Decreased Long-Term Survival of Patients With Newly Diagnosed Cancer Discharged Home After Unplanned ICU Admission: A Prospective Observational Study

• Published in: Critical care explorations 2024-08, Vol.6 (8), p.e1136
• Main Authors: Praça, Ana Paula Agnolon, Nassar Junior, Antônio Paulo, Ferreira, Alexandre Miras, Caruso, Pedro
• Format: Article
• Language: English
• Subjects: Adult; Aged; Brazil - epidemiology; Female; Humans; Intensive Care Units - statistics & numerical data; Male; Middle Aged; Neoplasms - diagnosis; Neoplasms - mortality; Neoplasms - therapy; Observational Study; Patient Admission - statistics & numerical data; Patient Discharge - statistics & numerical data; Proportional Hazards Models; Prospective Studies; Risk Factors; Survival Analysis
• ISSN: 2639-8028; 2639-8028
• DOI: 10.1097/CCE.0000000000001136

To compare the 18-month survival between patients with newly diagnosed cancer discharged home after early unplanned ICU admission and those without early unplanned ICU admission; we also evaluated the frequency and risk factors for early unplanned ICU admission. Observational study with prospectively collected data from September 2019 to June 2021 and 18 months follow-up. Single dedicated cancer center in São Paulo, Brazil. We screened consecutive adults with suspected cancer and included those with histologically proven cancer from among 20 highly prevalent cancers. None. The exposure was early unplanned ICU admission, defined as admission for medical reasons or urgent surgery during the first 6 months after cancer diagnosis. The main outcome was 18-month survival after cancer diagnosis, and the main analysis was Cox's proportional hazards model adjusted for confounders and immortal time bias. Propensity score matching was used in the sensitivity analysis. We screened 4738 consecutive adults with suspected cancer and included 3348 patients. Three hundred twelve (9.3%) had early unplanned ICU admission, which was associated with decreased 18-month survival both in the unadjusted (hazard ratio, 4.03; 95% CI, 2.89-5.62) and adjusted (hazard ratio, 1.84; 95% CI, 1.29-2.64) models. The sensitivity analysis confirmed the results because the groups were balanced after matching, and the 18-month survival of patients with early ICU admission was lower compared with patients without early ICU admission (87.0% vs. 93.9%; p = 0.01 log-rank test). Risk factors for early unplanned ICU admission were advanced age, comorbidities, worse performance status, socioeconomic deprivation, metastatic tumors, and hematologic malignancies. Patients with newly diagnosed cancer discharged home after early unplanned ICU admission have decreased 18-month survival compared with patients without early unplanned ICU admission.