Effect of PEEP titration on the incidence of elevated driving pressure in laparoscopic bariatric surgery: a randomized superiority trial Authors: Saullo Queiroz Silveira, Sérgio M. Pereira, Leopoldo Muniz da Silva, Rafael Souza Fava Nersessian, Helidea de Oliveira Lima, Desiree Stieven Machado, João Pedro Borghi Moreira, Fernando Nardy Bellicieri, Glenio B. Mizubuti, Joaquim Edson Vieira Published: 2026 ABSTRACT BackgroundSuboptimal positive end-expiratory pressure (PEEP) settings may lead to cyclic lung overstress, especially in obese patients undergoing general anesthesia. We hypothesized that recruitment maneuver (RM) followed by driving pressure–guided PEEP titration – before and during pneumoperitoneum – would reduce the incidence of elevated driving pressure (> 15 cmH₂O) during pneumoperitoneum.MethodsIn this single-center, randomized, superiority trial, obese patients undergoing laparoscopic bariatric surgery were allocated into three groups: (1) recruitment maneuver followed by driving pressure-guided PEEP titration (RMTP); (2) recruitment maneuver followed by fixed PEEP (6 cmH₂O) (RMFP); or (3) no recruitment maneuver with fixed PEEP (6 cmH₂O) (NRMFP). RM was performed with incremental PEEP (6→10→15→20 cmH₂O) and inspiratory pressure (26→30→35→40 cmH₂O), whereas PEEP titration was conducted in a decremental fashion. The primary outcome was the incidence of driving pressure > 15 cmH₂O.ResultsEighty-five patients were screened and 78 randomized. No significant between-group differences in the incidence of driving pressure > 15 cmH₂O were observed before pneumoperitoneum or after pneumoperitoneum reversal. During pneumoperitoneum before RM, driving pressure > 15 cmH₂O occurred less frequently with RMTP than with RMFP and NRMFP (7.7% vs. 46.2% and 46.2%; RR 6.00, 95% CI 1.49–24.20; P = 0.01). After RM, the incidence remained lower with RMTP (3.8%) compared with RMFP (30.8%; RR 8.00, 95% CI 1.08–59.50; P = 0.04) and NRMFP (46.2%; RR 12.00, 95% CI 1.68–85.73; P = 0.01). Secondary outcomes, including postoperative pulmonary complications and hemodynamic variables, were similar between groups.ConclusionsRecruitment maneuver combined with driving pressure-guided PEEP titration improved intraoperative respiratory mechanics and reduced the incidence of elevated driving pressure during pneumoperitoneum, although no differences in postoperative clinical outcomes were observed.Trial registrationBrazilian Registry of Clinical Trials, RBR-5tbdpy5. Registered on January 29, 2024.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12871-026-04026-0. Background General anesthesia and neuromuscular blockade promote a diaphragmatic cranial shift resulting in lung heterogeneities and atelectasis [1] which may be further exacerbated by conditions that lead to increased pleural pressures (e.g., obesity) [2, 3]. The most common strategy to prevent and revert intraoperative atelectasis is to combine recruitment maneuver (RM) with positive end-expiratory pressure (PEEP) [4]. Yet, inappropriate PEEP settings following RM can increase the risk of ventilator-induced lung injury (VILI), as high PEEP levels can cause lung overdistension, while low PEEP levels can lead to lung collapse. Therefore, PEEP individualization is clinically relevant [4]. Although there is no gold-standard method to individualize PEEP, recent experimental data suggests that driving pressure-guided PEEP may be used to optimize pulmonary biomechanics [5]. Driving pressure, calculated as plateau pressure minus PEEP in patients not initiating inspiratory efforts, is a key surrogate for assessing the risk of VILI [6]. Elevated driving pressures indicate cyclic lung overstress, generally caused by suboptimal ventilatory settings, and have been associated with increased mortality in patients with acute respiratory distress syndrome [7], and higher incidence of postoperative pulmonary complications (PPC) [8]. Notably, driving pressure-guided PEEP titration is simple and does not require further expertise or costly equipment compared to other methods (e.g., esophageal catheter-guided transpulmonary pressure, lung ultrasound, electrical impedance tomography) often purported to promote optimal lung aeration and/or protective lung ventilation. We designed a three-arm randomized study to evaluate the effects of recruitment maneuver and driving pressure–guided PEEP titration on intraoperative respiratory mechanics in obese patients undergoing laparoscopic bariatric surgery. We hypothesized that individualized PEEP titration guided by driving pressure would result in lower driving pressures during pneumoperitoneum. The primary outcome was the incidence of driving pressure > 15 cmH₂O across prespecified intraoperative timepoints, with particular focus during pneumoperitoneum. Secondary outcomes included postoperative oxygen requirements, incidence of postoperative pulmonary complications, and intraoperative hemodynamic variables. Methods Study design This randomized, three-arm, single-center clinical trial was conducted between March and June 2024. The study was designed as a superiority trial to evaluate intraoperative physiological differences in driving pressure between ventilatory strategies. This investigation was approved by the Hospital São Luiz & Rede D´or and Affiliated Teaching Hospitals Research Ethics Committee (protocol 5.640.555, CAAE 63239722.4.0000.0087) and registered in the Brazilian Registry of Clinical Trials (RBR-5tbdpy5) on January 29, 2024. Written informed consent was obtained from all participants as appropriate. This study adhered to CONSORT guidelines. Inclusion and exclusion criteria Patients 18–65 years-old, classified as American Society of Anesthesiologists’ Physical Status (ASA-PS) 2–3, with body mass index (BMI) ≥ 30 Kg/m2, undergoing elective laparoscopic bariatric surgery were included. Patients with BMI ≥ 50 Kg/m2 (‘super-obese’), congestive heart failure, chronic kidney failure, history of alcohol/drug abuse, allergy to study components, and/or undergoing degastrectomy or combined surgeries were excluded. Randomization and blinding Patients were screened during the pre-anaesthetic evaluation. Before randomization, baseline demographic, clinical, and anthropometric data were recorded. Eligible subjects meeting inclusion criteria were randomized in a 1:1:1 ratio to one of three groups: (1) recruitment maneuver with titrated PEEP (RMTP); (2) recruitment maneuver with fixed PEEP of 6 cmH₂O (RMFP); or (3) no recruitment maneuver with fixed PEEP of 6 cmH₂O (NRMFP). Randomization was performed by an independent statistician using a computer-generated permuted-block algorithm with variable block sizes of three and six (randomly allocated across 16 blocks). The allocation sequence was sealed in 78 sequentially numbered opaque envelopes, which were opened only after participant enrollment. This procedure yielded an exact distribution of 26 participants per group. Patients and outcome assessors were blinded to group allocation. Blinding of the anesthesiologist responsible for implementing the ventilatory strategy was not feasible given the procedural nature of the intervention. Anesthesia protocol Intraoperative monitoring included pulse oximetry (SpO2), electrocardiogram, non-invasive blood pressure (NIBP), nasopharyngeal temperature, end-tidal carbon dioxide (ETCO2), Bispectral Index (BIS), and train-of-four (TOF). Our institutional