Authors
Emine Arman Fırat, İlhami Çelik
Published in
Turkish journal of anaesthesiology and reanimation. Volume 54. Issue 4. Pages 274-281. Aug 28, 2026.
Abstract
Post-sepsis outcomes remain poorly defined, with recurrent infections and late mortality posing significant clinical challenges. To identify clinical and laboratory predictors of 90-day post-sepsis mortality and to evaluate infection site and pathogen similarity in recurrent sepsis.
This retrospective cohort study included 300 adult patients hospitalized with sepsis or septic shock at a tertiary care center (2018-2022). Demographic, clinical, laboratory, and microbiological data were analyzed. Multivariate bootstrap logistic regression identified independent predictors of mortality, and receiver operating characteristic analysis assessed diagnostic performance. Recurrent episodes were examined for infection site and for microbial concordance with the index hospitalization.
Of 183 discharged patients, 60 (32.8%) died within 90 days. Non-survivors were older, had higher Charlson Comorbidity Index scores, had longer intensive care unit (ICU) stays, had lower albumin and procalcitonin levels, and were less likely to achieve microbiological clearance by day 7. In multivariate analysis, only hypoalbuminemia independently predicted mortality (P=0.012). Receiver operating characteristic analysis showed the highest area under the curve for 1/procalcitonin (0.694), followed by age (0.665) and 1/albumin (0.647). Among recurrent sepsis cases, 55.9% had infections at the same site, but identical pathogens were detected in only 5.9%.
Advanced age, comorbidity burden, prolonged ICU stay, delayed microbiological response, and hypoalbuminemia are associated with increased 90-day post-sepsis mortality. Although pathogen concordance was uncommon, a low culture yield indicates that some recurrent episodes may be relapses caused by undetected similar pathogens. When culture-negative episodes are also considered, the true similarity rate between initial and recurrent infections is likely even higher.
PMID:
42664420
Bibliographic data and abstract were imported from PubMed on 29 Aug 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 9
- Comments 0