Comparison of Pressure-Controlled Ventilation with Volume Guarantee and Volume-Controlled Ventilation mode in Patients under Laparoscopic Surgery: A Meta-Analysis of Randomized Controlled Trials

Background(cid:0)In traditional Volume-Controlled Ventilation (VCV) mode, the creation of pneumoperitoneum during laparoscopic surgery may lead to Ventilator-Associated Lung Injury (VALI). Pressure-Controlled Ventilation with Volume Guarantee (PCV-VG) mode ensures providing adequate oxygen supply to patients while reducing the risk of lung injury. Methods(cid:0)Eligible randomized clinical trials (RCTs) were searched in Medline, Embase, the Cochrane Central Register of Controlled Trials (CENTRAL), Sino Med, China National Knowledge Infrastructure (CNKI) and Wan-Fang MED without language restriction up to March 2019. The primary outcome of this meta-analysis was airway peak pressure (Ppeak) at 30, 60 or 90 mins after complete CO2 insuation. This meta-analysis was followed the recommendations of the PRISMA statement. Results(cid:0)Finally, 9 articles were included. The Ppeak in the PCV-VG group was lower than that in the VCV group, and the difference was statistically signicant at 30mins [Mean Difference (MD)= -3.55, 95% Condence Interval (CI)= -5.13 to -1.98, I2=83%], 60mins [MD= -5.76, 95%CI= -8.15-3.36, I2=93%], 90 mins [MD= -4.59, 95%CI= -5.43-3.74, I2=30%] after complete CO2 insuation. Meanwhile, PCV-VG mode could effectively reduce airway mean pressure (Pmean) and improve dynamic compliance (Cdyn) of patients after complete CO2 insuation in laparoscopic surgery compared to VCV mode. However, no signicant difference was found in PetCO2, HR, MAP, PH, PaO2, and PaCO2 between the two-ventilation modes. Conclusions: PCV-VG mode are superior to VCV mode in providing adequate oxygenation at lower airway peak pressure and greater dynamic compliance in patients under laparoscopic surgery.


Background
In the past two decades, the Enhanced Recovery After Surgery (ERAS) concept and laparoscopic technology have promoted the enormous progress of surgical operations. Due to the rapid development of the technique of laparoscopy, laparoscopic surgery has attracted more attention [1] and gradually become the preferred treatment for some surgical operations with the advantages of small trauma, less postoperative pain and faster recovery, which can signi cantly shorten the hospital stay of inpatients [2].
Studies have shown that laparoscopic cholecystectomy can effectively reduce cardiovascular complications six-fold compared with traditional open surgery [3].
However, during laparoscopic surgery, pneumoperitoneum pressure of about 11-15mmHg would be created to facilitate surgical access. Due to the creation of pneumoperitoneum, a series of pathophysiological changes in cardiorespiratory systems would be generated in patients during the operation, such as the decline in lung compliance [4], increase in airway pressure [5], and reduction in functional residual capacity (FRC). In the severe cases, it could lead to the imbalance of ventilation/perfusion ratio, decreased arterial oxygen levels, lung collapse etc., and promote the occurrence of postoperative pulmonary complications (PPC) [6]. Multiple studies have shown that the application of lung ventilation protection strategy and positive end-expiratory pressure in laparoscopic surgery can reduce the pathophysiological changes caused by pneumoperitoneum [7,8]. However, it should not be neglected that improper setting of the ventilator is likely to cause VALI. Therefore, reasonable ventilation mode to reduce VALI is particularly important during laparoscopic surgery.
VCV mode is a commonly used conventional mode in laparoscopic surgery, which follows a constant ow pattern to deliver a preset tidal volume during the preset inspiratory time for ensuring constant minute ventilation [9,10]. However, this ow pattern generates high airway peak pressure, which can lead to barotrauma and volutrauma to the alveoli [11]; VCV mode would also lead to micro atelectasis as some alveoli with low lung compliance may turn to exhalation status before opening [12].
PCV-VG mode is a new type of intelligent ventilation mode, combined with the advantages of volume control and pressure control ventilation modes [13], delivering tidal volume at a preset pressure and inspiratory time. After setting the target tidal volume, the ventilator would automatically adjust the ow rates based on continuous monitoring of airway pressure and lung compliance [14], aimed at attaining the tidal volume at lower peak inspiratory pressures. Several studies have shown that [9,12,15,16], compared with VCV ventilation mode, PCV-VG ventilation mode has a smaller impact on respiratory mechanics.
Although some RCTs have indicated that PCV-VG pattern could reduce the impact on the respiratory mechanics of patients undergoing laparoscopic surgery and even have certain lung protection effects.
However, due to the small sample size and single-center of current clinical study, no comprehensive conclusions so far have been formed to guide clinical practice. According to some studies, we hypothesized that PCV-VG pattern would reduce airway pressure and increase lung compliance compared with VCV mode, which is more suitable for patients undergoing laparoscopic surgery. Therefore, we conducted this meta-analysis to explore the effect of the PCV-VG mode on patients undergoing laparoscopic surgery compared with the VCV mode.

