hemolysis-associated priapism in sickle cell disease

Vikki G. Nolan, Diego F. Wyszynski, Lindsay A. Farrer and Martin H. Steinberg


Priapism, although uncommon in the general population, is one of the many serious complications associated with sickle cell disease (SCD). Few studies have described the clinical and hematologic characteristics of individuals with priapism and SCD. Using data from the Cooperative Study for Sickle Cell Disease, we assembled 273 case subjects with priapism and 979 control subjects. Case subjects, compared with control subjects, had significantly lower levels of hemoglobin; higher levels of lactate dehydrogenase, bilirubin, and aspartate aminotransferase; and higher reticulocyte, white blood cell, and platelet counts. These findings suggest an association of priapism with increased hemolysis. Hemolysis decreases the availability of circulating nitric oxide, which plays an important role in erectile function.


Sickle cell disease (SCD) is an inherited condition caused by a point mutation in the β-globin gene (HBB), resulting in the substitution of valine for glutamic acid at position 6 of the β-globin chain (Glu6Val). This mutation results in the abnormal sickle hemoglobin (HbS). When deoxygenated, HbS polymerizes, damaging the sickle erythrocyte.1 These sickle cells are short lived and can interact with endothelial cells, leukocytes, platelets, and other plasma components to initiate the vasoocclusive manifestations associated with SCD.

Priapism is a sustained, painful, and unwanted erection of the penis that pathophysiologically is the result of either increased arterial inflow (ie, high flow) or, more commonly, the failure of venous outflow (ie, low flow), resulting in blood trapping within the erectile bodies. Although uncommon in the general population,2 priapism was recognized as a serious complication of SCD as early as 1934.3 Since then, many researchers have estimated its incidence and prevalence in various populations and the choices for its treatment.2-16

In general, about 30% of males with SCD under the age of 20 years reported having had at least one episode of priapism,8,11,17 whereas frequencies of 30% to 45% are estimated for adult men.18 Among patients with priapism, about 75% had their first occurrence before the age of 20 with the mean age being around 12 to 15 years.17,18

Few published studies, however, have described the clinical characteristics of their participants with priapism. Using data from the Cooperative Study of Sickle Cell Disease (CSSCD), we assembled a large cohort of patients with priapism to study their clinical, laboratory, demographic, and social characteristics.

Patients, materials, and methods

Patient database

The CSSCD enrolled and followed more than 4000 patients with SCD who had visited 1 of 23 participating clinical centers across the United States between 1978 and 1998.19,20 One of the primary aims of the CSSCD was to collect data on the clinical course of SCD from birth to death. Because of the clinical variability of SCD, data were collected on a large number of possible manifestations of the disease, including priapism. All incident events of priapism causing a participant to seek medical care were carefully recorded, and patients' medical records were reviewed by trained clinicians. These studies were approved by the Institutional Review Board of Boston University/Boston Medical Center.

Selection of case and control subjects

Among the more than 4000 patients enrolled in the CSSCD, there were 1737 males with sickle cell anemia (HbSS), HbSS-α thalassemia (HbSSα; defined as patients with either 1 or 2 deleted α-globin genes), or HbSC disease (compound heterozygosity for HbS and HbC) from which case and control subjects were drawn. A case is defined as a male CSSCD participant who had at least one episode of a painful erection of the penis which lasted longer than 1 hour and that was severe enough for the patient to seek medical care. This definition, therefore, includes both stuttering priapism (multiple episodes of erection lasting less than 3 hours, unrelated to sexual stimulation, and resolving spontaneously without loss of erectile function)21 and acute prolonged priapism. Control subjects were the remainder of male CSSCD participants who had never experienced an episode of priapism. For case subjects younger than 20 years of age, 5 age-matched control subjects were chosen randomly and the rest was excluded from analysis to avoid an overrepresentation of very young control subjects.

Statistical analyses

Clinical, hematologic, and sociodemographic characteristics were examined using multivariate logistic and linear regressions. Age-adjusted odds ratios (ORs) and their 95% confidence intervals (CIs) were determined for categoric variables, and age-adjusted means were calculated for continuous measures. Multivariate analyses were performed using backward, forward, and stepwise logistic regression. All statistical analyses were done using SAS software (version 8.2; SAS Institute, Cary, NC).


Two hundred seventy-three patients had a confirmed history of at least 1 episode of priapism or were observed to have an episode of priapism (Table 1). Nine hundred seventy-nine males without a reported episode of priapism served as control subjects. Case subjects were on average 4 years older than control subjects (P < .001). Therefore, all of the comparisons reported in Tables 1 and 2 are age adjusted. A higher proportion of case subjects than of control subjects had HbSS (case subjects, 72%; control subjects, 54%); however, HbSS with coincident α thalassemia and HbSC disease were more frequent among control subjects (SSα: case subjects, 16%; control subjects, 19%; HbSC; case subjects, 11%; control subjects, 27%), indicating that patients with HbSS with coincident α thalassemia and patients with HbSC are significantly less likely to have priapism when compared with patients with HbSS without coincident α thalassemia. The haplotype of the β-like globin gene cluster was not associated with priapism. Priapism was, however, found to be associated with other clinical manifestations of SCD, such as ischemic stroke, avascular necrosis, acute chest syndrome, and acute painful episodes.

