Radiologic measurement of brain swelling in patients with large hemispheric infarctions during targeted temperature management

Abstract

Brain herniation is most often the result of severe brain swelling and can rapidly lead to death or brain death. We retrospectively identified radiologic indicators to evaluate the effects of targeted temperature management (TTM) on the extent of cerebral edema and determine the cutoff values that best predict TTM outcomes in patients with large hemispheric infarction. We retrospectively reviewed brain computed tomography (CT) scans of 21 patients with large hemispheric infarctions, who were treated with TTM. We excluded 4 patients whose CT scans were inadequate for evaluation, which left 17 patients. We divided the patients into success and failure groups. TTM failure was defined as death or the need for decompressive hemicraniectomy (DHC) after TTM. Infarction size was measured as the total restricted area in diffusion-weighted imaging that was performed on admission. CT scans were obtained on the first and second days after TTM initiation and then every 2 days. We measured septum pellucidum shifts (SPS) and pineal gland shifts (PGS) on CT scans. The median time from symptom onset to TTM initiation was 14.5 hours. Ten patients were successfully treated with TTM, six patients died, and one patient underwent a DHC. Initial infarction sizes were not significantly different between the success and failure groups (p = 0.529), but the SPS and PGS at 36-72 hours after TTM initiation were (mean SPS: 5.0 vs. 14.9 mm, p = 0.001; mean PGS: 2.3 vs. 7.9 mm, p = 0.001). The sensitivity and negative predictive value for TTM failure caused by cerebral edema (SPS >= 9.25mm and PGS >= 3.70 mm) at 36-72 hours after TTM initiation were both 100%. The SPS and PGS on CT scans taken 36-72 hours after TTM initiation may help to estimate the effect of TTM on cerebral edema and guide further treatment.