Background: Gastrointestinal perforation is a common surgical problem in paediatric surgical ward in our country. It carries a high morbidity and mortality in children. The predictors of postoperative morbidity and mortality are still not well established. Objectives: The objective of this study was to identify the predictors of morbidity and mortality following surgery for gastrointestinal perforation in children. Methods: This was a single centre cross sectional type of descriptive observational study carried out in the Department of Paediatric Surgery, Mymensingh Medical College Hospital with 16-month duration from 1st January, 2017 to April 30, 2018, in children aged from 2-12 years. Predictors related to postoperative morbidity and mortality were investigated. Results: A total of 40 patients were selected for this study with mean age of 8.92±2.61 years. Male: female was 1.5:1. Analysis indicated that factors associated with postoperative morbidity and mortality were low BMI, poor nutritional status, low serum total protein and albumin, delay in seeking treatment and contaminated peritoneal collection. Appendicular perforation was the highest number perforation in 23(57.5%) patients. Ileal perforation was the second highest number 8(20%). Appendectomy was the highest surgical procedure 21(52.5%), followed by resection-anastomosis and wound excision & repair 3(7.5%) each. Serum total protein and albumin were below normal limit among 15(37.5%) and 35(87.5%) patients respectively. 13(32.5%) patients received treatment 72 hours after onset of symptom and 18(45%) patients undergone operation within 24-48 hours of admission. Postoperative complications developed in 30 patients. Among them SSI were in 20(66.66%), SIRS in 9(30%) and leakage of repair in 1(3.33%) patient. Mortality rate was 5%. Conclusion: Poor nutritional status, low total serum protein and albumin, delay in initiation of treatment, and contaminated peritoneal fluid are the predictors of morbidity and mortality of gastrointestinal perforation in children.