Transcranial Doppler (TCD) is used to detect children with sickle cell anemia (SCA) who are at risk for stroke, and transfusion programs significantly reduce stroke risk in patients with abnormal TCD. We describe the predictive factors and outcomes of cerebral vasculopathy in the Cré teil newborn SCA cohort (n ‫؍‬ 217 SS/ S␤ 0 ), who were early and yearly screened with TCD since 1992. Magnetic resonance imaging/magnetic resonance angiography was performed every 2 years after age 5 (or earlier in case of abnormal TCD). A transfusion program was recommended to patients with abnormal TCD and/or stenoses, hydroxyurea to symptomatic patients in absence of macrovasculopathy, and stem cell transplantation to those with human leukocyte antigengenoidentical donor. Mean follow-up was 7.7 years (1609 patient-years). The cumulative risks by age 18 years were 1.9% (95% confidence interval [95% CI] 0.6%-5.9%) for overt stroke, 29.6% (95% CI 22.8%-38%) for abnormal TCD, which reached a plateau at age 9, whereas they were 22.6% (95% CI 15.0%-33.2%) for stenosis and 37.1% (95% CI 26.3%-50.7%) for silent stroke by age 14. Cumulating all events (stroke, abnormal TCD, stenoses, silent strokes), the cerebral risk by age 14 was 49.9% (95% CI 40.5%-59.3%); the independent predictive factors for cerebral risk were baseline reticulocytes count (hazard ratio 1.003/L ؋ 10 9 /L increase, 95% CI 1.000-1.006; P ‫؍‬ .04) and lactate dehydrogenase level (hazard ratio 2.78/1 IU/mL increase, 95% CI1.33-5.81; P ‫؍‬ .007). Thus, early TCD screening and intensification therapy allowed the reduction of stroke-risk by age 18 from the previously reported 11% to 1.9%. In contrast, the 50% cumulative cerebral risk suggests the need for more preventive intervention. (Blood. 2011;117(4):1130-1140)

513 Background: The accelerated hemolysis in SCA increases bilirubin excretion and the susceptibility to pigment gallstone formation. Uridine diphosphate-glucuronosyltransferase (UGT1A1) gene polymorphism and alpha-thalassemia are factors known to modify this risk. However, their relative influence and other factors such as SCA disease severity and the use of ceftriaxone for acute febrile episodes remain to be explored in a newborn cohort. The goal of this study was to define the cumulative incidence and risk factors for the occurrence of gallstones in a SCA-newborn-cohort. Patients and Methods: Alpha-genes, beta-globin haplotypes, UGT1AI genotype, G6PD-enzymatic activity were recorded; baseline biological parameters were determined during the second year of life, away from crises and transfusions as previously described (Blood 2011,117, 4). Annual check-ups including abdominal sonography were performed as soon as the age of 1 year. Cholecystectomy was systematically performed in the 6 months following the discovery and confirmation of presence of gallstones. All hospitalizations and their causes were prospectively recorded. Patients were censored on the date of gallstone occurrence or last abdominal sonography; event rates/year (pain crises, ACS, febrile episodes) were calculated, and KM-estimated cumulative risk for gallstones and Cox regression were used to assess the predictive factors for gallstones. Results: SCA-patients (364: 353 SS, 8 Sb0, 3 SDPunjab; 182 F,182 M) born before February 2010 and followed until the age of 18–20 years were included in this study. Alpha-thalassemia was present in 46% patients (available in 323; 2 genes: n=33; 3 genes: n=116; 4 genes: n=172; 5 genes: n=2); beta-haplotypes (available in 291) were Car/Car in 42.6%, Ben/Ben in 19.9%, Sen/Sen in 9.6%, and others in 27.8%. G6PD activity (available in 309) was deficient in 11.6%. UGTA1 polymorphism (available in 175) showed Gilbert genotype in 19.8%, i.e., 7/7 in 13.4% and 7/8 in 6.4%. The frequency of the other genotypes was: 5/6: 6.4%; 5/7:7%; 5/8:0.6%; 6/6:23.3%, 6/7:34.3%; and 6/8:8.1%. Unconjugated bilirubin values at baseline were significantly correlated with the total number of UGTA1A1 repeats (r=0.415, p<0.001). Mean rates (SD) of pain crises, ACS and acute febrile episodes were 0.56 (0.68), 0.12 (0.18), 0.34 (0.54), respectively. Gallstones were observed in 98/364 patients at the median age of 9.2 (range: 2.5 to 17.9 years). The cumulative risk of gallstones was 5% by age 5, 20.7% by age 10 and 35.6 % by age 15 and was not significantly different in patients with Gilbert polymorphism (Log Rank p=0.4) but was significantly related to the number of alpha-genes (Log-Rank p=0.039). However, the risk was significantly increased in patients homozygous or heterozygous for (TA8) (Log Rank p=0.016). Univariate Cox regression analysis showed that the number of alpha-genes was a significant risk factor (p=0.014) whereas G6PD deficiency, beta-haplotypes and Gilbert genotype were not significant risk factors; however, the presence of at least one allele (TA8) was a significant risk factor (p=0.025). Among the baseline biological parameters, hemoglobin, WBC, neutrophils, platelets, MCV, and bilirubin were not significant factors whereas HbF level (p=0.028), reticulocyte count (p<0.001) and LDH (p=0.020) significantly increased the risk for gallstones. Among clinical events, pain crises, ACS and febrile episodes significantly increased the risk (p<0.001). Multivariate Cox analysis including all the significant risk factors in the univariate analysis retained the deletion of 2 alpha genes (HR=4.66; 95% CI:1.11-19.52,p=0.035) which decreases the risk, the presence of at least one allele (TA8) (HR=2.26:95% CI:1.07-4.78, p=0.032), which increases the risk and the baseline reticulocytes count per 1×109/L increase(HR:1.001;1.000-1.002;p=0.005), as independent and significant predictive factors for gallstones. Conclusion: This prospective SCA newborn-cohort study shows that gallstones may begin as soon as 2.5 years of age and that the cumulative risk reaches a high 35.6% by age 15. The deletion of 2 alpha genes, the presence of at least one UGT1A1 (TA8) allele and high baseline reticulocyte count are independent and significant predictive factors showing that the risks are multifacturial and highly and primarily dependent on hemolysis Disclosures: No relevant conflicts of interest to declare.

