Effect of driving pressure-guided individualized PEEP in children undergoing right lateral thoracotomy cardiac surgery with cardiopulmonary bypass on postoperative atelectasis and oxygenation: a randomized controlled trial Authors: Xinyue Chen, Haicheng Song, Mengxuan Zang, Jiahui Zhou, Yumei Liu, Yueyi Ren Published: 2026 ABSTRACT BackgroundRight lateral thoracotomy cardiac surgery induces postoperative atelectasis due to cardiopulmonary bypass (CPB), patient positioning, and mechanical ventilation. Fixed positive end-expiratory pressure (PEEP) may fail to meet pediatric ventilation requirements. This study aims to investigate the effects of driving pressure-guided individualised PEEP on postoperative atelectasis and oxygenation.MethodsChildren aged 6 months to 5 years undergoing right lateral thoracotomy cardiac surgery with CPB were randomly assigned to two groups. Group C received conventional protective ventilation with fixed PEEP of 5 cmH2O in the lateral decubitus position until the end of ventilation. Group D had driving pressure-guided individualized PEEP titration following both the lateral decubitus position and the CPB, with ventilation maintained at the titrated PEEP. Both groups received lung recruitment maneuvers. Observations were conducted at the following time points: after intubation (T0), after lateral positioning (T1), before CPB (T2), after CPB (T3), at the end of surgery (T4), before leaving the operating room (T5), after admission to the ICU (T6), and 1 h after extubation (T7). Lung ultrasonography was performed for each patient at T0 and T5. The primary outcomes were the incidence of postoperative atelectasis and lung ultrasound scores (LUS). Secondary outcomes included PaO2/FiO2 ratio (P: F ratio), incidence of post-CPB hypoxemia and postoperative pulmonary complications (PPCs) within 3 days, peak airway pressure (Ppeak), dynamic pulmonary compliance (Cdyn), PEEP, driving pressure, extubation time, length of ICU stay, length of hospital stay, heart rate and mean arterial pressure.ResultsAt T5, the incidence of atelectasis and LUS were lower in Group D than in Group C (P < 0.05). P: F ratio in Group D was higher than in Group C at T2, T4, and T6 (P < 0.05). The incidence of post-CPB hypoxemia and extubation time were lower in Group D (P < 0.05). Group D showed superior intraoperative respiratory mechanics parameters. However, there were no significant differences in the incidence of PPCs within 3 days, intraoperative hemodynamics, ICU length of stay, or hospital length of stay between the two groups (P > 0.05).ConclusionsDriving pressure-guided individualized PEEP ventilation reduces the incidence of postoperative atelectasis in children undergoing right lateral thoracotomy cardiac surgery with CPB. It also improves oxygenation and intraoperative respiratory mechanics. However, there was no significant difference in the incidence of PPCs within 3 days.Trial registrationChiCTR, ChiCTR2500098036. Registered on February 28, 2025, www.chictr.org.cn. Introduction Right lateral thoracotomy offers advantages in treating children with uncomplicated congenital heart disease due to its smaller incision, aesthetic appeal, and faster recovery [1]. Atelectasis is a common postoperative complication, primarily caused by reduced lung volume and airway closure, and is a major contributor to intraoperative and postoperative hypoxemia and PPCs [2–4]. The immature respiratory system in children is characterized by low functional residual capacity, high closing volume, and elevated oxygen consumption. Combined with the effects of surgical manipulation, inflammatory responses after CPB, and mechanical ventilation, these factors predispose pediatric patients to early postoperative respiratory insufficiency [5, 6]. Lung-protective ventilation strategy integrating tidal volume, lung recruitment, and PEEP is a key focus in perioperative mechanical ventilation management. It optimizes intraoperative ventilation settings to provide effective lung protection and mitigate pulmonary injury, with its application during surgery now widely recognized [7]. However, due to the heterogeneity of pediatric respiratory systems, traditional fixed PEEP settings struggle to meet individualized patient needs. Individualized PEEP effectively reduces lung injury by maintaining alveolar recruitment, increasing functional residual capacity, and improving the ventilation-perfusion ratio [8, 9]. According to a previous study, driving pressure serves as a key mediator influencing the occurrence of PPCs during protective ventilation [10]. Current studies indicate that driving pressure-guided individualized PEEP holds significant potential for perioperative lung protection during mechanical ventilation, optimizing intraoperative respiratory mechanics, improving oxygenation, and reducing postoperative atelectasis and the incidence of PPCs [11–13]. However, there are few reports on the application of this method in protective ventilation during right lateral thoracotomy pediatric cardiac surgery. This randomized controlled study aims to evaluate the effects of driving pressure-guided individualized PEEP on postoperative atelectasis and oxygenation in children undergoing right lateral thoracotomy cardiac surgery with CPB. Methods This was a single center, investigator-initiated, assessor-blinded randomized clinical trial conducted in the Qingdao Women and Children’s Hospital in China. This study was approved by the Ethics Committee of Qingdao Women and Children’s Hospital (Approval No. QFELL-KY-2024-12) and registered with the China Clinical Trial Registry (ChiCTR2500098036). The study