protocol for bariatric surgery has been previously described [9]. Predicted body weight (PBW) [50 + 0.91×(height − 152.4 cm) for males; 45.5 + 0.91×(height − 152.4 cm) for females] was used to base medication dosages, except for methadone and sugammadex (based on actual body weight) [10]. Following pre-hydration with intravenous crystalloid, anaesthesia was induced with propofol, remifentanil, rocuronium, methadone, magnesium sulphate, dexmedetomidine, dexamethasone, and parecoxib. Patients underwent orotracheal intubation and were positioned in a 30-degree head-up positioning before establishment of pneumoperitoneum (15–17 mmHg). Anaesthesia was maintained with sevoflurane targeting BIS values between 40 and 60. Remifentanil and rocuronium infusions were titrated according to haemodynamic parameters and neuromuscular monitoring, respectively. Intraoperative fluid therapy and temperature management were standardised, and metaraminol was administered when mean arterial pressure was < 65 mmHg. At the end of surgery, antiemetic prophylaxis and metamizole were administered, and neuromuscular blockade was reversed with sugammadex. In the post-anaesthesia care unit (PACU), patients received supplemental oxygen if/as needed to keep SpO2 > 90%. Postoperative analgesia and antiemetic therapy were administered according to institutional protocols. All patients were encouraged to ambulate on postoperative day 1 and underwent daily respiratory physiotherapy. Study protocol Lung-protective ventilation strategy was initiated following tracheal intubation under volume-controlled ventilation (VCV) using the Fabius Premium (Dräger Medical, Germany) anesthesia machine with FiO2 = 0.5, tidal volume (VT) = 7 mL/kg of PBW, PEEP = 6 cmH2O, respiratory rate (RR) = 15 breaths/minute, inspiratory pause = 20%, and inspiratory-to-expiratory ratio (I: E) = 1:2. RR was adjusted to 10 or 20 breaths/minute if ETCO2 < 30 or > 45 mmHg, respectively. If, in 10 min, these initial changes were insufficient to normalize ETCO2, further RR adjustments (to 8 or 25 breaths/minute, respectively) were implemented. For data collection, RR was maintained at 15 breaths/minute and was readjusted, based on previous ETCO2 levels, 5 min later at the studied timepoints (Suppl. Figure 1): (a) before pneumoperitoneum − 5 min after tracheal intubation and after implementation of the assigned ventilatory strategy, before pneumoperitoneum insufflation (b) during pneumoperitoneum before recruitment maneuver − 5 min after pneumoperitoneum insufflation and head-up positioning, before recruitment maneuver; (c) during pneumoperitoneum after recruitment maneuver − 5 min after recruitment maneuver following pneumoperitoneum insufflation and head-up positioning; (d) after pneumoperitoneum before recruitment maneuver − 5 min after pneumoperitoneum deflation in the horizontal supine position, before recruitment maneuver; and (e) after pneumoperitoneum after recruitment maneuver − 5 min after recruitment maneuver following pneumoperitoneum deflation. Patients randomized to the NRMFP group were not recruited at any timepoint; therefore, respiratory variables were repeatedly recorded during and after pneumoperitoneum. Similarly, patients randomized to RMTP and RMFP were recruited thrice and the RMTP group had PEEP titrated on these same occasions. When applicable, RM preceded PEEP titration (RMTP group). The RM was performed under pressure-controlled ventilation (PCV), using incremental PEEP and inspiratory pressure to minimize biological impact on lungs [10], with a maximum PEEP of 20 cmH2O and peak inspiratory pressure of 40 cmH2O (Fig. 1). Each pressure level was maintained for 10 respiratory cycles (40 s), with a RR of 15 breaths/minute. At 20 cmH2O of PEEP, a decremental PEEP titration maneuver was performed in VCV, decreasing PEEP by 2 cmH2O every 10 respiratory cycles, and keeping VT = 7 mL/kg, RR = 15 breaths/minute, inspiratory pause of 20%, and I: E = 1:2 [11]. Driving pressure of the last respiratory cycle was recorded at each PEEP level. The ideal PEEP was defined as the one that resulted in the lowest driving pressure. If different PEEP levels resulted in the same driving pressure, the lowest level was considered the optimal titrated PEEP and maintained throughout the procedure (Fig. 1). Fig. 1A Ventilatory maneuver in a representative patient from the Recruitment Maneuver Fixed Positive End-Expiratory Pressure (PEEP) (RMFP) group B Ventilatory maneuver in a representative patient from the Recruitment Maneuver Titrated PEEP (RMTP) group. In both patients, the recruitment maneuver was conducted under pressure-controlled ventilation, with a gradual increase in PEEP (6→10→15→20 cm H₂O) and inspiratory pressure (InspP) (26→30→35→40 cm H₂O) every 40 s. After the recruitment maneuver, a fixed PEEP of 6 cm H₂O was set for patients in the RMFP group (A). In the RMTP group, a decremental PEEP titration trial was performed under volume-controlled ventilation, with a gradual decrease in PEEP from 20 to 6 cm H₂O in 2 cm H₂O decrements every 40 s (B). For the representative patient (B), the optimal individualized PEEP was 12 cm H₂O (i.e., the lowest PEEP resulting in the lowest driving pressure) and was maintained until the next ventilatory maneuver. During the entire ventilatory intervention (recruitment maneuver and PEEP titration), a fractional inspired oxygen (FiO2) of 0.5 and a respiratory rate (RR) of 15 breaths per minute were utilised. During volume-controlled ventilation, a tidal volume (VT) of 7 mL/kg was employed A Ventilatory maneuver in a representative patient from the Recruitment Maneuver Fixed Positive End-Expiratory Pressure (PEEP) (RMFP) group B Ventilatory maneuver in a representative patient from the Recruitment Maneuver Titrated PEEP (RMTP) group. In both patients, the recruitment maneuver was conducted under pressure-controlled ventilation, with a gradual increase in PEEP (6→10→15→20 cm H₂O) and inspiratory pressure (InspP) (26→30→35→40 cm H₂O) every 40 s. After the recruitment maneuver, a fixed PEEP of 6 cm H₂O was set for patients in the RMFP group (A). In the RMTP group, a decremental PEEP titration trial was performed under volume-controlled ventilation, with a gradual decrease in PEEP from 20 to 6 cm H₂O in 2 cm H₂O decrements every 40 s (B). For the representative patient (B), the optimal individualized PEEP was 12 cm H₂O (i.e., the lowest PEEP resulting in the lowest driving pressure) and was maintained until the next ventilatory maneuver. During the entire ventilatory intervention (recruitment maneuver and PEEP titration), a fractional inspired oxygen (FiO2) of 0.5 and a respiratory rate (RR) of 15 breaths per minute were utilised. During volume-controlled ventilation, a tidal volume (VT) of 7 mL/kg was employed Patients were weaned using pressure-support ventilation (PSV). FiO2 was adjusted to 50% or lower to keep SpO2 > 92% and PEEP was maintained according to patients’ randomization (i.e., titrated PEEP for patients in the RMTP group and PEEP = 6 cmH2O for patients in RMFP and NRMFP groups). Thirty minutes after PACU admission, any supplemental oxygen was removed for 5 min and the SpO2 recorded, at the end of which oxygen supplementation was reinstated if SpO2<90%. Additional information is provided in the Supplementary material. Primary and secondary outcomes Our primary outcome was the incidence of potentially injurious mechanical