Introduction: Sickle cell disease (SCD) is characterized by deoxygenation-induced polymerization of hemoglobin in red blood cells (RBCs), leading to altered RBC rheology, vaso-occlusion, and hemolytic anemia. The chronic hemolytic anemia, experienced in varying degrees by patients with SCD, reduces oxygen-carrying capacity and leads to multiple clinical manifestations, including fatigue and multiorgan damage.

Methods: Systematic literature searches of Excerpta Medica (EMBASE) and MEDLINE (via PubMed) databases were conducted (January 1, 1998, to August 29, 2017) to identify studies reporting on the association of hemoglobin with clinical outcomes in patients with SCD. Abstracts from the previous 2 years of 5 scientific conferences that present studies related to SCD were also evaluated. Citations identified from the databases and grey literature searches were reviewed in a 2-step process by a single reviewer: first-pass title/abstract evaluation followed by full-text assessment. Three quality assessment (QA) scales were utilized to evaluate all included studies.

For quantitative analyses, study findings were separated into categories of outcomes related to hemoglobin. For each category, study findings were aggregated to perform separate meta-analyses assessing the overall magnitude of the association. The summary measure was weighted by study size. Given the variation in hemoglobin levels, we stratified our findings by levels of hemoglobin to examine heterogeneity. Pooled results were analyzed using random effects models to control for within- and between-study variability.

To derive risk ratios associated with hemoglobin change, ratios of means from select studies that reported hazard and odds ratios in meta-analyses by change in hemoglobin between groups for outcomes related to hemoglobin were combined. When studies reported separate values for hemoglobin outcomes in bivariate and multivariate analyses, these were combined within the study, forming a single value prior to the meta-analysis overall for studies in an outcome category. Analyses were performed using STATA 15.2.

Results: A total of 1434 records were initially screened, and 139 studies were identified for inclusion. Results reported are on four key clinical outcomes where the literature was most robust.

Ten studies representing 3397 patients and reporting on hemoglobin- and stroke-related outcomes demonstrated an association between lower hemoglobin levels and a history of stroke, silent cerebral infarct, or increased transcranial Doppler (TCD)-measured cerebral artery velocity; statistical significance was reached in 7 studies. Lower hemoglobin levels were significantly associated with the presence of microalbuminuria among 1579 pediatric patients with SCD in 7 of 8 studies. Twelve studies representing 1465 patients reported an association between lower hemoglobin and elevated estimated pulmonary artery systolic pressure (PASP), defined as elevated tricuspid regurgitant jet velocity (TRV ≥2.5 m/s). Six studies accounting for 3196 patients reported on hemoglobin and mortality among SCD patients and showed hemoglobin was lower across all but one study for deceased vs. living patients.

Meta-analyses conducted to determine aggregate differences across groups among each of these 4 clinical outcomes demonstrated that hemoglobin was statistically significantly lower by approximately 0.4 to 1 g/dL among those experiencing a negative clinical outcome (Table).

In our risk reduction meta-analysis, the risk for negative clinical outcomes decreased at all modeled levels of hemoglobin increase, most notably with improvements in hemoglobin of 1 g/dL or greater (Figure).

Conclusions: In conclusion, comprehensive evaluation of peer-reviewed literature published over the past 20 years demonstrates a significant relationship between chronic anemia and worse clinical outcomes in individuals with SCD. Meta-analyses further demonstrate that even relatively modest differences in hemoglobin are important and support hemoglobin increase by ≥1 g/dL as a relevant therapeutic target. These results suggest that interventions that reduce hemolytic anemia may confer clinical benefit in this patient population.


Ataga:Novartis: Consultancy, Honoraria; Bioverativ: Honoraria, Membership on an entity's Board of Directors or advisory committees; Global Blood Therapeutics: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Modus Therapeutics: Honoraria; Pfizer: Research Funding. Allen:Global Blood Therapeutics: Consultancy. Colby:Xcenda, LLC: Employment. Gittings:Xcenda, LLC: Employment. Agodoa:Global Blood Therapeutics: Employment, Other: Stakeholder.

Author notes


Asterisk with author names denotes non-ASH members.