was conducted according to the Consolidated Standards of Reporting Trials statement. Researchers assessed pediatric patients’ eligibility, obtained informed consent from their families, and secured signed informed consent forms. The study cohort comprised 80 children with congenital heart disease who underwent right lateral thoracotomy cardiac surgery with CPB at our hospital from March 2025 to June 2025. Inclusion criteria No restriction on gender; aged 6 months to 5 years; classified as American Society of Anesthesiologists (ASA) physical status II ~ III; scheduled to undergo right lateral thoracotomy cardiac surgery with cardiopulmonary bypass. Exclusion criteria Refusal of the child’s legal guardian to participate in this clinical trial; need for emergency surgery or conversion to open thoracotomy during the procedure; use of vasoactive drugs for hemodynamic support before anesthesia; history of respiratory tract infection within two weeks prior to elective surgery; prior thoracic or cardiac history such as genetic disorders, respiratory diseases, airway or chest wall abnormalities; abnormal preoperative chest X-ray or CT findings including pleural effusion, pneumothorax, atelectasis, or pneumonia. Anesthesia Anesthesia was induced with sufentanil (0.5 ~ 1 µg/kg), propofol (2 ~ 3 mg/kg), and rocuronium (0.6 mg/kg), and maintained with continuous infusions of propofol(2 ~ 6 mg/kg/h), remifentanil༈0.2 ~ 0.3 µg/kg/h༉, and dexmedetomidine༈0.6 µg/kg/h༉, supplemented with intermittent sevoflurane and rocuronium as needed. Doses were titrated to maintain hemodynamic stability and adequate anesthesia depth. All patients received standardized intraoperative monitoring and postoperative care in the ICU. Following extubation, administer 1 L/min of pure oxygen via nasal cannula. Perioperative ventilation settings and study protocol During preoxygenation, all pediatric patients received mask-assisted ventilation with pure oxygen. After intubation, volume-controlled ventilation was employed with a tidal volume (VT) of 6 ~ 8 ml/kg. Configure the inspiratory/expiratory ratio at 1:1.5 ~ 2, FiO2 at 50%, and PEEP at 2 cmH2O. Adjust the fresh gas flow rate to 1.5 L/min and modify respiratory rate to keep PETCO2 between 35 and 45 mmHg. After positioning the pediatric patient in the lateral decubitus position, PEEP was set according to the case group assignment. In this study, both groups received the same lung recruitment maneuver, which was performed after lateral positioning and before weaning from CPB. The alveolar recruitment maneuver used the vital capacity method, with peak airway pressure maintained at 30 cmH2O for 10 s, repeated three times. In Group C, after the lung recruitment maneuver, PEEP was adjusted from 2 cmH2O to 5 cmH2O and maintained until the end of ventilation. In Group D, driving pressure-guided PEEP titration was performed after the lung recruitment maneuver, which was conducted following lateral positioning and completion of CPB. PEEP titration began at 2 cmH2O and increased incrementally by 1 cmH2O until reaching 10 cmH2O. Each PEEP level was maintained for 10 respiratory cycles, with the driving pressure(Driving pressure = Plateau Pressure – PEEP) recorded during the final cycle. Upon completion of titration, the PEEP level yielding the lowest driving pressure was selected for subsequent ventilation. If identical driving pressures were achieved at different PEEP levels, the lower PEEP value was chosen. After the child has been weaned from CPB and is hemodynamically stable, repeat the PEEP titration using the method described above. If interference occurred during surgical procedures, additional titration was performed. Prior to weaning from CPB, all patients underwent alveolar recruitment maneuvers with peak airway pressure maintained at 30 cmH2O for 15 s. Ventilation was interrupted during cardiopulmonary bypass in both groups, while the ventilator circuit remained connected to the patient. Lung ultrasound assessment All pediatric patients underwent pulmonary ultrasound examinations both after induction and prior to leaving the operating room. These examinations were performed according to Monastesse’s protocol [14] by an anesthesiologist with over three years of pulmonary ultrasound experience who was not involved in this study. The Philips IE33 ultrasound machine (Amsterdam, Netherlands) with a 7 to 15 MHz linear transducer was used. During the examination, the child was placed in the supine position with both hands raised overhead. The thoracic cavity is divided into six regions on each side using the parasternal line, anterior axillary line, posterior axillary line, and the line connecting the two nipples, totaling twelve regions bilaterally. Each region is scored using the Lung Ultrasound Score (LUS), with a range of 0 to 3 points per region. Specific scoring criteria are as follows: (0) 0–2 B-lines; (1) ≥ 3 B-lines; (2) clustered B-lines or small subpleural consolidation; (3) evidence of pulmonary consolidation. If multiple ultrasound features are present within the same segment, the highest score is recorded. The total score assesses the degree of impaired lung ventilation, with values positively correlated to the severity of atelectasis. A score of ≥ 2 points in any of the twelve regions indicates positive atelectasis [15], and the incidence of postoperative atelectasis is calculated accordingly. Arterial blood gas monitoring Perioperative arterial blood gas analysis was performed at five time points: after induction (T0), before CPB (T2), at the end of surgery (T4), after admission to the ICU (T6), and 1 h after extubation (T7). The P: F ratio, defined as the PaO2/FiO2 ratio and calculated using the standard formula PaO2/FiO2, was used to determine the occurrence of post-CPB hypoxemia. Hypoxemia was defined as P: F ratio ≤ 300. PPCs This study followed up on the occurrence of PPCs within 3 days in pediatric patients, using the composite outcome measure defined by the European Society of Anaesthesiology [16]. PPCs primarily included respiratory tract infection, respiratory failure, pleural effusion, atelectasis, and pneumothorax. Respiratory tract infection was defined as suspected respiratory infection treated with antibiotics and meeting at least one of the following criteria: new onset or altered sputum characteristics; new or progressive pulmonary infiltrates; fever > 38℃; Respiratory failure was defined as postoperative PaO2 < 60 mmHg without supplemental oxygen, an P: F ratio < 300 mmHg, or SpO2 < 90% requiring oxygen therapy. Pleural effusion, atelectasis, and pneumothorax were defined based on postoperative chest imaging findings. The complete protocol is illustrated in Fig. 1. Collect the following data from each patient: age, height, weight, gender, diagnosis, duration of anesthesia during surgery, duration in lateral position, duration of aortic cross-clamping, duration of cardiopulmonary bypass, intraoperative heart rate and mean arterial pressure, respiratory mechanics, extubation time, ICU stay duration, and length of hospital stay. Fig. 1Research flowchart Research flowchart Primary outcomes Incidence of atelectasis at T5. Incidence of atelectasis at T5. Incidence of atelectasis at T5. Secondary outcomes PEEP values after lateral positioning and after CPB weaning.LUS at T5.P: F ratio: Perform arterial blood gas analysis at T0, T2, T4, T6, and T7, calculate the P: F ratio, and determine the incidence of post-CPB hypoxemia.Respiratory mechanics: Ppeak, Cdyn, and driving pressure at T0, T1, T2, T3, T4, and T5; PEEP values determined by two intraoperative titrations.Postoperative follow-up indicators: Incidence of PPCs within 3 days, extubation time, ICU length of stay, Hospital length of stay.Hemodynamic indicators: HR and MAP at T0, T1, T2, T3, T4, and T5. PEEP values after lateral positioning and after CPB weaning. PEEP values after lateral positioning and after CPB weaning. LUS at T5. LUS at T5. P: F ratio: Perform arterial blood gas analysis at T0, T2, T4, T6, and T7, calculate the P: F ratio, and determine the incidence of post-CPB hypoxemia. P: F ratio: Perform arterial blood gas analysis at T0, T2, T4, T6, and T7, calculate the P: F ratio, and determine the incidence of post-CPB hypoxemia. Respiratory mechanics: Ppeak, Cdyn, and driving pressure at T0, T1, T2, T3, T4, and T5; PEEP values determined by two intraoperative titrations. Respiratory mechanics: Ppeak, Cdyn, and driving pressure at T0, T1, T2, T3, T4, and T5; PEEP values determined by two intraoperative titrations. Postoperative follow-up indicators: Incidence of PPCs within 3 days, extubation time, ICU length of stay, Hospital length of stay. Postoperative follow-up indicators: Incidence of PPCs within 3 days, extubation time, ICU length of stay, Hospital length of stay. Hemodynamic indicators: HR and MAP at T0, T1, T2, T3, T4, and T5. Hemodynamic indicators: HR and MAP at T0, T1, T2, T3, T4, and T5. Sample size calculation Given the limited literature on pediatric cardiac surgery with contralateral thoracotomy, we conducted a pilot study involving 15 cases per group (30 cases total) to estimate the required sample size. We defined atelectasis as a LUS ≥ 2. The pilot study showed prior to leaving the operating room (T5), the incidence of atelectasis in Group C was 73.33% (11/15) compared to 40.00% (6/15) in Group D, representing an absolute difference of 33.33%. Using PASS.15 software for calculations, with α = 0.05 and power = 80%, and accounting for a 10% dropout rate, a total of 40 patients per group were required for the study. Randomization and blinding Patients were randomly assigned in a 1:1 ratio to either the fixed PEEP group (Group C) or the driving pressure-guided individualized PEEP group (Group D) using the envelope method based on a computer-generated randomization list. After allocation, group assignments and intervention details were placed in identical opaque envelopes and numbered to ensure concealment. On the day of surgery, a researcher opened the envelope and administered the intervention according to the protocol inside, while recording data. Throughout the study period, the ultrasound assessor was positioned with the ventilator settings interface out of their direct line of sight and occluded to maintain blinding. Blinding was also applied to patients’ families, postoperative follow-up personnel, and data analysts. Statistical analysis Statistical analysis was performed using SPSS 27.0 software. Normally distributed quantitative data were expressed as mean ± standard deviation (mean ± SD), with intergroup comparisons analyzed by independent samples t-test amd intragroup comparisons analyzed by repeated measures ANOVA. Both intergroup and intragroup multiple comparisons were corrected using the Bonferroni method. Non-normally distributed quantitative data were expressed as median (M) and interquartile range (IQR), with intergroup comparisons analyzed using the Mann-Whitney U test. Count data were expressed as counts and percentages, and intergroup comparisons were performed using the chi-square test or Fisher’s exact test. P value < 0.05 was considered statistically significant. Results Participant characteristics This study initially enrolled 85 children, with 5 cases excluded. Ultimately, 