ventilation, defined as driving pressure > 15 cmH2O before, during, and after pneumoperitoneum insufflation [8]. The threshold of 15 cmH₂O was selected a priori based on prior perioperative data associating driving pressure above this level with increased PPC risk [8]. The prespecified unadjusted analysis of the primary outcome was considered confirmatory. Adjusted regression models, repeated measures mixed effects analyses, and secondary outcome analyses were considered exploratory and interpreted accordingly. The primary endpoint was analysed according to the intention-to-treat principle, including all randomized participants (n = 78) in the groups to which they were originally assigned. As two participants had missing primary outcome data, intention-to-treat analyses were conducted under extreme-case assumptions. Specifically, best-case and worst-case imputation scenarios were applied, in which missing outcomes were alternatively considered as driving pressure ≤ 15 cmH₂O or > 15 cmH₂O, respectively [12]. The effect of the intervention on the primary outcome is reported as number and percentage and estimated with risk ratios and 95% CIs that were calculated using the Wald method. Secondary respiratory outcomes included: (i) postoperative oxygen requirements in the post-anesthesia care unit (PACU); and (ii) postoperative pulmonary complications (PPC) assessed up to postoperative day 7. PPC were predefined as respiratory infection, respiratory failure, pleural effusion, atelectasis requiring intervention, pneumothorax, bronchospasm, or aspiration pneumonitis [13], and each component was recorded and analyzed individually. Secondary hemodynamic outcomes included: (i) intraoperative hypotension (defined as mean arterial pressure < 65 mmHg, reported as number of episodes); (ii) need for vasopressor support; and (iii) total vasopressor dose administered. Secondary outcomes were analyzed per protocol and considered exploratory. Statistical analysis The sample size was determined a priori to detect a clinically relevant difference in intraoperative driving pressure between groups, which was defined as the primary endpoint of this superiority trial. Sample size was calculated using G*POWER [14] based on previous data [15] where mean intraoperative driving pressure was 18.8 ± 4.1 for fixed PEEP = 4–5 cmH2O, 14.4 ± 5.1 for fixed PEEP = 12 cmH2O, and 9.8 ± 1.4 cmH2O for individualized PEEP, assuming an effect size of 0.38. We estimated that 78 participants (26 per group) would result in 80% of power and significance level of 5% using a one-way analysis of variance (ANOVA) with fixed effects and considering 10% dropout rate. Because the study was designed to evaluate physiological differences in intraoperative respiratory mechanics, the sample size calculation was based on continuous intraoperative driving pressure values. The threshold of driving pressure > 15 cmH₂O was prespecified based on prior perioperative studies associating elevated driving pressure with postoperative pulmonary complications [8] and was used to provide a clinically interpretable categorical assessment of potentially injurious ventilation (Suppl. Figure 2). Continuous variables were expressed as mean (SD) or median (interquartile range), according to data distribution, and compared using parametric or non-parametric tests as appropriate. Categorical variables were expressed as number (%) and compared using χ²-based tests. The primary endpoint (incidence of driving pressure > 15 cmH₂O) was analysed across groups and timepoints using appropriate repeated categorical data methods with adjustment for multiple comparisons. For continuous repeated ventilatory variables, mixed-effects models were used to evaluate temporal changes in ventilatory parameters across groups and intraoperative timepoints. Models included group, timepoint, and group-by-time interaction as fixed effects, with subject-level random effects to account for repeated measurements. Firth penalized logistic regression analyses were performed to evaluate the independent association between group allocation and the occurrence of driving pressure > 15 cmH₂O during pneumoperitoneum before and after recruitment maneuver. Multivariable models included prespecified clinically relevant covariates, including body mass index, type of surgery, ARISCAT score, STOP-BANG score, prior diagnosis of obstructive sleep apnoea, and group allocation. Adjusted odds ratios (OR) with 95% confidence intervals were reported. The Firth penalized likelihood approach was used to reduce potential small-sample bias related to the limited number of outcome events [16]. As a sensitivity analysis, multivariable linear regression models using continuous driving pressure values during pneumoperitoneum as the dependent variable were additionally performed with the same prespecified covariates. Statistical significance was set at P < 0.05. Additional details regarding statistical modelling and post hoc analyses are provided in the Supplementary Material. All analyses were conducted using R software version 4.1.2 (R Foundation for Statistical Computing). NCSS 2021 Statistical Software (NCSS, LLC, Kaysville, Utah, USA) was used to generate the block randomisation list. Results Eighty-five patients were assessed for eligibility, and 78 were randomized [mean age 35.8 (9.0) years; 55 females; mean BMI 40.1 (3.6) kg/m²]. Two patients were excluded from per-protocol analyses after randomization, one due to unplanned intraoperative extubation and another due to protocol deviation related to incorrect initial ventilatory settings (Fig. 2). Baseline demographic and clinical characteristics of the per-protocol population (n = 76) are presented in Table 1, including 26 patients in the RMTP group and 25 patients in both the RMFP and NRMFP groups. Fig. 2CONSORT flow diagram. Abbreviations: BMI: Body mass index; PEEP: Positive End-Expiratory Pressure CONSORT flow diagram. Abbreviations: BMI: Body mass index; PEEP: Positive End-Expiratory Pressure Table 1Baseline demographic and clinical characteristics of patients included in the per-protocol analysis (n = 76)VariableOverall(n = 76)Study groupRMTP(n = 26)RMFP(n = 25)NRMFP(n = 25)Agea (years)35.87± 9.0437.1± 9.837± 8.833.5± 8.3Sexb Female55(72.4)20(76.9)18(72)17(68) Male21(27.6)6(23.1)7(28)8(32)ASA-PS classificationb II34(44.7)12(46.1)12(48)10(40) III42(55.3)14(53.8)13(52)15(60)BMI (kg/m− 2)a40.1± 3.639.8± 3.540.5± 3.940± 3.5Obesity classification2 (BMI) Grade I5(6.6)3(11.5)1(4)1(4) Grade II31(40.8)10(38.5)11(44)10(40) Grade III40(52.6)135013(52)14(56)Highest lifetime body weighta (Kg)116.5± 18.3112.3± 17120± 22.5117.4± 14.4Predicted weight1 (Kg)56.9[53.1–66.9]55.1[50.8–66.7]56.9[55–66]59.6[56.9–66.9]Difference between actual weight and predicted weighta (Kg)52.35± 10.950.5± 9.953.92± 12.952.7± 9.8ARISCAT scoreC15.00[15–15]15[15–15]15[15–15]15[15–15]ARISCAT risk stratificationb Low risk68(89.5)23(88.5)21(84)24(96) Intermediate risk8(10.5)3(11.5)4(16)1(4)STOP-BANG scorec4.00[2.7-5]3[2.2-4]4[3–5]4[3–5]STOP-BANG risk stratificationb Low risk33(43.4)14(53.8)10(40)9(36) Intermediate risk24(31.6)5(19.2)9(36)10(40) High risk19-257(26.9)6(24)6(24)Prior diagnosis of obstructive sleep apnoeab14(18.4)6(23.1)2(8)6(24)Predictors of difficult airwayb8(10.5)4(15.4)1(4)3(12)Hypertension224(31.6)6(23.1)11(44)7(28)Chronic obstructive pulmonary diseaseb1(1.3)0(0)1(4)0(0)Asthmab8(10.5)3(11.5)3(12)2(8)Diabetesb9(11.8)3(11.5)5(20)1(4)Smokingb9(11.8)5(19.2)2(8)2(8)Type