Predictors of index admission mortality and morbidity in contemporary esophageal atresia patients
jpedsurg.org shows its articles on its own site.
Read the article on jpedsurg.org ↗Article · Mar 2020 · 1 min read
In brief
In brief
This retrospective study of 253 EA/TEF patients validates the Spitz classification for mortality prediction while proposing a new morbidity risk algorithm. Major congenital heart disease emerged as the sole mortality predictor, while prenatal diagnosis, pre-operative ventilation, prematurity, and male sex predicted composite morbidity outcomes.
Written by the GCMD Library team from the article.
Abstract
Background/Purpose
The Spitz classification for esophageal atresia with/without tracheoesophageal fistula (EA/TEF) predicts mortality. This study evaluates the contemporary relevance of the Spitz classification and investigates predictors of morbidity.
Methods
EA/TEF patients born between 1995 and 2018 at two centers were retrospectively reviewed. Clinical variables including sex, prenatal diagnosis, birth weight, prematurity, major congenital heart disease (MCHD), and pre-operative mechanical ventilation (POMV) were collected. Index admission composite morbidity was considered positive if: length-of-stay >90th percentile (139 days), ventilation days >90th percentile (24 days), and/or gastrostomy was used for long-term feeding. Multivariable regression determined predictors of index admission mortality and composite morbidity. A composite morbidity predictive algorithm was created.
ROC curves evaluated model discrimination.
Results
Of 253 patients, 13 (5.1%) experienced index admission mortality. Of the patients not suffering mortality, 74 (31.6%) experienced composite morbidity. Only MCHD predicted mortality (p = 0.001); birth weight did not (p = 0.173). There was no difference between the Spitz classification and MCHD alone in predicting mortality risk (p = 0.198); both demonstrated very good discrimination. Prenatal diagnosis, POMV, prematurity, and male sex predicted composite morbidity risk (p < 0.001; p = 0.008; p = 0.009; p = 0.05). An algorithm incorporating these predictors demonstrated good discrimination (AUC = 0.784; 95% CI: 0.724, 0.844).
Conclusions
The Spitz classification maintains contemporary relevance for mortality risk, though birth weight can be de-emphasized. A new morbidity risk algorithm is proposed for early postnatal counseling.
Type of Study
Prognosis study.
Level of Evidence
Level IV.