A retrospective cohort study examines the long-term risk of ischemic stroke in patients with perioperative atrial fibrillation following cardiac and noncardiac surgery. The following data are reported: Which of the following best describes the study results?
A)Cardiac surgery carries a lower risk of atrial fibrillation than noncardiac surgery
B)Patients with perioperative atrial fibrillation following cardiac surgery should not receive anticoagulation
C)Perioperative atrial fibrillation increases the risk of stroke with either type of surgery
D)Perioperative atrial fibrillation is not predictive of stroke in patients undergoing cardiac surgery
E)The risk of stroke after noncardiac surgery is significantly higher compared to that after cardiac surgery

Question

A retrospective cohort study examines the long-term risk of ischemic stroke in patients with perioperative atrial fibrillation following cardiac and noncardiac surgery.  The following data are reported: Which of the following best describes the study results?
A)Cardiac surgery carries a lower risk of atrial fibrillation than noncardiac surgery
B)Patients with perioperative atrial fibrillation following cardiac surgery should not receive anticoagulation
C)Perioperative atrial fibrillation increases the risk of stroke with either type of surgery
D)Perioperative atrial fibrillation is not predictive of stroke in patients undergoing cardiac surgery
E)The risk of stroke after noncardiac surgery is significantly higher compared to that after cardiac surgery

Quizplus · Accepted Answer

Explanation: In this scenario, noncardiac surgery patients with perioperative atrial fibrillation had a higher rate of stroke in the year following their hospitalization compared to those without perioperative atrial fibrillation (1.37% versus 0.32%); the hazard ratio (HR) for this was 1.9. Similarly, cardiac surgery patients with perioperative atrial fibrillation had a higher rate of stroke compared to those without perioperative atrial fibrillation (0.91% versus 0.82%) with a HR of 1.3. HR is a measure of effect used in survival or time-to-event analysis. It has almost the same interpretation as relative risk; therefore, these study results can be interpreted in terms of risk. The null value for HR is 1 (no difference in risk between 2 groups). HR <1 indicates a protective effect, but a value >1 indicates a detrimental effect. In this study, the HR of stroke in the setting of perioperative atrial fibrillation (compared to no perioperative atrial fibrillation) was 1.9 for noncardiac and 1.3 for cardiac surgery patients; both values are >1 with 95% confidence intervals (CIs) that do not cross the null value (dark gray in table below). Therefore, there was a statistically significant increased risk of stroke following perioperative atrial fibrillation in both types of surgery. Unlike HRs, the rates of stroke (light gray in table below) are not measures of risk or association. For this reason, the CIs for cumulative rates of stroke do not have the same interpretation in terms of risk as the CIs for the HRs. They are related to sampling. For example, cardiac surgery patients with perioperative atrial fibrillation had a 0.91% rate of stroke with a 95% CI of 0.88%-1.07%. This means that if the study were repeated 100 times, the rate found would be in the range 0.88%-1.07% in 95 out of the 100 instances. (Choices A and B) There is no information provided regarding risk of atrial fibrillation or benefit of anticoagulation based on surgery type. The values given relate to stroke and not perioperative atrial fibrillation. (Choice D) The HR of 1.3 with a statistically significant CI means that among patients who underwent cardiac surgery, the risk of stroke was higher in those who developed perioperative atrial fibrillation compared to those who did not. (Choice E) Both cardiac and noncardiac surgery groups had an increased risk of stroke in the setting of perioperative atrial fibrillation (given HRs >1 with CIs that do not cross the null). However, the CIs overlap (1.1-1.7 and 1.5-2.4), so it is difficult to determine whether this increased stroke risk is significantly higher in one of the groups compared to the other. Educational objective: Hazard ratios are used in survival analysis and have a similar interpretation to relative risk. A hazard ratio <1 indicates a protective effect, while a value >1 indicates a detrimental effect.

A Retrospective Cohort Study Examines the Long-Term Risk of Ischemic