Cerebral Sinovenous Thrombosis in the Asphyxiated Cooled Infants: A Prospective Observational Study.

BACKGROUND

Cerebral sinovenous thrombosis is unusual in the asphyxiated cooled infants, but reliable data regarding the incidence of this comorbidity are lacking. We assessed the incidence of sinovenous thrombosis in a population of asphyxiated cooled infants by performing routine brain magnetic resonance venography.

METHODS

All asphyxiated infants who underwent therapeutic cooling at our institution completed brain magnetic resonance venography after rewarming. Assessing the incidence of cerebral sinovenous thrombosis was the primary goal. Secondary analyses included group comparisons for laboratory tests and monitored parameters, relationship between variables, logistic regression models, and receiver operating characteristic curve for cerebral sinovenous thrombosis prediction.

RESULTS

Cerebral sinovenous thrombosis was detected in 10 of 37 infants (27%), most commonly affecting the superior sagittal sinus (eight of ten). These infants manifested higher blanket (P < 0.001) and lower esophageal temperatures (P = 0.006), lower platelet counts (P = 0.045), and received more red blood cell transfusions (P = 0.038) than the cooled infants without thrombosis. Blanket temperature was independently associated with cerebral sinovenous thrombosis (P = 0.049), and 32°C/hour was the optimal cutoff value to predict the event (sensitivity, 90%; specificity, 88.5%).

CONCLUSIONS

High incidence or cerebral sinovenous thrombosis in neonates treated with therapeutic hypothermia suggests that magnetic resonance venography may be reasonable in many of these children. High blanket temperature may be one variable that helps identify patients at higher risk.