Literature search
To obtain relevant original studies, we searched the following databases: Medline, Embase, CENTRAL, Sino Med, CNKI and Wan-Fang MED without language restriction up to March 2019. We also searched The Clinical Trials Registry (www.clinicaltrials.com), references of included studies and Google Scholar to identify more trials. The process was carried out by two researchers. The literature retrieval was based on the following three aspects: Participants: participants who underwent laparoscopic surgery; Intervention group: PCV-VG mode; Control group: VCV mode. The complete PubMed retrieval strategy is shown in Appendix 1.

Selection criteria
The study selection process was independently performed by two reviewers. By removing duplicate articles, reading the title and abstract for initial literature screening, and careful reading the full text, the remaining studies were included in our meta-analysis. Some disagreements arising during this period were resolved by consultation with a third reviewer. Studies that met all the following criteria were included in this meta-analysis: (1) RCTs; (2) Patients underwent elective laparoscopic surgery, including transperitoneal, retroperitoneal laparoscopic surgery; (3) Ages ≥18 years old; (4) Intervention measures: PCV-VG mode; (4) the ventilation pattern of Control group: VCV mode. (5) Studies contain at least one of the following measured time points: 30min, 60min or 90min after complete CO2 insu ation; (6) Studies contain at least one of the following measured outcomes: Ppeak, Pmean, Cdyn, PetCO2, HR, MAP, PH, PaO2, PaCO2. Exclusion criteria:(1) Literature review, systematic review, case report, cohort study, etc.; (2) Patients were not intubated; (3) Patients underwent one-lung ventilation; (4) Studies with no target data; (5) Operation duration was less than 1h; (6) BMI ≥ 35kg/m2.

Data collection
Before the start of data collection, one author designed the data extraction form that included article title, author, publication year, country, language, group, number of patients in each group, ventilation mode, ventilation setting, surgery type, laparoscopic type, outcome. For the required data unavailable, we would contact with the author for relevant information. The process was carried out independently by two researchers. The controversial sections were discussed by two researchers with a third researcher to reach a consensus. The primary outcome of this meta-analysis was Ppeak at 30, 60 or 90mins after complete CO2 insu ation; The secondary outcomes were Pmean, Cdyn, HR, MAP, PH, PaO2, PaCO2 at the above measure time points.
Risk of bias assessment 11 original articles were selected according to the search strategy. The Review Manager 5.3 software was applied and the quality of the included RCTs was assessed with reference to the Cochrane Reviewer Handbook. Evaluation indicators include: (1) Whether the random sequence is generated correctly; (2) Whether the allocation hidden scheme is performed (3) Whether patients and researchers were blinded (4) Whether the assessor was blind (5) Whether the integrity of the outcome data was described (6) Whether there is selective reporting (7) Other biases. Through the above seven aspects, the quality of included studies was assessed and summarized as low, medium or high risk.

Statistical methods
In this study, we used Review Manager 5.3 for the corresponding statistical analysis. For binary outcomes, the Relative Risk (RR) with 95% CI was computed by an inverse variance method. In continuous outcomes, the Mean Difference (MD) with 95% CI was computed using the method of Mantel-Haenzsel. Heterogeneity was calculated by X2 and I2 tests [17]. X2 <0.10 or a quantitative assessment of the I2 test >50% with a P value < 0.01 were regarded as obviously heterogeneous, and the results required cautious interpretation. If the heterogeneity is more obvious, we would conduct a sensitivity analysis and analyze the remaining studies by excluding one study at a time to determine whether the results are robust. We conducted subgroup analysis according to the type of laparoscopy, namely whether it is retroperitoneal laparoscopy because it has been reported in studies [19][20][21] that retroperitoneal laparoscopic surgery has a greater impact on respiratory mechanics. Funnel plot analysis was performed to determine whether publication bias existed.

Study selection
In this meta-analysis, we conducted study selection according to the PRISMA statement and a total of 31 articles were retrieved (Fig.1). After removing duplicate and irrelevant studies, 2 studies were excluded and 9 studies were included in nal quantitative analysis.
Study characteristics Table.1 shows the basic information of the included studies.

