Two researchers (FXP/KEL) independently searched the Embase, Web of Science, PubMed, and Cochrane Library databases from their inception to April 10, 2025. The search strategy incorporated a combination of Medical Subject Headings (MeSH) and free-text terms. MeSH terms included Colorectal Neoplasm, Laparoscopy, and Stomach Neoplasm. Free-text terms comprised Colorectal Tumor, Colorectal Cancer, Colorectal Carcinoma, Gastric Neoplasm, Gastric Cancer, Stomach Cancer, Peritoneoscopy, Celioscopy, Laparoscopic Assisted Surgery, Laparoscopic Surgical Procedure, Pneumoperitoneum, abdominal pressure, and intraabdominal pressure. Boolean operators (AND/OR) were utilized to link MeSH terms with free-text terms, ensuring comprehensive coverage of relevant studies.

Inclusion Criteria: i) Patients aged >18 years, diagnosed with gastrointestinal diseases and undergoing laparoscopic surgery. ii) Intervention: Low pneumoperitoneum pressure; control group: standard or high pneumoperitoneum pressure. iii) Complete outcome data reported. Exclusion Criteria: i) Animal studies. ii) Meta-analyses. iii) Single-arm studies. iv) Letters or commentaries. v) Case reports.

The quality assessment was independently conducted by two researchers (FXP/KEL). For the included RCTs, the Cochrane Risk of Bias tool (14) was used to evaluate risks of bias, including selection bias, performance bias, detection bias, attrition bias, reporting bias, and other biases. For non-randomized studies, the Newcastle-Ottawa Scale (NOS) (15) was applied to assess bias, covering the following criteria: representativeness of the exposed cohort, selection of the non-exposed cohort, ascertainment of intervention, demonstration that the outcome of interest was absent at the study’s initiation, comparability of cohorts based on design or analysis, outcome assessment, adequacy of follow-up duration for outcomes to occur, and completeness of cohort follow-up. Any discrepancies between the two researchers were resolved through discussion or adjudication by a third researcher.

Zotero software was utilized for managing retrieved literature. Two researchers independently screened the literature by reviewing titles, abstracts, and full texts. Data from the ultimately included studies were extracted, encompassing the following information: first author, publication year, study type, sample size, patient age, BMI, ASA classification, and additional data including primary and secondary outcomes.

Primary outcomes: Postoperative resting pain; pain in PACU; activity-related pain. Secondary outcomes: Operative time; intraoperative blood loss; surgical field visibility; time to first postoperative flatus; length of hospital stay; anastomotic leakage; postoperative complications; Intra−operative pressure changes. Surgical field visibility refers to the clarity of the operative area under endoscopic visualization and the ease of surgical manipulation. Evaluation Criteria: The Leiden Surgical Rating Scale (L-SRS) was employed for assessment, which evaluates three key parameters: (i) clarity of the surgical field, (ii) instrument maneuverability, and (iii) degree of intraoperative interference. The L-SRS utilizes a 5-point scoring system, where 1 indicates extremely poor visibility, 2 denotes poor, 3 represents acceptable, 4 signifies good, and 5 corresponds to optimal surgical field conditions.

Statistical analyses were performed using RevMan 5.3.1 software (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration) (16). Dichotomous variables were expressed as odds ratios (OR) with 95% confidence intervals (CI), while continuous outcomes were reported as standardized mean differences (SMD) with 95% CI. Results were visualized using forest plots. Heterogeneity across studies was assessed using Higgins I² statistics. I² ≤ 50%: Indicates low heterogeneity, and a fixed-effects model was applied. I² > 50%: Indicates substantial heterogeneity, and a random-effects model was used. If significant heterogeneity was detected, subgroup or sensitivity analyses were conducted to explore potential sources of heterogeneity. The GRADE quality assessment for the primary outcomes and some of the secondary outcomes was conducted using the GRADEpro GDT software (version 3.6).

A total of 425 records were identified through database searches. After removing 65 duplicates, 360 records remained for screening. Following title and abstract screening, 305 records were excluded, resulting in 55 full-text articles assessed for eligibility, upon full-text review, 43 articles were excluded due to irrelevance or unmet criteria. Finally, 12 studies were included in the meta-analysis ( Figure 1 ).

A total of 12 studies (7–10, 17–24) were included, comprising 8 RCTs (8, 17–19, 21–24), 2 retrospective studies (10, 20), 1 prospective cohort study (7), and 1 propensity score-matched analysis (9). The pooled cohort consisted of 1,626 patients, with 771 cases (47.42%) allocated to the LPP group (8–10 mmHg) and 855 cases (52.58%) to the SPP group (12–15 mmHg). While minor protocol variations existed across studies regarding pneumoperitoneum pressure thresholds, the majority adhered to predefined pressure ranges. Three studies specifically evaluated laparoscopic gastric procedures (9, 10, 24), whereas the remaining nine focused on colorectal surgeries. Notably, 75% of the included studies (9 out of 12) were published post-2020, reflecting heightened research interest in optimizing pneumoperitoneum pressures. Geographically, contributions were equally distributed, with 6 studies originating from Asian countries and 6 from European nations ( Table 1 ).