80 participants were included, with 40 cases in each group (Fig. 2). There were no statistically significant differences between the two groups in terms of gender, age, BMI, ASA classification, left ventricular ejection fraction (LVEF), anesthesia duration, lateral decubitus time, CPB time, aortic cross-clamp time, or surgical type (P > 0.05) (Table 1). Fig. 2Enrollment flowchart Enrollment flowchart Table 1Comparison of general characteristics between two groups of patientsGroup CGroup DP-valueMale/Female (n)19/2120/200.823Age (months)27.05 ± 16.0927.08 ± 15.350.994BMI (kg/m2)14.62 ± 1.7414.17 ± 1.770.259ASA (Ⅱ/Ⅲ)2/383/370.531LVEF (%)64.22 ± 2.0664.93 ± 1.930.120Duration of anesthesia (min)189.53 ± 34.08193.15 ± 34.390.638Duration of lateral decubitus position (min)160.75 ± 29.22166.50 ± 35.740.433Duration of CPB (min)61.78 ± 20.1364.60 ± 22.550.556Duration of aortic cross-clamp (min)40.63 ± 17.2644.98 ± 18.020.274Tidal volume (ml/kg)7.02 ± 0.227.00 ± 0.280.676Surgical type (A/B/C/D)17/21/1/112/25/2/10.748A: Atrial septal defect repair; B: Ventricular septal defect repair with or without patent foramen ovale closure; C: Combined ventricular and atrial septal defect repair; D: Atrial septal defect repair with partial correction of pulmonary vein anomaliesP value < 0.05 was considered statistically significant Table 1Comparison of general characteristics between two groups of patientsGroup CGroup DP-valueMale/Female (n)19/2120/200.823Age (months)27.05 ± 16.0927.08 ± 15.350.994BMI (kg/m2)14.62 ± 1.7414.17 ± 1.770.259ASA (Ⅱ/Ⅲ)2/383/370.531LVEF (%)64.22 ± 2.0664.93 ± 1.930.120Duration of anesthesia (min)189.53 ± 34.08193.15 ± 34.390.638Duration of lateral decubitus position (min)160.75 ± 29.22166.50 ± 35.740.433Duration of CPB (min)61.78 ± 20.1364.60 ± 22.550.556Duration of aortic cross-clamp (min)40.63 ± 17.2644.98 ± 18.020.274Tidal volume (ml/kg)7.02 ± 0.227.00 ± 0.280.676Surgical type (A/B/C/D)17/21/1/112/25/2/10.748A: Atrial septal defect repair; B: Ventricular septal defect repair with or without patent foramen ovale closure; C: Combined ventricular and atrial septal defect repair; D: Atrial septal defect repair with partial correction of pulmonary vein anomaliesP value < 0.05 was considered statistically significant Comparison of general characteristics between two groups of patients A: Atrial septal defect repair; B: Ventricular septal defect repair with or without patent foramen ovale closure; C: Combined ventricular and atrial septal defect repair; D: Atrial septal defect repair with partial correction of pulmonary vein anomalies P value < 0.05 was considered statistically significant Primary outcome Incidence of atelectasis: The incidence of atelectasis at T0 (after induction) was 72.5% and 75% in the two groups, respectively, with no statistically significant difference (P > 0.05). The incidence of atelectasis in group D was significantly lower than that in group C at T5 (before leaving the operating room) (P < 0.05) (Fig. 3a). Incidence of atelectasis: The incidence of atelectasis at T0 (after induction) was 72.5% and 75% in the two groups, respectively, with no statistically significant difference (P > 0.05). The incidence of atelectasis in group D was significantly lower than that in group C at T5 (before leaving the operating room) (P < 0.05) (Fig. 3a). Incidence of atelectasis: The incidence of atelectasis at T0 (after induction) was 72.5% and 75% in the two groups, respectively, with no statistically significant difference (P > 0.05). The incidence of atelectasis in group D was significantly lower than that in group C at T5 (before leaving the operating room) (P < 0.05) (Fig. 3a). Fig. 3Primary outcome between the two groups Primary outcome between the two groups Secondary outcomes PEEP values: Following the lateral position and CPB, the median individualized PEEP values obtained through titration in Group D were 6 cmH2O and 7 cmH2O, respectively, which were higher than the 5 cmH2O setting in Group C (P < 0.001) (Fig. 4a).Lung ultrasound scores: There was no statistically significant difference in LUS between Group C (9.43 ± 2.67) and Group D (9.70 ± 2.61) at T0 (P > 0.05). At T5, the LUS in Group D (6.58 ± 2.37) was significantly lower than that in Group C (7.68 ± 2.21) (P < 0.05) (Fig. 4b).P: F ratio: Compared with Group C, there was no significant statistical difference in P: F ratio between groups at T0 and T7 (P > 0.05), while group D showed significantly higher P: F ratio than group C during T2 to T6 (P < 0.05). Compared with T0, Group C showed significant reductions at T2 to T6, while Group D showed a significant decrease only at T2 and increased at T7 (P < 0.05). The incidence of hypoxemia after CPB was significantly reduced in Group D (Fig. 4c; Table 2).Respiratory mechanics: From T0 to T5, the difference in Ppeak between the two groups was not statistically significant (P > 0.05). Compared with T0, Ppeak increased in both groups from T1 to T5 (P < 0.001). Compared to T0, Group C exhibited increased driving pressure from T1 to T3 and decreased Cdyn from T1 to T4, while Group D showed increased driving pressure and decreased Cdyn from T1 to T3. However, at T5, Group D demonstrated decreased driving pressure and increased Cdyn. Compared to Group C, Group D exhibited significantly reduced driving pressure from T2 to T5 and demonstrated markedly superior Cdyn at T2, T4, and T5 (P < 0.05) (Fig. 4d, e,f). PEEP values: Following the lateral position and CPB, the median individualized PEEP values obtained through titration in Group D were 6 cmH2O and 7 cmH2O, respectively, which were higher than the 5 cmH2O setting in Group C (P < 0.001) (Fig. 