Impact of intensive care-related factors on outcome in stroke patients, results from the population-based Brest Stroke Registry: an observational study

Abstract

Background Little is known on the burden of ICU care for stroke patients. The aim of this study was to provide a description of management strategies, resource use, complications and their association with prognosis of stroke patients admitted to ICU. Methods Using a population-based stroke registry, we analyzed consecutive stroke patients admitted to 3 ICU with at least one organ failure between 2008 and 2017. The study period was divided into two periods corresponding to the arrival of mechanical reperfusion technique. Predictors of ICU mortality were separately assessed in two multivariable logistic regression models, a “clinical model” and an “intervention model”. The same analysis was performed for predictors of functional status at hospital discharge. Results 215 patients were included. Stroke etiology was ischemia in 109 patients (50.7%) and hemorrhage in 106 patients (49.3%). Median NIHSS score was 20.0 (9.0; 40.0). The most common reason for ICU admission was coma (41.2%) followed by acute circulatory failure (41%) and respiratory failure (27.4%). 112 patients (52%) died in the ICU and 20 patients (11.2%) had a good functional outcome (mRS≤3) at hospital discharge. In the "clinical model,” factors independently associated with ICU mortality were: age (OR = 1.03 [95%CI, 1.0 to 1.06]; p=0.04) and intracranial hypertension (OR = 6.89 [95%CI, 3.55 to 13.38]; p<0.0001). In the “intervention model,” the need for invasive mechanical ventilation (OR = 7.39 [95%CI, 1.93 to 28.23]; p=0.004), the need for vasopressor therapy (OR = 3.36 [95%CI, 1.5 to 7.53]; p=0.003) and decision of withholding life support treatments (OR = 19.24 [95%CI, 7.6 to 48.65]; p<0.0001) were associated with bad outcome. Conclusion Our study showed the very poor prognosis of acute stroke patients admitted to ICU. These results also suggest that the clinical evolution of these patients during ICU hospitalization may provide important information for prognostication.