Recruitment Maneuver to Reduce Postoperative Pulmonary Complications after Laparoscopic Abdominal Surgery: A Systematic Review and Meta-Analysis

Background: Lung-protective ventilation strategies are recommended for patients undergoing mechanical ventilation. However, there are currently no guidelines to follow regarding recruitment maneuvers (RMs). We attempted to identify the effects of RMs on patients undergoing laparoscopic abdominal surgery. Methods: We searched for randomized controlled trials (RCTs) in PubMed, the Cochrane Library databases, Embase, Web of Science and the ClinicalTrials.gov registry for trials published up to December 2021. The primary outcome was postoperative pulmonary complications (PPCs). The secondary outcomes consisted of the static lung compliance, driving pressure (DP), intraoperative oxygenation index (OI), OI in the post-anesthesia care unit (PACU), mean arterial pressure (MAP) and heart rate (HR). Seventeen RCTs with a total of 3480 patients were examined. Results: Patients who received RMs showed a considerable reduction in PPCs (risk ratio (RR) = 0.70; 95% confidence interval (CI): 0.62 to 0.79; p < 0.01), lower DP (weighted mean difference (WMD) = −3.96; 95% CI: −5.97 to −1.95; p < 0.01), elevated static lung compliance (WMD = 10.42; 95% CI: 6.13 to 14.71; p < 0.01) and improved OI (intraoperative: WMD = 53.54; 95% CI: 21.77 to 85.31; p < 0.01; PACU: WMD = 59.40; 95% CI: 39.10 to 79.69; p < 0.01) without substantial changes in MAP (WMD = −0.16; 95% CI −1.35 to 1.03; p > 0.05) and HR (WMD = −1.10; 95% CI: −2.29 to 0.10; p > 0.05). Conclusions: Recruitment maneuvers reduce postoperative pulmonary complications and improve respiratory mechanics and oxygenation in patients undergoing laparoscopic abdominal surgery. More data are needed to elucidate the effect of recruitment maneuver on the circulatory system.


Introduction
Laparoscopic surgery is becoming more and more common due to its advantages of minimal incisions, clear surgical views and reduced postoperative hospital stays [1,2]. However, the pneumoperitoneum and Trendelenburg position cause cephalad displacement of the diaphragm, which reduces pulmonary compliance and functional residual capacity (FRC) and greatly increases the risk of postoperative pulmonary complications (PPCs) [3]. PPCs have been reported to be associated with increased early postoperative mortality, ICU readmission and length of hospital stay [4,5]. Therefore, it is critical to prevent PPCs in the perioperative period.
Pulmonary-protective ventilation strategies, including low tidal volume (TV) ventilation, positive end-expiratory pressure (PEEP) ventilation and the recruitment maneuver (RM), are among the beneficial means for reducing PPCs that many researchers have studied [6]. 2 of 24 There remains controversy and a lack of guidelines to follow regarding the RM. The RM can reverse pulmonary atelectasis to some extent and maintain the alveolar opening by increasing the airway pressure. Depending on the fluctuation of airway pressure, RMs can be divided into the sustained RM and stepwise RM. The stepwise RM comprises a stepwise increase in TV and stepwise increase in PEEP [7]. Previous systematic reviews have reported that the RM in patients undergoing general anesthesia improves oxygenation and reduces PPCs [8]. However, the study did not distinguish between laparoscopic and open surgery, and the number of included publications was limited. Another large multicenter randomized controlled trial (RCT) showed that the open-lung ventilation strategy was not effective in reducing the incidence of PPCs compared to conventional protective ventilation [4]. Therefore, we performed this meta-analysis of RCTs to discuss the effect of the RM on PPCs, the respiratory mechanics and the hemodynamics during laparoscopic abdominal surgery.

Materials and Methods
We report the results of this meta-analysis in compliance with the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines [9]. This study is registered in the International Prospective Register of Systematic Reviews (PROSPERO) with registration number CRD42022315969.

