Impact of disease genotype and severity of anemia on myocardial extracellular volume fraction in sickle cell disease (SCD) and non-SCD hemolytic anemias

Abstract Background: Sickle cell disease (SCD) is associated with a unique cardiomyopathy characterized by diastolic dysfunction superimposed on an anemia-related high-output state. Diffuse myocardial fibrosis is a key, pathological feature of this cardiomyopathy, which can be quantified by the extracellular volume fraction (ECV) derived from cardiac magnetic resonance imaging (CMR). Although ECV strongly correlates with histologic collagen fraction, it can also be increased by myocardial edema and amyloidosis. Further, while ECV correlates with the severity of anemia in sickle cell anemia, the effect of anemia itself on ECV, potentially independent of fibrosis, is little studied. Published ECV data in SCD are available only for individuals with sickle cell anemia (SS) and sickle-β0-thalassemia (Sβ0). It is not known whether diffuse myocardial fibrosis also occurs in other genotypes of SCD or non-SCD hemolytic anemias. Objective: Study a broad sample of patients with different SCD genotypes and non-SCD hemolytic anemias using CMR to evaluate the effects of diagnosis (specific type of hemolytic anemia) and degree of anemia on ECV. Methods: We studied three groups of individuals based on type of hemolytic anemia: (1) SS and Sβ⁰, (2) sickle-hemoglobin C disease and sickle-β⁺-thalassemia (SC/Sβ⁺), and (3) non-SCD forms hemolytic anemias (e.g., hereditary spherocytosis, hereditary xerocytosis, autoimmune hemolytic anemia, HbH disease). CMR was performed with gadolinium-enhanced T1 mapping to quantify ECV. Hematocrit (Hct) was measured at the time of imaging. Results: We studied 65 individuals (mean age 18.4 years [range: 5.5-57.5]; 54% female): 24 with SS/Sβ0, 22 with SC/Sβ⁺, and 19 with non-SCD forms of hemolytic anemia. In the SS/Sβ0 group, 96% had abnormally increased ECV (normal <25), compared to 86% in SC/Sβ⁺, and 58% in non-SCD. ECV values differed significantly across all groups (P<0.0001); SS/Sβ⁰: 32.2 ± 3.7% (mean ± SD); SC/Sβ⁺: 28.3 ± 2.8%; and non-SCD: 26.0 ± 3.1%. All pairwise comparisons were significantly different: SS/Sβ0 vs SC/Sβ⁺: mean difference –3.89%, P=0.0003; SS/Sβ0 vs non-SCD: –6.15%, P<0.0001; and SC/Sβ⁺ vs non-SCD: –2.26%, P=0.017. Native T1 values, which also reflect myocardial fibrosis but are measured independently of Hct, differed significantly across groups and followed the same gradient observed with ECV: SS/Sβ⁰: 1034 ± 35 ms; SC/Sβ⁺: 1014 ± 38 ms; and non-SCD: 1001 ± 31 ms (P = 0.008). Hct also differed significantly across groups (P<0.0001): SS/Sβ⁰: 28.8±4.2% (mean ± SD); SC/Sβ⁺: 31.4 ± 5.1%; and non-SCD: 36.9 ± 5.3%. Pairwise comparisons were significant except between SS/Sβ⁰ and SC/Sβ⁺ (P=0.18). To determine whether ECV varied by both the degree of anemia and type of hemolytic anemia, we used linear regression with an interaction term (Hct × diagnosis group). In SC/Sβ⁺ (used as the reference group), higher Hct was associated with lower ECV (β = –0.33, P=0.001). There was a significant interaction for SS/Sβ0 (interaction term β = –0.31, P=0.046), indicating a steeper increase in ECV with decreasing Hct compared to SC/Sβ⁺ group. No significant interaction was found for non-SCD (P=0.342). Together, these results indicate that as Hct decreases, ECV increases more steeply in SS/Sβ0 than in SC/Sβ⁺or non-SCD. That is, individuals with SS/Sβ⁰ have a disproportionately higher ECV for any given decrement in Hct compared to the other disease groups. Conclusion: Abnormally increased ECV values, indicative of diffuse myocardial fibrosis, occur in individuals with a range of hemolytic anemias (both SCD and non-SCD). However, the prevalence and degree of ECV elevation are greatest among individuals with more severe SCD genotypes (SS/Sβ0 > SC/Sβ⁺ > non-SCD). Hct correlated inversely with ECV indicating that degree of anemia influences ECV. However, the slope of this relationship was steeper in SS/Sβ0, indicating that ECV values reflect both a disease-specific effect related to sickle hemoglobin and an effect of hemolytic anemia. Whether the effect of anemia arises from hemolysis or pathophysiologic adaptations (with potentially non-fibrotic components) to chronic anemia is not known. Regardless of its causes, increased ECV reflects pathologic myocardial remodeling that may be of prognostic significance. Further studies are planned to elucidate the biological mechanisms of myocardial fibrosis in SCD and to test anti-fibrotic therapies.