Venovenous extracorporeal membrane oxygenation initiation and reduction in vasopressor requirements

To the Editor,

We read with great interest the article by Nagler et al. entitled “Vasopressor requirements after initiation of venovenous extracorporeal membrane oxygenation in patients with severe respiratory failure” [1]. The authors demonstrated a significant reduction in Vasoactive-Inotropic Score (VIS) after V-V ECMO initiation, identifying arterial pH correction as the primary determinant of improved haemodynamic stability.

Acute respiratory distress syndrome (ARDS) is frequently complicated by acute cor pulmonale and right ventricular (RV) dysfunction, both independently associated with adverse outcomes [2]. Hypercapnia and acidosis increase pulmonary vascular resistance, impair RV–pulmonary arterial coupling, and reduce systemic vascular responsiveness to catecholamines. In this context, the finding that arterial pH was independently associated with VIS reduction is physiologically coherent and consistent with the established pathophysiological framework of RV–lung interactions. Restoration of acid–base equilibrium may simultaneously reduce RV afterload and enhance vasopressor sensitivity, thereby contributing to circulatory stabilization. However, several aspects merit further discussion.

First, the interpretation of pH as the dominant driver of vasopressor reduction warrants methodological caution. Arterial pH, PaCO₂, and bicarbonate concentration are tightly interdependent variables. Modelling them simultaneously may introduce collinearity and influence coefficient estimates in multivariable regression. Furthermore, the peri-cannulation period is characterized by time-varying therapeutic interventions. Fluid resuscitation, sedation titration, ventilatory adjustments, and clinician-driven vasopressor weaning frequently occur in parallel with ECMO initiation. Although the mixed-effects model adjusted for selected covariates, residual confounding related to treatment indication or dynamic haemodynamic management cannot be entirely excluded [3].

Second, the timing of haemodynamic assessment deserves particular attention. In the present study, the primary analysis focused on changes occurring within 24 h after ECMO initiation. The first 24 h, however, represent a highly dynamic clinical window. During this period, substantial modifications in preload status, vasoactive support, ventilatory strategy, and systemic inflammation may occur as part of the natural course of critical illness. Consequently, haemodynamic changes observed at 24 h may partially reflect broader stabilization processes rather than the immediate physiological effect of extracorporeal support.

In contrast, very early haemodynamic modifications—within the first hours after cannulation—are more likely to reflect the direct consequences of carbon dioxide removal, pH normalization, and RV afterload reduction. From a physiological standpoint, the haemodynamic response to extracorporeal carbon dioxide removal is expected to be near-immediate if mediated by pulmonary vascular tone modulation. In our experience, we observed measurable reductions in vasopressor requirements as early as 2 and 12 h after V-V ECMO initiation, accompanied by improvements in RV performance indices [4].

Third, the potential contribution of RV unloading extends beyond acid–base correction alone. V-V ECMO may reduce RV afterload through improved oxygenation, attenuation of hypoxic pulmonary vasoconstriction, and enhanced mixed venous oxygen content. Experimental and clinical observations have demonstrated that initiation of V-V ECMO can rapidly improve RV pulmonary artery coupling and reduce RV systolic pressure [5]. The absence of systematic echocardiographic or invasive haemodynamic monitoring in the present study limits the ability to distinguish between acid–base-mediated vasopressor sensitivity and true mechanical unloading of the right ventricle.

Finally, while the reduction in VIS was statistically significant, the magnitude of change in the overall cohort was modest. The more pronounced effect observed in patients with higher baseline noradrenaline requirements is clinically meaningful and likely identifies a subgroup in whom respiratory acidosis and RV afterload mismatch represent dominant contributors to shock physiology [2]. Conversely, persistent vasopressor dependency despite acid–base correction may signal intrinsic myocardial dysfunction or distributive shock mechanisms.

These distinctions have practical implications. High vasopressor requirements at the time of cannulation are often interpreted as an indication to consider venoarterial support. The present findings reinforce the concept that V-V ECMO may exert relevant circulatory effects even in vasopressor-dependent patients, particularly when hypercapnia and acidosis are prominent. Careful haemodynamic phenotyping—including early assessment of RV function and response to carbon dioxide removal—may therefore be more informative than vasopressor dose alone when guiding ECMO configuration.

Overall, Nagler et al. provide important mechanistic insight into the haemodynamic consequences of V-V ECMO initiation. Prospective studies incorporating ultra-early haemodynamic assessment and standardized evaluation of right ventricular function would further clarify the mechanisms underlying haemodynamic improvement after V-V ECMO initiation in patients with severe respiratory failure.

Authors' contributions

RL drafted the manuscript. FS and GL critically revised the manuscript. All authors read and approved the final version.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Funding

None.

Data availability

Not applicable.

Declaration of competing interest

The authors declare that they have no competing interests.