The relationship between the severity of hemolysis, clinical manifestations and risk of death in 415 patients with sickle cell anemia in the US and Europe

Abstract

The intensity of hemolytic anemia has been proposed as an independent risk factor for the development of certain clinical complications of sickle cell disease, such as pulmonary hypertension, hypoxemia and cutaneous leg ulceration. A composite variable derived from several individual markers of hemolysis could facilitate studies of the underlying mechanisms of hemolysis. In this study, we assessed the association of hemolysis with outcomes in sickle cell anemia. A hemolytic component was calculated by principal component analysis from reticulocyte count, serum lactate dehydrogenase, aspartate aminotransferase and total bilirubin concentrations in 415 hemoglobin SS patients. Association of this component with direct markers of hemolysis and clinical outcomes was assessed. As primary validation, both plasma red blood cell microparticles and cell-free hemoglobin concentration were higher in the highest hemolytic component quartile compared to the lowest quartile (P≤0.0001 for both analyses). The hemolytic component was lower with hydroxyurea therapy, higher hemoglobin F, and alpha-thalassemia (P≤0.0005); it was higher with higher systemic pulse pressure, lower oxygen saturation, and greater values for tricuspid regurgitation velocity, left ventricular diastolic dimension and left ventricular mass (all P<0.0001). Two-year follow-up analysis showed that a high hemolytic component was associated with an increased risk of death (hazard ratio, HR 3.44; 95% confidence interval, CI: 1.2-9.5; P=0.02). The hemolytic component reflects direct markers of intravascular hemolysis in patients with sickle cell disease and allows for adjusted analysis of associations between hemolytic severity and clinical outcomes. These results confirm associations between hemolytic rate and pulse pressure, oxygen saturation, increases in Doppler-estimated pulmonary systolic pressures and mortality (Clinicaltrials.gov identifier: NCT00492531).

Figure 1.

Relationships among the hemolytic component and the markers from which it is derived in all sickle cell anemia patients. Pearson’s correlation coefficient is provided for each relationship. All P<0.0001.

Figure 2.

Distribution of plasma RBC microparticle counts, a direct marker of intravascular hemolysis, by extreme quartiles of hemolytic component in the subset of sickle cell anemia patients without detectable hemoglobin A. (A) Flow cytometric approach to identify plasma RBC microparticles. The 4-quadrant dot plots show two population events: glycophorin A^high (GPA^high) Annexin V⁺ events (in red) and glycophorin A^intermediate (GPA^int) Annexin V⁺ events (in blue) in both the lowest and highest quartile samples. The GPAhigh Annexin V⁺ and GPA^int Annexin V⁺ events reflect two discrete populations that met the definition of RBC microparticles. The lower 2 panels indicate the side scatter (SSC) and forward scatter (FSC) distribution of the 7.6 micron counting bead (black), 3 micron (violet), 0.9 micron (orange), and 0.5 micron (green) beads relative to RBC microparticles (red and blue populations). Note that both y and x axes are in log scale. The number of events for each population is indicated in the corresponding color, with acquisition set to achieve 1000 events of 3 micron beads in each sample for an added measure of standardization. Calculation of RBC microparticle counts per microliter (μL) were obtained using [the absolute number of 7.6 micron beads added to each sample (ie. 116,000) X (ratio of 7.6 micron bead events: microparticle events)] divided by 10 as indicated in the Design and Methods section. (B) Distribution of GPA^high and GPA^int RBC microparticles per microliter (microparticles/μL) across the lowest and highest quartile of hemolytic component values. *, P=0.0001. Each point represents an individual sample and lines indicate the median and interquartile range for each group. Points represented in red reflect the 2 samples illustrated in the dot plots above in A.

Figure 3.

Distribution of plasma cell-free hemoglobin, a direct interquartile marker of intravascular hemolysis, by extreme quartiles of hemolytic component in the subset of sickle cell anemia patients without detectable hemoglobin A. Plasma cell-free hemoglobin concentrations are shown for the lowest and highest quartiles in microM heme. The box plots indicate the median and interquartile range. The whiskers indicate 1.5 times the interquartile range from the nearest quartile.

Figure 4.

Distribution of hemolytic component in all patients with sickle cell anemia according to variables known to influence or reflect hemolysis. (A) Hydroxyurea treatment (mean in each group shown by blue cross). (B) Alpha thalassemia genotype. (C) Percentage of hemoglobin F. (D) Hemoglobin concentration.

Figure 5.

Survival in all patients with sickle cell anemia according to the hemolytic component. (A) According to duration of follow up. (B) According to age at death.