Table 1.

Phenotypic characteristics of case subjects with priapism and control subjects in a population of patients with SCD, presented as frequency (% of nonmissing)

Table 2.

Laboratory characteristics of cases with priapism and controls in a population of patients with SCD, presented as age-adjusted means ± standard error

There was evidence for a modest but statistically significant association between folate use and the occurrence of priapism. Similarly, alcohol consumption yielded slightly significant evidence of an association, whereas smoking showed no association. Taking supplemental iron, blood-pressure medication, and/or heart medication was not found to be associated with the presence of priapism. Agents that might increase nitric oxide levels such as statins and angiotensin-converting enzyme (ACE) inhibitors were not used in these patients at the time of study, and too few patients used nitroglycerin for analysis to be done.

Patient ages, blood pressure, and selected laboratory data are shown in Table 2. Patients with priapism had lower hemoglobin levels, higher levels of lactate dehydrogenase (LDH), bilirubin, and aspartate aminotransferase (AST); and higher reticulocyte, white blood cell, and platelet counts than did control subjects. All of these differences were highly statistically significant (P < .001, except for AST, P < .01). Mean systolic and diastolic blood pressure, and levels of fetal hemoglobin (HbF), creatinine, and alanine aminotransferase (ALT) were not associated with priapism. When only patients with HbSS were analyzed, individuals with priapism continued to have lower hemoglobin levels (P < .004) and higher levels of LDH (P < .04), reticulocytes (P = .01), and platelets (P = .03) than did control subjects. Multivariate analyses of both the complete cohort of case and control subjects and the patients with HbSS alone showed that LDH, reticulocyte count, and platelet count were most significantly associated with priapism.


Few studies have attempted to understand the vascular basis of priapism, a common complication of SCD. We studied the association of priapism with selected clinical and laboratory features of sickle hemoglobinopathies using data collected in the CSSCD.19,20 This database included individuals with HbSS, HbSSα, and HbSC disease. Priapism was 1.4 times more common among individuals with HbSS than among those with HbSSα and 2.7 times more common than among individuals with HbSC disease.

Our data suggest that priapism in SCD appears to be affected by the rate of hemolysis. Hemolytic anemia varies in intensity among the different genotypes comprising SCD and is regulated by cell density and membrane injury. Patients with HbSS are more severely anemic than individuals with HbSSα who, in turn, are usually more anemic than patients with HbSC disease.22-25 Even among patients with a single genotype, the hemoglobin concentration is variable. In HbSS, the most common and clinically severe form of the disease, red cell 51Cr survival ranges between 2 and 21 days and is reflected in similarly wide variations of hemoglobin concentration, reticulocyte count, bilirubin level, and LDH, all of which are clinical markers of hemolysis.22,24 Recent work suggests that hemolytic anemia is the driving force behind some complications of SCD because of its effects on nitric oxide (NO) bioavailability.26 NO, by binding soluble guanylate cyclase, converts cyclic guanosine triphosphate (cGTP) to cyclic guanosine monophosphate (cGMP), relaxing vascular smooth muscle and causing vasodilation. Plasma hemoglobin liberated from intravascularly destroyed sickle erythrocytes consumes NO-producing methemoglobin while catalyzing the production of oxidant radicals and adhesion molecules such as vascular cell adhesion molecule 1.26 The normal balance of vasoconstriction/vasodilation is therefore skewed toward vasoconstriction.

Castro et al27 and Gladwin et al28 found that markers of hemolysis and reduced NO availability, such as total hemoglobin concentration, bilirubin level, LDH, and the arginine-ornithine ratio, were associated with sickle cell pulmonary hypertension which affects about 30% of patients with HbSS and is a major risk factor for premature death. Among the vasoocclusive complications of SCD, only priapism and stroke were more common in patients with pulmonary hypertension.28

Penile NO, originating from both neurons and endothelium, modulates penile vascular and smooth muscle relaxation.29 Low-flow priapism, the most common form in SCD, may be the result of venous outflow occlusion because of reduced NO availability.30-32 Priapic activity and pronounced erectile responses to stimulation are seen in transgenic mice in which both neuronal (Nos1) and endothelial (Nos3) genes are knocked out and in sickle transgenic mice expressing HbS and lacking any murine hemoglobin.33 This was a consequence of dysregulated activity of downstream signaling in the phosphodiesterase type 5 (Pde5a) pathway. Depleted NO may cause Pde5a downregulation, less enzyme available for hydrolysis of cGMP, a reduction in its effector protein kinase G, and supersensitization to local cGMP from other sources. In SCD, NO deficiency secondary to hemolysis may downregulate PDE5A and its downstream effectors so that non–NO-mediated uninhibited increases in cGMP result in uncontrolled erection. Sildenafil and similar PDE5A inhibitors may paradoxically alleviate or prevent priapism in these states. Like pulmonary hypertension and priapism, stroke, another subphenotype of sickle cell anemia, is associated with hemoglobin concentration.34,35