4a). PEEP values: Following the lateral position and CPB, the median individualized PEEP values obtained through titration in Group D were 6 cmH2O and 7 cmH2O, respectively, which were higher than the 5 cmH2O setting in Group C (P < 0.001) (Fig. 4a). Lung ultrasound scores: There was no statistically significant difference in LUS between Group C (9.43 ± 2.67) and Group D (9.70 ± 2.61) at T0 (P > 0.05). At T5, the LUS in Group D (6.58 ± 2.37) was significantly lower than that in Group C (7.68 ± 2.21) (P < 0.05) (Fig. 4b). Lung ultrasound scores: There was no statistically significant difference in LUS between Group C (9.43 ± 2.67) and Group D (9.70 ± 2.61) at T0 (P > 0.05). At T5, the LUS in Group D (6.58 ± 2.37) was significantly lower than that in Group C (7.68 ± 2.21) (P < 0.05) (Fig. 4b). P: F ratio: Compared with Group C, there was no significant statistical difference in P: F ratio between groups at T0 and T7 (P > 0.05), while group D showed significantly higher P: F ratio than group C during T2 to T6 (P < 0.05). Compared with T0, Group C showed significant reductions at T2 to T6, while Group D showed a significant decrease only at T2 and increased at T7 (P < 0.05). The incidence of hypoxemia after CPB was significantly reduced in Group D (Fig. 4c; Table 2). P: F ratio: Compared with Group C, there was no significant statistical difference in P: F ratio between groups at T0 and T7 (P > 0.05), while group D showed significantly higher P: F ratio than group C during T2 to T6 (P < 0.05). Compared with T0, Group C showed significant reductions at T2 to T6, while Group D showed a significant decrease only at T2 and increased at T7 (P < 0.05). The incidence of hypoxemia after CPB was significantly reduced in Group D (Fig. 4c; Table 2). Respiratory mechanics: From T0 to T5, the difference in Ppeak between the two groups was not statistically significant (P > 0.05). Compared with T0, Ppeak increased in both groups from T1 to T5 (P < 0.001). Compared to T0, Group C exhibited increased driving pressure from T1 to T3 and decreased Cdyn from T1 to T4, while Group D showed increased driving pressure and decreased Cdyn from T1 to T3. However, at T5, Group D demonstrated decreased driving pressure and increased Cdyn. Compared to Group C, Group D exhibited significantly reduced driving pressure from T2 to T5 and demonstrated markedly superior Cdyn at T2, T4, and T5 (P < 0.05) (Fig. 4d, e,f). Respiratory mechanics: From T0 to T5, the difference in Ppeak between the two groups was not statistically significant (P > 0.05). Compared with T0, Ppeak increased in both groups from T1 to T5 (P < 0.001). Compared to T0, Group C exhibited increased driving pressure from T1 to T3 and decreased Cdyn from T1 to T4, while Group D showed increased driving pressure and decreased Cdyn from T1 to T3. However, at T5, Group D demonstrated decreased driving pressure and increased Cdyn. Compared to Group C, Group D exhibited significantly reduced driving pressure from T2 to T5 and demonstrated markedly superior Cdyn at T2, T4, and T5 (P < 0.05) (Fig. 4d, e,f). Outcomes measures between the two groups. Compared with group C, *P<0.05, **P<0.01, ***P<0.001; compared with T0, #P<0.05, ##P<0.01, ###P<0.001. The P values have been corrected using the Bonferroni method. P value<0.05 was considered statistically significant Outcomes measures between the two groups. Compared with group C, *P<0.05, **P<0.01, ***P<0.001; compared with T0, #P<0.05, ##P<0.01, ###P<0.001. The P values have been corrected using the Bonferroni method. P value<0.05 was considered statistically significant. a Titration results at two time points in group C and group D; b LUS scores at T₀ and T₅ in group C and group D; c P/F ratio over different time points; d Ppeak over different time points; e Driving pressure over different time points; f Cdyn over different time points (5)Postoperative follow-up indicators: The overall incidence of PPCs within 3 days was 52.5% in Group C and 47.5% in Group D, with no statistically significant difference between groups (P > 0.05). The time to extubation was significantly shorter in Group D than in Group C (P < 0.05), while there was no statistically significant difference in ICU length of stay or hospital stay between groups (P > 0.05) (Table 2). (5)Postoperative follow-up indicators: The overall incidence of PPCs within 3 days was 52.5% in Group C and 47.5% in Group D, with no statistically significant difference between groups (P > 0.05). The time to extubation was significantly shorter in Group D than in Group C (P < 0.05), while there was no statistically significant difference in ICU length of stay or hospital stay between groups (P > 0.05) (Table 2). Postoperative follow-up indicators: The overall incidence of PPCs within 3 days was 52.5% in Group C and 47.5% in Group D, with no statistically significant difference between groups (P > 0.05). The time to extubation was significantly shorter in Group D than in Group C (P < 0.05), while there was no statistically significant difference in ICU length of stay or hospital stay between groups (P > 0.05) (Table 2). Table 2Incidence of PPCs and prognosis duration in childrenGroup CGroup DP-valuePost-CPB Hypoxemia19 (47.5%)9 (22.5%)0.003PPCs21 (52.5%)19 (47.5%)0.655Pulmonary Infection19 (47.5%)16 (40.0%)0.499Pneumothorax/Pneumopericardium6 (15.0%)4 (10.0%)0.735Atelectasis5 (12.5%)4 (10.0%)1.000Pleural Effusion3 (7.50%)2 (4.00%)1.000Extubation time (h)3.25 ± 0.812.84 ± 0.830.027ICU stay (h)42.39 ± 9.7241.12 ± 6.990.506Hospital stay (d)8.80 ± 1.778.82 ± 1.550.947Data are presented as mean (SD) or n (%). P value < 0.05 was considered