TOPLINE

Among critically ill patients aged 90 years or older with sepsis, those with non-urinary tract infections had higher risk for early mortality than those with urinary tract infections, despite similar illness severity at admission. Beyond 6 months, mortality risk did not differ significantly between the two groups.

METHODOLOGY

• Researchers conducted a retrospective cohort study using data from a large intensive care registry with clinical data from Australia and New Zealand, to assess differences in outcomes between sepsis of urinary tract and non-urinary tract origin among nonagenarians and centenarians admitted to the ICU.
• A total of 1669 patients aged 90 years or older admitted to ICUs with sepsis between 2010 to 2023 were included; after 1:1 propensity score matching, 1130 patients (565 per group) were analyzed.
• Patients were classified as having urinary tract sepsis (n = 573) or non-urinary tract sepsis (n = 1096) using Acute Physiology and Chronic Health Evaluation (APACHE) III-J diagnostic codes.
• Primary outcome was all-cause mortality within 180 days and beyond 180 days; secondary outcomes included ICU mortality, hospital mortality, ICU and hospital length of stay.

TAKEAWAY

• Non-urinary tract sepsis was associated with a 66.7% higher risk for mortality within 180 days than urinary tract sepsis (hazard ratio [HR], 1.667; P < .001).
• Beyond 180 days, risk for mortality was not significantly different between the groups.
• APACHE III score, which assessed the severity of illness, was the strongest independent predictor of mortality, with each 1-point increase corresponding to a 2.7% increase in risk for death (HR, 1.027; P = .001).
• ICU and hospital length of stay were similar between groups; higher Sequential Organ Failure Assessment score (geometric mean ratio [GMR], 1.036; P = .010) and APACHE III score (GMR, 1.005; P = .006) were associated with longer ICU stay.

IN PRACTICE

"[These findings] suggests that conventional severity scores alone may not fully capture prognostic differences in this population. Infection source may be considered alongside established clinical and physiological factors during early risk stratification and goals-of-care discussions for patients of very advanced age," the authors of the study wrote.

SOURCE

The study was led by Je Min Suh, Austin Health in Heidelberg, Victoria, Australia. It was published online on August 3 in BMJ Open.

LIMITATIONS

Microbiological data including pathogen type and antimicrobial resistance profiles, adequacy of source control, and pre-ICU treatment delays were unavailable. Infection source was identified using diagnostic coding rather than microbiological confirmation, which may introduce misclassification bias. The study was confined to high-resource healthcare settings in Australia and New Zealand, potentially limiting generalizability to other health systems.

DISCLOSURES

No specific funding was reported. The authors reported no relevant conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.