No abstract

No abstract

263 Early occurrence of dactylitis, severe anemia and leukocytosis have been identified as risk factors for adverse outcome in children with SCA (Miller NEJM 2000), but the impact of acute splenic sequestrations (ASS) on disease severity has not yet been reported. The goal of this study was to define the predictive factors of ASS and to evaluate if this complication was a risk factor for disease severity. Methods. SS patients of the newborn cohort, seen at our SCA Center in Créteil before 3 months of age and older than 2 years at the last visit (n=197), were included in this study. Alpha and beta genotypes, and G6PD activity were determined. Blood parameters were recorded at steady state during the second year of life. ASS was defined as an Hb-decrease of ≥20% associated with a spleen-increase ≥2cm. Vaso-occlusive crises (VOC) requiring hospitalizations and acute chest syndromes (ACS) were recorded. The yearly rates of VOC and ACS were calculated for the entire follow-up and for the period before any intensification (transfusion program, hydroxyurea or SCT). Results. Median follow-up was 7.9 yr (range 2–24), providing 1820 patient-years. Alpha-Thal was present in 70/162 (43.2%); beta-haplotypes, available in 140 patients, were Car/Car in 62 (44.3%), Ben/Ben in 35 (25%), Sen/Sen in 7 (5%), and others in 34 (25.7%). G6PD deficiency was present in 24/171 (14%). At baseline, mean (SD) blood parameters were: Hb: 8g1/dL (1.2); Ht: 24.6% (3.9); MCV: 79.2 fL(8.3); HbF: 15.7% (8.2); LDH: 919 UI/L (369) and reticulocytes: 296 (114); WBC: 14.3 (5.2), neutrophils 5.8 (3.1) × 109/L. Among the 197 SS-patients, 64 (32%) experienced their first ASS at the median age of 2.0 yr (range 0.1–12.9) and 36/64 recurred. ASS occurred before 1 yr of age in 10 (5%) and before 2 yr in 33/197 (17%). Splenectomy was performed in 24 of them at the median age of 5.1 yr (range 2.7–12.9). The KM-estimated cumulative risk of ASS occurrence at age 7 was 32% (95%CI: 25–39%). The Cox-regression analysis showed that gender, G6PD deficiency, beta haplotypes were not risk factors whereas alpha-Thal significantly increased the risk of ASS occurrence (HR: 1.9, 95% CI:1.1-3.3, p=0.021). Among blood parameters, multivariate Cox analysis retained low platelet (HR: 1.4 per 1×107/L decrease; 95%CI:1.1-1.8, p=0.014) and high reticulocyte counts (HR=2.0 per 1×107/L increase; 95%CI:1.6-2.6) as independent significant predictive factors for ASS. Comparison of patients with or without ASS history showed no significant difference in the rate of ACS but significantly higher rates of VOC during the entire follow-up (0.68±0.58 vs 0.47±0.55; p=0.014) and in the number of hospitalizations (1.64±0.83 vs 1.29±0.92; p=0.01) in patients with ASS history. Rates of VOC before intensification were 1.1±0.7 and even 2.2±2.0 in those having experienced ASS before age 1 vs 0.6±0.7 (p=0.001) in those with no ASS history. Conclusion. Results from this study in a newborn cohort show that ASS are more frequent in patients with alpha-Thal and are associated with a higher rate of VOC. Splenectomy is usually recommended in case or recurrent ASS; however, our previous results show that geno-identical SCT can cure 95%SCA children and partly restore splenic function. We suggest that this procedure be proposed to those with available donor before considering total splenectomy. Disclosures: No relevant conflicts of interest to declare.

