Fire-Related Smoke Inhalation Injury as an Independent Predictor of Mortality in Hospitalized Burn Patients: A Systematic Review and Meta-Analysis of Adjusted Estimates

Abstract

Background: Smoke inhalation injury is the main fire-related respiratory injury in burn patients and is included in most burn mortality scores, but whether it predicts death independently of burn size and age has been disputed. Objective: To estimate the multivariable-adjusted association between inhalation injury and mortality in hospitalized burn patients and the attenuation produced by adjustment. Methods: PubMed, Europe PMC, OpenAlex and Scopus were searched without date or language limits, with citation searching. Cohort studies reporting an adjusted estimate for inhalation injury were included. Adjusted odds ratios (ORs) were pooled with random-effects models (DerSimonian–Laird, Hartung–Knapp); risk of bias was assessed with the Quality In Prognosis Studies (QUIPS) tool and certainty with the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach. The protocol was not registered. Results: 53 cohorts (338,479 patients, 28 countries) were included. The pooled adjusted OR from 44 cohorts was 3.23 (95% confidence interval [CI] 2.67–3.90) (I² 72.3%; 95% prediction interval 1.43–7.30); adjusted hazard ratios gave 2.46 (1.45–4.19) (7 cohorts). The crude OR was 9.45 (6.40–13.95). In 23 cohorts reporting both, adjustment reduced the OR from 9.61 to 2.92. Estimates were similar in high-income (3.20) and low- and middle-income countries (3.38; p = 0.76). No small-study effects were found (Egger p = 0.93); certainty was moderate. Conclusion: Inhalation injury was associated with about threefold higher adjusted odds of death after burns, a smaller association than crude comparisons suggest. It should remain a core element of early burn triage and risk adjustment, including where bronchoscopic confirmation is unavailable.