Reply: Using different statistical tools to see forests and trees.

Reply to the Editor:

The authors reported no conflicts of interest.

The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.

We read with interest the commentary provided by Van den Eynde and colleagues regarding the article by Guariento and colleagues and their examination of outcomes with a modulated renewal analysis in a group of truncus arteriosus patients at Boston Children's Hospital. Van den Eynde and colleagues provide additional insight about the statistical question and answer the conundrum investigators face when analyzing complex clinical data. Their discussion of major adverse cardiac events in patients with HIV makes clear that reliance on Kaplan-Meier methods and the accompanying log-rank testing can result in clinical events being overlooked when only examining the time to a first event.

We completely agree with and applaud Guariento and colleagues' application of the modulated renewal analysis to examine problems where multiple reinterventions are likely or, as is the case with truncus, inevitable. Congenital heart surgeons are always thinking about—and planning for—the next operation and the statistics we apply should match this clinical behavior. The list of congenital heart problems for which this type of analysis would apply includes coarctation, pulmonary atresia, tetralogy of Fallot, Ebstein anomaly, single ventricle pathology, congenital aortic stenosis, and transposition of the great arteries; basically, most congenital cardiac lesions where patients are monitored for life and issues arise that may require reintervention.

It appears that Guariento and colleagues may have cleaned the lenses of our statistical binoculars. We have to continue to use Kaplan-Meier to view the forest of overall and long-term survival while employing modulated renewal analysis to peer within the forest for clues that may better inform how we take care of patients with lifelong diseases. The next step will be to train surgeons to understand the difference in methodologies and use the treatment tailored to each individual question.