Background: Both severe sepsis and septic shock are the terminal events of all infectious diseases including diarrhea and often associated with fatal outcome. However, death is highest in septic shock even in high resource centre in developed countries. Thus, identification of factors associated with septic shock from severe sepsis is critically important. Nevertheless, data are scarce on the clinical predictors of septic shock in under-five children presenting with severe sepsis especially in resource poor settings. We evaluated the factors associated with septic shock and their outcome in such population. Methods: This study involved the analysis of retrospective data in diarrheal children which had been extracted from the hospital management system (SHEBA), an online data base of the Dhaka Hospital of icddr,b. All under-five diarrheal children, admitted to the Dhaka Hospital of the International Canter for Diarrhoeal Diseases Research, Bangladesh (icddr,b) having severe sepsis between October`2010 and September`2011 were studied. Severe sepsis defined as tachycardia plus hypo (≤35.0° C) or hyperthermia (≥38.5° C), or abnormal WBC count plus presence of infection with poor peripheral perfusion (age specific hypotension and/or absent peripheral pulses or delayed capillary refilling time (CRT) in absence of dehydration. Patient unresponsive to fluid (normal saline/cholera saline) boluses (20 ml/kg; maximum 60 ml/kg and 40 ml/kg in well nourished and malnourished children respectively) and required inotrop(s) categorized as septic shock. Children with (cases=88) and without septic shock (controls=116) wee compared. Results: Median (inter-quartile range) age (months) was comparable among the cases and the controls [5.3 (3.2, 12.0) vs. 5.6 (2.7, 10.0); p = 0.515). Case-fatality-rate was significantly higher among the cases than the controls (67% vs. 14%; p