Background: The mortality rate of severe sepsis remains unacceptably high. It is difficult to make advances in the treatment of this problematic and increasingly frequent medical condition. In severe sepsis, hypothermia can be recognized as an important feature. The present study investigated the role of hypothermia in the prognosis of the colon ascendens stent peritonitis (CASP) model. Methods: We employed the CASP model for wild-type C57BL/6 mice. We compared physiologic indices in survivor and non-survivor groups after CASP to test whether low temperature might be a helpful predictor in sepsis. To certify this hypothesis, we examined the survival rate, peritoneal leukocytes, and organ damage. We also measured the bacterial burden and inflammatory cytokine levels at different times. Results: The temperature varied dramatically in the survivors' group compared with the non-survivors' group at 18 h. We divided the CASP models into a mild group and a severe group, based on temperatures above or below 32 degrees C at 18 h. Mice in the severe group had a lower survival rate (0% versus 87.5%), more peritoneal leukocytes, more bacterial culture results, higher expressions of cytokines, and more classical features in pathology compared with the mild group. Conclusions: Hypothermia (below 32 degrees C at 18 h) might be a predictor of prognosis in CASPinduced sepsis. (C) 2013 Elsevier Inc. All rights reserved.