Search Strategy
PubMed, Embase, the Cochrane Library databases, Web of Science and the Clinical-Trials.gov registry were searched, and we included literature published before December 2021. We used Medical Subject Headings (MeSH) terms and multiple combinations related to "Abdomen", "Laparoscopy" and "Hand-Assisted Laparoscopy" for retrieval. With no MeSH terms associated with the RM, we used "recruitment maneuver", "recruitment maneuvers", "RM", "open lung", "protected ventilation" or "protective ventilation" for the search based on previous literature [10,11]. The study type was restricted to RCTs. There were no language restrictions. Finally, the above findings were combined to produce our results. The general search strategy is provided in Table 1.

Selection Criteria
Studies were selected for inclusion based on the following criteria. The screening process was performed independently by SP and WW.

•
The subjects were adult patients subjected to laparoscopic abdominal surgery requiring general anesthesia and mechanical ventilation.

•
The included studies were required to compare RM groups with non-RM groups (or control groups).

•
The included studies had to plainly state the mechanical ventilation strategies, and inclusion and exclusion criteria. Postoperative pulmonary complications had to be reported. • Studies containing patients who were minors or had previous lung disease were excluded.

Data Extraction
Two researchers (YP and JW) independently collected the following information from the original texts: the first author, publication year, ASA grading, age, gender, sample size, body mass index (BMI), surgery type, ventilation settings (the TV, airway pressure, PEEP and RM), hemodynamic parameters (mean arterial pressure (MAP) and heart rate (HR)), respiratory indicators (the incidence of PPC, static lung compliance, driving pressure (DP), intraoperative oxygenation index (OI) and OI in the post-anesthesia care unit (PACU)). We calculated the OI as the arterial partial pressure of oxygen/the inspiratory oxygen fraction (PaO 2 /FiO 2 ). The DP was computed as (airway plateau pressure-PEEP), while the static lung compliance was measured as TV/(airway plateau pressure-PEEP). If the patients were divided into multiple groups in the article, only data from the RM group (followed by PEEP) and conventional ventilation group (without RM) were recorded. Any disputes were adjudicated by SY and HX.
Continuous data and dichotomous data were expressed as the means ± standard deviations (SDs) and numbers, respectively. If continuous data were provided as medians, interquartile ranges or ranges, we transformed them to means and SDs on the basis of the Cochrane Collaboration recommendations [12].

Statistical Analysis
We used Review Manager 5.3 (Cochrane Collaboration, Oxford, UK) and Stata17.0 (StataCorp, College Station, TX, USA) to aggregate the data in accordance with the PRISMA standards [13]. The inverse-variance and Mantel-Haenszel methods were performed separately to assess continuous and dichotomous variables among merged trials. We calculated the weighted mean differences (WMDs) and 95% confidence intervals (CIs) for continuous variables, while for dichotomous variables, we derived the risk ratios (RRs) and 95% CIs.
The heterogeneity was assessed using Cochrane's Q test. p > 0.10 indicated that heterogeneity was not detected, and the fixed-effects model was used to calculate the combined statistics. Additionally, p < 0.10 suggested significant heterogeneity, and the

Grading Evidence Quality
The results of the assessment of the evidence quality using GRADEpro are presented in Table 3. Based on the risk of bias, inconsistency, indirectness, imprecision and publication bias, we classified the evidence quality into four levels: high, moderate, low and very low. In terms of the risk of bias, we ranked the risk for all 18 indicators assessed as not serious. The inconsistency for the static lung compliance, driving pressure, intraoperative OI and OI in the PACU was rated as severe due to I 2 > 50%, which indicates unacceptable heterogeneity. The indirectness and imprecision for all the indicators were classified as not serious because all the studies made direct comparisons between RMs and control groups with adequate sample sizes. No publication bias was found according to Egger's test and Begg's test. Due to the RR being less than 0.5, the quality of evidence for single RMs, sustained RMs, recruited pressure < 40 and comparisons to ZEEP was improved. Finally, we had moderate confidence in the outcomes for the static lung compliance, driving pressure, intraoperative OI and OI in the PACU, while we had high confidence in the rest of the results.

Incidence of PPC
Seventeen studies with a total of 3480 patients reported PPCs whose general incidence was about 21.9% (448/1734 in the non-RM group and 314/1746 in the RM group). RMs significantly reduced PPCs, with low heterogeneity, compared to the control group ( RR = 0.70; 95% CI: 0.62 to 0.79; p < 0.01; p for heterogeneity > 0.10; I 2 = 28%) (Figure 3).