of surgeryb Gastric bypass66(86.8)24(92.3)21(84)21(84) Gastric sleeve10(13.2)2(7.7)4(16)4(16) Duration of surgeryc (min)85[70–100]85[70-93.7]90[70–110]80[65–95]Abbreviations:ASA-PS American Society of Anesthesiologists’ Physical Status classification,BMI Body mass index,RMTP Recruitment maneuver and positive end-expiratory pressure (PEEP) titration,RMFP Recruitment maneuver and fixed PEEP,NRMFP No recruitment maneuver and fixed PEEPaData expressed as mean (standard deviation), One-way ANOVA test, bValues expressed as n (%), χ2 test cData expressed as median (IQR [interquartile range]), Kruskal Wallis test Table 1Baseline demographic and clinical characteristics of patients included in the per-protocol analysis (n = 76)VariableOverall(n = 76)Study groupRMTP(n = 26)RMFP(n = 25)NRMFP(n = 25)Agea (years)35.87± 9.0437.1± 9.837± 8.833.5± 8.3Sexb Female55(72.4)20(76.9)18(72)17(68) Male21(27.6)6(23.1)7(28)8(32)ASA-PS classificationb II34(44.7)12(46.1)12(48)10(40) III42(55.3)14(53.8)13(52)15(60)BMI (kg/m− 2)a40.1± 3.639.8± 3.540.5± 3.940± 3.5Obesity classification2 (BMI) Grade I5(6.6)3(11.5)1(4)1(4) Grade II31(40.8)10(38.5)11(44)10(40) Grade III40(52.6)135013(52)14(56)Highest lifetime body weighta (Kg)116.5± 18.3112.3± 17120± 22.5117.4± 14.4Predicted weight1 (Kg)56.9[53.1–66.9]55.1[50.8–66.7]56.9[55–66]59.6[56.9–66.9]Difference between actual weight and predicted weighta (Kg)52.35± 10.950.5± 9.953.92± 12.952.7± 9.8ARISCAT scoreC15.00[15–15]15[15–15]15[15–15]15[15–15]ARISCAT risk stratificationb Low risk68(89.5)23(88.5)21(84)24(96) Intermediate risk8(10.5)3(11.5)4(16)1(4)STOP-BANG scorec4.00[2.7-5]3[2.2-4]4[3–5]4[3–5]STOP-BANG risk stratificationb Low risk33(43.4)14(53.8)10(40)9(36) Intermediate risk24(31.6)5(19.2)9(36)10(40) High risk19-257(26.9)6(24)6(24)Prior diagnosis of obstructive sleep apnoeab14(18.4)6(23.1)2(8)6(24)Predictors of difficult airwayb8(10.5)4(15.4)1(4)3(12)Hypertension224(31.6)6(23.1)11(44)7(28)Chronic obstructive pulmonary diseaseb1(1.3)0(0)1(4)0(0)Asthmab8(10.5)3(11.5)3(12)2(8)Diabetesb9(11.8)3(11.5)5(20)1(4)Smokingb9(11.8)5(19.2)2(8)2(8)Type of surgeryb Gastric bypass66(86.8)24(92.3)21(84)21(84) Gastric sleeve10(13.2)2(7.7)4(16)4(16) Duration of surgeryc (min)85[70–100]85[70-93.7]90[70–110]80[65–95]Abbreviations:ASA-PS American Society of Anesthesiologists’ Physical Status classification,BMI Body mass index,RMTP Recruitment maneuver and positive end-expiratory pressure (PEEP) titration,RMFP Recruitment maneuver and fixed PEEP,NRMFP No recruitment maneuver and fixed PEEPaData expressed as mean (standard deviation), One-way ANOVA test, bValues expressed as n (%), χ2 test cData expressed as median (IQR [interquartile range]), Kruskal Wallis test Baseline demographic and clinical characteristics of patients included in the per-protocol analysis (n = 76) Abbreviations:ASA-PS American Society of Anesthesiologists’ Physical Status classification,BMI Body mass index,RMTP Recruitment maneuver and positive end-expiratory pressure (PEEP) titration,RMFP Recruitment maneuver and fixed PEEP,NRMFP No recruitment maneuver and fixed PEEP aData expressed as mean (standard deviation), One-way ANOVA test, bValues expressed as n (%), χ2 test cData expressed as median (IQR [interquartile range]), Kruskal Wallis test Ventilatory parameters In the per-protocol analysis (n = 76), ventilatory parameters including driving pressure, plateau pressure, PEEP, and peak pressure are summarized in Table 2. Measurements obtained before pneumoperitoneum were recorded after initiation of the allocated ventilatory strategy and therefore do not represent pre-intervention baseline values. Driving pressure was consistently lower in the RMTP group compared with RMFP and NRMFP across all intraoperative timepoints (P < 0.05), whereas no significant differences were observed between RMFP and NRMFP during pneumoperitoneum. Plateau and peak pressures were generally higher in the RMTP group, reflecting the higher PEEP levels used in the titration strategy (P < 0.05). Peak pressure was significantly higher in the RMTP group at all timepoints except during pneumoperitoneum before recruitment maneuver, where no between-group difference was observed. Static and dynamic compliance were consistently higher in the RMTP group throughout the intraoperative period (P < 0.05) (Suppl. Table 1). Detailed post hoc pairwise comparisons are provided in Supplementary Table 2. Table 2Intraoperative ventilatory parameters of patients included in the per-protocol analysis across prespecified timepoints (n = 76)Ventilatory parametersStudy groupP-valueRMTP(n = 26)RMFP(n = 25)NRMFP(n = 25)Initial Tidal Volumea (mL)388[352.5-465.5]390[380–460]417[390–462.0]0.21Before PneumoperitoneumPlateau pressurea (cmH2O)16[15–20]14[13–15]16[16–18]< 0.001Peak pressurea (cmH2O)20[18–23]18[16–19]20[19–21]0.003PEEPa (cmH2O)10[10–12]6[6–6]6[6–6]< 0.001Driving pressurea6[5.3-7]8[7–9]10[10–12]< 0.001During Pneumoperitoneum Before Recruitment ManeuverRespiratory ratea (breaths/min)15[15–15]15[15–15]15[15–15]0.75Plateau pressurea (cmH2O)22[20-24.8]20[18–21]20[20–22]0.03Peak pressurea (cmH2O)24[22.3–28.5]24[21–25]24[23–26]0.16Driving pressurea (cmH2O)12[10–13]14[12–15]14[14–16]< 0.001During Pneumoperitoneum After Recruitment ManeuverPlateau pressurea (cmH2O)24[21.3–26.8]19[18–21]20[20–22]< 0.001Peak pressurea (cmH2O)27[25.2–29.7]23[20–24]24[23–26]< 0.001PEEPa (cmH2O)16.[14–18]6[6–6]6[6–6]< 0.001Driving pressurea (cmH2O)8[7–10]13[12–15]14[14–16]< 0.001After Pneumoperitoneum Before Recruitment ManeuverRespiratory ratea (ipm)20[15–20]15[15–20]15[15–20]0.13Plateau pressurea (cmH2O)23[21-25.8]17[16–18]18[16–18]< 0.001Peak pressurea (cmH2O)26[24–29]20[19–22]21[20–22]< 0.001Driving pressurea (cmH2O)7.5[7–9]11[10–12]12[10–12]< 0.001After Pneumoperitoneum After Recruitment ManeuverPlateau pressurea (cmH2O)17[15–19]14[13–15]18[16–18]< 0.001Peak pressurea (cmH2O)20.5[18–22]17[16–18]21[20–22]< 0.001PEEPa (cmH2O)11[10–12]6[6–6]6[6–6]< 0.001Driving pressurea (cmH2O)6[5.3-7]8[7–9]12[10–12]< 0.001Significant pairwise post hoc comparisons are provided in Supplementary Table 2. Analyses of ventilatory parameters were performed per protocol due to incomplete physiologic data in two randomized participantsaData expressed as median (IQR [interquartile range]), Kruskal Wallis test followed by a Dunn post-hoc test if the P-value was < 0.05 Table 2Intraoperative ventilatory parameters of patients included in the per-protocol analysis across prespecified timepoints (n = 76)Ventilatory parametersStudy groupP-valueRMTP(n = 26)RMFP(n = 25)NRMFP(n = 25)Initial Tidal Volumea (mL)388[352.5-465.5]390[380–460]417[390–462.0]0.21Before PneumoperitoneumPlateau pressurea (cmH2O)16[15–20]14[13–15]16[16–18]< 0.001Peak pressurea (cmH2O)20[18–23]18[16–19]20[19–21]0.003PEEPa (cmH2O)10[10–12]6[6–6]6[6–6]< 0.001Driving pressurea6[5.3-7]8[7–9]10[10–12]< 0.001During Pneumoperitoneum Before Recruitment ManeuverRespiratory ratea (breaths/min)15[15–15]15[15–15]15[15–15]0.75Plateau pressurea (cmH2O)22[20-24.8]20[18–21]20[20–22]0.03Peak pressurea (cmH2O)24[22.3–28.5]24[21–25]24[23–26]0.16Driving pressurea (cmH2O)12[10–13]14[12–15]14[14–16]< 0.001During Pneumoperitoneum After Recruitment ManeuverPlateau pressurea (cmH2O)24[21.3–26.8]19[18–21]20[20–22]< 0.001Peak pressurea (cmH2O)27[25.2–29.7]23[20–24]24[23–26]< 0.001PEEPa (cmH2O)16.