Abstract
Dear Editor,
We read with great interest the article by Novelleto et al, “In-Hospital Mortality in Children Undergoing Congenital Heart Surgery in North Brazil: A 9-Year Analysis From a Regional Referral Center”. 1 The authors should be commended for addressing the outcomes of congenital heart surgery in an underserved region of Brazil. Congenital heart disease remains a major contributor to childhood morbidity and mortality worldwide, and in many low- and middle-income countries, a substantial proportion of affected children remain undiagnosed or untreated. 2 In this context, the development and strengthening of regional cardiac programs are essential to improving access to care and reducing health disparities.
We would, however, like to raise several methodological and interpretative issues that may influence the concluisons drawn from the study. First, there appears to be an inconsistency in the reported number of deaths across the tables. In Table 3 from the original paper, the total number of deaths is 131, and the authors correctly state that Rashkind septostomies were excluded. However, in Tables 2, 4, and 5 from the original paper, the total number of deaths is reported as 135. This suggests that the four patients who underwent Rashkind procedures may have been included in the overall analysis of surgical outcomes. The inclusion of Rashkind atrial septostomy among the analyzed procedures and its apparent consideration within the Risk Adjustment for Congenital Heart Surgery (RACHS-1) categories may introduce heterogeneity into the dataset and affect the interpretation of results. Balloon atrial septostomy is a catheter-based intervention rather than a surgical procedure and is not included in the RACHS-1 classification system.
Second, the study does not provide the total number of procedures performed in each RACHS-1 category. The authors appropriately acknowledge this limitation, noting that the absence of these data preclude calculation of category-specific mortality rates and limits comparison with international benchmarks. Nevertheless, Table 5 juxtaposes expected mortality values from the RACHS-1 literature with the proportion of deaths observed in each category within the study population. As these represent fundamentally different metrics—expected mortality rates versus the distribution of deaths—presenting them side by side may inadvertently lead readers to conclude that observed mortality exceeds expected values; a conclusion that cannot be supported without appropriate denominators.
Furthermore, the analyses presented focus exclusively on the subgroup of patients who died during hospitalization. While this approach provides valuable descriptive insights, the absence of a comparison with survivors limits the ability to identify factors independently associated with mortality.
We are confident that these issues were not intentional and recognize the important contribution of the authors in bringing attention to congenital cardiac care in the Brazilian Amazon. We hope that clarifying these methodological aspects may help avoid potential misinterpretation of the data and further stimulate high-quality research on surgical outcomes in resource-limited settings.
Sincerely,
