Risk factors and simple scoring system for predicting postoperative nutritional status of Hirschsprung's disease.

BACKGROUND

Nutritional problem after surgery for Hirschprung's disease (HSCR) was not optimistic. This study aimed to analyze the risk factors of postoperative undernutrition for patients with HSCR and establish a scoring system for predicting postoperative undernutrition.

METHODS

Retrospective review of 341 patients with HSCR who received Laparoscopic-assisted pull-through surgery in a tertiary-level pediatric hospital was conducted with assessments of clinical data. Univariate/multivariate Logistic regression analysis was used to identify independent factors of postoperative undernutrition, and establish a scoring system for predicting postoperative nutritional status based on the sum of adjusted odds ratios (ORs).

RESULTS

The postoperative undernutrition of 341 patients with HSCR was 29.9%. Multivariate Logistic regression analysis showed that non-breast feeding (mixed: OR = 6.116, artificial: OR = 12.00), preoperative undernutrition (risk of malnutrition: OR = 7.951, malnutrition: OR = 8.985), non-parental caregivers (OR = 3.164), long-segment HSCR (OR = 12.820), postoperative complications within 30 days (grade 1 ~ 2: OR = 2.924, Grade 3 ~ 4: OR = 6.249), and surgery for other systemic malformation (OR = 5.503) were risk factors for postoperative undernutrition (all  < 0.05), and scoring system was developed based on these determinants. The area under the receiver operator characteristic curve of the derivation sample was 0.887 (95% confidence interval [CI]: 0.839-0.934) and that of the validation sample was 0.846 (95% CI: 0.772 ~ 0.920) with the optimal cut-off value of 12; calibration curves of the derivation sample showed considerable predictive performance for postoperative undernutrition.

CONCLUSION

Risk factors identified affecting postoperative undernutrition should be taken seriously in patients with HSCR. We successfully developed a desirable scoring system to predict postoperative nutritional status, which might be helpful for clinical practice.