[14–18]6[6–6]6[6–6]< 0.001Driving pressurea (cmH2O)8[7–10]13[12–15]14[14–16]< 0.001After Pneumoperitoneum Before Recruitment ManeuverRespiratory ratea (ipm)20[15–20]15[15–20]15[15–20]0.13Plateau pressurea (cmH2O)23[21-25.8]17[16–18]18[16–18]< 0.001Peak pressurea (cmH2O)26[24–29]20[19–22]21[20–22]< 0.001Driving pressurea (cmH2O)7.5[7–9]11[10–12]12[10–12]< 0.001After Pneumoperitoneum After Recruitment ManeuverPlateau pressurea (cmH2O)17[15–19]14[13–15]18[16–18]< 0.001Peak pressurea (cmH2O)20.5[18–22]17[16–18]21[20–22]< 0.001PEEPa (cmH2O)11[10–12]6[6–6]6[6–6]< 0.001Driving pressurea (cmH2O)6[5.3-7]8[7–9]12[10–12]< 0.001Significant pairwise post hoc comparisons are provided in Supplementary Table 2. Analyses of ventilatory parameters were performed per protocol due to incomplete physiologic data in two randomized participantsaData expressed as median (IQR [interquartile range]), Kruskal Wallis test followed by a Dunn post-hoc test if the P-value was < 0.05 Intraoperative ventilatory parameters of patients included in the per-protocol analysis across prespecified timepoints (n = 76) Significant pairwise post hoc comparisons are provided in Supplementary Table 2. Analyses of ventilatory parameters were performed per protocol due to incomplete physiologic data in two randomized participants aData expressed as median (IQR [interquartile range]), Kruskal Wallis test followed by a Dunn post-hoc test if the P-value was < 0.05 Mixed-effects analysis revealed a significant group-by-time interaction for driving pressure, plateau pressure, PEEP and peak pressure (Fig. 3; Suppl. Table 3). In the RMTP group, driving pressure was significantly reduced during pneumoperitoneum, both before and after recruitment maneuver (P < 0.05), while no such reduction was observed in the RMFP and NRMFP groups (P > 0.05). Both static and dynamic compliance increased after pneumoperitoneum deflation and recruitment maneuver in the RMTP and RMFP groups (Suppl. Figure 3; Suppl. Table 3). Fig. 3Ventilatory parameters across intraoperative timepoints: driving pressure (A), peak pressure (B), and plateau pressure (C). A significant group × time interaction was observed for all variables (P < 0.05). Post hoc pairwise comparisons were adjusted using the Tukey-Kramer method. *P < 0.05 compared Before PN vs. During PN before RM. †P < 0.05 compared During PN before RM vs. During PN after RM. ‡P < 0.05 compared During PN after RM vs. After PN before RM. §P < 0.05 After PN before RM vs. After PN after RM. Abbreviations: PN, pneumoperitoneum; RM, recruitment maneuver; RMTP, recruitment maneuver with driving pressure-guided PEEP titration; RMFP, recruitment maneuver with fixed PEEP; NRMFP, no recruitment maneuver with fixed PEEP. Absolute between-group differences in driving pressure at significant timepoints were as follows: before pneumoperitoneum, RMTP versus RMFP = − 2 cmH₂O and RMTP versus NRMFP = − 4 cmH₂O; during pneumoperitoneum before recruitment maneuver, RMTP versus RMFP = − 2 cmH₂O and RMTP versus NRMFP = − 2 cmH₂O; during pneumoperitoneum after recruitment maneuver, RMTP versus RMFP = − 5 cmH₂O and RMTP versus NRMFP = − 6 cmH₂O; after pneumoperitoneum before recruitment maneuver, RMTP versus RMFP = − 3.5 cmH₂O and RMTP versus NRMFP = − 4.5 cmH₂O; after pneumoperitoneum after recruitment maneuver, RMTP versus RMFP = − 2 cmH₂O and RMTP versus NRMFP = − 6 cmH₂O Ventilatory parameters across intraoperative timepoints: driving pressure (A), peak pressure (B), and plateau pressure (C). A significant group × time interaction was observed for all variables (P < 0.05). Post hoc pairwise comparisons were adjusted using the Tukey-Kramer method. *P < 0.05 compared Before PN vs. During PN before RM. †P < 0.05 compared During PN before RM vs. During PN after RM. ‡P < 0.05 compared During PN after RM vs. After PN before RM. §P < 0.05 After PN before RM vs. After PN after RM. Abbreviations: PN, pneumoperitoneum; RM, recruitment maneuver; RMTP, recruitment maneuver with driving pressure-guided PEEP titration; RMFP, recruitment maneuver with fixed PEEP; NRMFP, no recruitment maneuver with fixed PEEP. Absolute between-group differences in driving pressure at significant timepoints were as follows: before pneumoperitoneum, RMTP versus RMFP = − 2 cmH₂O and RMTP versus NRMFP = − 4 cmH₂O; during pneumoperitoneum before recruitment maneuver, RMTP versus RMFP = − 2 cmH₂O and RMTP versus NRMFP = − 2 cmH₂O; during pneumoperitoneum after recruitment maneuver, RMTP versus RMFP = − 5 cmH₂O and RMTP versus NRMFP = − 6 cmH₂O; after pneumoperitoneum before recruitment maneuver, RMTP versus RMFP = − 3.5 cmH₂O and RMTP versus NRMFP = − 4.5 cmH₂O; after pneumoperitoneum after recruitment maneuver, RMTP versus RMFP = − 2 cmH₂O and RMTP versus NRMFP = − 6 cmH₂O Mean driving pressure-guided PEEP was 11.1(2.7) cmH2O before pneumoperitoneum, 15.4(3.3) cmH2O during pneumoperitoneum, and 11.2(2.8) cmH2O after pneumoperitoneum; it was also higher during pneumoperitoneum than other timepoints (P < 0.001) (Suppl. Figure 4). Primary outcome In the intention-to-treat analysis including all randomized participants (n = 78), missing primary outcome data were imputed as driving pressure ≤ 15 cmH₂O for the primary analysis (Table 3). No between-group differences were observed before pneumoperitoneum or after pneumoperitoneum reversal. However, during pneumoperitoneum, the incidence of driving pressure > 15 cmH₂O was significantly lower in the RMTP group compared with both RMFP and NRMFP, before and after recruitment maneuver, whereas no significant differences were observed between RMFP and NRMFP. Sensitivity analyses using the alternative extreme-case scenario, in which missing outcomes were imputed as driving pressure > 15 cmH₂O, yielded similar effect estimates (Suppl. Table 4). Per-protocol analyses (n = 76) demonstrated consistent findings, with significant between-group differences observed only during pneumoperitoneum before and after recruitment maneuver (Suppl. Table 5; Suppl. Figure 5). Table 3Incidence of driving pressure > 15 cmH₂O across prespecified intraoperative timepoints in patients included in the intention-to-treat analysis (n = 78)TimepointGroupEvents / TotalRisk, n/N (%)Relative Risk (95% CI)Risk Difference*P-value**Before PNRMTP0 / 260.0Reference––RMFP0 / 260.01.00 (0.02–38.60)0.001.00NRMFP1 / 263.83.00 (0.13–50.42)3.80 (− 6.4–15.5)0.49During PN Before RMRMTP2 / 267.7Reference––RMFP11 / 2642.35.50 (1.35–22.42)34.60 (10.7–56.6)0.01NRMFP11 / 2642.35.50 (1.35–22.42)34.60 (10.7–56.6)0.01During PN After RMRMTP1 / 263.8Reference––RMFP7 / 2626.97.00 (1.92–42.96)23.10 (2.7–42.5)0.04NRMFP11 / 2642.311.00 (1.53–59.17)38.50 (17.4–57.0)0.01After PN Before RMRMTP0 / 260.0Reference––RMFP1 / 263.83.00 (0.13–60.42)3.80 (− 6.4–15.5)0.49NRMFP2 / 267.75.00 (0.25–49.35)7.70 (− 3.9–20.6)0.29After PN After RMRMTP0 / 260.0Reference––RMFP0 / 260.01.00 (0.02–38.60)0.001.00NRMFP2 / 267.75.00 (0.25–79.35)7.70 (− 3.9–20.6)0.29Data are presented as number of events/total and percentage. Relative risks (RRs) and 95% confidence intervals (CIs) were calculated using the RMTP group as the reference category. Missing primary outcome data (n = 2) were imputed as driving pressure ≤ 15 cmH₂O for the main intention-to-treat analysisAbbreviations: RMTP Recruitment maneuver and Positive End-Expiratory Pressure (PEEP) titration, RMFP Recruitment maneuver and fixed PEEP, NRMFP No recruitment maneuver and fixed PEEP, PN Pneumoperitoneum, RM Recruitment