statistically significant Table 2Incidence of PPCs and prognosis duration in childrenGroup CGroup DP-valuePost-CPB Hypoxemia19 (47.5%)9 (22.5%)0.003PPCs21 (52.5%)19 (47.5%)0.655Pulmonary Infection19 (47.5%)16 (40.0%)0.499Pneumothorax/Pneumopericardium6 (15.0%)4 (10.0%)0.735Atelectasis5 (12.5%)4 (10.0%)1.000Pleural Effusion3 (7.50%)2 (4.00%)1.000Extubation time (h)3.25 ± 0.812.84 ± 0.830.027ICU stay (h)42.39 ± 9.7241.12 ± 6.990.506Hospital stay (d)8.80 ± 1.778.82 ± 1.550.947Data are presented as mean (SD) or n (%). P value < 0.05 was considered statistically significant Incidence of PPCs and prognosis duration in children Data are presented as mean (SD) or n (%). P value < 0.05 was considered statistically significant (6)Hemodynamic indicators: No statistically significant differences in MAP or HR were observed between the two groups from T0 to T5. Compared with T0, MAP significantly increased in both groups from T2 to T5 (P < 0.05), while HR significantly increased at T4 and T5 (P < 0.001) (Table 3). (6)Hemodynamic indicators: No statistically significant differences in MAP or HR were observed between the two groups from T0 to T5. Compared with T0, MAP significantly increased in both groups from T2 to T5 (P < 0.05), while HR significantly increased at T4 and T5 (P < 0.001) (Table 3). Hemodynamic indicators: No statistically significant differences in MAP or HR were observed between the two groups from T0 to T5. Compared with T0, MAP significantly increased in both groups from T2 to T5 (P < 0.05), while HR significantly increased at T4 and T5 (P < 0.001) (Table 3). Table 3Hemodynamic parameters between the two groupsIndicatorGroup CGroup DP-valueHR(times/min) T0109.57 ± 11.71109.15 ± 9.460.859 T1107.55 ± 13.37107.25 ± 7.940.903 T2127.65 ± 13.95128.65 ± 11.940.731 T3136.58 ± 12.92137.80 ± 11.070.650 T4126.23 ± 12.84125.97 ± 10.210.923 T5124.65 ± 12.45124.33 ± 10.410.900MAP(mmHg) T059.77 ± 5.0059.22 ± 5.140.629 T159.57 ± 5.8959.20 ± 5.980.783 T258.61 ± 6.7658.68 ± 6.830.961 T357.73 ± 7.6657.98 ± 8.490.887 T463.18 ± 6.4363.28 ± 6.540.945 T562.43 ± 6.0462.60 ± 7.730.911Data are presented as mean (SD). P value < 0.05 was considered statistically significant Table 3Hemodynamic parameters between the two groupsIndicatorGroup CGroup DP-valueHR(times/min) T0109.57 ± 11.71109.15 ± 9.460.859 T1107.55 ± 13.37107.25 ± 7.940.903 T2127.65 ± 13.95128.65 ± 11.940.731 T3136.58 ± 12.92137.80 ± 11.070.650 T4126.23 ± 12.84125.97 ± 10.210.923 T5124.65 ± 12.45124.33 ± 10.410.900MAP(mmHg) T059.77 ± 5.0059.22 ± 5.140.629 T159.57 ± 5.8959.20 ± 5.980.783 T258.61 ± 6.7658.68 ± 6.830.961 T357.73 ± 7.6657.98 ± 8.490.887 T463.18 ± 6.4363.28 ± 6.540.945 T562.43 ± 6.0462.60 ± 7.730.911Data are presented as mean (SD). P value < 0.05 was considered statistically significant Hemodynamic parameters between the two groups Data are presented as mean (SD). P value < 0.05 was considered statistically significant Discussion This study demonstrates that in pediatric cardiac surgery with lateral thoracotomy assisted by CPB, compared with fixed PEEP, a driving pressure-guided individualized PEEP strategy reduced the incidence of immediate postoperative atelectasis but did not decrease the overall incidence of pulmonary complications within three days after surgery. In addition, compared with Group C, Group D had higher oxygenation indices during and early after surgery, lower lung ultrasound scores at the end of surgery, and improved driving pressure and dynamic lung compliance. Research by Amato et al. [17] suggested that driving pressure is the key mediator for the benefits of various interventions, with the advantages of VT and PEEP only being beneficial when associated with a reduction in driving pressure. A meta-analysis incorporating 17 randomized controlled studies indicated that driving pressure is the significant mediator influencing the development of pulmonary complications in protective ventilation strategies [10]. This study therefore titrated PEEP based on driving pressure as the target parameter to identify the optimal PEEP level. Driving pressure-guided individualized PEEP titration is typically performed using either an incremental or decremental method [18]. Considering that excessively high PEEP may lead to intraoperative circulatory depression and pulmonary hyperinflation, and that a single titration is often insufficient to adapt to the dynamic physiological changes throughout the surgical procedure [19, 20], this study adopted an incremental approach. Titrations were conducted at two key stages of the surgery, with an upper PEEP limit set at 10 cmH₂O. Previous studies have indicated that at least 5 cmH₂O of PEEP is necessary during mechanical ventilation to confer lung protective effects [21]. Therefore, Group C in this study received a fixed PEEP of 5 cmH₂O. During the two titration procedures in this study, the median optimal PEEP measured for Group D was 6 cmH₂O and 7 cmH₂O, respectively, both higher than that for Group C. This difference may be attributed to physiological characteristics of pediatric patients, such as weaker structural support of the chest wall, an underdeveloped respiratory system, low functional residual capacity, and susceptibility to post-cardiopulmonary bypass inflammatory responses, making them more prone to postoperative hypoxemia, atelectasis, and pneumothorax after general anesthesia [6, 22]. This also suggests that the actual PEEP level required by this population at different surgical stages is higher than conventional settings [23, 24]. The implementation of this strategy during surgery did not cause significant hemodynamic fluctuations. Given the high