Incidence of PPC
Seventeen studies with a total of 3480 patients reported PPCs whose general incidence was about 21.9% (448/1734 in the non-RM group and 314/1746 in the RM group). RMs significantly reduced PPCs, with low heterogeneity, compared to the control group (RR = 0.70; 95% CI: 0.62 to 0.79; p < 0.01; p for heterogeneity > 0.10; I 2 = 28%) (Figure 3).   Only one study included subjects who were all elderly patients (age ≥ 65 years). Six studies, with a total of 1823 patients, included non-elderly subjects (age < 65 years). The results showed that RMs reduced the incidence of PPCs in non-elderly patients but were not effective in elderly patients (age ≥ 65: RR= 0.82; 95% CI: 0.43 to 1.58; age < 65: RR= 0.77; 95% CI: 0.63 to 0.95; p < 0.05; p for heterogeneity > 0.10; I 2 = 5%). We should be cautious

Subgroup Analysis of PPCs by Age
Only one study included subjects who were all elderly patients (age ≥ 65 years). Six studies, with a total of 1823 patients, included non-elderly subjects (age < 65 years). The results showed that RMs reduced the incidence of PPCs in non-elderly patients but were not effective in elderly patients (age ≥ 65: RR= 0.82; 95% CI: 0.43 to 1.58; age < 65: RR= 0.77; 95% CI: 0.63 to 0.95; p < 0.05; p for heterogeneity > 0.10; I 2 = 5%). We should be cautious regarding the effect of RMs on the elderly due to the insufficient number of studies ( Figure 5).

Subgroup Analysis of PPCs by Age
Only one study included subjects who were all elderly patients (age ≥ 65 years). Six studies, with a total of 1823 patients, included non-elderly subjects (age < 65 years). The results showed that RMs reduced the incidence of PPCs in non-elderly patients but were not effective in elderly patients (age ≥ 65: RR= 0.82; 95% CI: 0.43 to 1.58; age < 65: RR= 0.77; 95% CI: 0.63 to 0.95; p < 0.05; p for heterogeneity > 0.10; I 2 = 5%). We should be cautious regarding the effect of RMs on the elderly due to the insufficient number of studies ( Figure  5).

Subgroup Analysis of PPCs by the Number of RMs
Five studies used single RMs during the procedures. The remaining 12 studies used repeated RMs. The results showed that a single RM significantly reduced the incidence of PPCs with no heterogeneity (RR = 0.36; 95% CI 0.21 to 0.64; p < 0.01; p for heterogeneity > 0.10; I 2 = 0%). The use of repeated RMs also reduced the incidence of PPCs with acceptable heterogeneity (RR = 0.73; 95% CI: 0.64 to 0.83; p < 0.01; p for heterogeneity > 0.10; I 2 = 29%). A single RM may be more efficient than repeated RMs (p for heterogeneity < 0.05; I 2 = 81.6%) ( Figure 6).

Subgroup Analysis of PPCs by the Number of RMs
Five studies used single RMs during the procedures. The remaining 12 studies used repeated RMs. The results showed that a single RM significantly reduced the incidence of PPCs with no heterogeneity (RR = 0.36; 95% CI 0.21 to 0.64; p < 0.01; p for heterogeneity > 0.10; I 2 = 0%). The use of repeated RMs also reduced the incidence of PPCs with acceptable heterogeneity (RR = 0.73; 95% CI: 0.64 to 0.83; p < 0.01; p for heterogeneity > 0.10; I 2 = 29%). A single RM may be more efficient than repeated RMs (p for heterogeneity < 0.05; I 2 = 81.6%) ( Figure 6).

Subgroup Analysis of PPCs by Recruited Pressure
Rothen et al. [33] reported that a recruited pressure greater than 40 cm H2O is required to ensure opening in pulmonary atelectasis. In Figure 8, we divide the included studies into two groups according to the recruited pressure at 40 cm H2O. There are five studies with recruited pressures ≥ 40 cm H2O, while four studies had recruited pressures < 40 cm H2O. The results showed that the incidence of PPCs was reduced when the recruited pressure was less than 40 cm H2O, while a recruited pressure ≥ 40 cm H2O was not beneficial for improving outcomes (recruited pressure ≥ 40 cm H2O: RR = 0.50; 95% CI: 0.24 to 1.04; p > 0.05; p for heterogeneity > 0.10; I 2 = 21%; recruited pressure < 40 cm H2O: RR= 0.41; 95% CI: 0.27 to 0.61; p < 0.01; p for heterogeneity > 0.10; I 2 = 0%). The heterogeneity was reduced to 0% in sensitivity analysis by excluding the study of Nestler et al. [24] from the subgroup with recruited pressures ≥ 40 cm H2O (RR = 0.37; 95% CI: 0.16 to 0.84; p < 0.05; p for heterogeneity > 0.10; I 2 = 0%).