Risk of bias
The bias results are shown in Fig.2. The quality of the included studies was evaluated by the Review Manager 5.3 software. The risk of bias in the included RCTs was mainly caused by the absence of random allocation of concealment schemes in multiple studies. Due to the obvious ventilator ventilation mode, the blind method could not be set and the risk of bias could be implemented.
The Secondary Outcomes

Publication bias
We analyzed the included RCTs based on the Ppeak before the initiation of pneumoperitoneum to determine whether there was a publication bias. The bias map showed that the asymmetry between the two was due to a small sample size.

Sensitivity analysis
We also did a sensitivity analysis and the synthesized outcome remained unchanged after excluding one study each time. The sensitivity analysis results also demonstrated that the outcome result was robust.

Discussion
In the general anesthesia with mechanical ventilation, a better ventilation mode should provide patients with enough oxygen while minimizing the risk of lung injury, without causing the drastic changes in hemodynamics [12,31]. Our meta-analysis showed that under the condition of pneumoperitoneum, the principal observation was that the Ppeak was signi cantly lower in the PCV-VG modes compared to the VCV mode; Meanwhile, PCV-VG mode could effectively reduce Pmean and improve Cdyn of patients after the formation of pneumoperitoneum in laparoscopic surgery compared with VCV mode. However, no signi cant difference was found between the PetCO2, HR, MAP, PH, PaO2, and PaCO2 compared with the VCV mode.
Since the conception of ERAS has been put forward in 1990, the effectiveness of a series of interventions in favor of ERAS across the perioperative period has been recognized in clinical practice. Recently, the indepth development of minimally invasive surgical technology has further contributed to the implementation and promotion of ERAS. Laparoscopic surgery has become the preferred surgical method in some abdominal surgeries and urological surgeries because of its small trauma, light postoperative pain, enhanced postoperative recovery, which signi cantly short inpatients' hospital stay and lowered the incidence of postoperative complication and mortality. However, laparoscopic surgeries require a pneumoperitoneum pressure of around 11-15 mmHg, which can cause a series of pathophysiological changes in the respiratory and circulatory systems. For example, the compliance of the lungs and chest wall decreased by about 25%, and the functional residual capacity decreased, eventually leading to imbalance of lung ventilation/perfusion ow, arterial oxygenation, lung collapse, etc., which may lead to postoperative pulmonary complications [4][5][6].
While ensuring normal ventilation during general anesthesia, mechanical ventilation is also an invasive and non-physiological procedure. The ventilator can perform mechanical ventilation for patients under general anesthesia, maintain the oxygen supply to the patient's tissues during anesthesia, and control the patient's breathing movement by adjusting various parameters. However, the improper setting is likely to cause ventilator-related lung injury, including volume injury, barometric injury, atelectasis injury, and biological injury. Therefore, to reduce ventilator-related lung injury, reasonable adjustments of parameters will be particularly important in mechanical ventilation. PCV-VG mode is a new type of intelligent ventilation mode, combined with the advantages of both volume control and pressure control modes, adjusting automatically according to respiratory dynamic parameters. Decelerating ow is adopted to reduce airway pressure while ensuring the preset tidal volume. By continuously monitoring airway pressure and lung compliance, airway pressure can be automatically adjusted to increase lung compliance [12,32].
Ppeak is the maximum pressure during the ventilator's active in ation to the lungs. It needs to overcome the thoracic elastic resistance, lung elastic resistance and airway resistance. The normal range should not exceed 30-35cmH2O. Maeda [33] found that the higher the peak pressure of the airway, the greater the risk of lung injury. Eisner et al. [34] found that Ppeak height is a major risk factor for early barotrauma in patients with lung injury. In our study, we found that the PCV-VG mode signi cantly reduced the Ppeak after pneumoperitoneum formation compared to the VCV mode. Boules [35] studied the effects of both PCV-VG and VCV ventilation modes on respiratory mechanics and hemodynamic parameters in singlelung ventilation (OLV), which results were like our study that the Ppeak of PCV-VG was signi cantly lower than that of VCV after pneumoperitoneum formation (P < 0.05). Dion [36] observed 20 morbidly obese patients who underwent laparoscopic surgery with VCV, PCV, or PCV-VG mechanical ventilation and found that compared with VCV, the Ppeak values in PCV and PCV-VG modes were lower, which was the same as our research results. At the meantime, we obtained similar results in the subgroup analysis of airway peak pressure according to whether it's laparoscopic or retroperitoneal laparoscopic surgery, and found that 60 minutes after the formation of pneumoperitoneum, the PCV-VG mode group had a lower air peak pressure. However, there was no statistical signi cance between the two groups in the subgroup analysis 30 min after the formation of pneumoperitoneum, which may be caused by the insu cient sample size.
