Failed Treatment Now Prevents ARDS! *

* In China.

In a long-ago age, before most of you were born (as doctors), there was a magical and hopeful time in critical care, when everyone but everyone believed that ARDS, and sepsis too, would soon be dramatically mitigated by pharmacologic treatments made possible by the impressive and regular basic science breakthroughs emerging from NIH-funded research.

Many compounds were instilled, infused, or sprayed into patients with lung infiltrates and severe hypoxemia receiving mechanical ventilation: aspirin, budesonide, beta-agonists (both IV and inhaled), corticosteroids, statins, N-acetylcysteine, nitric oxide, prostacyclin, ketoconazole, lisofylline, keratinocyte growth factor, stem cells, interferon, and surfactants, to name a lot of them.

Except for corticosteroids, which might have maybe helped and thus ended up on some-but-not-all critical care societies’ recommended treatments for ARDS, no medicine demonstrated significant benefits in mortality or other clinically important endpoints.

---

But there was another agent: sivelestat. An inhibitor of neutrophil elastase (aka leukocyte elastase) that protected twelve 1990s-vintage hamsters from a specific type of endotoxin-induced ARDS.

Neutrophil elastase has many effects in humans, some of which are destructive (e.g., direct tissue damage) and others likely beneficial (e.g., regulation of inflammation) during critical illness. Its role in ARDS is unclear, but there’s abundant circumstantial evidence to implicate it in at least some subtypes of acute lung injury/ARDS.

STRIVE Trial

In the early 2000s, at 105 ICUs in the U.S., Europe, and Australasia, 492 mechanically ventilated patients with ARDS were randomized to receive either sivelestat at a dose of 0.16 mg/kg/hr or placebo for up to 14 days of mechanical ventilation. Eli Lilly, makers of sivelestat, funded the trial.

The trial was stopped early (before target enrollment of n=620) because of the statistical near-impossibility of a reduction in ventilator-free days or 28-day mortality (the co-primary endpoints, which occurred virtually identically between groups).

More concerningly, at six months, there was significantly higher mortality observed in the sivelistat arm (40%) than placebo (31%; p=0.006).

Reported in 2004, the STRIVE trial seemed to end sivelistat’s prospects in the U.S., where the drug was never FDA-approved.

However, STRIVE left open the possibility that although sivelistat might be ineffective at treating ARDS, it might prevent lung injury, especially if given earlier.

Looking East: Sivelistat as Periop Prevention

Sivelistat had already received regulatory approval in 2002 in Japan, and later in South Korea and China, based on earlier low-quality studies showing improvements in lung function. The neutrophil elastase inhibitor has continued to be produced and marketed in Asia as Elaspol® by Ono Pharmaceuticals and its partners.

Two cohort studies conducted in China and published in 2023 and 2024 suggested that sivelistat administered preventively to cardiac bypass patients significantly reduced ARDS (to zero, vs 55% in propensity-matched controls in one study!).

A 2023 meta-analysis of 15 studies, including STRIVE, but mostly in Asia, many non-randomized, and at moderate risk of bias, concluded that sivelistat clearly improved mortality, ARDS, and ICU stays.

A randomized trial followed.

Nanjing Hospital sivelistat trial

At a single tertiary cardiac bypass surgery center (Nanjing Drum Tower Hospital) in China, 424 patients were randomized to receive sivelistat or placebo within 6 hours of ICU admission in 2024-2025. Rigorous trial processes of randomization, blinding, and outcome measurement were described in the manuscript.

The authors report that sivelestat roughly halved the incidence of postoperative ARDS (16.8% vs 31.2%) and significantly reduced 90-day all-cause mortality (1.1% vs 5.2%), compared with placebo. There was no increase in mortality at 180 days.

Cases of ARDS in the sivelestat arm that did occur were milder, and there were no severe cases. Sivelistat-treated patients also had lower rates of adjudicated pneumonias and lower levels of numerous inflammatory markers. The results were published in JAMA Network Open in 2026.

Discussion

Keep in mind, this is not your grandfather’s ARDS, nor your fellowship attending’s. Retrospective U.S.-based cohorts estimate the incidence of postoperative ARDS after cardiopulmonary bypass to be 1-5%, and 15% in the very highest-risk cohorts. Those represent the more clinically significant, unequivocal cases.

More recent prospective studies, especially out of Asia, commonly found ARDS rates to be much higher (20-30%), and this trial’s was among the very highest (31% in the placebo arm). The higher prospective incidences may be due to closer surveillance and stricter application of definitions, including the inclusion of patients on HFNO immediately post-extubation, for example (endorsed by the newest Berlin definitions).

---

The prophylactic postoperative infusions of sivelistat in the Nanjing Drum Tower Hospital trial can’t be compared with STRIVE, which delivered sivelistat to medical patients who had already developed ARDS.

That said, dramatic reported reductions in a condition like ARDS that has already proved to be highly resistant to intervention should raise skepticism, especially in a single-center trial. Triple the skepticism when it’s accompanied by an 80% mortality reduction. Add in a skepticism multiplier when it comes from a nation that has produced a disproportionate share of research findings that were later retracted.

---

There is nothing in the NDTH trial that suggests fraud or misconduct, and it is deeply unfair that the majority of honest, good-faith researchers in any country should labor under a cloud of skepticism based on their countrymen’s past behavior. But the truth is that neither we (nor the editors and reviewers at JAMA, for that matter) have any reliable way to vet the research coming out of China.

Suffice it to say that the finding that preventively-administered sivelistat dramatically reduces ARDS and mortality after cardiopulmonary bypass surgery would be a remarkable and practice-changing development—if later replicated in multicenter randomized trials conducted in the U.S., Europe, and Australasia.

Sivelistat is not approved in the U.S. today for any indication. At least one other randomized trial testing sivelistat further for the prevention of ARDS is in process (in China).

References

Pan T, Xu C, Wang YP, et al. Sivelestat and Incidence of Acute Respiratory Distress Syndrome After Cardiovascular Surgery. JAMA Network Open. 2026;9(3):e260390. doi:https://doi.org/10.1001/jamanetworkopen.2026.0390

‌Lee WL, Slutsky AS. Preventing ARDS, Not Treating It—Lessons From Sivelestat After Cardiopulmonary Bypass. JAMA Network Open. 2026;9(3):e261201. doi:https://doi.org/10.1001/jamanetworkopen.2026.1201

Neutrophil Elastase Inhibition in Acute Lung Injury: Results of the STRIVE Study. Critical Care Medicine. 2004. Zeiher BG, Artigas A, Vincent JL, et al.‌

Sivelestat for Septic Patients With Acute Respiratory Distress Syndrome: A Systematic Review and Meta-Analysis of a Deadly Duo. Frontiers in Medicine. 2025. Zheng WH, Hu YG, Yu DX, Huang HB.

Effect of Sivelestat Sodium in Patients With Acute Lung Injury or Acute Respiratory Distress Syndrome: A Meta-Analysis of Randomized Controlled Trials. BMC Pulmonary Medicine. 2017. Pu S, Wang D, Liu D, et al.

Therapeutic Effect and Mechanism of Sivelestat Sodium on Acute Lung Injury: A Randomized Controlled Trial. Medicine. 2025. Zhou Y, Chen G, Xu J, et al.

Neutrophil Elastase Inhibitor (Sivelestat) in the Treatment of Acute Respiratory Distress Syndrome Induced by COVID-19: A Multicenter Retrospective Cohort Study. Respiratory Research. 2025. Li Y, Zhao J, Wei J, et al.