Sinnige JS, Buiteman-Kruizinga LA, Horn J, Paulus F, Schultz MJ, Serpa Neto A; for the ACTiVE Investigators and the Protective Ventilation Network. Effect of Automated Closed-Loop Ventilation vs Protocolized Conventional Ventilation on Ventilator-Free Days in Critically Ill Adults: A Randomized Clinical Trial. JAMA. Published online December 8, 2025. doi:10.1001/jama.2025.24384
Automated closed-loop ventilation systems have generated considerable interest in critical care, promising real-time optimization of ventilator settings through continuous physiologic feedback. INTELLiVENT adaptive support ventilation (INTELLiVENT-ASV, Hamilton Medical AG) is among the most advanced of these systems, capable of managing patients from initiation of ventilation through extubation. Earlier studies suggested improvements in ventilation quality, reduced manual adjustments, and enhanced adherence to lung-protective strategies. But whether these physiologic advantages translate into better patient-centered outcomes has remained an open question — until now.
The ACTiVE Trial
The ACTiVE (Effects of Automated Closed-Loop Ventilation vs Conventional Ventilation on Duration and Quality of Ventilation) trial was an investigator-initiated, international, multicenter, randomized superiority trial conducted at 7 ICUs across the Netherlands and Switzerland. Between October 2020 and June 2025, 1,514 adult patients were randomized within 1 hour of initiating invasive mechanical ventilation, with 1,201 included in the primary analysis (602 closed-loop, 599 conventional). Patients were expected to require ventilation for 24 hours or longer. Final follow-up was completed at 90 days.
Patients in the closed-loop group received INTELLiVENT-ASV, which continuously adjusted tidal volume, respiratory rate, FiO₂, and PEEP based on the Otis equation and Mead formula, targeting optimal minute ventilation and end-tidal CO₂ while automatically adjusting oxygenation parameters via SpO₂ targets. Weaning in this group — including transitions from controlled to assisted modes — was fully automated. The conventional group received volume- or pressure-controlled ventilation with protocolized lung-protective strategies, including thrice-daily weaning assessments and standardized SBTs. Importantly, both groups followed the same sedation, ventilator management, and liberation protocols, creating a rigorous head-to-head comparison.
Primary Outcome
The primary outcome was ventilator-free days at day 28. The median was 16.7 days (IQR, 0.0–26.1) in the closed-loop group and 16.3 days (IQR, 0.0–26.5) in the conventional group (odds ratio 0.91; 95% CI, 0.77–1.06; P = .23). This difference was not statistically significant, and no treatment × subgroup interactions were identified across six prespecified subgroups including admission type, neurological admission, cardiac arrest, hypoxemic respiratory failure, BMI, and severity of illness.
Secondary Outcomes and Safety
There were no significant differences in 28-day mortality (37.2% vs 36.5%), duration of ventilation among survivors (3.3 vs 2.6 days), ICU or hospital length of stay, or 90-day mortality between groups. Ventilation complications — ARDS, VAP, atelectasis, and pneumothorax — were similarly distributed.
Where the closed-loop group did show a signal was in ventilation quality. In a subset of 152 patients with granular breath-by-breath data, closed-loop ventilation was associated with higher odds of being in a more favorable ventilation zone (OR 1.50; 95% CI, 1.43–1.57; P < .001), spending more time in the optimal zone (48.4% vs 36.7%) and less time in the critical zone (11.6% vs 35.8%). Severe hypercapnia and hypoxemia were also less frequent in the closed-loop group, and fewer patients required rescue therapies — primarily prone positioning (9.2% vs 13.9%) — though these differences did not reach statistical significance after multiplicity adjustment.
Clinical Relevance
This is the largest and most methodologically rigorous randomized trial to date comparing automated closed-loop ventilation with conventional protocolized ventilation. The central takeaway is clear: early initiation of INTELLiVENT-ASV did not improve ventilator-free days at day 28 when compared against well-protocolized conventional care.
That said, the ventilation quality data are noteworthy. The closed-loop system consistently kept patients in more protective ventilation zones, and the trends toward fewer rescue therapies and less severe gas exchange derangements suggest real physiologic advantages — even if these did not translate into a detectable difference in the primary endpoint. This raises the question of whether the benefit of automation may be more apparent in settings with less structured ventilator management, fewer resources, or less experienced staff, where adherence to best practices is more variable.
Limitations
Several important caveats deserve attention. First, both groups received highly protocolized, resource-intensive care with thrice-daily weaning assessments — a standard that exceeds what many ICUs achieve in routine practice. This may have narrowed the performance gap between strategies. Second, the trial did not capture data on clinician workload or staffing requirements, which could be meaningful differentiators in real-world practice. Third, the ventilation quality analysis was limited to a small, non-systematically sampled subset and relied on subjective classification thresholds. Fourth, the heterogeneous patient population may have diluted effects in specific subgroups, and the relatively short ventilation courses for most patients may have limited the time available for automation to demonstrate a cumulative benefit. Finally, the study evaluated a single closed-loop platform (INTELLiVENT-ASV), and results may not generalize to other automated systems.
Bottom Line
The ACTiVE trial provides strong evidence that automated closed-loop ventilation does not increase ventilator-free days compared with protocolized conventional ventilation in a well-resourced, protocol-driven ICU environment. However, the observed improvements in ventilation quality and trends toward fewer rescue therapies suggest that the conversation around automation in mechanical ventilation is far from over. Future trials should examine whether these systems offer greater benefit in less controlled settings, and whether improvements in process measures like ventilation quality and clinician workload ultimately matter for patients.
Reference: Sinnige JS, Buiteman-Kruizinga LA, Horn J, et al. Effect of Automated Closed-Loop Ventilation vs Protocolized Conventional Ventilation on Ventilator-Free Days in Critically Ill Adults: A Randomized Clinical Trial. JAMA. Published online December 8, 2025. doi:10.1001/jama.2025.24384
Trial Registration: ClinicalTrials.gov Identifier: NCT04593810
I think you are missing the point here. Where Closed loop automated ventilation will shine is in small hospitals or remote hospitals that don’t have ventilation experts. Not every hospital has the kind of expertise or staffing, found in teaching centres. Automated ventilation will significantly improve outcomes in rural or remote hospitals. These modes can be initiated by nursing or a family physician who is covering emergency for the weekend. Automated ventilation may not be better than the experts running the ventilators in a downtown centre of excellence, but I’ll take just as good in a small hospital setting any day.
Did I miss the point? or did you skip the section on clinical relevance? 😊
This post is a summary of the actual study. Not an opinion piece. This summary addresses your comments in both the Clinical Relevance and Bottom Line sections, because the actual study did as well.