Is Bacteremic Sepsis Associated with Higher Mortality in Transplant Recipients than in Non-Transplant Patients? A Matched Case-Control Propensity-Adjusted Study.
Clin Infect Dis. 2014 Oct 9;
Authors: Kalil AC, Syed A, Rupp ME, Chambers H, Vargas L, Maskin A, Miles CD, Langnas A, Florescu DF
Abstract
BACKGROUND: Sepsis is a serious solid organ transplant (SOT) complication. Evidence on survival differences between SOT and non-SOT patients with sepsis is lacking.
METHODS: Matched, case-control propensity-adjusted study. Conditional logistic regression was performed for risk factor analysis; Cox proportional hazards regression for survival analysis.
RESULTS: A total of 369 patients (123 cases; 246 controls) diagnosed with blood culture proven sepsis were matched 1:2 by age, gender, and hospital location. The distribution of allografts: 36.6% kidney, 34.1% liver, 13% kidney-pancreas, 7.3% small bowel/liver, 5.7% heart/lung, and 3.3% multivisceral. The conditional logistic regression showed that the following factors were significantly more frequently associated with SOT compared to non-SOT: higher number of comorbidities OR=8.2 [95%CI 1.48 to 45.44] (p=0.016); higher SOFA score OR=1.2 [95%CI 1.07 to 1.32] (p=0.001); presence of nosocomial infection OR=36.3 [95%CI 9.71 to 135.96] (p<0.0001); inappropriate initial antibiotics OR=0.04 [95%CI 0.006 to 0.23] (p<0.0001); and lower white blood cell count OR=0.93 [95%CI 0.89 to 0.97] (p<0.0001). Cox proportional hazards regression showed that after all adjustments for clinical presentation, severity of illness and types of infection, SOT recipients with sepsis had a significantly lower risk of death at 28 days: HR=0.22 [95%CI 0.09 to 0.54] (p=0.001), and at 90 days: HR=0.43 [95%CI 0.20 to 0.89] (P=0.025).
CONCLUSIONS: The 28-day and 90-day mortality were significantly decreased for transplant recipients compared with non-transplant patients. These findings suggest that the immunosuppression associated with transplantation may provide a survival advantage to transplant recipients with sepsis through modulation of the inflammatory response.
PMID: 25301215 [PubMed - as supplied by publisher]