Page 64 - Cardiac abnormalities after aneurysmal subarachnoid hemorrhage

Chapter 4
Discussion
WMA after aSAH may be present on admission, but can also develop during the course of the aSAH without a preceding rebleeding. Abnormal admission ECG (sinus tachycardia, ST-segment elevation or –depression) and positive troponin T are risk factors for early WMA, and a myocardial infarct pattern on the admission ECG and positive troponin T also predicts late WMA. The generally accepted hypothesis is that following the initial bleed a catecholamine release by the sympathetic nerve endings in the myocardium is responsible for the observed cardiac abnormalities.8, 17-19 Myocardial ischemia or infarction as a cause for the abnormalities is unlikely as normal myocardial perfusion was reported after aSAH.17 Risk factors for CAD such as smoking and hypertension were not associated with WMA in the univariable analyses, nor were the WMA associated with the perfusion territory of a single coronary artery. Nevertheless, we cannot exclude that at least in some patients the cardiac abnormalities were related to myocardial ischemia as some of the included patients, with risk factors for cardiovascular disease, may have had asymptomatic coronary artery disease (CAD). This is supported by our finding that an ECG compatible with an acute myocardial infarction predicts late WMA.
Our study is not unique in investigating prognostic significance of cardiac abnormalities after SAH. However, it distinguishes from other studies as only few studies investigated the time course of the WMA, 7, 10 and we could not find any studies that investigated risk factors for late WMA. The studies published so far have given contradicting results on the associations between troponin release, 3, 10, 20 BNP release 21 and WMA on admission. Differences in study design, definitions of determinants and outcomes, and also assay (e.g. NT-proBNP versus of BNP) might explain the discrepant findings between studies. Moreover, most of the previous studies were retrospective, monocenter, or had a smaller sample size than our study. The prospective studies that have been performed used admission determinants disregarding the dynamic nature of the WMA.
Although our study was prospective, multicenter, and used predefined definitions for the clinical determinants, it still has some potential limitations. The number of patients with late development of WMA is rather small, which limits the power for analysis on risk factors for late development of WMA. This is partly due to missing data of the last investigation, mostly because patients had died or were discharged. Another potential limitations lies in the study design as observational studies are prone for several forms of bias.22 However, by using strict definitions for all determinants
62