# Vasopressor Requirements after Initiation of Venovenous Extracorporeal Membrane Oxygenation in Patients with Severe Respiratory Failure

## TL;DR

This study found that starting V-V ECMO in patients with severe respiratory failure reduces the need for vasopressors, mainly by correcting acidosis.

## Contribution

The study identifies respiratory acidosis correction as the primary driver of reduced vasopressor use after V-V ECMO initiation.

## Key findings

- Vasopressor requirements decreased significantly after V-V ECMO initiation.
- Higher arterial pH was strongly associated with lower vasopressor scores.
- Increased lactate levels correlated with higher vasopressor requirements.

## Abstract

Patients with severe respiratory failure frequently suffer from concomitant haemodynamic compromise. By correcting respiratory acidosis and permitting reduced mechanical ventilation pressures, venovenous extracorporeal membrane oxygenation (V-V ECMO) may indirectly improve haemodynamics. The aim of this study was to assess how vasopressor requirements changed after V-V ECMO cannulation, and which factors were the primary drivers of this change.

This retrospective single-centre study included 107 consecutive adult recipients of V-V ECMO from 2010 to 2024 who required noradrenaline within 24 h before ECMO cannulation. The primary outcome was the change in Vasoactive-Inotropic Score (VIS) from Day 0 (24 h before) to Day 1 (24 h after ECMO initiation). Secondary outcomes included changes in fluid balance, ventilator settings, blood gas and laboratory parameters. A linear mixed-effects model was used to assess the effects of daily net fluid balance, mean airway pressure (Paw), mean daily pH, arterial partial pressure of oxygen (PaO2), arterial partial pressure of carbon dioxide (PaCO2), mean daily propofol dose, and lactate on the VIS over time (“Day −2” to “Day +3”).

From Day 0 to Day 1, the daily mean VIS significantly decreased from a median of 14 (IQR 6, 30) to a median of 12 (5, 22). This was accompanied by significant reductions in Paw and PaCO₂, and a significant increase in arterial pH (p < 0.001 for all). In the multivariate model, a higher arterial pH was significantly associated with a lower VIS (β = −9.2 per +0.1-unit, p < 0.001). Higher lactate was associated with higher VIS (β = 4.5, p < 0.001). Sensitivity analyses revealed more pronounced effects of pH increase on VIS reduction in patients with high noradrenaline requirements.

After initiation of V-V ECMO, a significant decrease in vasopressor requirements was observed, this benefit being directly attributable to the correction of severe respiratory acidosis.