• Wei-Hung Chen, Chyi-Huey Bai, Sheng-Jean Huang, Hou-Chang Chiu, and Li-Ming Lien.
• Department of Neurology, Shin Kong WHS Memorial Hospital, Taipei, Taiwan.
• Surg Neurol. 2007 Jan 1; 68 Suppl 1: S68-73; discussion S74.

BackgroundLarge hemispheric infarcts cause high mortality and morbidity. Understanding the clinical course and prognostic factors in patients with LHI, thereby enabling the identification of patients who will benefit from early aggressive intervention, is important. This study describes the clinical course of patients who had LHI and identifies the predictors for mortality.MethodsA retrospective collection of clinical and laboratory data in patients admitted to a neurologic intensive care unit of a medical center was examined. Large hemispheric infarct was defined as an infarct that involved at least 2 of the 3 (deep, superior, and posterior) MCA territories. Patients who received a hemicraniectomy were not included.ResultsFifty patients with radiologically confirmed LHI were analyzed. The 30-day mortality rate was 22%. Only patients who had massive infarcts (complete MCA territory infarcts and beyond) died, whereas none with i-MCAs died (P < .001). For the 26 patients with massive infarcts, the 30-day mortality was 42.3%. Early deterioration, ipsilateral pupil dilation, and a low GCS were associated with mortality. Further analysis revealed that an age less than 70 years (OR 24.5, 95% CI 2.3-262.6) and a GCS less than 10 at the second day of stroke (OR 15, 95% CI 1.5-149.5) predicted a fatal outcome among patients with massive infarcts. A GCS less than 12 at the first day of stroke and early CT findings of hypodensity more than one half of the MCA territory were associated with massive infarct.ConclusionsThe extent of infarction is a crucial factor for mortality. The consciousness level may identify patients at risk for massive infarct at the first day of stroke and predict a fatal outcome as early as the second day. Early identification of the extent of infarction and close monitoring of the consciousness level help predict outcome.

37 keywords...

hide…