We conducted a retrospective study on newborns with sickle-cell disease (SCD), born 1995-2009, followed in a multicentre hospital-based network. We assessed patient outcomes, medical care and compliance with the national guidelines published in December 2005. Data from 1033 patients (742 SS/Sβ°-thalassaemia) with 6776 patient-years of follow-up were analysed (mean age 7·1 ± 3·9 years). SCD-related deaths (n = 13) occurred only in SS-genotype patients at a median age of 23·1 months, mainly due to acute anaemia (n = 5, including 2 acute splenic sequestrations) and infection (n = 3). Treatment non-compliance was associated with a 10-fold higher risk of SCD-related death (P = 0·01). Therapeutic intensification was provided for all stroke patients (n = 12), almost all patients with abnormal transcranial Doppler (TCD) (n = 76) or with >1 acute chest syndrome/lifetime (n = 64) and/or ≥3 severe vaso-occlusive crises/year (n = 100). Only 2/3 of patients with baseline haemoglobin <70 g/l received intensification, mainly for other severity criteria. Overall, hydroxycarbamide was under-prescribed, given to 2/3 of severe vaso-occlusive patients and 1/3 of severely anaemic patients. Nevertheless, introduction of the on-line guidelines was concomitant with an improvement in medical care in the 2006-2009 cohort with a trend towards increased survival at 5 years, from 98·3% to 99·2%, significantly increased TCD coverage (P = 0·004) and earlier initiation of intensification of therapy (P ≤ 0·01).

Our study shows that continuous MET has comparable efficacy to the automated method in terms of stroke prevention, decrease in HbS, and iron overload prevention. It is feasible in all hospital settings and is often combined with AET successively over time.

Summary In patients with sickle cell anaemia (SCA), concomitant glucose‐6‐phosphate dehydrogenase (G6PD) deficiency is usually described as having no effect and only occasionally as increasing severity. We analysed sequential clinical and biological data for the first 42 months of life in SCA patients diagnosed by neonatal screening, including 27 G6PD‐deficient patients, who were matched on sex, age and parents' geographic origin to 81 randomly selected patients with normal G6PD activity. In the G6PD‐deficient group, steady‐state haemoglobin was lower (−6·2 g/l, 95% confidence interval (CI), [−10·1; −2·3]) and reticulocyte count higher (247 × 109/l, 95%CI, [97; 397]). The acute anaemic event rate was 3 times higher in the G6PD‐deficient group (P < 10−3). A higher proportion of G6PD‐deficient patients required blood transfusion (20/27 [74%] vs. 37/81 [46%], P < 10−3), for acute anaemic events, and also vaso‐occlusive and infectious events. No significant between‐group differences were found regarding the rates of vaso‐occlusive, infectious, or cerebrovascular events. G6PD deficiency in babies with SCA worsens anaemia and increases blood transfusion requirements in the first years of life. These effects decrease after 2 years of age, presumably as the decline in fetal haemoglobin levels leads to increased sickle cell haemolysis and younger red blood cells with higher G6PD activity.

Children with sickle cell disease (SCD) have a significant vascular morbidity, especially cerebral macrovasculopathy (CV), detectable by transcranial Doppler. This study aimed to identify risk factors for CV using longitudinal biological and clinical data in a SCD newborn cohort followed at the Robert Debre Reference centre (n = 375 SS/Sβ(0) ). Median follow-up was 6·8 years (2677 patient-years). Among the 59 children presenting with CV, seven had a stroke. Overall, the incidence of CV was 2·20/100 patient-years [95% confidence interval (95% CI): 1·64-2·76] and the incidence of stroke was 0·26/100 patient-years (95% CI: 0·07-0·46). The cumulative risk of CV by age 14 years was 26·0% (95% CI: 20·0-33·3%). Risk factors for CV were assessed by a Cox model encompassing linear multivariate modelling of longitudinal quantitative variables. Years per upper-airway obstruction [Hazard ratio (HR) = 1·47; 95% CI: 1·05-2·06] or bronchial obstruction (HR = 1·76; 95% CI: 1·49-2·08) and reticulocyte count (HR = 1·82 per 50 × 10(9) /l increase; 95% CI: 1·10-3·01) were independent risk factors whereas fetal haemoglobin level (HR = 0·68 per 5% increase; 95% CI: 0·48-0·96) was protective. Alpha-thalassaemia was not protective in multivariate analysis (ancillary analysis n = 209). Specific treatment for upper or lower-airway obstruction and indirect targeting of fetal haemoglobin and reticulocyte count by hydroxycarbamide could potentially reduce the risk of CV.