Nebulized Heparin in Adults With Acute Respiratory Failure: A Meta-Analysis of Randomized Trials

Bottom line

In a meta-analysis of 16 RCTs (1,620 patients), nebulized unfractionated heparin was associated with lower all-cause mortality and more ventilator-free days, without a significant increase in major bleeding, though the mortality analysis drew on only 10 of the 16 pooled trials.

PICO at a glance

• Population: Adult patients with acute respiratory failure, whether invasively mechanically ventilated or breathing spontaneously (16 RCTs, 1,620 patients pooled).
• Intervention or exposure: Nebulized unfractionated heparin, a median dose of 25,000 IU per administration, a median daily total of 75,000 IU, and a median maximum treatment duration of 10 days; device findings were exploratory.
• Comparator: Standard care or placebo.
• Outcomes: Primary: all-cause mortality at longest follow-up. Secondary: ventilation-free days by day 28, major bleeding, minor bleeding, heparin-induced thrombocytopenia (HIT).
• Study design: Random-effects meta-analysis of randomized controlled trials.

Clinical question and design

Acute respiratory failure involves dysregulated pulmonary coagulation alongside inflammation, and nebulized unfractionated heparin has been proposed to disrupt this local coagulopathy while carrying a theoretical anticoagulant-sparing safety profile compared with IV heparin because it is delivered directly to the airway. Individual randomized trials have been small and inconsistent, so this meta-analysis pooled 16 RCTs to ask whether nebulized heparin moves the needle on survival and at what bleeding cost.

Published in Critical Care Medicine (2026;54(7):1755-1766; DOI 10.1097/CCM.0000000000007161), investigators systematically searched PubMed, Embase, and the Cochrane Central Register of Controlled Trials (CENTRAL). Two independent investigators extracted trial design, setting, etiology of respiratory failure, heparin dosing regimens, follow-up duration, and outcomes, resolving discrepancies by consensus. Sixteen trials met inclusion criteria, contributing 787 patients randomized to nebulized heparin and 833 to control (1,620 total). Six of the 16 trials (38%) were multicenter, 5 focused specifically on COVID-19-associated respiratory failure, and 12 enrolled ICU patients.

Intervention and outcomes measured

Across included studies, the median dose was 25,000 IU per administration, the median daily dose was 75,000 IU, and the median maximum treatment duration was 10 days. These are separate descriptive summaries of varied trial regimens; they do not establish a universal three-times-daily schedule or a validated ten-day course. The primary outcome was all-cause mortality at longest follow-up.

Results

Across the subset of 10 trials that reported the primary endpoint, nebulized heparin reduced all-cause mortality versus control: 110 of 645 patients (17.1%) versus 157 of 711 patients (22.1%); risk ratio, 0.79 (95% CI, 0.66-0.95), a statistically significant reduction. The trial-level denominators (645 plus 711 equals 1,356) confirm that only 10 of the 16 pooled trials contributed mortality data. Mortality estimates therefore apply to the trials contributing that outcome.

Safety and additional findings

Nebulized heparin was also associated with more ventilation-free days by day 28 (mean difference, +4.85 days; 95% CI, 1.47-8.24). Major bleeding was rare in both arms (1.1% vs. 0.7%) and the difference was not statistically significant (RR, 1.48; 95% CI, 0.42-5.18, crossing 1); no minor bleeding events or heparin-induced thrombocytopenia were reported in either arm.

Uncertainty and limitations

The retained full meta-analysis includes 16 studies (1,620 participants). The median dose was 25,000 IU per administration, the median daily total was 75,000 IU, and the median maximum treatment duration was 10 days. Mortality was reported in 10 trials: 110/645 with inhaled heparin versus 157/711 controls (RR 0.79, 95% CI 0.66-0.95; I² 0%; NNT 20), but GRADE certainty was low for risk of bias and suspected publication bias. Trim-and-fill adjustment changed the mortality estimate to RR 0.83 (95% CI 0.68-1.01). Ventilator-free days increased by 4.85 days (95% CI 1.47-8.24; moderate certainty); major bleeding was reported in six studies (6/548 versus 4/545). Device and COVID subgroup estimates are exploratory and should not select a device or protocol.

Literature review and evidence synthesis

This meta-analysis adds pooled outcome data to a nebulized-heparin literature that has, until now, relied largely on small single-center trials. A 2022 Cochrane rapid review on anticoagulants for people hospitalized with COVID-19 addresses systemic anticoagulant strategies in COVID-19-related respiratory failure, the same population that made up 5 of this analysis's 16 included trials, but via a categorically different route and mechanism: systemic anticoagulation for thromboprophylaxis versus local, nebulized anti-inflammatory and antifibrotic heparin effect.

How this fits with the broader evidence

Nebulized heparin and systemic anticoagulation are different interventions and should not be treated as interchangeable evidence. The apparent mortality benefit in this review has low certainty, and the trim-and-fill sensitivity estimate includes no benefit. Rare bleeding events and imprecise safety estimates do not establish absence of harm. The findings can inform local evidence review, but they do not validate a universal regimen or mandate protocol adoption.

ACPE UAN: 0683-0000-26-036-H01-P

---

Study source: PMID:42148826