Abstract
Sickle cell anemia (SS) is a phenotypically variable disease whose course is difficult to predict. The Cooperative Study of Sickle Cell Disease (CSSCD) found that dactylitis in the 1st year of life predicted adverse outcomes in later childhood. We aimed to determine whether early vaso-occlusive complications, including dactylitis, were prognostic in the Dallas Newborn Cohort. We studied all cohort members with SS or sickle-β0-thalassemia who were <1 yr of age at their first clinic visit, ≥5 yrs of age at last follow-up, and who had complete records. We defined 3 potential “early” (occurring in the first 3 yrs of life) predictors: any hospitalization for (1) pain crisis (non-dactylitis), (2) dactylitis, or (3) acute chest syndrome (ACS). We studied the associations of these predictors with the following “late” (occurring on or after the 3rd birthday) outcomes: death of any cause; overt stroke; use of hydroxyurea (HU), chronic transfusion (CT), or stem cell transplantation (SCT); and mean number of hospitalizations for late pain crisis and ACS. Late pain and ACS episodes were enumerated for each patient between the 3rd birthday and the last clinical encounter or the start date of a disease-modifying therapy (HU, CT, or SCT), whichever was first. Mean number of pain and ACS events was analyzed for the late follow-up period in total and in 2-yr intervals. Outcomes up to age 20 were included. Two-sided Fisher exact and t-tests were used appropriately. There were 264 subjects (256 SS; 54.9% male). Mean age at first visit was 4.1±2.3 mos (±S.D.) and mean follow-up was 12.1±4.3 yrs. The following early hospitalizations occurred: 53 subjects (20.1%) had pain crisis; 16 (6.1%) had dactylitis, and 85 (32.9%) had ACS. There were 5 deaths and 30 overt strokes. Sixty-six subjects were treated with HU (37), CT (40), and/or SCT (1). We found that subjects who had early pain, dactylitis, or ACS (compared with those who did not) were not more likely to die (1.7 vs. 2.1%; P>0.99) or have a stroke (12.2 vs. 10.3%; P=0.69). However, the use of a disease-modifying therapy was more common among subjects who had early pain (37.7 vs. 19.9%; P=0.01) and ACS (37.6 vs. 16.2%; P<0.001), but not dactylitis (18.8 vs. 23.6%; P>0.99). This prediction held only for HU use when the treatments (HU, CT, or SCT) were analyzed separately. Subjects who experienced early pain or ACS had on average a 2.2-fold (P=0.02) or 2.1-fold (P=0.01), respectively, higher number of late pain crises between ages 3 and 11, but not beyond (all P>0.05). Dactylitis did not predict a higher number of late painful events at any age (all P>0.05). Likewise, neither early pain nor dactylitis was associated with a higher number of hospitalizations for late ACS (all P>0.05). However, subjects who had early ACS had a 1.7 to 3.6-fold higher mean number of ACS events throughout all late age groups (all P<0.05). In summary, early hospitalization for pain, dactylitis, or ACS did not predict death or stroke. Early pain and ACS were associated with use of HU in later childhood, but not CT or SCT. Early pain and ACS predicted an increased number of hospitalizations for pain until age 11, but not beyond. Early ACS was a strong predictor of recurrent ACS throughout childhood. Notably, we found that hospitalization for dactylitis had no particular prognostic significance, unlike the CSSCD. In conclusion, the prognostic significance of early vaso-occlusive complications is limited.
Prognostic significance of plasma bilirubin in sickle cell diseases