maneuver*Risk difference represents the absolute difference in incidence between each intervention group and the RMTP reference group at the same timepoint, **P values were derived from Fisher’s exact test due to sparse event counts. When zero counts were present in any cell, the Haldane–Anscombe correction was applied to stabilize the variance Table 3Incidence of driving pressure > 15 cmH₂O across prespecified intraoperative timepoints in patients included in the intention-to-treat analysis (n = 78)TimepointGroupEvents / TotalRisk, n/N (%)Relative Risk (95% CI)Risk Difference*P-value**Before PNRMTP0 / 260.0Reference––RMFP0 / 260.01.00 (0.02–38.60)0.001.00NRMFP1 / 263.83.00 (0.13–50.42)3.80 (− 6.4–15.5)0.49During PN Before RMRMTP2 / 267.7Reference––RMFP11 / 2642.35.50 (1.35–22.42)34.60 (10.7–56.6)0.01NRMFP11 / 2642.35.50 (1.35–22.42)34.60 (10.7–56.6)0.01During PN After RMRMTP1 / 263.8Reference––RMFP7 / 2626.97.00 (1.92–42.96)23.10 (2.7–42.5)0.04NRMFP11 / 2642.311.00 (1.53–59.17)38.50 (17.4–57.0)0.01After PN Before RMRMTP0 / 260.0Reference––RMFP1 / 263.83.00 (0.13–60.42)3.80 (− 6.4–15.5)0.49NRMFP2 / 267.75.00 (0.25–49.35)7.70 (− 3.9–20.6)0.29After PN After RMRMTP0 / 260.0Reference––RMFP0 / 260.01.00 (0.02–38.60)0.001.00NRMFP2 / 267.75.00 (0.25–79.35)7.70 (− 3.9–20.6)0.29Data are presented as number of events/total and percentage. Relative risks (RRs) and 95% confidence intervals (CIs) were calculated using the RMTP group as the reference category. Missing primary outcome data (n = 2) were imputed as driving pressure ≤ 15 cmH₂O for the main intention-to-treat analysisAbbreviations: RMTP Recruitment maneuver and Positive End-Expiratory Pressure (PEEP) titration, RMFP Recruitment maneuver and fixed PEEP, NRMFP No recruitment maneuver and fixed PEEP, PN Pneumoperitoneum, RM Recruitment maneuver*Risk difference represents the absolute difference in incidence between each intervention group and the RMTP reference group at the same timepoint, **P values were derived from Fisher’s exact test due to sparse event counts. When zero counts were present in any cell, the Haldane–Anscombe correction was applied to stabilize the variance Incidence of driving pressure > 15 cmH₂O across prespecified intraoperative timepoints in patients included in the intention-to-treat analysis (n = 78) Data are presented as number of events/total and percentage. Relative risks (RRs) and 95% confidence intervals (CIs) were calculated using the RMTP group as the reference category. Missing primary outcome data (n = 2) were imputed as driving pressure ≤ 15 cmH₂O for the main intention-to-treat analysis Abbreviations: RMTP Recruitment maneuver and Positive End-Expiratory Pressure (PEEP) titration, RMFP Recruitment maneuver and fixed PEEP, NRMFP No recruitment maneuver and fixed PEEP, PN Pneumoperitoneum, RM Recruitment maneuver *Risk difference represents the absolute difference in incidence between each intervention group and the RMTP reference group at the same timepoint, **P values were derived from Fisher’s exact test due to sparse event counts. When zero counts were present in any cell, the Haldane–Anscombe correction was applied to stabilize the variance In exploratory multivariable logistic regression analyses, using RMTP as the reference group, higher odds of driving pressure > 15 cmH₂O were observed in both RMFP and NRMFP during pneumoperitoneum before recruitment maneuver (Suppl. Table 6). Similar findings were observed after recruitment maneuver (Suppl. Table 7). Higher ARISCAT and STOP-BANG scores were also associated with increased odds of the outcome at both timepoints. Sensitivity analyses using continuous driving pressure as the dependent variable yielded consistent results, with group allocation remaining independently associated with lower driving pressure during pneumoperitoneum (Supplementary Tables 8–11; Suppl. Figure 6–7). Secondary outcomes Table 4 summarizes all secondary outcomes. Among predefined postoperative pulmonary complications, only respiratory failure and pleural effusion were observed. No cases of pneumonia, bronchospasm, reintubation, atelectasis requiring intervention, or need for noninvasive ventilation were recorded. There were no differences in any secondary outcomes evaluated. Postoperative oxygen supplementation and SpO2 were similar in PACU and in the inpatient unit. Moreover, the incidence of SpO2 > 89% or > 91% 30 min after extubation was comparable between groups. Twenty-one patients developed respiratory failure or pleural effusion: 3 (11.5%) in RMTP, 8 (32%) in RMFP, and 10 (40%) in NRMFP, with no differences observed between groups. Hemodynamic management (i.e., need for vasopressor during surgery, total vasopressor use, and number of hypotension episodes) did not differ between groups. Additional respiratory and hemodynamic parameters as well as pain and nausea/vomiting parameters in PACU (including comparisons between groups) are shown in Suppl. Tables 12 and 13. Table 4Secondary outcomes: postoperative oxygen use, postoperative pulmonary complications, and hemodynamic managementVariableStudy groupP-valueRMTP(n = 26)RMFP(n = 25)NRMFP(n = 25)Postoperative Oxygen UseSupplemental oxygen upon admission to PACUa7(26.9)7(28)7(28)0.99Supplemental oxygen during the hospital staya11(42.3)13(52)16(64)0.29Supplemental oxygen maintenance at dischargea3(11.5)7(28)9(36)0.11Supplemental oxygen maintenance in the inpatient unita3(11.5)8(32)9(36)0.10SpO2 30 min after extubationb93.7(3.8)92.3(3.9)91.7(4.4)0.19SpO2 > 89% 30 min after extubationa23(88.5)19(76)18(72)0.32SpO2 > 91% 30 min after extubationa21(80.8)17(68)14(56)0.16Sat/FiO2 ratio in PACUb446.1(18.3)439.6(18.3)436.5(20.9)0.19SpO2 6 h after extubationc96[95–97]95.0[95–96]96[94–96]0.45SpO2 12 h after extubationc95.2(2.2)95.9(2.1)95(1.5)0.29Postoperative Pulmonary ComplicationsRespiratory Failurea3(11.5)8(32)9(36)0.10Pleural Effusiona0(0)0(0)1(4)NAHemodynamic ManagementNeed for vasopressor during surgerya0.18 No8(30.8)12(48)6(24) Yes18(69.2)13(52)19(76)Use of metaraminol during surgerya0.23 No8(30.8)7(35)3(13.6) Yes18(69.2)13(65)19(86.4)Metaraminolc (mg)2.5[2-3.4]1.5[1-3.5]2.5[1-4.5]0.54Number of hypotension episodesc4[0–6]1[0–3]3[1–8]0.15Total volume of crystalloid infusedc (mL)800[750–1000]800[750- 000]1000[750–1000]0.65Postoperative pulmonary complications were predefined as respiratory failure, pleural effusion, pneumonia, bronchospasm, atelectasis requiring intervention, reintubation, and need for noninvasive ventilation at postoperative day 7. Only respiratory failure and pleural effusion occurred during the study period. Secondary outcomes were analyzed per protocol and should be interpreted as exploratory. The study was not powered to detect differences in secondary clinical outcomesaValues expressed as n (%), χ2 test, bData expressed as mean (standard deviation), One-way ANOVA test followed by a Tukey post-hoc test if theP-value was < 0.05, cData expressed as median (IQR [interquartile range]), Kruskal Wallis test followed by a Dunn post-hoc test if theP-value was < 0.05. NA – Not applicable Table 4Secondary outcomes: postoperative oxygen use, postoperative pulmonary complications, and hemodynamic managementVariableStudy groupP-valueRMTP(n = 26)RMFP(n = 25)NRMFP(n = 25)Postoperative Oxygen UseSupplemental oxygen upon admission to PACUa7(26.9)7(28)7(28)0.99Supplemental oxygen during the hospital staya11(42.3)13(52)16(64)0.29Supplemental