Subgroup Analysis of PPCs by Recruited Pressure
Rothen et al. [33] reported that a recruited pressure greater than 40 cm H 2 O is required to ensure opening in pulmonary atelectasis. In Figure 8,

Subgroup Analysis of PPCs by ZEEP or PEEP Used in Control Group
In the open-lung strategy, the RM is usually used in combination with PEEP. Some control groups of the included studies used PEEP, while the rest used zero end-expiratory pressure (ZEEP). We performed subgroup analysis based on whether PEEP was used in the control group. The results showed that there was a significant difference in the incidence of PPCs, with no heterogeneity, regardless of whether PEEP was used in the control group (compared to ZEEP: RR = 0.48; 95% CI: 0.37 to 0.64; p < 0.01 p for heterogeneity > 0.10; I 2 = 0%; compared to PEEP: RR = 0.78; 95% CI: 0.68 to 0.90; p < 0.01; p for heterogeneity > 0.10; I 2 = 0%). The protective effect in comparison with ZEEP was more pronounced (p for subgroup differences < 0.01) (Figure 9).

Subgroup Analysis of PPCs by ZEEP or PEEP Used in Control Group
In the open-lung strategy, the RM is usually used in combination with PEEP. Some control groups of the included studies used PEEP, while the rest used zero end-expiratory pressure (ZEEP). We performed subgroup analysis based on whether PEEP was used in the control group. The results showed that there was a significant difference in the incidence of PPCs, with no heterogeneity, regardless of whether PEEP was used in the control group (compared to ZEEP: RR = 0.48; 95% CI: 0.37 to 0.64; p < 0.01 p for heterogeneity > 0.10; I 2 = 0%; compared to PEEP: RR = 0.78; 95% CI: 0.68 to 0.90; p < 0.01; p for heterogeneity > 0.10; I 2 = 0%). The protective effect in comparison with ZEEP was more pronounced (p for subgroup differences < 0.01) (Figure 9).

Static Lung Compliance
Seven studies involving a total of 628 patients reported static lung compliance, and the data suggest that the RM is beneficial in enhancing lung compliance but is highly heterogeneous (WMD: 10.42; 95% CI: 6.13 to 14.71; p < 0.01; p for heterogeneity < 0.10; I 2 = 95%) ( Figure 10).

Static Lung Compliance
Seven studies involving a total of 628 patients reported static lung compliance, and the data suggest that the RM is beneficial in enhancing lung compliance but is highly heterogeneous (WMD: 10.42; 95% CI: 6.13 to 14.71; p < 0.01; p for heterogeneity < 0.10; I 2 = 95%) (Figure 10).

Driving Pressure
The driving pressure was reported in seven trials with a total of 2603 individuals, and the findings showed that the RM was useful in reducing the DP but was very heterogeneous (WMD: −3.96; 95% CI: −5.97 to −1.95; p < 0.01; p for heterogeneity < 0.10; I 2 = 96%) ( Figure 11).

Intraoperative Oxygenation Index
The intraoperative OIs were reported for 1285 patients from 11 studies. The global data suggested that the RM could improve the intraoperative OI but with high heterogeneity (WMD: 53.54; 95% CI: 21.77 to 85.31; p < 0.01, p for heterogeneity < 0.10; I 2 = 96%) ( Figure 12). Seven studies examined the postoperative OIs in patients who underwent laparoscopic abdominal surgery. The RM group had higher OIs than the control group (WMD: Figure 11. Forest plot for driving pressure between RM and control groups [17][18][19]21,24,25,27].