1067 Background: Elevation of tricuspid regurgitant jet velocity (TRJV) predicts high systolic pulmonary artery pressure and early mortality in adults with sickle cell anemia (SCA). To date, few studies have reported the concomitant analysis of lung alterations and high TRJV in SCA-children. Objective: To evaluate the relationship between lung function, TRJV and hematologic parameters in SCA-children. Patients and Methods: SCA-children of the Creteil-CHIC cohort were assessed at steady state on the same day by cardiac echocardiography, pulmonary function tests (PFT), clinical and biological parameters. All data were recorded in the CHIC database. Forced vital capacity (FVC), forced expiratory volume in 1 sec (FEV1), total lung capacity (TLC) and diffusing capacity for carbon monoxide (DLCO) were measured with the single-breath technique. DLCO was adjusted to hemoglobin and to the alveolar volume (KCOc). Univariate and multivariate linear regression analyses were performed to evaluate factors correlating with TRJV and KCOc and logistic regression was applied to evaluate the risk factors associated with elevated TRJV (≥ 2.5 m/s). Results: A total of 228 check-ups with cardiac and lung assessment, performed in 163 SCA-patients (160 SS, 3 SB0) was analyzed. Check-ups were performed at the median age of 13.2 years (range: 5.7–19.9). Among the 228 check-ups, 151 were performed after intensification (59 on hydroxyurea, 57 on transfusion program and 38 after hematopoietic stem cell transplantation (HSCT). TRJV (median value: 2.2 m/s, range 1.4–3.1) was significantly positively correlated with age (r=0.164, p=0.013), was not correlated with TLC but was significantly and negatively correlated with FVC before b2 (r=−0.145, p=0.03) and after b2 (r=−0.184, p=0.008), FEV1 before b2 (r=−0.213, p=0.001) and after b2 (r=−0.178, p=0.012), FEV1/FVC before b2 (r=23?0.165, p=0.013) and positively correlated with KCOc (r=0.379, p<0.001). Multiple linear regression analysis including age and all significant PFT data retained only KCOc as a significantly correlated factor (standardized beta: 0.390, p<0.001). Among biological parameters, TRJV was not correlated to HbF %, platelets, LDH and proBNP but significantly correlated to hemoglobin, hematocrite, reticulocytes, WBC, neutrophils, HbS %, ASAT and bilirubin. Multiple linear regression including all these biological significant biological parameters retained only reticulocytes as significantly associated with TRJV (standardized beta: 0.361, p=0.008). The addition of KCOc to the model retained reticulocytes (standardized beta: 0.198, p=0.010) and KCOc (standardized beta: 0.327, p<0.001) as significantly correlated. Among the biological parameters significantly correlated with KCOc (hemoglobin, reticulocytes, LDH, WBC, platelets, HbS%, ASAT, bilirubin), multiple linear regression analysis retained HbS% (standardized beta: 0.190, p=0.027) and reticulocytes (standardized beta: 0.212, p=0.014) as significantly and positively independent correlated factors. Elevated TRJV (≥ 2.5 m/s) was observed in 39 patients. Reticulocytes per 1×109/L increase (OR: 1.005; 95% CI:1.001–1.009; p=0.007) and KCOc per 1% increase (OR: 1.022; 95% CI:1.007–1.037; p=0.004) were retained as independent and significant factors associated to the risk of TRJV ≥ 2.5 m/s by multivariate logistic regression. When comparing the check-ups in patients without intensification, only those performed after HSCT showed significantly lower KCOc (p<0.001), reticulocytes (p<0.001) and TRJV (p=0.035) Conclusion: In this study, we confirm that SCA-children have an elevated gas transfer per unit lung volume (KCOc) correlated to hemolysis and HbS%. We show for the first time that an increase in KCOc significantly raises the risk for TRJV ≥ 2.5 m/s, even after adjustment for reticulocytes. Only HSCT, which resulted in significantly lower reticulocytes and KCOc, is significantly associated with lower TRJV. These data are encouraging and suggest that HSCT could be recommended to patients with elevated TRJV Disclosures: No relevant conflicts of interest to declare.

