Intensive care patient survival after limiting life‐sustaining treatment—The FINNEOL national cohort study

• Published in: Acta anaesthesiologica Scandinavica Vol. 64; no. 8; pp. 1144 - 1153
• Main Authors: Adamski, Jan, Weigl, Wojciech, Lahtinen, Pasi, Reinikainen, Matti, Kaminski, Tadeusz, Pietiläinen, Laura, Musialowicz, Tadeusz
• Format: Journal Article
• Language: English
• Published: Copenhagen Wiley Subscription Services, Inc 01-09-2020
• Subjects: Cohort analysis; Decision analysis; Intensive care; Patients; Physical fitness; Regression analysis; Survival
• ISSN: 0001-5172; 1399-6576; 1399-6576
• DOI: 10.1111/aas.13612

Background
Few studies have examined survival in intensive care unit (ICU) patients after the restriction of life‐sustaining treatment (LST). We aimed to analyse independent factors associated with hospital and 12‐month survival rates in ICU patients after treatment restrictions.
Methods
This retrospective observational study examined all patients treated in adult ICUs from 1 January 2016 until 31 December 2016 included in the Finnish ICU Registry. Multivariable logistic regression analysis was performed to explain the effect on survival.
Results
Decisions to limit LST were made for 2444 patients (13.7%; 95% CI 13.2‐14.2). ICU, hospital, and 12‐month survival rates were 71% (95% CI 69‐73), 49% (95% CI 47‐51), and 24% (95% CI 22‐26), respectively. In patients for whom life support was withheld, increased 12‐month survival rates were associated with admission from the operating theatre (OR 1.9, 95% CI 1.1‐3.4), good pre‐hospital physical fitness (OR 4.7, 95% Cl 1.2‐16.8) and being housed at home (OR 2.0, 95% Cl 1.4‐2.8). Decreased survival rates were associated with admission from a hospital ward (OR 0.67, 95% Cl 0.5‐0.9), higher comorbidity (OR 0.6, 95% Cl 0.4‐0.9), cancer (OR 0.4, 95%CI 0.2‐0.9), greater illness severity (SAPS II; OR 0.98, 95% Cl 0.98‐0.99), and higher care intensity (TISS‐76; OR 0.93, 95% Cl 0.92‐0.95).
Conclusion
Survival among ICU patients with limited treatment was higher than expected. Advanced age was not associated with higher mortality, potentially because treatment restrictions may be set more easily for older patients.