oxygen maintenance at dischargea3(11.5)7(28)9(36)0.11Supplemental oxygen maintenance in the inpatient unita3(11.5)8(32)9(36)0.10SpO2 30 min after extubationb93.7(3.8)92.3(3.9)91.7(4.4)0.19SpO2 > 89% 30 min after extubationa23(88.5)19(76)18(72)0.32SpO2 > 91% 30 min after extubationa21(80.8)17(68)14(56)0.16Sat/FiO2 ratio in PACUb446.1(18.3)439.6(18.3)436.5(20.9)0.19SpO2 6 h after extubationc96[95–97]95.0[95–96]96[94–96]0.45SpO2 12 h after extubationc95.2(2.2)95.9(2.1)95(1.5)0.29Postoperative Pulmonary ComplicationsRespiratory Failurea3(11.5)8(32)9(36)0.10Pleural Effusiona0(0)0(0)1(4)NAHemodynamic ManagementNeed for vasopressor during surgerya0.18 No8(30.8)12(48)6(24) Yes18(69.2)13(52)19(76)Use of metaraminol during surgerya0.23 No8(30.8)7(35)3(13.6) Yes18(69.2)13(65)19(86.4)Metaraminolc (mg)2.5[2-3.4]1.5[1-3.5]2.5[1-4.5]0.54Number of hypotension episodesc4[0–6]1[0–3]3[1–8]0.15Total volume of crystalloid infusedc (mL)800[750–1000]800[750- 000]1000[750–1000]0.65Postoperative pulmonary complications were predefined as respiratory failure, pleural effusion, pneumonia, bronchospasm, atelectasis requiring intervention, reintubation, and need for noninvasive ventilation at postoperative day 7. Only respiratory failure and pleural effusion occurred during the study period. Secondary outcomes were analyzed per protocol and should be interpreted as exploratory. The study was not powered to detect differences in secondary clinical outcomesaValues expressed as n (%), χ2 test, bData expressed as mean (standard deviation), One-way ANOVA test followed by a Tukey post-hoc test if theP-value was < 0.05, cData expressed as median (IQR [interquartile range]), Kruskal Wallis test followed by a Dunn post-hoc test if theP-value was < 0.05. NA – Not applicable Secondary outcomes: postoperative oxygen use, postoperative pulmonary complications, and hemodynamic management Postoperative pulmonary complications were predefined as respiratory failure, pleural effusion, pneumonia, bronchospasm, atelectasis requiring intervention, reintubation, and need for noninvasive ventilation at postoperative day 7. Only respiratory failure and pleural effusion occurred during the study period. Secondary outcomes were analyzed per protocol and should be interpreted as exploratory. The study was not powered to detect differences in secondary clinical outcomes aValues expressed as n (%), χ2 test, bData expressed as mean (standard deviation), One-way ANOVA test followed by a Tukey post-hoc test if theP-value was < 0.05, cData expressed as median (IQR [interquartile range]), Kruskal Wallis test followed by a Dunn post-hoc test if theP-value was < 0.05. NA – Not applicable Discussion This randomized controlled trial evaluated whether recruitment maneuver combined with driving pressure-guided PEEP titration reduced the incidence of driving pressure > 15 cmH₂O in obese patients undergoing laparoscopic bariatric surgery compared with fixed PEEP strategies with or without recruitment maneuver. The RMTP group showed lower rates of elevated driving pressure, particularly during pneumoperitoneum, whereas no differences were observed between RMFP and NRMFP groups. Driving pressure-guided PEEP levels also varied substantially according to pneumoperitoneum status, reflecting dynamic intraoperative changes in respiratory mechanics. The effects of driving pressure-guided PEEP titration were most evident during pneumoperitoneum, when the RMTP group showed a lower incidence of elevated driving pressure compared with fixed PEEP strategies, both before and after recruitment maneuver. Pneumoperitoneum increases chest wall elastance and mechanical load on the respiratory system, contributing to higher driving pressures during laparoscopic surgery in obese patients [17]. In this setting, individualized PEEP titration improved intraoperative respiratory mechanics during the period of greatest respiratory system constraint [7, 18]. Although the absolute reduction in driving pressure was modest (approximately 2–5 cmH₂O), previous perioperative studies in higher-risk patients suggest that even small reductions may be associated with lower rates of postoperative pulmonary complications [19]. However, whether this relationship applies to predominantly low-risk surgical populations remains uncertain. Interpretation of these findings also requires consideration of the threshold used to define potentially injurious ventilation. The cut-off of driving pressure > 15 cmH₂O was derived primarily from studies in critically ill and acute respiratory distress syndrome populations and has not been specifically validated in low-risk obese surgical patients [7]. It is possible that the threshold associated with clinically relevant ventilator-induced lung stress differs in lower-risk perioperative populations. Lower thresholds may increase detection of potentially injurious ventilation at the expense of specificity, whereas higher thresholds currently lack robust validation in the perioperative literature. Results from the IMPROVE-2 trial [20], in which targeting a driving pressure below 13 cmH₂O was associated with a significantly higher incidence of reintubation and requirement for curative non-invasive ventilation, suggest that a less restrictive threshold may represent a safer intraoperative target. In addition, this trial was not powered to detect differences in clinical outcomes, and the predominantly low-risk population likely contributed to the low incidence of postoperative pulmonary complications observed across groups. Previous studies have evaluated individualized PEEP strategies using adjunct monitoring techniques such as electrical impedance tomography or respiratory compliance measurements [11, 21, 22]. In the present study, PEEP titration was guided by driving pressure, a parameter readily derived from standard ventilatory variables and easily applicable during routine intraoperative monitoring. Notably, previous studies enrolling obese patients have reported a wide range of optimal PEEP values, varying from 9 to 18.5 cmH₂O, potentially reflecting differences in BMI across study populations [15, 19–22]. Recent meta-analytic evidence published after the design and conduct of the present trial also suggested potential benefits of fixed PEEP levels ≥ 10 cmH₂O in obese patients undergoing laparoscopic bariatric surgery [23]. However, variability among included studies regarding ventilatory protocols, recruitment maneuver, obesity severity, and perioperative management reflects the complexity of defining an optimal fixed PEEP strategy in this population. It is also worth noting that the use of 6 cmH₂O as fixed PEEP in the control groups, below the optimal range identified in a meta-analysis of obese patients (≥ 10 cmH₂O), may have amplified the observed between-group differences [23]. In the present study, optimal PEEP differed by approximately 4 cmH₂O between pneumoperitoneum and desufflation phases, highlighting the dynamic nature of intraoperative respiratory mechanics and the potential need for repeated PEEP titration. Recruitment maneuver without individualized PEEP titration did not reduce the incidence of elevated driving pressure at any intraoperative timepoint, including during pneumoperitoneum. The three-arm design allowed assessment of the isolated effect of recruitment maneuver and suggests that recruitment alone may have limited effects on respiratory mechanics in