Aims & Objectives: To determine the potentially preventable deaths in our pediatric intensive care unit (PICU). Methods Retrospective cross-sectional Setting: PICU of a tertiary care hospital Subjects: All patients aged 1 month to 16 years who died in PICU form January 2013 to December 2014 Interventions: None Measurements and Main results: Medical records review of all patients who died in PICU over 2 years period was done by two independent reviewers to find delayed recognition, slow intervention, unintended/unexpected harm, medication error, transfusion reaction and nosocomial infection which could have resulted in unexpected mortality. Preventability was labelled on a 6 point scale used in earlier studies. Results labelled on a 6 point scale used in earlier studies. Total 92/690 patients died during the 2 years (mortality rate 133/1000 admissions) with median age of 60 months (interquartile range, IQR, 114). Males were 52 (56%) and the median PRISM III score was 17 (IQR, 6). Admission diagnosis included infection/sepsis (n=29, 35%), neurological illnesses (n=16, 17%), hematology/oncological diseases (n=6, 6%), cardiovascular diseases (n=4, 4%) and others. No mortality was identified with definitive or strong evidence of preventability. Three (4%) children were identified as possibly preventable (>50/50, category 4), twelve children (15%) possibly preventable (<50/50, category 3), thirty (32%) as slight evidence of preventability, and forty-seven (48%) were identified as definitely not preventable. Delayed recognition of the patients deteriorating condition (21%), slow intervention (53%) and nosocomial infections (26%) were found to be related to potentially preventable deaths. Conclusions There is small but important opportunity of mortality prevention which can help improve outcomes.