Comparison of Dynamic Compliance and Airway Resistance in Pressure-Controlled Versus Volume-Controlled Ventilation in Children Undergoing Corrective Heart Surgery

Children with acyanotic left-to-right intracardiac shunt lesions (atrial septal defect or ventricular septal defect) with raised pulmonary artery pressure underwent corrective surgery on cardiopulmonary bypass and were allocated to pressure-controlled or volume-controlled ventilation.

Allocation used pair matching on age, sex and lesion type. One member of each matched pair was assigned by concealed coin toss, and the other was assigned to the alternate group.

Sample size was calculated from the expected difference in mean airway resistance between ventilation modes reported by Tan et al., assuming a standard deviation of 5.82, a between-group difference of 3.9, 5% type I error and 80% power, giving 35 participants per group. Two additional participants were enrolled to allow for protocol deviations or incomplete datasets, giving 36 per group.

Anaesthesia was delivered by a single experienced paediatric cardiac anaesthesiologist so that protocol execution stayed uniform. All intraoperative and postoperative data were recorded by an independent observer blinded to allocation, and analysis used anonymised datasets. The anaesthesiologist managing ventilation could not be blinded to allocation.

Ventilation was standardised across both groups: target tidal volume 6-8 mL/kg, end-tidal carbon dioxide maintained at 35-45 mmHg, positive end-expiratory pressure 4-5 cm H2O, and an inspiratory-to-expiratory ratio of 1:2. Respiratory rate was adjusted to maintain normocapnia and fraction of inspired oxygen was guided by oxygenation status. In the volume-controlled group, tidal volume was set at 6-8 mL/kg. In the pressure-controlled group, inspiratory pressure was titrated to achieve the same tidal volume while keeping peak inspiratory pressure at or below 30 cm H2O.

Respiratory mechanics were recorded at eight predefined perioperative timepoints chosen to represent distinct cardiopulmonary physiological states:

T0: immediate post-intubation T1: ten minutes post-intubation T2: pre-sternotomy T3: pre-bypass T4: immediate post-bypass T5: five minutes after protamine administration T6: post-sternal closure T7: pre-weaning from mechanical ventilation T0 to T3 represented the pre-cross-clamp phase, and T4 to T7 the post-bypass phase. T4 and T5 measurements were taken only after successful separation from bypass, reinstitution of ventilation, restoration of normothermia and haemodynamic stability.

Continuous variables were assessed for normality using the Shapiro-Wilk test. Most perioperative respiratory and haemodynamic variables were non-normally distributed and were expressed as median with interquartile range. Because allocation was pair-matched, matched-pair analysis was used throughout: the Wilcoxon signed-rank test for matched continuous variables and McNemar's test for matched categorical variables, with Benjamini-Hochberg false discovery rate correction applied across repeated perioperative comparisons.

Longitudinal trajectories were additionally modelled using linear mixed-effects models with random patient intercepts, assessing effects of ventilation mode, timepoint, diagnosis, age, body weight, and mode-by-time interaction. All tests were two-tailed with significance at p less than 0.05.

Note I've replaced the en-dashes with hyphens, the subscript ₂ with plain "2", and Shapiro-Wilk/Benjamini-Hochberg dashes with hyphens. PRS is inconsistent with non-ASCII characters and they sometimes come out as mojibake in the public record. Check Preview after saving.