Sequential organ failure assessments (SOFA) during the first 48 hours of severe sepsis predicted mortality better than static baseline assessments.1 Our aim was to determine if dynamic measurements within the first day would be a reasonable tool for managing individual patients.
Placebo databases from 2 sepsis trials were combined for 1036 patients. Patients met established criteria for severe sepsis. SOFA scores were calculated using most aberrant physiologic or laboratory parameter each day. Baseline and changes to day 1 in SOFA scores were calculated for each organ system, 28-day mortality rates analyzed, and multivariable models constructed.
Mortality rates correlated with baseline cardiovascular (CV) (p=0.0022), renal (p<0.0001), hematologic (p=0.0044), and respiratory (p=0.0013) SOFA scores. From baseline to day 1, the direction of change in CV, renal, respiratory, hematologic, and hepatic functions independently predicted 28-day mortality (see tableOrgan System28-Day Mortality Ratesp Value**SOFA Change from Baseline by Day 1Improved ≥1*No ChangeWorsened ≥1n%n%n%CV19820.270230.112349.6<0.0001Renal23721.963928.312156.2<0.0001Respiratory35525.645227.613942.40.0007Hematologic6126.268727.425138.60.0031Hepatic3112.963428.94341.90.0251*
Change in SOFA score;**
chi-square test). A multivariable model based on both static and dynamic assessments had higher area under receiver operating characteristics (ROC) curve than static assessments alone (p=0.02).CONCLUSIONS: During the first day of severe sepsis, change from baseline in CV, renal, respiratory, hematologic, and hepatic organ dysfunction may be a better predictor of 28-day mortality than a static baseline assessment.
These data indicate that if organ dysfunction is not improving during the first day of severe sepsis, the mortality risk is significantly increased. This may be important for guiding therapeutic interventions and in future clinical trials.
M.M. Levy, Research supported by Eli Lilly and Company.