Invited Commentary 

Cardiology

Preventing ARDS, Not Treating It—Lessons From Sivelestat After Cardiopulmonary Bypass

Warren L. Lee, Arthur S. Slutsky

JAMA Netw Open 2026;9;(3):e261201. doi:10.1001/jamanetworkopen.2026.1201

Pan and colleagues¹ report the findings of a large, blinded, placebo-controlled single-center randomized clinical trial of sivelestat for the prevention of acute respiratory distress syndrome (ARDS) among patients undergoing elective cardiovascular surgery. Sivelestat, an inhibitor of neutrophil elastase, was administered within 6 hours of admission to the intensive care unit (ICU) after surgery and continued for 7 days or until discharge from the ICU. The primary end point was the incidence of ARDS. The study randomized 424 patients, of whom 382 completed the trial; 42 patients withdrew before receiving any intervention. The authors observed a 31.2% incidence of postoperative ARDS in the placebo group compared with 16.8% in the sivelestat group. In addition to being fewer in number, cases of ARDS in the sivelestat arm were less severe (eg, fewer cases of moderate severity and no severe cases). The sivelestat arm exhibited a significantly lower 90-day mortality rate than the placebo arm (1.1% vs 5.2%) as well as a lower rate of pneumonia. These improvements in clinical end points were accompanied by a panoply of significantly lower inflammatory markers at all time points measured.

The beneficial findings may at first seem surprising to many clinicians, for whom neutrophil elastase in ARDS is probably remembered as a failed therapeutic target. Some will recall the STRIVE trial (Sivelestat Trial in ALI Patients Requiring Mechanical Ventilation) published in 2004, which randomized almost 500 patients with ARDS to a continuous infusion of sivelestat or placebo.² In that study, there was no effect of sivelestat on 28-day mortality, but there was an increase in 180-day all-cause mortality in the sivelestat group compared with the placebo group.

The failure of sivelestat in STRIVE largely extinguished enthusiasm for elastase inhibition in ARDS, particularly in North America, in spite of compelling preclinical data implicating neutrophil elastase in the pathogenesis of ARDS.³ However, supportive retrospective analyses and a recent meta-analysis⁴ have prompted another look at sivelestat. At present, sivelestat is approved for use in parts of Asia but not in North America or Europe.

How do we reconcile the results of the study by Pan et al¹ with STRIVE? The simple answer is that the studies are fundamentally different. STRIVE randomized patients with ARDS to the drug, on average, 28 hours after the start of mechanical ventilation, and most patients had an infectious etiology. In contrast, Pan and colleagues¹ administered sivelestat prophylactically, after surgery within 6 hours of admission to the ICU. This was a preventive rather than a therapeutic study that benefitted from a clearly defined risk factor for ARDS (ie, cardiopulmonary bypass) whose time of onset was unambiguous. The patient populations in the 2 studies were also quite different. In STRIVE, patients came from the US, Canada, Belgium, Spain, Australia, and New Zealand; in the present study, all of the patients came from China.

There is much to appreciate about this study. It was rigorously conducted, incorporating protocolized ICU care for lung-protective mechanical ventilation, fluid administration, and antibiotic use. The diagnosis of ARDS was based on the Berlin Definition, and the authors collected extensive clinical and biological data throughout the study. The main results were highly statistically significant and robust to extensive sensitivity analyses that supported the primary outcome. The authors also documented biological efficacy of the drug, demonstrating significant reductions in neutrophil elastase levels and accompanying reductions in other biomarkers, such as interleukin 6 (IL-6) and procalcitonin. Given the prior STRIVE trial, the authors note that mortality at 90 days, which was significantly lower among patients who received sivelestat, was identical to 180-day mortality.

However, it is the generalizability of the results to other cohorts of patients undergoing cardiovascular surgery that remains a critical question. The incidence of ARDS in this study was substantially higher—by at least an order of magnitude—than reported in large retrospective cardiovascular surgery cohorts.⁵ Pan and colleagues¹ attribute the elevated rate of ARDS to the high-risk nature of their patient population, but clinicians may reasonably question whether this elevated rate aligns with their own experience. Although the study excluded patients estimated to be at high risk of postoperative mortality (based on a EuroSCORE II [European System for Cardiac Operative Risk Evaluation II]⁶ estimated mortality rate >3%), the observed mortality rate in the placebo group was more than 5%, consistent with a high-risk population.

Why did this group of patients develop such a high incidence of ARDS? Could the increased incidence be due to differences in the operative approaches between Chinese vs US or European cohorts? Could genetic differences be contributing?⁷ We can only speculate, but the unexpectedly high incidence of postoperative ARDS in this study will make it difficult to generalize its results. Along the same lines, other exclusion criteria for the trial will also limit generalizability. For example, patients exhibiting evidence of inflammation preoperatively (eg, elevated IL-6, procalcitonin, or C-reactive protein levels), as well as those requiring emergency surgery, were not enrolled. Nonetheless, the findings raise the interesting question of whether sivelestat could be beneficial prophylactically in other situations predisposing to ARDS, where relatively early treatment may be feasible, such as major trauma or ischemia-reperfusion injury.

In conclusion, this study provides provocative evidence that postoperative inhibition of neutrophil elastase can reduce the incidence of ARDS and mortality after elective cardiovascular surgery in a Chinese cohort. We concur with the authors that confirmation in larger, multicenter trials will be necessary. Taking a broader perspective, this study adds to accumulating data that anti-inflammatory therapy has an important place among selected patients either for the prevention or the treatment of ARDS. Choosing the right patients, however, is critical.