# Effects of two ventilator-weaning methods on lung volume and ventilation distribution by electrical impedance tomography in post-cardiac surgery patients: a prospective cohort study

## TL;DR

This study compares two ventilator-weaning methods in post-cardiac surgery patients using electrical impedance tomography to assess lung volume and ventilation distribution.

## Contribution

The study introduces a novel use of EIT to evaluate ventilator weaning methods and identifies patterns of volume loss associated with weaning success or failure.

## Key findings

- Failure group showed higher pendelluft and greater EELVloss during T-piece SBT.
- Success group had greater EELVloss during PSV-SBT, indicating an abnormal pattern.
- T-volume loss group had higher weaning failure rates and reduced baseline dorsal ventilation.

## Abstract

The effect of different spontaneous breathing trial (SBT) methods on lung volume and ventilation distribution has not been well clarified in post-cardiac surgery patients.

In this prospective observational study, patients underwent 30 min of pressure-support ventilation (PSV)-SBT [PS 8 cmH2O, zero positive end-expiratory pressure (ZEEP)], followed by a 30-min T-piece trial if tolerated. Electrical impedance tomography (EIT) was used to continuously monitor regional lung ventilation and end-expiratory lung volume (EELV) at baseline, PSV-SBT 3 min, PSV-SBT 30 min, T-piece SBT 3 min and T-piece SBT 30 min. EELVloss = [VTbaseline/tidal impedance variation (TIV)baseline] × ΔEELI. EELVloss PSV was defined as volume loss at 30 min of PSV-SBT and EELVloss T-piece was defined as volume loss during T-piece SBT.

In 60 patients who complied with both SBT steps, 43 succeeded (71.7%) and 17 failed (28.3%) the T-piece SBT. Compared to the success group, the failure group exhibited a higher incidence of pendelluft (52.9% vs. 23.3%, p = 0.045) and significantly greater EELVloss at T-piece SBT 30 min (623[459,746] ml vs. 511[376,702]ml, p = 0.003). However, the success group showed greater EELVloss PSV than the failure group (322[247,459] ml vs. 199[166, 269] ml, p < 0.001), which was an abnormal pattern. Notably, the failure group had lower TIV (2102[1769,2562] vs. 2742[2153,3872], p = 0.005) and a higher respiratory rate (RR) than baseline at PSV-SBT 30 min (20[17,24] vs. 16[12,18], p < 0.001). Furthermore, we classified all patients into two groups based on the predominant reduction of volume loss: P-volume loss group (N = 37, EELVloss PSV > EELVloss T-piece) and T-volume loss group (N = 23, EELVloss T-piece > EELVloss PSV). In addition, the T-volume loss group had a higher weaning failure rate than the P-volume loss group (52.2% [12/23] vs. 13.5% [5/37], p < 0.001) and was associated with reduced baseline dorsal ventilation (39%[37%,43%] vs. 44%[41%,50%], p = 0.023). ROC analysis suggested that a dorsal ventilation threshold of 40.5% was associated with T-volume loss.

The successful weaning patients had a higher reduction of EELVloss PSV and a lower reduction of EELVloss T-piece. In the weaning failure patients, the paradox of lower EELVloss PSV that was accompanied by a high RR and low VT might be associated with air trapping. Attention should be paid to using EELVloss PSV to identify weaning failure.

The online version contains supplementary material available at 10.1186/s40560-026-00850-1.