Intraoperative Oxygenation Index
The intraoperative OIs were reported for 1285 patients from 11 studies. The global data suggested that the RM could improve the intraoperative OI but with high heterogeneity (WMD: 53.54; 95% CI: 21.77 to 85.31; p < 0.01, p for heterogeneity < 0.10; I 2 = 96%) ( Figure 12).

Intraoperative Oxygenation Index
The intraoperative OIs were reported for 1285 patients from 11 studies. The global data suggested that the RM could improve the intraoperative OI but with high heterogeneity (WMD: 53.54; 95% CI: 21.77 to 85.31; p < 0.01, p for heterogeneity < 0.10; I 2 = 96%) ( Figure 12).

Oxygenation Index in Post-Anesthesia Care Unit
Seven studies examined the postoperative OIs in patients who underwent laparoscopic abdominal surgery. The RM group had higher OIs than the control group (WMD: 59.40; 95% CI: 39.10 to 79.69; p < 0.05; p for heterogeneity < 0.10; I 2 = 96%) ( Figure 13).

Heart Rate
Six studies, with a total of 1692 patients, reported HR. Overall, there was no significant difference in the effect of the RM on HR compared to control (WMD: −1.10; 95% CI: −2.29 to 0.10; p > 0.05; p for heterogeneity > 0.10; I 2 = 0%) ( Figure 15).

Discussion
This meta-analysis included 17 RCTs comparing RMs and conventional mechanical ventilation in patients undergoing laparoscopic abdominal surgery. The types of procedures included robot-assisted laparoscopic radical prostatectomy (RARP), laparoscopic colorectal cancer resection, laparoscopic gastric cancer radical surgery, laparoscopic total hysterectomy and laparoscopic bariatric surgery. Patients undergoing laparoscopic abdominal surgery are at high risk for PPCs. The RM is an effective method for improving pulmonary atelectasis. However, there are few systematic reviews or meta-analyses re- Figure 14. Forest plot for mean arterial pressure between RM and control groups [17,19,24,26,29,30,32].

Heart Rate
Six studies, with a total of 1692 patients, reported HR. Overall, there was no significant difference in the effect of the RM on HR compared to control (WMD: −1.10; 95% CI: −2.29 to 0.10; p > 0.05; p for heterogeneity > 0.10; I 2 = 0%) ( Figure 15).

Discussion
This meta-analysis included 17 RCTs comparing RMs and conventional mechanical ventilation in patients undergoing laparoscopic abdominal surgery. The types of procedures included robot-assisted laparoscopic radical prostatectomy (RARP), laparoscopic colorectal cancer resection, laparoscopic gastric cancer radical surgery, laparoscopic total hysterectomy and laparoscopic bariatric surgery. Patients undergoing laparoscopic abdominal surgery are at high risk for PPCs. The RM is an effective method for improving pulmonary atelectasis. However, there are few systematic reviews or meta-analyses reporting the effects of RMs on patients undergoing laparoscopic abdominal surgery. There- Figure 15. Forest plot for subgroup analysis of heart rate between RM and control groups [17,24,26,29,30,32].