Pmean is related to the volume at the end of inspiration. The higher the pressure is, the larger the lung volume at the end of inspiration will be. Excessive expansion of alveoli can damage the alveolar membrane and the basement membrane of pulmonary blood vessels, leading to volume-related damage. In this study, the average airway pressure of laparoscopic patients was also lower under PCV-VG mode.
Lung compliance means the changes in lung volume caused by the change of unit pressure, and it expresses the effect of chest pressure on lung volume, which is equivalent to the elasticity of lung tissue. Cdyn refers to lung compliance measured during the respiratory cycle when air ow is not blocked, and which is negatively correlated with Ppeak. Studies [12,37] indicated that Cdyn decreased by 25%-40% in patients with mechanical ventilation after pneumoperitoneum. In addition, lateral decubitus reduces lung compliance, increase airway resistance, limits diastolic contraction of the lung, and leading to the imbalance of ventilation and blood supply, etc., all of which could result in partial alveolar collapse and failure to participate in ventilation, limiting the removal of CO2 and posing a threat to the perioperative safety of patients. Retroperitoneal laparoscopic surgery, as an extension of the abdominal cavity mirror operation, is widely used in urology presently. The so-called posterior peritoneal cavity is not a truly anatomical cavity. In the retroperitoneal laparoscopic surgery, a certain pressure of CO2 gas is injected into the posterior peritoneal cavity based on the patient's lateral folding position, and a potential cavity is opened for surgical operation [38]. Three special points of retroperitoneal laparoscopic surgery: lateral position, folding position, and CO2 pneumoperitoneum. The combination of the three points is more likely to cause the mismatch between ventilator parameters and the patient's physiological parameters, leading to ventilator-related lung injury and hypercapnia, affecting the human respiratory and circulatory systems.
In severe cases, it will increase blood pressure, increase heart rate and pulmonary vascular resistance, leading to decreased pulmonary compliance and intrapulmonary shunt [24,27]. Therefore, "individualized ventilation" was adopted according to patients' vital signs and respiratory mechanics in post-laparoscopic surgery, aiming to reduce the occurrence degree of ventilator-related lung injury. This study found that in the retroperitoneal laparoscopic surgery, the Ppeak of the PCV-VG group was lower than that of the VCV group.
In addition, in our study, it was found that after the formation of pneumoperitoneum, PCV-VG mode could not signi cantly reduce PetCO2 and PaCO2 compared with VCV mode, which was different from our previous hypothesis. In previous studies [12], studies indicated that PetCO2 and PaCO2 in the PCV-VG group decreased signi cantly during and after pneumoperitoneum. This may be due to inadequate sample size. At the same time, there was no statistically signi cant difference between the two ventilation modes of PH and PaO2, which was consistent with the studies of El-Ramely and Rachel [32,39]. The above results indicated that PCV-VG mode had no obvious advantages in improving oxygenation and ventilation compared with VCV mode.
In this meta-analysis, we included patients who were non-morbidly obese patients without preoperative lung injury. According to previous studies [12], experiments based on the healthy population can more accurately re ect the in uence of laparoscopic pneumoperitoneum on patients. We did not include patients with restrictive and obstructive pulmonary disease, so the direct impact of ventilation patterns could be better assessed. However, there was also has some limitations in our article. Firstly, due to the limited data of the trials, we did not analyze the respiratory mechanics, hemodynamic blood gas analysis and postoperative complications of patients after pneumoperitoneum; Secondly, we only included adult patients with double-lung ventilation, and the included patients did not include pediatric patients and patients with single-lung ventilation; Finally, we included patients who were non-morbidly obese and had no preoperative lung injury. The results of these patients may be different from the results of this metaanalysis. Therefore, more high-quality, multi-center randomized controlled trials are expected to address these issues.

Conclusions
PCV-VG mode are superior to VCV mode in providing adequate oxygenation at lower airway peak pressure and greater dynamic compliance in patients under laparoscopic surgery. Table 1 Due to technical limitations, Table 1 is only available as a download in the supplemental les section.           Meta-analysis of the PH, PaO2 and PaCO2 in patients under PCV-VG and VCV mode during laparoscopic surgery.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.