Design: Systematic review and meta-analysis of 105 observational studies
Population: Over 20 million adults with sepsis or septic shock
Primary outcome: All-cause mortality by end point, with vs without malignancy
Effect: In-hospital OR 1.90 (95% CI 1.69–2.13); ICU OR 1.21 (0.94–1.56); unadjusted

A meta-analysis of 105 studies including over 20 million people compared mortality in sepsis or septic shock with and without cancer, separating mortality end points.

Malignancy was associated with higher in-hospital mortality (OR 1.90, 95% CI 1.69–2.13), early mortality (1.81) and longer-term mortality (OR 2.28 at intermediate and 2.41 at extended follow-up). ICU mortality was not clearly raised (1.21, 0.94–1.56). The excess was most consistent for haematological cancers; for metastatic disease it concentrated early.

The estimates are unadjusted, so they describe prognosis, not the independent effect of cancer. Clinically, they support early, honest goals-of-care conversations in septic patients with active cancer, without assuming that ICU care is futile — ICU mortality itself was not clearly higher.

Expect higher hospital mortality in septic patients with active cancer, especially haematological
Do not assume ICU care is futile — ICU mortality was not clearly raised
Start goals-of-care conversations early alongside full sepsis treatment
Involve the oncology team early to inform prognosis

Why it matters

Prognosis shapes the conversation, but not a reason to withhold initial resuscitation.

The statistics, in plain English

Odds ratios near 2 sound like doubled risk, but when deaths are common, odds ratios overstate relative risk; the authors note risk ratios were smaller.