Questioning the classification of “high blood flow” versus “low blood flow” ECCO₂R in ultra-low tidal volume ventilation studies: a call for functional classification

Dear Editor:

We read with great interest the recent study by Monet et al. [1] investigating the feasibility and safety of ultra-low tidal volume ventilation (≤ 3 mL/kg) combined with ECCO₂R in acute respiratory failure. While the study provides valuable insights into lung-protective strategies, we wish to highlight a critical point in the authors' classification of ECCO₂R devices into “high blood flow” (HBF) and “low blood flow” (LBF) groups, which may undermine the validity of their conclusions.

Arbitrary thresholds compromise comparability

The study defines HBF as “blood flow ≥ 1000 mL/min” and LBF as “blood flow < 500 mL/min” without citing standardized criteria. This dichotomy ignores two key issues:

Threshold variability: existing literature uses conflicting cutoffs (e.g., HBF as > 800 mL/min in SUPERNOVA study [2]).

Functional disconnection: blood flow alone poorly predicts CO₂ clearance. For example, the Prismalung + ®(classified as LBF in the study) achieves 90 mL/min above CO₂ removal at 400–450 mL/min blood flow [3], surpassing some “HBF” devices at 500 mL/min with smaller membrane surfaces.

By prioritizing blood flow over CO₂ extraction rate (mL/min) and membrane efficiency (CO₂ clearance per L blood flow), the authors risk misclassifying device performance. A device with 500 mL/min above flow but low membrane efficiency may be functionally inferior to a 400 mL/min device with optimized design, yet both would be grouped differently in this analysis.

Confounding by device heterogeneity

The HBF/LBF grouping aggregates fundamentally distinct technologies: HBF group includes pumpless arteriovenous devices (e.g., iLA Activve®) while LBF group combines with roller pump system (e.g., PrismaLung +®) and centrifugal pump system (e.g., Hemolung Respiratory Assist System®), ignoring their divergent hemodynamic impacts [4].

This heterogeneity introduces unmeasured confounding. For instance, the reported “no significant safety difference” between groups could mask device-specific risks (e.g., hemolysis in centrifugal pumps vs. thrombosis in pumpless systems).

Overlooked interactions with ventilation strategy

The study’s primary endpoint—feasibility of ultra-low tidal volume ventilation—depends on precise CO₂ control, which is determined by ECCO₂R efficiency (CO₂ clearance/mL blood flow), not absolute flow rates. A functional classification based on CO₂ extraction capacity would possibly better predict the ability to maintain pH and PaCO₂ targets.

Therefore, to advance future ECCO₂R research, we propose:

Standardized functional metrics: report CO₂ extraction rate (mL/min) normalized to blood flow (mL/min) and membrane surface area (m²).

Device-specific subgroup analyses: compare outcomes by technology type (e.g., centrifugal vs. roller pump systems) rather than arbitrary flow categories.

Dynamic performance assessment: incorporate real-time CO₂ clearance data during dose titration, as static flow thresholds cannot capture device responsiveness to metabolic demands.

While Monet et al. [1] contribute importantly to the field, re-evaluating ECCO₂R classification strategies is essential to avoid misleading conclusions and guide evidence-based device selection.