Discussion
This meta-analysis included 17 RCTs comparing RMs and conventional mechanical ventilation in patients undergoing laparoscopic abdominal surgery. The types of procedures included robot-assisted laparoscopic radical prostatectomy (RARP), laparoscopic colorectal cancer resection, laparoscopic gastric cancer radical surgery, laparoscopic total hysterectomy and laparoscopic bariatric surgery. Patients undergoing laparoscopic abdominal surgery are at high risk for PPCs. The RM is an effective method for improving pulmonary atelectasis. However, there are few systematic reviews or meta-analyses reporting the effects of RMs on patients undergoing laparoscopic abdominal surgery. Therefore, a comprehensive analysis of previous RCTs was necessary. Our results showed that, for patients undergoing laparoscopic abdominal surgery, RMs reduced the incidence of PPCs and the driving pressure and improved the oxygenation and static lung compliance compared with controls, without significant differences in the MAP and HR. The heterogeneity was higher for the static lung compliance, DP, intraoperative OI and OI in the PACU, while less heterogeneity was found for PPCs, the MAP and the HR. Heterogeneity may arise from several sources. First, the enrolled patients had a wide age range and underwent different laparoscopic abdominal procedures. Second, the intraoperative ventilation strategy is highly variable. The tidal volume, RM and PEEP can affect the oxygenation and respiratory mechanics. Third, the DP and OI are directly provided in some articles. For studies where the data are not available, we calculated them using equations.
Obese patients are likely to undergo laparoscopic bariatric surgery; they usually have reduced functional residual capacity (FRC), impaired oxygen reserves and comorbidities [34,35]. Pulmonary atelectasis, which plays an important role in PPCs [36], is further aggravated under the influence of general anesthesia, pneumoperitoneum and the Trendelenburg position. The role of RMs in obese patients is still worth discussing. Several studies have demonstrated that RMs can ameliorate PPCs. Reinius et al. [37] concluded that RMs alone were not sufficient to maintain improved respiratory function. We performed subgroup analysis based on BMI and found that RMs reduced PPCs in both obese and non-obese patients, with no significant difference between the two subgroups. This was contrary to the finding of Cui et al. [38], whose meta-analysis indicated that RMs did not improve PPCs in obese patients. However, the fact that there were only two studies with a total of 70 patients in the obese group and high heterogeneity lent low credibility to their findings.
The majority of patients undergoing laparoscopic radical prostatectomy and tumor resection are elderly. With increasing age, elderly patients have compromised respiratory compliance, increased closing volumes and impaired airway protective reflexes. These changes make them more prone to abnormal gas exchange and pulmonary atelectasis. Our subgroup analysis based on age showed that RMs reduced the incidence of PPCs in non-elderly patients but were not effective in elderly patients. However, there was only one study in the elderly group, containing 62 patients, which made the results less reliable. The meta-analysis by Cui et al. [38] showed that RMs reduced the incidence of PPCs in elderly patients undergoing general anesthesia. However, Cui et al. classified patients as elderly or non-elderly based on age 60, whereas our study used 65 as the cut-off. In addition to patient characteristics, the RM itself is worthy of further discussion. Some experiments used single RM [18,23,26,29,30], while others employed repeated RMs [16,17,[19][20][21][22]24,25,27,28]. The results showed that both methods reduced PPCs. Unexpectedly, single RM had even lower risk ratios and a statistically significant difference compared to the other subgroup, which indicated a more pronounced effect. Although the RM is considered to be an effective means of reducing pulmonary atelectasis and preventing PPCs, repeated RMs are accompanied by an increased risk of lung hyperinflation and hemodynamic instability in normal lungs. The single RM in the included studies was administered after intubation or pneumoperitoneum, a phase with a higher incidence of pulmonary atelectasis and greater risk of the development of hemodynamic instability due to medications, positive pressure ventilation and pneumoperitoneum. There is no high-quality evidence to recommend routine RMs after tracheal intubation for patients undergoing general anesthesia, and anesthesiologists need to assess the patient's risk-benefit ratio to tailor treatment. Continuous hemodynamic and SpO2 monitoring is necessary during RMs.
RMs are usually classified as sustained RMs and stepwise RMs. A sustained RM involves setting the airway pressure at a high value and ventilating continuously for a period of time. This is commonly achieved by adjusting the airway-pressure-limiting valve on the ventilator and squeezing the air reservoir. The sustained RM is easy to perform and is widely used in clinical settings. However, when switching back to machinecontrolled mode, there is a risk that the alveoli will re-collapse. A stepwise RM gradually boosts the airway pressure by stepping up the tidal volume or PEEP. The stepwise RM is ventilator driven and can be followed by PEEP titration. The operation is complicated and time consuming. As shown in Figure 7, subgroup analysis showed that the risk ratio was lower in the sustained RM group, and the difference was considered statistically significant compared to the stepwise RM group. No heterogeneity was found in either subgroup. This is consistent with the findings of Cui et al. The incidence of PPC in patients receiving stepwise RM was 22.6% while