Abstract
Abstract 1624
Previous studies showed that: (1) increased blood levels of the omega-3 (n3) fatty acids (FA) eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) are associated with increased steady-state Hb levels and reduced prevalence of the complications of sickle cell disease (SCD); (2) EPA and DHA reduced the number of crisis in a pilot group of 5 HbSS patients. The primary aim of this study was to assess the effects of EPA and DHA in a larger number of patients. SCD causes recurrent vaso-occlusion, ischemia-reperfusion injury and inflammation (Osarogiagbon et al. Blood. 2000, 96: 314–320). A metabolite of DHA (10,17S-docosatriene) protects against ischemia-reperfusion injury by inhibiting leukocyte infiltration and expression of pro-inflammatory genes (Marcheselli et al, J Bio Chem. 2003; 278: 43807–43817. Our hypothesis was that anti-inflammatory properties contribute to the mechanism(s) of action of DHA/EPA in SCD. The secondary objective was to test this hypothesis.
Following Institutional Review Board approval and informed consent, 20 HbSS patients (10 Males, 10 Females) were enlisted to receive oral EPA 15mg /kg/day and DHA 10 mg/kg/day with standard treatment of SCD, and reviewed at 2-monthly intervals for 6 months. Standard treatment included folic acid 5 mg daily, pyrimethamine-sulfadoxine (25mg/500mg) 2–3 tablets monthly for malaria prophylaxis, and appropriate management of acute clinical events. To assess the placebo effect of regular clinical monitoring, another group of 8 HbSS patients (2 Males, 6 Females) were enlisted to be reviewed 2-monthly for 6 months, but observed on standard treatment only. Inclusion criteria for both groups were 3 or more crisis/yr and age>5yrs. Exclusion criteria: pregnancy, hydroxyurea or regular transfusion therapy, any other disease e.g diabetes. To reduce the confounding effect of individual variability in severity of SCD, the study was so designed that each participant was his/her own control. Using the Mann-Whitney test, we compared pre- and post-treatment numbers of crisis over 6 months, steady state levels of Hb and plasma unconjugated bilirubin. To test the hypothesis that anti-inflammatory effects contribute to the mode of action DHA/EPA, we compared 3 indices of inflammation pre-and post-treatment: plasma level of interleukin-6 measured by ELISA, peripheral blood neutrophil and platelet counts determined with an automated cell counter.
In the omega-3 FA group 4 participants dropped out or were withdrawn from the study: 1 developed recurrent transient ischemic attacks and started regular blood transfusion to prevent stroke; 1 had recurrent severe crisis and began hydroxyurea therapy; 2 were not compliant with follow-up and omega-3 FA therapy. For the 16 evaluable participants (7M and 9F aged 9–33 yrs) the median number of crisis over 6 months reduced from 3 to 0 (p<0.0001) and steady state unconjugated bilirubin from 14.95 umol/l to 8.6 umol/l (p = 0.03). There were no significant differences between pre and post-treatment median neutrophil count (4.94 × 109/l, 4.36 × 109/l, p = 0.32) platelet count (318 × 109/l, 255 × 109/l, p= 0.057) Hb level (7.7 g/dl, 7.5 g/dl, p=1.0) and IL-6 (10.5 pg/ml, 12 pg/ml, p = 0.59).
In the standard treatment group, 3 participants (2M, 1F) were not compliant with follow-up. The evaluable 5 female HbSS patients, age 17–26 yrs, had no significant change between months 0 and 6 median number of crisis (1.5, 1, p = 0.12) steady state Hb (7.6 g/dl, 6.6 g/dl, p = 0.54) unconjugated bilirubin 24 umol/l, 30 umol/l, p = 0.84), neutrophil count (5.45 × 109/, 4.66 × 109/, p = 0.15) and platelet count 359 × 109/, 411 × 109/, p = 0.84). Months 0 and 6 plasma IL-6 concentrations could be measured in only 3 participants in the standard treatment group. Since this number is too small for statistical analysis, the respective 0 and 6 month values for the 3 individuals are provided: 8 pg/ml and 7 pg/ml, 9 pg/ml and 9.6 pg/ml, 18 pg/ml and 22 pg/ml.
The data suggest that EPA and DHA reduce the number of crisis and steady state hemolysis in SCD, and do not support the hypothesis that these effects are mediated by inhibiting inflammation.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
