Patients with cancer represent a substantial and growing proportion of intensive care unit (ICU) admissions. However, how consistently they are explicitly included, reported, and analytically represented in randomized controlled trials (RCTs) of ICU interventions remains unclear. We conducted a systematic review of adult ICU RCTs published from January 1, 2000 through the date of the final database search (January 19, 2026). Eligible studies enrolled adult patients (≥ 18 years) managed as critically ill, prespecified all-cause mortality as a primary or co-primary outcome, included at least 100 participants, and were published in English-language journals indexed in the National Library of Medicine Core Clinical Journals subset. Two independent reviewers screened records and extracted trial characteristics, intervention domains, and reporting of cancer status. Trials were classified according to whether patients with cancer were explicitly included, explicitly excluded, or not reported. Interventions were grouped into five domains: pharmacologic/immunomodulatory therapies, hemodynamic/resuscitation strategies, mechanical ventilation/oxygenation, renal/metabolic support, and general ICU management. A total of 77 RCTs comprising 144,548 critically ill adults met the inclusion criteria. Twenty-four trials (31.1%) explicitly reported inclusion of patients with cancer, nine (11.6%) explicitly excluded them, and 44 (57.1%) did not report cancer status. Among studies reporting prevalence, the proportion of enrolled patients with cancer ranged from 3% to 33%, with only four trials including ≥ 20% oncology patients. Inclusion varied across intervention domains, ranging from 18.2% in general ICU management trials to 40% in pharmacologic or immunomodulatory intervention trials. None of the included studies reported mortality outcomes stratified according to cancer status. In adult ICU randomized trials that prespecified mortality as a primary or co-primary outcome, cancer status was frequently not reported and explicit inclusion of patients with cancer was limited. The absence of oncology-specific outcome reporting restricts the ability to assess the applicability of trial findings to this growing ICU population. Future ICU trials should report cancer status more consistently and consider prespecified oncology-focused analyses when clinically justified.
Melo et al. (Mon,) studied this question.