Right ventricular free wall longitudinal strain during weaning from mechanical ventilation using high-flow or conventional oxygen treatment: a pilot study

To the Editor,

Previous studies on cardiovascular system alterations due to transition from positive to negative intrathoracic pressure because of withdrawal of mechanical ventilation (MV) and resumption of spontaneous breathing focused on left ventricular function [1,2,3]. Relevant studies on right ventricular (RV) function are scarce [4,5,6]. Advancements in echocardiography, namely two-dimensional speckle-tracking echocardiography, provide novel parameters, such as RV free wall longitudinal strain (RVFWSL) that may outperform conventional measurements [7]. There is no information on RVFWSL during weaning from MV. We investigated the effect of spontaneous breathing trial (SBT), supported by high-flow oxygen treatment (HFOT), or by T-piece, on RV function in patients with prolonged MV.

Nine tracheostomized patients on MV (assist controlled mode) underwent a 30-min SBT receiving oxygen either via T-piece, or by HFOT, through tracheostomy with a flow rate of 60 L/min, followed by a washout period of 15 min on MV and 30 min with the other modality in a randomized crossover manner. Transthoracic echocardiography was performed on MV (baseline) and at the end of each SBT. Images were acquired in cine-loop format from several consecutive beats and analyzed offline (EchoPAC Version 204; GE Healthcare, Chicago, IL). A repeated-measures analysis of variance (ANOVA) was used to compare echocardiographic and physiological parameters during MV, HFOT, and T-piece. All statistical tests were two tailed; significance was defined as p < 0.05.

Fourteen SBT sessions were completed; five patients were studied on two different days with alternative sessions and four patients once. Three patients were successfully weaned from MV (Table 1). There was no significant difference in RVFWSL values measured during MV, HFOT, and T-piece (22.3%, 25.0%, and 24.7%, respectively, p = 0.415). Higher baseline RVFWSL values were associated with successful weaning (25.9 vs 20.4%, p = 0.045, Table 2).

Table 1 Baseline characteristics and outcomes of the included patients

Table 2 Echocardiographic and physiological parameters of included patients undergoing 14 sessions of spontaneous breathing trial

Our finding of a well-preserved RV response to successful SBTs, as opposed to MV, is consistent with earlier reports using pulmonary artery catheterization [4,5,6]. Since positive pressure ventilation and particularly positive end-expiratory pressure (PEEP), in most situations, reduce venous return and may increase pulmonary vascular resistance (i.e., RV afterload) [3], one would expect that MV or a modality that applies PEEP could adversely affect the RV function, compared to spontaneous breathing. Nevertheless, this may occur when PEEP induces lung overdistention [8] or in the presence of RV dysfunction [9]. We found no difference in RVFWSL between T-piece and MV, possibly due to the low level of PEEP applied during MV (Table 2). Also, we found no difference in RVFWSL between T-piece and HFOT. Although HFOT via nasal cannula generates some PEEP, HFOT via tracheostomy possibly provides lower degree of PEEP, even with the highest flow [10, 11], attributable to the fact that tracheostomy bypasses the upper airways [11, 12]. Thus, the effect of HFOT via tracheostomy on RV seems negligible. On the contrary, our finding that higher baseline RVFWSL values were associated with successful weaning suggests a role of RV function in weaning outcome. Finally, our finding that right atrial reservoir strain increased after discontinuation of MV (Table 2) possibly signifies improvement of right atrial filling secondary to decreased intrathoracic pressure.

In summary, in tracheostomized patients with prolonged MV, RVFWSL was maintained during successful SBTs, supported by HFOT or T-piece. Higher baseline RVFWSL values were associated with successful weaning.