Abstract
Introduction
About 80 % of sickle cell disease (SCD) patients live in sub-Saharan Africa whereas most studies on SCD are performed in Europe and America. Hyperhemolysis is thought to play a major role in the pathological process of SCD vasculopathy. Nevertheless, the paradigm of hyperhemolysis is controversial and has never been studied in the African context. We aim to analyze the association between hemolysis and clinical vascular complications among SCD African patients.
Methods
CADRE is a multinational cohort of SCD African patients aged 3 years and older, included prospectively and explored in a steady state. All patients from the CADRE cohort in Ivory Coast, Cameroun and Mali were included in the present study. Patients were classified in SS-Sβ0 phenotype or SC-Sβ+ phenotype. SCD vascular complications (pulmonary arterial hypertension (PAH), microalbuminuria, leg ulcers, priapism, stroke and osteonecrosis) were assessed using clinical examination, laboratory exams and echocardiography. Hemolysis was measured by a composite score, created with a principal component analysis, that included LDH, hemoglobin and bilirubin rates and clinical icterus. The association between hyperhemolysis (upper quartile of the hemolysis score) and the vascular complications was assessed using multivariate regression analysis in the whole population with further stratification by hemoglobin phenotype.
Results
We included 2409 patients among which 1751 SS-Sβ0 patients and 658 SC-Sβ+ patients. Compared to SC-Sβ+ patients, SS-Sβ0 patients were younger (15 years old versus 21 years old) and exhibited more leg ulcers (166 (9.5%) versus 24 (3.7%)) and microalbuminuria (573 (42.4%) versus 119 (20.0%)). The hemolysis score was higher in SS-Sβ0 patients as compared to SC-Sβ+ patients (median 1.18 versus 0.43, Figure). After adjustment for age, sex and country, hyperhemolysis was associated with microalbuminuria (OR = 1.59 [1.17-2.16]) and priapism (OR=1.58 [1.01-2.49]) (Table). In SS-Sβ0 patients, hyperhemolysis was associated with microalbuminuria (OR=1.54 [1.10-2.15]) and there was a trend to an association with PAH (OR=1.65 [0.91-2.98]) and priapism (OR=1.50 [0.95-2.39]). In SC-Sβ+ patients, only the association between hyperhemolysis and microalbuminuria remained significant. Sensibility analysis in the adult population showed that hyperhemolysis was significantly associated with all the SCD vascular complications, except for osteonecrosis.
Conclusion
In African SCD patients, associations between hyperhemolysis and SCD vascular complications were statistically significant but of modest magnitude, and depended on the hemoglobin phenotype. These results suggest that in the African context, hemolysis is not a major determinant of the development of SCD vasculopathy.
. | . | % of patients with the complication . | Multivariate analysis . | |||
---|---|---|---|---|---|---|
N event/N total | Quartiles 1 to 3 | Quartile 4 | OR | IC 95% | P value | |
PAH* | 115/431 | 6.14 | 7.08 | 1.58 | 0.89-2.83 | 0.123 |
Urine albumin/creatinine > 30mg/g | 724/1991 | 24.88 | 45.36 | 1.59 | 1.17-2.16 | 0.004 |
Leg ulcer, lifetime | 190/2409 | 6.80 | 11.01 | 1.20 | 0.81-1.78 | 0.355 |
Priapism, lifetime ** | 160/1095 | 13.27 | 17.90 | 1.58 | 1.01-2.49 | 0.046 |
Stroke, lifetime | 25/2409 | 0.79 | 1.77 | 1.32 | 0.49-3.54 | 0.578 |
Osteonecrosis, lifetime | 277/2409 | 11.22 | 12.28 | 1.00 | 0.70-1.43 | 1.000 |
. | . | % of patients with the complication . | Multivariate analysis . | |||
---|---|---|---|---|---|---|
N event/N total | Quartiles 1 to 3 | Quartile 4 | OR | IC 95% | P value | |
PAH* | 115/431 | 6.14 | 7.08 | 1.58 | 0.89-2.83 | 0.123 |
Urine albumin/creatinine > 30mg/g | 724/1991 | 24.88 | 45.36 | 1.59 | 1.17-2.16 | 0.004 |
Leg ulcer, lifetime | 190/2409 | 6.80 | 11.01 | 1.20 | 0.81-1.78 | 0.355 |
Priapism, lifetime ** | 160/1095 | 13.27 | 17.90 | 1.58 | 1.01-2.49 | 0.046 |
Stroke, lifetime | 25/2409 | 0.79 | 1.77 | 1.32 | 0.49-3.54 | 0.578 |
Osteonecrosis, lifetime | 277/2409 | 11.22 | 12.28 | 1.00 | 0.70-1.43 | 1.000 |
PAH: Pulmonary Arterial Hypertension. * Patients from Mali and Cameroun only ** Males only
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.