Evolution of Clinical Characteristics and Outcomes of Critically Ill Patients 90 Years Old or Older Over a 12-Year Period: A Retrospective Cohort Study

Abstract

Objectives: The global population is aging, and the proportion of very elderly patients 90 years old or older in the ICU is expected to increase. The changes in the comorbidities and outcomes of very elderly patients hospitalized in the ICU that have occurred over time are unknown. Design: Retrospective observational cohort study. Setting: ICUs at a single academic hospital in Germany. Patients: Ninety years old or older and admitted to the ICU between January 1, 2008, and April 30, 2019. Interventions: None. Measurements and Main Results: Of the 92,958 critically ill patients, 1,108 were 90 years old or older. The study period was divided into two halves: January 1, 2008–August 30, 2013, and September 1, 2013–April 30, 2019. The number of patients 90 years old or older increased from the first period (n = 391; 0.90% of total admissions) to the second period (n = 717; 1.44%). The patients’ demographic characteristics were similar between the both time periods. The median Charlson Comorbidity Index was higher during the first period (1 [interquartile range, 1–3]) than compared with the second time period (1 [0–2]; p = 0.052). The Simplified Acute Physiology Score (SAPS) II was higher during the first time period (38 [29–49]) than during the second period (35 [27–45]; p = 0.005). Vasopressor therapy was necessary in 40% (n = 158) and 43% (n = 310) of patients in each time period, respectively (p = 0.363). Invasive mechanical ventilation was administered in 37% (n = 146) and 34% (n = 243) of patients in each time period, respectively (p = 0.250). The median length of the ICU stay was significantly lower in the first time period than in the second time period (1.4 vs. 1.7 d; p = 0.002). The ICU (18% vs. 18%; p = 0.861) and hospital (31% vs. 29%; p = 0.395) mortality rates were comparable between the two groups. The 1-year mortality was significantly lower during the second time period than during the first time period (61% vs. 56%; p = 0.029). Cox regression analysis revealed that the SAPS II, medical cause of admission, mechanical ventilation requirement, and vasopressor use were associated with 1-year mortality. Conclusions: The number of patients 90 years old or older who were treated in the ICU has increased in recent years. While the patients’ clinical characteristics and short-term outcomes have not changed significantly, the long-term mortality of these patients has improved in recent years.