sensitivity and specificity of ultrasound in assessing perioperative atelectasis in pediatric patients [25], this study used lung ultrasound and performed postoperative lung ultrasound scoring according to the Monastesse protocol [14] to calculate the incidence of immediate postoperative atelectasis. At the end of surgery, the incidence of atelectasis was 65% in Group C and 42.5% in Group D, and the LUS score in Group D (6.58 ± 2.37) was significantly lower than that in Group C (7.68 ± 2.21). Similar benefits were observed in another study investigating the application of driving pressure-guided dynamic PEEP titration in children at intermediate-to-high risk of developing PPCs undergoing laparoscopic surgery [26]. In addition, the management strategy used in this study during cardiopulmonary bypass, which involved complete interruption of mechanical ventilation with the ventilator circuit remaining connected to the patient, may have exacerbated the development of atelectasis, particularly in the dependent lung regions. Whether the application of continuous positive airway pressure during cardiopulmonary bypass could further enhance the protective effect of driving pressure-guided PEEP remains to be investigated in future studies. It is worth emphasizing that the atelectasis assessed by LUS immediately after surgery was mainly due to compressive or resorptive mechanisms, whereas the atelectasis defined within the PPCs referred to radiological evidence of atelectasis detected within three days after surgery. The difference in incidence between the two is related to their distinct pathological mechanisms, observation time windows, and clinical implications. Therefore, the improvement in LUS observed in this study did not directly translate into a reduction in PPCs within three days after surgery. Currently, the minimally invasive lateral thoracotomy approach is widely used due to its advantages such as excellent surgical field exposure, favorable cosmetic outcomes, and rapid postoperative recovery [2, 27]. However, the retraction of the right lung inherent to this approach, combined with factors such as cardiopulmonary bypass, localized hypothermia, lateral positioning, and mechanical ventilation, can all lead to impaired gas exchange. Children undergoing cardiac surgery with CPB often develop severe respiratory dysfunction due to systemic inflammation and ischemia-reperfusion injury induced by CPB, and prolonged mechanical ventilation [28, 29]. The P: F ratio is a key parameter for assessing pulmonary gas exchange function and is generally inversely correlated with the degree of lung injury. CPB as a major physiological insult during cardiac surgery can reduce pulmonary surfactant and significantly increase the risk of postoperative hypoxemia. Studies have shown that a driving pressure-guided individualized PEEP strategy can improve intraoperative oxygenation without exacerbating the inflammatory response [30, 31]. Additionally, higher driving pressure is independently associated with prolonged mechanical ventilation in children [6]. Similar benefits were observed in this study: although the P: F ratio decreased at T2 compared to T0 in both groups, the group D exhibited significantly higher P: F ratio at T2, T4, and T6 than the group C. Furthermore, the incidence of post-bypass hypoxemia in this group was only 22.5%, significantly lower than the 47.5% in the control group. Although this study observed a shorter extubation time in Group D compared with Group C, this finding should be interpreted with caution and considered exploratory, because postoperative ICU management may have been influenced by bias related to P: F ratio assessment and ventilator management. Moreover, respiratory management during the ICU phase was not standardized in this study, so extubation time may be subject to confounding from multiple factors. Driving pressure, Ppeak, and Cdyn can dynamically reflect intraoperative respiratory function changes, illustrating the overall impact of lung tissue elasticity, airway resistance, and intrathoracic pressure on pediatric respiratory physiology. Compared to the group C, the group D exhibited significantly lower driving pressure at time points T2 to T5, along with significantly higher lung dynamic compliance at T2, T4, and T5. By titrating the optimal PEEP targeting the minimum driving pressure, this strategy maintained lower driving pressure and better lung compliance without significantly increasing Ppeak. Theoretically, this approach better aligns with the physiological heterogeneity of pediatric patients, contributing to maintaining alveolar stability, increasing end-expiratory lung volume, reducing airway closure and alveolar collapse, thereby optimizing intraoperative respiratory mechanics [32]. Although this study observed that the application of driving pressure-guided individualized PEEP reduced the incidence of immediate postoperative atelectasis and improved oxygenation indices during and early after surgery, the follow-up for the incidence of PPCs within three days after surgery showed rates of 52.5% in Group C and 47.5% in Group D, with no significant difference between the two groups. Previous studies have found that although a driving pressure-guided individualized PEEP ventilation strategy can effectively reduce the incidence of PPCs in non-cardiac and non-thoracic surgeries [12, 13], in cardiac surgery, this strategy often fails to significantly reduce the overall incidence of PPCs or may only lower the risk of early postoperative PPCs [33, 34]. To analyze the possible reasons: First, although children undergoing elective