Age At Surgical Repair Reflects Preoperative Severity and Predicts Early Outcomes After Tetralogy of Fallot Repair in A Resource-Limited Setting: A Retrospective Cohort Study

Abstract

Objective: Children with tetralogy of Fallot (ToF) in resource-limited settings are often repaired years later than the age recommended in high-income countries. We set out to describe the preoperative severity profile of such a cohort and to test whether age at repair predicts low cardiac output syndrome (LCOS) and 30-day mortality independently of hypoxemia, nutritional status, and cardiac anatomy.

Materials and Methods: We conducted a retrospective cohort study of 116 children who underwent ToF repair between 2021 and 2025, comparing preoperative severity between children younger and older than the cohort’s median age of 4 years. Logistic regression was used to estimate how age, preoperative oxygen saturation (SpO₂), weight, nutritional status, McGoon ratio, Nakata index, ventricular septal defect (VSD) size, major aortopulmonary collateral arteries (MAPCA), and sex related to LCOS and 30-day mortality, and discrimination was assessed with the area under the receiver operating characteristic curve (AUC).

Results: Median age at repair was 4.0 years (interquartile range 2.0–7.0); no patient was younger than 1 year. LCOS developed in 42 patients (36.2%), and 42 (36.2%) died within 30 days, with 83.3% of deaths occurring in patients who had LCOS. Compared with children 4 years or older (n=66), those younger than 4 years (n=50) had lower preoperative SpO₂ (77.3 ± 7.1% vs 81.6 ± 8.0%, p=0.003) and smaller VSDs (11.1 vs 13.0 mm, p=0.003), while nutritional status, hemoglobin/hematocrit, McGoon ratio, Nakata index, and MAPCA prevalence were similar between the two groups. Younger age was independently associated with both LCOS (OR per year younger 1.23, 95% CI 1.07–1.40) and mortality (OR 1.25, 95% CI 1.09–1.44); this association held after adjustment for VSD size and undernutrition, and again after adjustment for SpO₂ (age <4 years: OR 3.78, 95% CI 1.65–8.64 for mortality; OR 3.24, 95% CI 1.43–7.35 for LCOS). SpO₂, nutritional status, McGoon ratio, Nakata index, and MAPCA showed no independent association with either outcome. Age alone discriminated mortality with an AUC of 0.71 (95% CI 0.61–0.80), which rose to 0.73 once VSD size was added.

Conclusion: Younger age at repair — rather than hypoxemia, undernutrition, or pulmonary/septal anatomy on its own — emerged as the strongest and most consistent preoperative predictor of LCOS and early mortality in this cohort. This likely reflects both a genuine physiological vulnerability in younger, smaller children and a tendency for the most severely cyanotic infants and toddlers to be referred more urgently. Where health-system capacity allows, these findings support prioritizing earlier, elective repair, and where it does not, they support closer perioperative vigilance for young children.