Short-Term Results of Patients With Spontaneous Subarachnoid Hemorrhage in Intensive Care Unit: Single-Center Experience

Abstract

Objective: Few studies have evaluated patients with spontaneous subarachnoid haemorrhage (sSAH) from an intensivist perspective. This study aimed to report the results of patients with sSAH in a high-volume centre monitored by a team experienced in the fields of brain surgery, interventional radiology and intensive care. Materials and Methods: Data of patients with sSAH followed up between January 2014 and July 2018 in the intensive care unit (ICU) were retrieved from ICU patient observation charts, file records and hospital automated information system. Results: This study enrolled 150 patients, of which 61 (40.7%) patients died despite receiving intensive care. Mortality rates between patients with (42.8%) and without (40%) vasospasm were comparable (p=0.917). Vasospasm developed in 37.8% of the 45 patients who underwent endovascular coiling and in 19.2% of those who underwent neurosurgical clipping (p=0.044). The median times that elapsed before endovascular or surgical procedures were 2.5 [interquartile range (IQR): 2-5] days in the surviving group and 2 (IQR: 1-5) days in the deceased group (p=0.164). Blood sodium and blood chloride levels were significantly higher in the deceased group from the third day onward. The median blood sodium level exceeded 142 mEq/L in the deceased group, but was lower than 142 mEq/L on the same day in the surviving group. Conclusion: The results of this study suggest that Glasgow coma scale (GCS) at admission to the ICU is one of the important factors that affect treatment success. GCS is an important independent factor in selecting the timing or type of treatment (surgical clipping/endovascular coiling) and medical treatments such as nimodipine in patients with sSAH requiring intensive care. In addition, the incidence of vasospasm was higher in patients who underwent endovascular coiling. Increased sodium and chloride values during follow-up are the only parameters significantly associated with mortality.