Fetal hemoglobin (HbF) is a major modifier of some subphenotypes of SCD, including osteonecrosis,36 acute chest syndrome,37 and acute painful episodes.38 HbF was not however associated with pulmonary hypertension, stroke, or priapism.28,35 This suggests that the role of HbF in reducing the hemolytic rate may be dominated by other determinants of hemolysis or that very high HbF concentrations are needed to reduce the likelihood of these clinical events. The lack of an association of HbF with priapism, pulmonary hypertension, and stroke may also be a result of an analytical approach that does not account for the interactions of many other genetic modifiers with HbF in modulating these phenotypes.39 Although it should not be discounted as a modulator of these subphenotypes, HbF appears to have a minimal direct protective effect.

Coincident α thalassemia has disparate effects in HbSS40 because it protects patients with HbSS from priapism and stroke35 while increasing the chance of developing osteonecrosis; some but not all studies have suggested an adverse effect on the incidence of acute chest syndrome and acute painful episodes.37,38,41,42 The risk of stroke, pulmonary hypertension, and priapism is likely to be decreased because α thalassemia reduces hemolysis.

Leukocyte counts were increased in patients with priapism compared with control subjects. Neutrophils are effectors and markers of inflammation and tissue damage.43 In sickle cell anemia, increased granulocyte counts are an adverse risk factor for survival and have been associated with events like acute chest syndrome and stroke in children.44 Activated neutrophils are present during acute painful episodes and have increased adherence to endothelial cells.45 Increased leukocyte and platelet counts in priapism case subjects compared with control subjects may reflect their more active hematopoiesis because of increased hemolysis. Sickle reticulocytes, whose numbers are increased in response to hemolysis, display receptors and ligands responsible for their adherence to endothelium, providing an additional link between hemolytic anemia and sickle vasculopathy.46-48

Hemolytic anemia-induced phenotypes are likely to be improved by transfusion and agents that increase NO bioavailability but helped to a lesser extent by drugs that induce HbF expression, unless the response is very robust. Administration of arginine, the substrate of the NO synthases, has been associated with a reduction in pulmonary artery pressure in sickle cell anemia; transfusions have been found to prevent stroke.49,50 Sildenafil, a PDE5 inhibitor, may also be effective in treating pulmonary hypertension.51,52

Hydroxyurea appears to have little effect in sickle cell pulmonary hypertension,28 and yet it may help prevent stroke in susceptible children53 and also has an effect in some instances of priapism. Five patients with stuttering priapism, whose average age was 20 years, experienced relief during hydroxyurea treatment.13 Their average increase in hemoglobin was 16 g/L (1.6 g/dL) while HbF increased by 8.6%. Hydroxyurea, which increases hemoglobin concentration in HbSS by increasing HbF concentration and reducing hemolysis,54 is also a NO donor and may directly activate cellular guanylate cyclase, increasing cGMP.55,56 The exact role of hydroxyurea in the management of priapism is unknown. Because young children have a superior HbF response to hydroxyurea than do adults, and, as the magnitude of this response is likely to influence its beneficial effects, hydroxyurea could be useful in the management of priapism in some individuals.

In summary, the vascular basis of priapism in SCD is poorly understood, and priapism may be one reflection of the vasculopathy of sickle cell disease. We studied the association of sickle cell priapism with selected clinical and laboratory features of SCD using data collected in the CSSCD19,20 and found priapism was most common in individuals with the greatest amount of hemolysis. This association suggests that priapism may be one vascular manifestation of reduced NO availability. In related work we found an association of polymorphisms and haplotypes of the Klotho (KL) gene with sickle cell priapism.57 KL encodes a membrane protein that regulates many vascular functions, including NO release.58 Inhaled NO, arginine, or even sildenafil59 may be worthy of study for the management of SCD-associated priapism.


We thank the investigators of the Cooperative Study of Sickle Cell Disease who obtained clinical and laboratory data on our patients and blood samples for DNA-based studies. Dr Stephen H. Embury performed the analysis of the β and α-globin genotype.


  • Reprints:
    Martin H. Steinberg, Center of Excellence in Sickle Cell Disease, E248, Boston Medical Center, 88 E Newton St, Boston, MA 02118; e-mail: mhsteinb{at}
  • Prepublished online as Blood First Edition Paper, June 28, 2005; DOI 10.1182/blood-2005-04-1594.

  • Supported by the National Heart, Lung, and Blood Institute (NHLBI) (grant HL R01 68970) (M.H.S.) and (T32 HL007501) (V.G.N.).

  • V.G.N. was responsible for drafting the paper, experiment design, data analysis, literature review, and revisions; D.F.W. partook in data analysis, manuscript revisions, and data analysis; L.A.F. was responsible for manuscript revision; and M.H.S. conceived of the work and aided in the initial draft and subsequent revisions, literature analysis. and study design.

  • An Inside Blood analysis of this article appears at the front of this issue.

  • The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 U.S.C. section 1734.

  • Submitted April 20, 2005.
  • Accepted June 22, 2005.


View Abstract