Writing the Discussion Section
5- 29
Example of a Well-Written Discussion (Clinical Study)
In a large cohort of patients referred for exercise stress testing, we found that ventricular ectopy after exercise (during the recovery phase) was a better predictor of increased risk of death than was ventricular ectopy during exercise only. [Conclusion] The occurrence of frequent ventricular ectopy during recovery was strongly predictive of an increased risk of death from all causes over a 5-year follow-up period, whereas the occurrence of frequent ventricular ectopy only during exercise was not. This association persisted even after propensity-based adjustment for clinical and exercise characteristics known to predict an increased risk of death. [Supporting findings] Until recently, it was thought that exercise-induced ventricular ectopy was not independently related to an increased risk of coronary heart disease, the extent of coronary artery disease, mortality from all causes, or the risk of major cardiac events. 4,8,16,31 However, one recent report showed that among over 6000 asymptomatic men, ventricular ectopy during exercise was associated with a relative risk of death from cardiovascular disease of approximately 3 when the cohort was followed for 23 years. 6 [How study fits with existing literature; restatement of the gap in knowledge] The current study clarifies these previous findings and extends them to a large cohort likely to be representative of patients seen in clinical practice. [How the study fills the gap] Because of the size of the study sample, we were able to examine carefully the prognostic importance of frequent ventricular ectopy during and after exercise in large numbers of subjects (more than 1000 patients in each group). The large cohort also made it possible for us to perform propensity matching, 27 thus allowing a more valid comparison of patients with and without frequent ventricular ectopy than would have been possible by standard regression techniques. 28 [Strengths of study, and interpretation of findings] Finally, our observations were consistent with our a priori hypothesis that frequent ventricular ectopy during recovery would be a stronger predictor of risk than ectopy during exercise, which had been based on the recognition of recovery as a period of rapid vagal reactivation. 11 Because the cohort was a heterogeneous one, including patients who underwent stress testing with electrocardiography only, with echocardiography, or with nuclear perfusion scintigraphy, we did not have systematic data on left ventricular systolic function and myocardial ischemia in all patients. Nonetheless, it is noteworthy that in the subgroup of 6421 patients for whom ejection-fraction data were available, a low ejection fraction (40 percent or less) was associated with frequent ventricular ectopy during recovery. Furthermore, both ventricular ectopy during recovery and a low ejection fraction were independent predictors of death. We focused on death from all causes and could not differentiate among deaths due to arrhythmias, those due to other cardiac causes, and those due to noncardiac causes. We and others have commented on this issue before, pointing out that only death from all causes can be considered a truly unbiased and objective end point that is also clinically relevant when arrhythmia-related outcomes are studied. 20,21 [A possible limitation of the study (and why it’s not a limitation), and interpretation of findings] How should the finding of an association between frequent ventricular ectopy during recovery from exercise and mortality from all causes be incorporated into clinical practice? Because this was a prospective, observational study, making treatment recommendations on the basis of our results is problematic. Nonetheless, it is clear that frequent ventricular ectopy during recovery is a marker of an increased risk of death. Accordingly, comprehensive risk-factor assessment and aggressive management of the risk factors identified may well be justified in patients with this finding. In addition, the association of asymptomatic left
Made with FlippingBook Digital Publishing Software