Elevation of troponin I in acute ischemic stroke

Background. Cardiac morbidities account for 20% of deaths after ischemic stroke and is the second commonest cause of death in acute stroke population. Elevation of cardiac troponin has been regarded as a prognostic biomarker of poor outcome in patients with acute stroke. Methods. This retrospective study enrolled 871 patients with acute ischemic stroke from August 2010 to March 2015. Data included vital signs, laboratory parameters collected in the emergency department, and clinical features during hospitalization. National Institutes of Health Stroke Scale (NIHSS), Barthel index, and modified Rankin Scale (mRS) were used to assess stroke severity and outcome. Results. Elevated troponin I (TnI) > 0.01 µg/L was observed in 146 (16.8%) patients. Comparing to patients with normal TnI, patients with elevated TnI were older (median age 77.6 years vs. 73.8 years), had higher median heart rates (80 bpm vs. 78 bpm), higher median white blood cells (8.40 vs. 7.50 1,000/m3) and creatinine levels (1.40 mg/dL vs. 1.10 mg/dL), lower median hemoglobin (13.0 g/dL vs. 13.7 g/dL) and hematocrit (39% vs. 40%) levels, higher median NIHSS scores on admission (11 vs. 4) and at discharge (8 vs. 3), higher median mRS scores (4 vs3) but lower Barthel index scores (20 vs. 75) at discharge (p < 0.001). Multivariate analysis revealed that age ≥ 76 years (OR 2.25, CI [1.59–3.18]), heart rate ≥ 82 bpm (OR 1.47, CI [1.05–2.05]), evidence of clinical deterioration (OR 9.45, CI [4.27–20.94]), NIHSS score ≥ 12 on admission (OR 19.52, CI [9.59–39.73]), and abnormal TnI (OR 1.98, CI [1.18–3.33]) were associated with poor outcome. Significant factors for in-hospital mortality included male gender (OR 3.69, CI [1.45–9.44]), evidence of clinical deterioration (OR 10.78, CI [4.59–25.33]), NIHSS score ≥ 12 on admission (OR 8.08, CI [3.04–21.48]), and elevated TnI level (OR 5.59, CI [2.36–13.27]). C-statistics revealed that abnormal TnI improved the predictive power of both poor outcome and in-hospital mortality. Addition of TnI > 0.01 ug/L or TnI > 0.1 ug/L to the model-fitting significantly improved c-statistics for in-hospital mortality from 0.887 to 0.926 (p = 0.019) and 0.927 (p = 0.028), respectively. Discussion. Elevation of TnI during acute stroke is a strong independent predictor for both poor outcome and in-hospital mortality. Careful investigation of possible concomitant cardiac disorders is warranted for patients with abnormal troponin levels