this setting. These results are consistent with the PROBESE trial, which failed to demonstrate clinical benefit with a strategy combining recruitment maneuver and high fixed PEEP compared with low PEEP without recruitment maneuver [24]. Previous studies evaluating individualized PEEP strategies in obese patients have reported variable optimal PEEP levels according to body habitus and intraoperative conditions [15, 25, 26] . Despite the observed intraoperative physiological improvements, no differences in PPC or oxygen requirements were observed. This absence of detectable clinical benefit should be interpreted cautiously, as the study was powered to detect differences in driving pressure rather than postoperative clinical outcomes. In addition, the predominantly low-risk population and the relatively modest absolute reductions in driving pressure may have limited the ability to detect clinically meaningful postoperative differences. Similar findings were reported by Kim et al. in patients undergoing laparoscopic or robotic abdominal surgery, in which driving pressure-guided individualized PEEP reduced intraoperative driving pressure without significantly decreasing postoperative pulmonary complications [27]. Likewise, the DESIGNATION and IMPROVE-2 trials also demonstrated reductions in driving pressure without improvements in postoperative pulmonary outcomes, results directionally consistent with ours [20, 28]. Notably, IMPROVE-2 also reported higher rates of reintubation or need for noninvasive ventilation in the intervention group [20]. However, important differences in surgical population and intraoperative conditions should be considered when comparing these studies, including the inclusion of open or emergency abdominal surgery and the absence of pneumoperitoneum in the DESIGNATION trial [28], which was precisely the intraoperative phase during which the intervention demonstrated the greatest effect in our study. In the present study, no impact on clinical outcomes was observed despite such measures having been applied to all participants. Higher ARISCAT and STOP-BANG scores were independently associated with elevated driving pressure during pneumoperitoneum. As both instruments are established predictors of perioperative respiratory risk, these results suggest that patients with greater baseline respiratory vulnerability may be more prone to elevated intraoperative mechanical stress [29, 30]. Whether individualized ventilatory strategies provide greater benefit in higher-risk populations remains uncertain and warrants further investigation. Although higher PEEP and recruitment maneuver may improve respiratory mechanics, they can also affect hemodynamic stability [31, 32]. No significant between-group differences were observed regarding hemodynamic management, including vasopressor requirement, total vasopressor use, and number of hypotensive episodes. However, the trial was not powered to detect differences in hemodynamic outcomes. The multicenter randomized PROBESE trial [24], which compared fixed high PEEP combined with recruitment maneuver versus fixed low PEEP, reported significant hemodynamic effects associated with the intervention strategy. In contrast, individualized PEEP titration may be expected to produce less pronounced hemodynamic compromise than fixed high PEEP approaches, particularly when lower effective PEEP levels are selected. Consistent with our findings, several studies evaluating individualized PEEP titration using different methodologies also reported no significant differences in hemodynamic outcomes [11, 19]. Conversely, deleterious hemodynamic effects were reported in the DESIGNATION and IMPROVE-2 trials [20, 28]. However, important methodological differences should be considered when comparing these studies with the present trial. IMPROVE-2 included exclusively emergency surgery patients, a population inherently more susceptible to perioperative hemodynamic instability [20]. In DESIGNATION trial, optimal PEEP was defined as the highest PEEP associated with the lowest driving pressure, whereas in the present study the lowest PEEP resulting in lowest driving pressure was selected [28]. This strategy may have reduced exposure to unnecessarily high intrathoracic pressures and consequently limited hemodynamic impairment. Study limitations This study has several limitations. First, although adequately powered to detect differences in physiological respiratory variables, the trial was neither designed nor powered to detect differences in postoperative clinical outcomes, including PPC and oxygenation endpoints. Consequently, the absence of significant differences in these outcomes should be interpreted cautiously. In addition, the relatively low number of outcome events resulted in wide confidence intervals in some adjusted analyses, limiting the precision of effect estimates despite the use of Firth penalized logistic regression to reduce small-sample bias. Second, objective assessment of perioperative atelectasis was not performed, as imaging modalities such as computed tomography, lung ultrasound, or electrical impedance tomography were not incorporated into the study protocol. Therefore, the mechanistic relationship between the observed reductions in driving pressure and the degree of lung aeration or alveolar recruitment could not be directly assessed. Third, the study population consisted predominantly of low-risk patients, with a median ARISCAT score of 15 and nearly 90% classified as low risk for postoperative pulmonary complications, in addition to excluding super-obese patients. Accordingly, the generalisability of these findings to higher-risk surgical populations or patients with more severe obesity may be limited. Finally, although the threshold of driving pressure > 15 cmH₂O was selected a priori based on previous perioperative and critical care literature associating elevated driving pressure with worse pulmonary outcomes, the optimal intraoperative threshold for low-risk surgical populations remains uncertain. Furthermore, the primary outcome was ultimately analysed as a binary variable despite the sample size calculation having been based on continuous driving pressure values, which may simplify interpretation of a continuous physiological parameter and potentially reduce statistical power. Therefore, the present trial should be interpreted primarily within a physiological and mechanistic framework rather than as evidence of direct clinical benefit. Conclusion In conclusion, recruitment maneuver combined with driving pressure-guided PEEP titration improved intraoperative respiratory mechanics and reduced the incidence of elevated driving pressure during pneumoperitoneum in obese patients undergoing laparoscopic bariatric surgery, including a predominantly low-risk and non-super-obese population. However, no differences were observed in hemodynamic management or postoperative clinical outcomes. In contrast, recruitment maneuver without individualized PEEP titration did not reduce the incidence of elevated driving pressure compared with lung-protective ventilation strategy without recruitment maneuver. Adequately powered trials are required to determine whether intraoperative optimization of driving pressure translates into meaningful improvements in patient-centered clinical endpoints. Supplementary Information Supplementary Material 1. Supplementary Material 1.