Mortality Risks and Causes in Previous Carbon Monoxide Poisoning: A Nationwide Population-Based Cohort Study.

Abstract

Objectives: Carbon monoxide (CO) poisoning can cause brain, heart, and kidney injuries. We aimed to determine the association of risks of all-cause and cause-specific mortality in patients with previous CO poisoning.

Design, setting, and patients: This population-based cohort study used data from the National Health Insurance Service database and the National Death Registry of Korea. Adult patients diagnosed with CO poisoning and controls between 2002 and 2020 were included. Patients were matched with controls on a 1:1 ratio, considering age, sex, insurance type, income level, residential location, smoking status, alcohol consumption, obesity status, medical and psychiatric illness history, and Charlson Comorbidity Index at the index date. The cohort was monitored from 2002 to 2022 or until death or emigration in terms of all-cause and cause-specific mortality.

Interventions: None.

Measurements and main results: A total of 48,600 patients with CO poisoning and matched controls were included. The cohort included 41.30% females, and the mean age was 48.05 years. Patients with CO poisoning exhibited a substantially elevated risk of all-cause mortality compared with those in the control group, with an adjusted hazard ratio (aHR) of 15.67 (95% CI, 12.58-19.51). The mortality associated with infectious (aHR, 6.71; 95% CI, 1.51-29.72), neoplasm/oncologic (aHR, 5.20; 95% CI, 3.39-7.99), endocrine (aHR, 13.44; 95% CI, 1.76-102.70), neurologic (aHR, 7.42; 95% CI, 2.91-18.90), cardiovascular (aHR, 8.97; 95% CI, 5.05-15.93), respiratory (aHR, 17.54; 95% CI, 5.48-56.17), and gastrointestinal (aHR, 24.72; 95% CI, 3.34-182.69) disorders was significantly greater in the former. Deaths due to external causes, including suicide, were significantly higher in the CO poisoning group (aHR, 50.07; 95% CI, 30.98-80.90).

Conclusions: Patients with CO poisoning exhibited a heightened risk of all-cause mortality compared with the matched controls. Additionally, the cause-specific mortality risk differed between the groups.