Longitudinal Follow-up From Newborn Screening Reveals Deletional Hemoglobin H Disease and Hemoglobin H Constant Spring Disease Are Distinct Thalassemia Syndromes

Alpha thalassemia disorders are rapidly increasing in North America. This has resulted in proposals for universal newborn screening (NBS) for hemoglobin H disease. However, the institution of routine newborn screening and construction of guidelines for early intervention requires longitudinal clinical data before setting national goals. Since 1995, California has performed universal screening for alpha thalassemia disorders. The longitudinal follow up of data from patients with hemoglobin H disorders diagnosed in the asymptomatic period provides essential information needed for formulating public health policy.

Hemoglobin H disorders were diagnosed by high performance liquid chromatography with multiplex GAP-PCR assay to determine deletional hemoglobin H disease (deletion of 3 α globin genes, HbH) and the non-deletional hemoglobin H Constant Spring (α⁰ thalassemia with Constant Spring mutation, HCS). Longitudinal clinical data for all patients from the Northern California Thalassemia Center were analyzed. Ethnicity, growth data, clinic visits, hospitalizations, complications including splenectomy, transfusion, and iron overload were monitored. Quantitative liver iron concentration was determined by ferritometer.

86 patients predominantly diagnosed through NBS were longitudinally followed. Out of these, 60 (70%) had HbH, 23 (27%) had HCS and 3 (3%) had other forms of hemoglobin H disease. The parental ethnicity in HbH was 79% Asian, 6% Hispanic, and 15% African-American (in one or both parents). All patients with HCS were of Asian ethnicity. Longitudinal data for hemoglobin revealed that anemia was more severe in HCS at all ages (p<0.001). Mean hemoglobin in HbH increased from 8.8 g/dL (6.9-10.6 g/dL) at 6 months to 9.4 g/dL (7.9-11.5 g/dL) at 5 years (p<0.001). However, mean hemoglobin in HCS remained unchanged from 7.4 g/dL (5.8-9.9 g/dL) at 6 months to 7.2 g/dL (3.8-8.7 g/dL) at 5 years (p=ns). There was no hemoglobin value <6.7 g/dL in 237 patient-years of observation of 60 patients with HbH. Compared to HbH, red blood cells in HCS had higher mean corpuscular hemoglobin (18.6 versus 16.6 pg, p<0.001) and mean corpuscular volume (65.2 versus 54.0 fL, p<0.001). The mean absolute reticulocyte count was 88.2 ×10³/μL in HbH versus 235.1 ×10³/μL in HCS (p<0.001), while the mean serum bilirubin was 0.56 mg/dL and 2.60 mg/dL, respectively (p<0.001).

Clinical severity and complications were markedly worse in HCS in contrast with HbH. Growth was delayed in HCS with mean weight-for-age Z-score -0.91 compared with -0.06 in HbH (p<0.001). The mean height-for-age Z-score was also lower in HCS (-1.29) compared with HbH (-0.43, p<0.001). The striking susceptibility to acute worsening of anemia with infections requiring urgent blood transfusion was observed in HCS, but not in HbH. The probability of receiving one or more blood transfusion by 20 years was 3% in HbH and 82% in HCS (p<0.001). Transfusions in HCS were required for 13% infants and median transfusion-free survival was 6 years. Splenectomy improved hemoglobin by 2.9 g/dL (0.4 to 4.0 g/dL, p=0.012) and reduced transfusions in HCS. Iron overload, measured by serum ferritin and liver iron concentration, developed during the first decade in HCS and increased during follow up. Median ferritin in HCS between 12 –17 years was 330 ng/mL (66-1420 ng/mL). Serum ferritin in HbH did not increase between 0–18 years (median 40 ng/mL, range 5–182 ng/mL), but older patients showed strong positive correlation between age and ferritin (p<0.001). In patients with HbH or HCS undergoing ferritometer examination, the degree of serum ferritin elevation underestimated the liver iron concentration.

Our data support the utility of a universal NBS program, particularly in areas where α^CS mutation is prevalent, since young infants with HCS can develop life-threatening anemia. HCS is a serious disease that needs close follow-up by a specialty thalassemia center to plan for emergency and elective transfusions, measure iron overload, monitor growth failure and evaluate the need for splenectomy. In contrast, HbH is asymptomatic during infancy and childhood; its complications are age-dependent, and monitoring for hemosiderosis and growth failure is more important in older children. In summary, HCS should be recognized as a thalassemia syndrome distinct from HbH with a different screening and treatment approach.

No relevant conflicts of interest to declare.