Abstract
Stroke is a serious complication of sickle cell anemia (SS). Steady-state hemoglobin (Hgb) desaturation, which is common in SS and often thought to be benign, could contribute to the risk of stroke because it perturbs endothelial function in SS (
Lancet 2003;362:1450
) and might limit oxygen delivery to the brain. One previous study of SS patients found an association between nocturnal oxygen desaturation and strokes, transient ischemic attacks, and seizures (Lancet 2001;357:1656
). Nocturnal oxygen saturation is cumbersome to measure, and the association between daytime Hgb saturation and stroke has not been reported. We hypothesized that daytime steady-state Hgb desaturation was associated with and preceded overt stroke in children with SS. We performed a nested case-control study of the Dallas Newborn Cohort (Blood 2004;103:4023
). Cases were defined as cohort subjects with SS or sickle-β0-thalassemia (Sβ0) who had clinically overt ischemic strokes in the preceding 12 years. Controls were cohort subjects with SS or Sβ0 but without overt stroke who were reported previously (Br J Haematol 2005;131:129
). We collected mean steady-state oxygen saturations by pulse oximetry (SpO2) and mean steady-state hematologic data for cases (pre-stroke) and controls. For cases we also identified one SpO2 measurement and one blood count that was recorded closest to but preceding each stroke (the proximate pre-stroke values). We further calculated a predicted SpO2 for cases using a model that includes age, sex, and mean steady-state values of Hgb and reticulocyte count (Br J Haematol 2005; 131:129
) and calculated odds ratios for stroke by logistic regression. We identified 22 cases (100% SS; 68% male; mean age 7.9 years) and 390 controls (98% SS; 55% male; mean age 9.5 years). The steady-state SpO2 of cases (94.2%, standard deviation [SD] 3.0) was significantly lower than controls (96.3%, SD 3.0) (P=0.0014). The proximate pre-stroke SpO2 of cases (93.7% SD 3.8) was even lower than steady-state and significantly lower than controls (P=0.0006). The predicted steady-state SpO2 for cases was 95.2% (SD 1.4), meaning that the observed SpO2 of cases (94.2%) was lower than would be expected for SS patients without stroke given age, sex, and steady-state values of Hgb and reticulocyte count. The odds ratio for stroke was 1.32 (95% confidence interval 1.15 – 1.51) for each unit (1%) decrease in SpO2 while simultaneously controlling for age, sex, and steady-state values of Hgb and reticulocyte count (P<0.0001). In summary, we found that children with SS who develop a stroke have lower pre-stroke steady-state SpO2 values than children with SS who do not develop a stroke. Moreover, in children who develop stroke, the measured SpO2 value decreases as the time interval to stroke narrows. The odds of stroke increase as the SpO2 decreases. In conclusion, steady-state Hgb desaturation is a risk factor for ischemic stroke in children with SS. Decline in steady-state SpO2 over time further increases this risk. Hgb desaturation is an easily measured, potentially modifiable risk factor that could be used to identify children with SS who have an increased risk of overt ischemic stroke.
Figure
