Implementation of hospital-based sickle cell newborn screening and follow-up programs in Haiti.

ABSTRACT: One in 120 children are born with sickle cell disease (SCD) in Haiti. However, health care challenges include isolated newborn screening (NBS) activities and lack of transcranial Doppler (TCD) ultrasound to assess stroke risk. The implementation activities of the Comparative Study of Children in Haiti and Miami with Sickle Cell Disease involved both NBS and TCD ultrasound implementations at 4 Haitian clinical sites. We hypothesized that hospital-based newborn SCD screening and follow-up programs would be feasible at Haiti. A traditional NBS laboratory method with dried blood samples was performed at 3 Port-au-Prince sites, and the traditional method plus point-of-care (POC) testing was used at the 2 northern sites. The rate of clinical follow-up for newborns with SCD as the outcome for the NBS intervention was compared with that of the NBS method. The NBS programs identified SCD in 0.77% of 8224 newborns over a 24-month period. In the rural hospital assigned to the combination screening, 56% of newborns identified with POC testing returned for follow-up, compared with 0% when POC was not available (P = .044). Newborns who tested positive for SCD and children aged <6 years with SCD at the clinical sites were eligible for study follow-up. Accrual was successful: 165 participants (mean age, 42 months; 53% males; 93% hemoglobin SS) were recruited and received oral penicillin. TCD ultrasound screening was hampered by poor internet connections and trained staff leaving Haiti, with only 1 active site conducting screening. Despite challenges, the implementation of NBS and sickle cell programs in Haiti is feasible. We are in the process of understanding how to mitigate implementation limitations.

Newborn Screening for Sickle Cell Disease Using Point-of-Care Testing in Low-Income Setting.

BACKGROUND: Newborn screening provides early diagnosis for children with sickle cell disease (SCD), reducing disease-related mortality. We hypothesized that rapid point-of-care (POC) Sickle SCAN would be reliable in Haiti and would assist newborn screening. METHODS: Dried blood specimens were obtained from infant heel sticks and analyzed by isoelectric focusing (IEF) at a public hospital in Cap-Haïtien during a 1-year period. A total of 360 Guthrie cards were also analyzed for quality assurance by high-performance liquid chromatography at the Florida Newborn Screening Laboratory. In addition, two-thirds of the infants were also screened by the POC to assess differences with the IEF. The hemoglobinopathy incidence and the specificity and sensitivity of the POC scan were assessed. RESULTS: Overall, 1.48% of the children screened positive for SCD. The specificity and the sensitivity of POC Sickle SCAN were 0.97 (confidence interval 0.95-0.99) and 0.90 (confidence interval 0.55-1.00), respectively, relative to high-performance liquid chromatography gold standard. The confirmatory testing rate was 75% before POC and improved to 87% after POC was added for dual screening. Confirmatory testing revealed that 0.83% of children screened had SCD. Children who screened positive for SCD by POC started penicillin earlier, had their first pediatric follow-up a median of 38 days earlier, and received antipneumococcal vaccination on time when compared with those who screened positive for SCD by IEF alone. CONCLUSIONS: The observational study revealed a high incidence of SCD among Haitian newborns. Sickle SCAN had excellent specificity and sensitivity to detect SCD during newborn screening and shortened health care access for children positive for SCD.

Increased prevalence of false positive hemoglobinopathy newborn screening in premature infants.

BACKGROUND: The objective was to investigate the specificity of the hemoglobinopathy newborn screening in premature neonates as compared to term neonates. PROCEDURE: The screening results from infants suspected to have hemoglobinopathy disease identified by the Florida Newborn Screening Program for years 2002-2007 were compared to the corresponding confirmatory testing. The risks for false positives for preterm and full term newborns were calculated by Chi-square or the Cochran-Armitage test for trend. Isoelectric focusing and HPLC were the methods of hemoglobin screening. RESULTS: Over 2,300 neonates (1/576 neonates born in Florida) were suspected to have hemoglobinopathy. The most common abnormal pattern in term and preterm infants (gestational age 22-36 weeks) suggesting disease at screening was FS. Overall, 93% of the children who screened positive for FCA and 64% of infants identified with FSA were later confirmed with trait. FSC was confirmed in 96% of the cases in both preterm and term infants. Compared to term newborns, preterm newborns were more likely to have a false positive result for FS or FC at screening. Twenty-three percent of preterms with FS and 59% of preterms with FC were diagnosed as traits by confirmatory testing, compared to only 2% and 6% for term infants (P < 0.001). CONCLUSIONS: As compared to term newborns, more preterm newborns with trait were misidentified as having sickle cell anemia or hemoglobin C at screening. We speculate that abnormal hemoglobins may precede the development of hemoglobin A during fetal life.

Recruitment of infants with sickle cell anemia to a Phase III trial: data from the BABY HUG study.

BACKGROUND: Protocol-eligible subjects may not be candidates for research participation or may decline. To determine factors that affected accrual, we evaluated enrollment in BABY HUG, a multi-center, randomized, placebo-controlled Phase III trial of hydroxyurea (HU) in infants with sickle cell anemia. METHODS: An anonymized registry of potential subjects served as the primary source of data. Study coordinators considered all infants less than age 18 months with a hemoglobin FS diagnosis on newborn screening. Data included the number of potentially eligible subjects, whether parents were approached, and reasons for participating or declining. RESULTS: Of 1106 potential participants, 28% were not approached for reasons such as prior poor adherence to clinical care. Interested families expressed willingness to contribute to medical knowledge (51%), hope of being randomized to receive hydroxyurea (51%), and desire for closer clinical care (51%) as reasons for participating. Disease severity or the perception that their child was ill had less impact on willingness to participate (16%). Parents who declined cited fear of research (19%), transportation problems (14%), and the demanding nature of the study (25%). Ultimately, 234 (21%) gave informed consent, with little variability of acceptance rates among sites. Importantly, the number of subjects enrolled correlated with the number of families that were approached. Sites that excluded patients based on clinical/psychosocial biases were not more successful in recruiting than those who approached all eligible subjects. CONCLUSION: Large, demanding clinical trials require an adequate pool of potential participants. Approaching all potentially eligible patients without predetermined biases enhances success in recruitment.