© 2016, Medical Association of Thailand. All rights reserved. Objective: To identify incidence, characteristics and outcomes of patients who were re-admitted to surgical intensive care units (SICUs). Material and Method: Multicenter prospective cohort study conducted in 9 university-affiliated surgical ICUs in Thailand (THAI-SICU study) from April 2011 to January 2013. Results: A total of 144 patients (3.1%) re-admitted to our surgical ICUs from 4,652 cases were recruited. Re-admission baseline characteristics were advanced age (mean = 71 years), low body mass index, and higher APACHE-II and SOFA score within 24 hours of first ICU admission. Many significant comorbidities were found in the re-admission group, including: hypertension, cardiovascular diseases, and respiratory diseases. ICU mortality and hospital mortality were higher in readmission group than those in the non re-admission group (20.1% vs. 9.3%, p < 0.001 and 27.8% vs. 11.3%, p < 0.001, respectively). The relative risk ratio for mortality between re-admission and non re-admission in ICU was 2.17 times and in hospital mortality was 2.46 times greater. Independent potential risk factors for re-admission were age (OR 1.028, 95% CI 1.001-1.051), emergency surgical intervention (OR 1.978, 95% CI 1.027-3.813), transfer back from general wards (OR 4.175, 95% CI 2.020-8.628), and respiratory failure needing mechanical ventilation (OR 2.167, 95% CI 1.065-4.407). Conclusion: Re-admission was found in 3.1% of cases in our surgical ICUs. This problem is associated with significantly higher ICU and hospital mortality. Risk factors of re-admission were patient age, emergency surgery, re-admission from general wards, and need for respiratory support.