Acute Coronary Syndrome After Aneurysmal Subarachnoid Hemorrhage: Incidence, Risk Factors and Impact on the Outcome

Background and Objectives: Development of acute coronary syndrome (ACS) after aneurysmal subarachnoid hemorrhage (aSAH) strongly affects further neuro-intensive care management. We aimed to analyze the incidence, risk factors and clinical impact of ACS in aSAH patients. Materials and Methods: This retrospective analysis included 855 aSAH cases treated between 01/2003 and 06/2016. The occurrence of ACS during 3 weeks of aSAH was documented. Patients' demographic, clinical, radiographic and laboratory characteristics at admission were collected as potential ACS predictors. The association between ACS and the aSAH outcome was analyzed as the occurrence of cerebral infarcts in the computed tomography scans and unfavorable outcome (modified Rankin scale > 3) at 6 months after aSAH. Univariable and multivariable analyses were performed. Results: ACS was documented in 28 cases (3.3%) in the final cohort (mean age: 54.9 years; 67.8% females). In the multivariable analysis, there was a significant association between ACS, an unfavorable outcome (adjusted odds ratio [aOR] = 3.43, p = 0.027) and a borderline significance with cerebral infarcts (aOR = 2.5, p = 0.066). The final prediction model for ACS occurrence included five independent predictors (age > 55 years [1 point], serum sodium < 142 mmol/L [3 points], blood sugar >= 170 mg/dL [2 points], serum creatine kinase >= 255 U/L [3 points] and gamma-glutamyl transferase >= 36 U/L [1 point]) and showed high diagnostic accuracy for ACS prediction (AUC = 0.879). Depending on the cumulative score value, the risk of ACS in the cohort varied between 0% (0 points) and 66.7% (10 points). Conclusions: ACS is a rare, but clinically very relevant, complication of aSAH. The development of ACS can reliably be predicted by the presented prediction model, which enables the early identification of aSAH individuals at high risk for ACS. External validation of the prediction model is mandatory.