(289) Characterizing the Profile and Mortality of Very Elderly Patients in the ICU: A Chilean Cohort

Introduction: Patients aged ≥ 80 years, known as "very elderly patients" has been steadily increasing in the ICU admissions. However, there is a lack of adequate characterization of this patients and the prognostic factors predicting clinical outcomes. Admitting very elderly patients to the ICU presents a significant clinical challenge, needing rapid assessment of prognostic factors, including acute clinical status, baseline patient condition, patient and family preferences, and survival probability, which may limit access for favorable responders.
This study aimed to analyze mortality in patients aged ≥ 80 years admitted to the adult ICU of a private high-complexity center.

Methods: A non-concurrent, single-center cohort study included patients aged ≥ 80 years admitted to the ICU during 2022 at a private high-complexity center. A 90-day follow-up period was defined to determine mortality at 28 and 90 days. The study protocol received approval from the UFT Ethics Committee.

Results: 51 patients. Demographics 51% male, median age 85 years (IQR 82-88.5). Mortality rates at 28 and 90 days were 23.5% and 33.3%. Diagnoses included 43% infectious diseases, 29% acute neurological/neurosurgical pathologies, and 27% other conditions. Associated comorbidities included arterial hypertension (45%), coronary heart disease (31%), chronic respiratory pathologies (14%), pre-existing neurological conditions (22%), and active oncological pathologies (18%).

Kaplan-Meier survival curves had overlapping curves. Subgroup analysis revealed higher mortality at 90 days in patients with chronic respiratory pathologies (p = 0.021) and pre-existing neurological conditions (p = 0.005).

Conclusions: Mortality rates of 23.5% at 28 days and 33.3% at 90 days exceeded the center's median rates, aligning with acceptable ICU mortality rates. Despite being very elderly, comorbidity rates were lower, potentially due to triage processes for ICU admission, reflecting a selected population. Notably, only chronic respiratory pathologies and pre-existing neurological conditions correlated with mortality, suggesting age, sex, and disease type insignificance in this subgroup.

Limitations include a non-concurrent, single-center cohort of 51 patients, necessitating consideration of availability and biases associated with retrospective data collection.