Quantitative analysis of erythrocytes containing fetal hemoglobin (F cells) in children with sickle cell disease

Variation in the level of fetal hemoglobin (HbF) accounts for much of the clinical heterogeneity observed in patients with sickle cell disease (SCD). The HbF level has emerged as an important prognostic factor in both sickle cell pain and mortality, and a % HbF of 10-20% has been suggested as a threshold level for diminished clinical severity. The number of erythrocytes that contain HbF (termed F cells) may also be critically important, as F cells resist intravascular sickling and have preferential in vivo survival. Since F cells can be enumerated with high accuracy using flow cytometry methods, we prospectively studied a cohort of 242 children with SCD. Children with HbS and hereditary persistence of fetal hemoglobin (S/HPFH) had essentially 100% F cells. In contrast, children with homozygous sickle cell anemia (HbSS), HbS/beta0 thalassemia, or HbS/beta+ thalassemia had significantly lower mean % F cell values (55.9, 61.6, and 51.3%, respectively; P < 0.001), and children with HbSC had even fewer F cells (27.0%; P < 0.001). There was a highly significant correlation between the % F cells and the log (% HbF), which was observed for the total population of children (r = 0.95, P < 0.001), as well as for each of the individual subgroups of children with HbSS (r = 0.94, P < 0.001), HbSC (r = 0.89, P < 0.001), or HbS/beta0 thalassemia and HbS/beta+ thalassemia (r = 0.95, P <0.001). This logarithmic correlation between % F cells and % HbF has not been previously described and has important implications for the pharmacologic manipulation of HbF in patients with SCD.