that in patients receiving sustained RM was 7.1%. Significantly, the included studies only compared the RM and control groups. No direct comparison of different RMs was performed. From the available data, we could not identify which RM was more effective. A study by Rothen et al. [33] based on CT imaging suggested that a recruited pressure of 40 cm H 2 O was efficient in reversing pulmonary atelectasis. We performed further subgroup analysis accordingly. Our results showed that the subgroup with a recruited pressure greater than 40 cm H 2 O showed a poor reduction in the incidence of PPCs, while that with a recruited pressure less than 40 cm H 2 O was good. Sensitivity analysis showed that the heterogeneity originated from the study of Nestler [24]. It was the only study in which the number of PPC cases was greater in the RM group than the control group. This may have been due to errors caused by the small samples, as only 25 patients per group were analyzed. After excluding this study, the results showed that a recruited pressure greater than 40 also reduced PPCs compared to the control group.
The PEEP should be manipulated following RMs to keep the alveoli open. Karsten et al. [39] demonstrated that the combination of the RM and PEEP guaranteed homogeneity in the local ventilation during laparoscopic surgery and enhanced the oxygenation and lung compliance based on electrical impedance tomography (EIT). In an observational study of 10,978 patients, Myrthe et al. [40] noted that mechanical ventilation combined with PEEP at 5-10 cm H 2 O was associated with fewer postoperative respiratory complications and shorter hospital stays in major abdominal surgery. Our subgroup analysis revealed that the incidence of PPCs was lower in the RM group, regardless of whether PEEP was performed in the control group. The protective effect of the RM coupled with PEEP was more apparent than that with ZEEP. This suggested that neither PEEP nor the RM alone was fully effective and that their combination was necessary to maximize the benefits.
We analyzed two indices of pulmonary function: the static lung compliance and driving pressure. The results showed that RMs improved static lung compliance while decreasing the DP. This may be the mechanism by which RMs reduce PPCs. It has been shown that, among patients undergoing mechanical ventilation during general anesthesia, an increased DP is associated with more PPCs, and a lower DP may be lung protective [40]. Christopher et al. [6] also noted that the pulmonary compliance and driving pressures should be examined after RMs to assess the effects. We evaluated the oxygenation indices of patients during surgery and in the PACU. The fact that the RM improves intraoperative OIs has been confirmed by most studies. The RM needs to be maintained with a ventilator. There is a risk of the alveoli re-collapsing after detachment from the circuit. However, our results suggested that the RM was equally beneficial for improving oxygenation in the PACU.
The hemodynamic stability during the RM is noteworthy. We evaluated two hemodynamic parameters: the MAP and HR. There was no significant difference between the control and RM groups. However, we cannot assume, on this basis, that the RM does not have any impact on the circulatory system. The point at which the data were recorded varied widely between studies, with some being recorded 60 min after pneumoperitoneum and others before the end of surgery. The parameters when carrying out the RM were not recorded. In fact, the increased transpulmonary pressure (TP) during RMs causes elevations of the central venous pressure (CVP), pulmonary vascular resistance index (PVRI) and pulmonary artery pressure (PAP), which raises the preload and afterload of the right ventricle, resulting in a transient decrease in the right and left ventricular ejection fraction (R/LVEF) during RMs. Celebi et al. [41] showed that the effect of the RM on the right ventricle was temporary and that the hemodynamics returned to normal with the release of the high airway pressure. Reis et al. [42] also demonstrated that the right ventricular work increases only during the first 2 min after intervention.
There are some limitations of this meta-analysis that need to be taken into account. First, the diagnostic criteria for PPCs varied among studies. Some studies reported the incidence of PPCs at 24 h postoperatively, while others reported PPCs at 5 or even 7 days following surgery. The PPCs were well defined in the high-quality studies but not explicitly stated in others. These factors may affect the accuracy of the conclusions. Second, the measurements for continuous data were conducted at different time points (e.g., 40, 50 or 60 min after pneumoperitoneum). Third, quantitative hemodynamic analysis is inadequate, and the safety of the RM remains to be further clarified. Fourth, the majority of the patients included in the study had normal cardiopulmonary function, so our conclusions may not be applicable to patients with severe cardiac or pulmonary disease.

Conclusions
Our systematic review and meta-analysis have demonstrated that the recruitment maneuver reduces postoperative pulmonary complications and improves respiratory mechanics and oxygenation in patients undergoing laparoscopic abdominal surgery. More data are needed to elucidate the effect of recruitment maneuver on the circulatory system. In general, the use of recruitment maneuver during mechanical ventilation maybe beneficial. Moreover, the long-term outcome parameters for the recruitment maneuver in patients and how to choose the optimal recruitment maneuver according to patient characteristics remain to be further explored.

Data Availability Statement:
No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Conflicts of Interest:
The authors declare no conflict of interest.