1536 Poster Board I-559 Background Pulmonary hypertension (PHT) is a widely recognized complication of sickle cell disease (SCD). It affects 32% of adults with SCD and is associated with an increased risk for early mortality. Screening of children with SCD by Doppler echocardiography has been recommended, using tricuspid regurgitant jet velocity (TRJV) to estimate pulmonary artery systolic pressure. However, the prevalence and risk factors of elevated TRJV in children with SCD at steady state have not been extensively defined. Methods SCD patients of the CHIC-cohort were prospectively assessed : alpha and beta genotype, G6PD, platelet-CD36 expression were determined; baseline blood parameters between 1 and 3 years of age were recorded and re-assessed every year at steady state with clinical data, TCD screening, abdominal sonography and every two years by Doppler echocardiography, evaluation of lung function and cerebral MRI/MRA. Pulmonary hypertension was defined as a TRJV ≥2.5 m/s. Univariate and multivariate logistic regression analyses were performed to evaluate risk factors for elevated TRJV. Results A total of 662 echocardiograms with available measure of TRJV were performed away from vaso-occlusive crises (VOC) and acute chest syndromes (ACS) in 320 SCD-patients (268 SS, 1 SDPunjab, 7 Sb0; 14 Sb+, 30 SC) between 12/98 and 07/09 and were included in the analyzes: 127/662 were performed in patients on hydroxurea (HU) therapy, 121 on transfusion program (TP) and 63 after stem cell transplantation (SCT). In SS/Sb0 patients, TRJV did not correlate with fractional shortening, ejection fraction, TCD velocities, systemic and diastolic blood pressure, cumulative number of VOC and ACS, LDH and SpO2, but correlated with age (r=0.258,p<0.001), height (SD) (r=-0.119, p=0.008), Hct (r=-0.091, p=0.028), MCV (r=0.109, p=0.009) and bilirubin (r=0.235, p<0.001). Elevated TRJV ≥2.5 m/s was found in 5/44 SC/Sb+ (11%) at the mean age of 12.6 ± 2.1 yr and in 57/276 SS/Sb0 (21%) (p=NS) at the mean age of 12.6 ± 4.5 yr (range 3.5-20). Univariate logistic regression analysis of the data from SS/Sb0 patients showed no significant association between elevated TRJV ≥2.5 m/s and the following variables: gender, alpha-thalassemia, G6PD deficiency, height and weight (SD), baseline blood parameters (recorded between 1 and 3 years of age), systolic/diastolic -blood pressure, cardiac rate, WBC and platelet count, HbF%, LDH, AST/ALT, history of abnormal TCD, overt and silent strokes and ongoing HU therapy and TP or a history of SCT. In contrast, CD36 deficiency (OR=4.55;CI95%:1.73-11.97,p=0.002), age (OR=1.087;CI95%:1.03-1.15,p=0.005), low SpO2 (OR=1.17;CI95%:1.05-1.32,p=0.005), Hb (OR=1.27;CI95%:1.07-1.51,p=0.007), high MCV (OR=1.03; CI95%: 1.01-1.05, p=0.008) and total bilirubin (OR=1.008; CI95%:1.002-1.014,p=0.006) were significantly associated with TRJV ≥2.5 m/s. After adjustment for age, SpO2, Hb, MCV, total bilirubin and CD36 deficiency remained significant risk factors. Conclusion In this pediatric cohort, early assessed by TCD and transfused in case of abnormal occurrence, no association was found between cerebral vasculopathy and elevated TRJV, which could be the result of early initiation of transfusion programs as soon as TCD became abnormal. The data also confirm the link between pulmonary hypertension, low oxygen saturation and hemolysis, previously reported by others. In addition, we show for the first time that CD36 deficiency is a risk factor for elevated TRJV in SCD patients. CD36 deficiency has been shown to be a common occurrence in subjects of African descent. The CD36 scavenger receptor is a multifunctional receptor, which is expressed on the surface of various cell types, and is involved in immunity, metabolism and angiogenesis. Previous studies suggested that CD36 deficiency did not modify the clinical course of SCD patients but echocardiograms had not been analyzed. Recent studies showed that hypoxia increases CD36 expression via its HIF-1 responsive promoter element and activation of the PI3K pathway, suggesting that CD36 deficient SCD patients could have an abnormal response to hypoxia. Additional studies will be needed to understand the relationship between PHT and CD36 deficiency and the role of hypoxia in these patients. Disclosures No relevant conflicts of interest to declare.

No abstract