Stroke Following Cardiac Surgery
October 1, 1999 3 minutes read
Stroke Following Cardiac Surgery
Abstract & commentary
Synopsis: Strokes after cardiac surgery are more common after initial normal neurologic recovery and atrial fibrillation was related to these later strokes only if a low cardiac output state was induced.
Source: Hogue CW, et al. Circulation 1999;100: 642-647.
Stroke after cardiac surgery continues to be a problem and is associated with high mortality. Thus, Hogue and colleagues sought to identify risk factors for early and later stroke in 2972 patients aged younger than 50 years in whom epicardial echocardiography was used to identify ascending aorta atherosclerosis. The surgical approach was modified to avoid atheroma during aorta manipulations. Stroke was defined as new permanent neurological defects that could not be attributed to metabolic or other problems, and they were classified as early (immediately after surgery) or later (occurring after initial normal neurologic recovery). The incidence of stroke was 1.61% (48/2972) and 65% were later. Multivariate analysis showed that prior neurologic events (odds ratio 12, P < 0.001), aortic atherosclerosis (OR = 2, P = 0.004), female sex (OR = 0.7, P = 0.004), and duration of cardiopulmonary bypass (OR = 1.1, P = 0.005) were independent predictors of early stroke. Later stroke risk factors also included diabetes (OR = 2.8, P = 0.008) and low cardiac output associated with atrial fibrillation (OR = 1.7, P = 0.033). In-hospital mortality was 41% in those with early strokes, 13% in those with late strokes, and 3.7% in the nonstroke patients (P < 0.001). Hogue et al conclude that strokes after cardiac surgery were more common after initial normal neurologic recovery and atrial fibrillation was related to these later strokes only if a low cardiac output state was induced. Also, all strokes were more common in women, those with prior strokes, and those with atherosclerosis.
Comment by Michael H. Crawford, MD
The major finding of this study was the low rate of stroke (1.6%) in an older cohort in whom epiaortic echo was used to guide aorta cannulation. Previous studies have found stroke rates of 3-6% in similar populations. Unfortunately, epiaortic echo was used in all the patients, rather than being randomly assigned vs. no echo assessment. Thus, we cannot be sure that the aortic echos made the difference or some other technical advance or population characteristic was the reason for the lower rates. This study and others strongly suggest that intraoperative ascending aorta evaluation may reduce preoperative stroke rates. The results are more impressive when you consider that some of the strokes may have been present preoperatively and missed because no formal preoperative neurological assessment was done, but post-operative patients suspected of stroke were evaluated by neurologists.
Another deficiency of the study was a failure to evaluate cognitive function before and after surgery. Previous studies have suggested that mild intellectual impairment occurs in up to 6% of post-operative patients, resulting in a total stroke plus milder impairment rate of 10-15%.
Much has been made of the morbidity and mortality associated with post-operative atrial fibrillation that occurs frequently after cardiac surgery. Yet in this study, atrial fibrillation alone was no more likely to be associated with stroke than no atrial fibrillation (1%). However, the presence of low cardiac output plus atrial fibrillation raised it to 4% (P = 0.004). On the other hand, we do not know how the patients with atrial fibrillation were managed in this study. It is possible that they were aggressively managed with heparin, transesophageal echo, and cardioversion, if appropriate. Aggressive treatment would reduce the rate of strokes from this cause.
In conclusion, it is clear that more attention needs to be paid to the condition of the ascending aorta in cardiac surgery patients. Whether preoperative echo, intraoperative echo, or experienced palpitation is the best approach is unknown. Also, if echo is employed, should it be transesophageal or epiaortic? In addition, women and patients with prior stroke are at higher risk of early and late strokes. Interestingly, diabetes and age were not predictive, which Hogue et al suggested may be due to assessing the patients for aortic atherosclerosis. In previous studies in which age and diabetes were risk factors, they may have been serving as markers for atherosclerosis. In descending order of importance, perioperative stroke seems related to a history of prior stroke, female sex, and aortic atherosclerosis.
Which is most true concerning stroke following cardiac surgery?
a. The majority occur after initial normal neurologic recovery.
b. Most are related to post-operative atrial fibrillation.
c. The incidence is 10-15%.
d. Preoperative neurological exam is predictive.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content