Among the 8 RCTs evaluated, three studies (17, 19, 21) exhibited a high risk of attrition bias due to incomplete outcome reporting. The trial by Schietroma M et al. (24) demonstrated methodological limitations, including insufficient documentation of random sequence generation, allocation concealment, and blinding procedures for investigators, participants, or outcome assessors. All other RCTs maintained adequate methodological quality without additional high-risk biases ( Figures 2 , 3 ). For the four observational studies included, quality assessment scores ranged between 7–8 stars (maximum 9-star scale), indicating moderate-to-high methodological reliability ( Table 2 ).

The evidence grading of the primary and some secondary outcomes indicated that the quality of evidence ranged from moderate to low. The downgrading was primarily due to inconsistencies in study design, small sample sizes, imprecision of results, substantial heterogeneity across studies, and the number of studies included is small ( Supplementary 1 ).

Five studies analyzed postoperative resting pain in patients, while only two studies evaluated pain in PACU and activity-related pain. The meta-analysis results demonstrated LPP laparoscopic surgery significantly reduced postoperative resting pain (SMD = -0.40, 95% CI: -0.68 to -0.12, P = 0.005, I² = 66%, Figure 4A ) and pain in PACU (SMD = -1.06, 95% CI: -1.65 to -0.47, P = 0.0004, I² = 0%, Figure 4B ). However, no significant difference was observed in postoperative activity-related pain between the two groups (SMD = -0.58, 95% CI: -1.44 to 0.28, P = 0.18, I² = 92%, Figure 4C ). A subgroup analysis of postoperative resting pain, stratified by geographic region (European vs. Asian countries), revealed reduced heterogeneity within subgroups. LPP significantly improved postoperative resting pain in European patients (SMD = -0.92, 95% CI: -1.22 to -0.63, P < 0.00001, I² = 45%) but showed no significant benefit for Asian patients (SMD = -0.18, 95% CI: -0.58 to 0.22, P = 0.38, I² = 0%). However, the limited number of studies in each subgroup may compromise the reliability of these findings.

Ten studies analyzed operative time, five studies evaluated intraoperative blood loss, and four studies assessed the surgical field. One study was excluded from the surgical field analysis due to dichotomous data that could not be pooled. Meta-analysis revealed no significant differences between LPP and SPP laparoscopy in operative time (SMD = -0.09, 95% CI: -0.20 to 0.02, P = 0.12, I² = 0%, Figure 5A ), intraoperative blood loss (SMD = -0.00, 95%CI: -0.13 to 0.13, P = 0.95, I² = 0%, Figure 5B ), or surgeon’s visual field (SMD = -0.03, 95% CI: -0.32 to 0.26, P = 0.84, I² = 64%, Figure 5C ). For the surgical field outcomes, significant heterogeneity was observed among studies. Sensitivity analysis excluding the study by Xia L (8) reduced heterogeneity (I² = 0%), but the limited number of remaining studies compromises the reliability of this conclusion.

Time to first postoperative flatus and hospital stay: Four studies analyzed the time to first postoperative flatus, and seven studies evaluated postoperative hospital stay. Meta-analysis demonstrated that patients undergoing LPP laparoscopy for gastrointestinal surgery had a significantly shorter time to first flatus compared to the SPP group (SMD = -0.27, 95% CI: -0.50 to -0.05, P = 0.02, I² = 29%, Figure 6A ). However, no significant difference was observed in postoperative hospital stay between the groups (SMD = 0.03, 95% CI: -0.1 to 0.15, P = 0.69, I² = 0%, Figure 6B ).

Postoperative complications and anastomotic leakage: Eight studies analyzed postoperative complications, and six studies evaluated anastomotic leakage. Meta-analysis revealed no significant differences between LPP and SPP laparoscopy in postoperative complications (OR = 0.78, 95% CI: 0.55 to 1.09, P = 0.14, I² = 7%, Figure 7A ) and anastomotic leakage (OR = 1.17, 95% CI: 0.50 to 2.69, P = 0.72, I² = 23%, Figure 7B ). Further subgroup analysis of postoperative complication grades between the two groups showed that LPP laparoscopy was associated with a reduced incidence of Clavien-Dindo grade II complications (OR = 0.56, 95% CI: 0.34 to 0.93, P = 0.03, I² = 0%, Figure 8 ), while no differences were observed between the two groups in Clavien-Dindo grade III and IV complications.

Intra−operative pressure changes: Seven studies analyzed the number of patients requiring intraoperative pressure changes. Meta-analysis indicated that significantly more patients in the LPP laparoscopy group underwent intraoperative pneumoperitoneum pressure adjustments compared to the SPP group (OR = 4.01, 95% CI: 2.48 to 6.50, P < 0.00001, I² = 46%, Figure 9 ).

Several limitations should be acknowledged. First, the analysis of certain outcomes—including pain in PACU, surgical field visibility, and activity-related pain—was based on a limited number of studies, with substantial heterogeneity observed (I² > 50%), which may compromise the reliability of pooled results. Second, variations in pneumoperitoneum pressure settings between intervention (LPP) and control (SPP) groups across studies introduced potential confounding, leading to instability in meta-analytical estimates. Finally, inconsistent application of nerve-sparing techniques between groups in some included studies may have influenced postoperative pain and functional recovery outcomes.