cardiac surgery are themselves at intermediate-to-high risk for PPCs, it is important to recognize that they differ from children with established acute respiratory distress syndrome (ARDS). ARDS is typically characterized by infection-driven diffuse alveolar damage, accompanied by surfactant dysfunction and persistent inflammation [35]. In contrast, the lung injury in our study population primarily results from CPB-induced systemic inflammation, intraoperative lung retraction, and ischemia-reperfusion injury. Therefore, the potential benefit from this strategy in these children may be limited. Second, we actively followed up on the composite outcome of PPCs within only three days postoperatively, which is a relatively short follow-up period capturing only very early PPC events. Moreover, the composite PPC endpoint was broadly defined, and some complications might have been transient and self-limiting. In particular, for the diagnosis of respiratory tract infection, some cases may have represented only early, mild infections that do not reflect clinically significant severe infection. Although this broad definition increases sensitivity, it may make it difficult to detect differences between groups. Third, postoperative respiratory management made by ICU physicians based on clinical experience may have attenuated the benefits of the intraoperative intervention. The improvement in lung aeration and oxygenation achieved intraoperatively, if not maintained by a sustained lung-protective ventilation strategy after transfer to the ICU, may be rapidly diluted or reversed. This phenomenon is a well-recognized challenge in the field of perioperative lung protection. Our results are consistent with the findings of some similar studies. For example, a study involving 216 children undergoing pulmonary lobectomy also indicated that compared to a fixed PEEP group (5 cmH₂O), the driving pressure-guided lung-protective ventilation strategy, while improving intraoperative lung compliance and oxygenation, did not show a statistically significant difference in PPCs incidence between the groups [36]. The physiological benefits of this strategy may not fully translate into a significant reduction in PPC incidence. This further prompts us to consider and explore more systematic and sustained lung-protective ventilation strategies for patients undergoing cardiac surgery, including continuous lung-protective ventilation support during cardiopulmonary bypass and the emergence period, as well as sequential oxygen therapy after extubation, in order to ultimately reduce the overall incidence of PPCs and improve long-term patient outcomes. Although the individualized PEEP strategy did not significantly reduce the overall incidence of PPCs within three days after surgery, its improvement in the rate of immediate postoperative atelectasis holds certain clinical significance. Second, the improvement in oxygenation and respiratory mechanics suggests that this strategy may better align with the heterogeneous physiological characteristics of pediatric patients, facilitating earlier recovery of respiratory function. Moreover, the implementation of driving pressure-guided incremental PEEP titration with an upper limit of 10 cmH2O during surgery did not cause significant hemodynamic fluctuations, indicating that it can be safely performed in pediatric cardiac surgery. Furthermore, we believe that even without a significant reduction in the overall incidence of PPCs, the implementation of this strategy still represents a valuable exploratory attempt in cardiac surgery for children at intermediate-to-high risk, and also suggests that future research directions should extend to the entire perioperative period. This study has several limitations. First, it is a single-center study with a small sample size, and larger-scale research is needed in the future for further validation. Second, this study included children aged 6 months to 5 years undergoing minimally invasive cardiac surgery for simple congenital heart disease in the lateral decubitus position. Future studies should consider age stratification at randomization. Furthermore, the applicability and benefits of this strategy in patients with complex congenital heart disease and those in other age groups remain to be further validated. Third, stabilizing each PEEP level for only 10 respiratory cycles in this study may not have been sufficient to achieve complete equilibration of lung mechanics. Although this short-duration titration protocol is commonly used in perioperative studies, it may have slightly affected the precision of identifying the optimal PEEP. Finally, the current conclusions are primarily based on clinical observations and physical signs. Subsequent studies could incorporate more precise imaging techniques and simultaneously monitor changes in lung injury-related biomarkers and inflammatory cytokine levels, thereby revealing the physiological and clinical effects of this ventilation strategy in a multi-dimensional and in-depth manner. Conclusions The driving pressure-guided individualized PEEP can be safely applied to pediatric patients with congenital heart disease undergoing right lateral thoracotomy cardiac surgery with cardiopulmonary bypass. Compared to conventional protective ventilation, this strategy effectively reduces the incidence of postoperative atelectasis and lowers lung ultrasound scores, improves intraoperative oxygenation and respiratory mechanics parameters, and decreases the occurrence of post-cardiopulmonary bypass hypoxemia. However, no significant improvement was observed in the overall incidence of PPCs.