The research was motivated by a persistent gap in pediatric cardiac care. Children undergoing surgery for congenital heart disease face substantial postoperative morbidity, including prolonged ventilation, unplanned reintubation, and extended hospital stays, yet cost-effective tools for preoperative risk stratification remain limited. In adults, inflammatory indices computed from routine complete blood counts—such as the neutrophil-to-lymphocyte ratio and the systemic immune-inflammation index—have repeatedly been shown to predict surgical outcomes. Children, however, are not small adults. Their immune physiology differs markedly, with evolving distributions of leukocyte populations across infancy and childhood, and there was no guarantee that adult-derived biomarkers would translate meaningfully to pediatric populations.

The research team, led by Yihong Li and Qiongyu Wu of the Department of Anesthesiology at Kunming Children’s Hospital, with corresponding authors Hongtao Wang and Liming Cheng, took an unusually rigorous approach. They assembled a cohort of 776 children younger than 18 years who underwent congenital heart disease surgery with cardiopulmonary bypass at a single tertiary pediatric hospital between 2018 and 2024. From each child’s preoperative complete blood count, they calculated six inflammatory indices: the systemic immune-inflammation index (SII), the systemic inflammation response index (SIRI), the neutrophil-to-lymphocyte ratio (NLR), the platelet-to-lymphocyte ratio (PLR), the monocyte-to-lymphocyte ratio (MLR), and the pan-immune-inflammation value (PIV). Each index was natural-log transformed and examined using multivariable logistic regression, with false discovery rate controlled by the Benjamini-Hochberg method—a statistical safeguard designed to prevent spurious positive findings when many biomarkers are tested simultaneously.

The primary endpoint was a composite of serious in-hospital adverse outcomes: in-hospital mortality, prolonged mechanical ventilation exceeding 72 hours, unplanned reintubation, or prolonged hospitalization beyond the 90th percentile. Of the 776 children, 153—19.7 percent—met this composite endpoint, a sobering reminder of the physical toll that even successful congenital heart surgery can take. When the investigators tested all six indices against this endpoint, every single one emerged as independently associated with adverse outcomes after full statistical adjustment and false discovery rate correction, with adjusted q-values all below 0.05.

The most intriguing result came not from the direction of the association but from its shape. Most biomarker studies assume a linear relationship: more biomarker, more risk. The team instead deployed generalized additive models and two-segment piecewise regression to explore whether the relationship might bend. For the log-transformed monocyte-to-lymphocyte ratio, it did—dramatically. The analysis revealed a U-shaped threshold association with an inflection point at a log-transformed value of −2.86, corresponding to a raw MLR of 0.057. Above this threshold, each log-unit increase in MLR nearly doubled the odds of an adverse outcome, with an odds ratio of 1.99 (95 percent confidence interval, 1.41 to 2.82; P < 0.001). Below the threshold, the association was not statistically significant, with an odds ratio of 0.15 and a P value of 0.16. In other words, once a child’s preoperative monocyte-to-lymphocyte ratio climbed past a specific cut point, risk escalated sharply—and the location of that inflection mattered more than the ratio’s overall magnitude.

The biological logic behind this threshold is plausible. Monocytes orchestrate innate inflammatory responses and tissue remodeling, while lymphocytes reflect adaptive immune competence. An elevated ratio may signal a preoperative state of immune imbalance in which innate inflammatory drive outweighs adaptive immune reserve—a state that cardiopulmonary bypass, with its well-known capacity to trigger systemic inflammatory response, could then amplify into clinically significant postoperative organ dysfunction. A U-shaped pattern, however, complicates simple interpretation: the fact that risk was not elevated below the threshold suggests that only beyond a certain degree of immune dysregulation does the ratio become informative, and that both very low and very high values in the adjusted model deserve scrutiny in future work.

Anemia emerged as a critical modifier of the relationship. When the analysis was stratified by anemia status, the association between elevated MLR and adverse outcomes was strong in non-anemic children, with an odds ratio of 2.18, but reversed direction in anemic children, with an odds ratio of 0.69, and the interaction was statistically significant (P for interaction = 0.006). This finding has immediate interpretive value for clinicians. Anemia alters the relative composition of blood counts and the inflammatory milieu itself, meaning that the same numerical MLR value may carry entirely different prognostic meaning depending on a child’s hemoglobin status. Any future use of inflammatory indices for risk stratification in this population, the authors argue, must account for anemia before the numbers are read.

Yet the study is equally defined by its honesty about limits. The monocyte-to-lymphocyte ratio achieved the highest area under the receiver operating characteristic curve of all six indices—but only 0.676, with a sensitivity of 63.4 percent and a specificity of 63.7 percent. In practical terms, an AUC of 0.676 reflects discriminative ability that is statistically detectable but clinically weak, closer to a coin that is modestly weighted than to a decisive diagnostic instrument. The authors are explicit: despite the statistically significant U-shaped association, MLR cannot serve as an independent clinical risk-stratification tool, and none of the six routine inflammatory indices achieved adequate standalone predictive capacity in this pediatric cardiac cohort.

The investigators position their findings as hypothesis-generating rather than practice-changing. They caution that the observed threshold effect and the anemia-related effect modification require external prospective validation before any clinical application of MLR-based preoperative risk evaluation can be considered. The single-center, retrospective design introduces the usual vulnerabilities: referral patterns, surgical case mix, local anesthetic and bypass protocols, and coding practices may all differ across institutions, and retrospective data collection cannot exclude residual confounding by illness severity. Still, the study’s statistical framework—false discovery rate control, nonlinear modeling, formal interaction testing, and calibration assessment of the prediction model, which performed well on Hosmer-Lemeshow and Brier score metrics—sets a standard that future biomarker studies in pediatric surgery would do well to follow.

The broader significance of the work lies in its methodological lesson as much as its clinical one. By refusing to assume linearity, the researchers uncovered a threshold pattern that a conventional linear regression would have obscured or mischaracterized. And by testing whether a single number behaves differently across patient subgroups, they exposed anemia as a factor capable of inverting a biomarker’s apparent meaning. For the hundreds of thousands of children worldwide who undergo surgery for congenital heart disease each year, the promise of a cheap, universally available blood-based risk marker remains compelling—but this study makes clear that the path to clinical utility runs through larger, multicenter, prospective validation, careful attention to nonlinear dose-response relationships, and a clear-eyed recognition of what a modest AUC can and cannot deliver. The monocyte-to-lymphocyte ratio has earned a place on the research agenda; whether it earns a place on the preoperative checklist is a question only future evidence can answer.

Subject of Research: Inflammatory blood count indices as predictors of adverse outcomes after pediatric congenital heart surgery with cardiopulmonary bypass

Article Title: A U-shaped threshold in the monocyte-to-lymphocyte ratio predicts adverse outcomes after pediatric cardiac surgery: a retrospective cohort study

Article References: Li, Y., Wu, Q., Jiang, S., Yang, R., Zhang, Y., Feng, B., Wang, H., & Cheng, L. (2026). A U-shaped threshold in the monocyte-to-lymphocyte ratio predicts adverse outcomes after pediatric cardiac surgery: a retrospective cohort study. BMC Pediatrics. https://doi.org/10.1186/s12887-026-07690-w

Image Credits: AI Generated

DOI: 10.1186/s12887-026-07690-w

Keywords: congenital heart disease, cardiopulmonary bypass, monocyte-to-lymphocyte ratio, pediatric cardiac surgery, inflammatory biomarkers, postoperative outcomes, risk stratification, complete blood count, anemia, U-shaped association, BMC Pediatrics, retrospective cohort study