Australasian Medical Publishing co. All rights reserved. Objective: Lack of management guidelines for life-threatening asthma (LTA) risks practice variation. This study aims to elucidate management practices of LTA in the intensive care unit (ICU). Design: A retrospective cohort study. Setting: Thirteen participating ICUs in Australia between July 2010 and June 2013. Participants: Patients with the principal diagnosis of LTA. Main outcome measures: Clinical history, ICU management, patient outcomes, ward education and discharge plans. Results: Of the 270 (267 patients) ICU admissions, 69% were female, with a median age of 39 years (interquartile range [IQR], 26–53 years); 119 (44%) were current smokers; 89 patients (33%) previously required ICU admission, of whom 23 (25%) were intubated. The median ICU stay was 2 days (IQR, 2–4 days). Three patients (1%) died. Seventy-nine patients (29%) received non-invasive ventilation, with 11 (14%) needing subsequent invasive ventilation. Sixty-eight patients (25%) were intubated, with the majority of patients receiving volume cycled synchronised intermittent mechanical ventilation (n = 63; 93%). Drugs used included ß 2 -agonist by intravenous infusion (n = 69; 26%), inhaled adrenaline (n = 15; 6%) or an adrenaline intravenous infusion (n = 23; 9%), inhaled anticholinergics (n = 238; 90%), systemic corticosteroids (n = 232; 88%), antibiotics (n = 126; 48%) and antivirals (n = 22; 8%). When suitable, 105 patients (n = 200; 53%) had an asthma management plan and 122 (n = 202; 60%) had asthma education upon hospital discharge. Myopathy was associated with hyperglycaemia requiring treatment (odds ratio [OR], 31.6; 95% CI, 2.1–474). Asthma education was more common under specialist thoracic medicine care (OR, 3.0; 95% CI, 1.61–5.54). Conclusion: In LTA, practice variation is common, with opportunities to improve discharge management plans and asthma education.