Comparison of low-pressure and standard-pressure pneumoperitoneum laparoscopic cholecystectomy in patients with cardiopulmonary comorbidities: a double blinded randomized clinical trial

Abstract

Background

The benefits of low-pressure laparoscopic cholecystectomy (LPLC) in patients with cardiopulmonary comorbidities remain unclear. This study aimed to explore the feasibility and pulmonary effects of LPLC in patients with cardiopulmonary comorbidities.

Methods

This was a multicenter, parallel, double-blind, randomized controlled trial. Eligible patients included patients with cardiac or pulmonary comorbidities, who were randomly assigned (1:1) to undergo LPLC (10 mmHg) or standard-pressure laparoscopic cholecystectomy (SPLC) (14 mmHg). The primary outcome was postoperative partial pressure of carbon dioxide (CO₂). Surgical safety variables, patient recovery, pulmonary function parameters, and surgeon comfort were also compared between groups.

Results

This study enrolled 144 participants, with 124 participants extracted for the final analysis (62 in LPLC and 62 in SPLC group, respectively). The median postoperative PaCO2 was similar in the LPLC (43.3 mmHg) and SPLC (43.0 mmHg) groups (p = 0.988). Pulmonary parameters including postoperative pH, PaCO2, HCO3, and lactate levels were similar between the two groups. Postoperative base excess was significantly higher in the LPLC group (− 0.6 mmol/L [− 6.9 ~ 7.5] vs. −1.9 mmol/L [− 6.6 ~ 5.4]; p = 0.031). There was no between-group difference regarding intraabdominal operative time, rate of intraoperative bile spillage, blood loss, surgeon comfort during surgery, and conversion rate. Moreover, postoperative major complication rates, the median time to the first flatus, postoperative hospital stay, or mean postoperative visual analog scale score for pain were similar in both groups.

Conclusions

This study found no reduction of partial pressure of CO₂ with LPLC compared with SPLC for patients with cardiopulmonary comorbidities. LPLC with a pneumoperitoneum pressure of 10 mmHg may be safe and feasible for these patients when performed by experienced surgeons, although it does not improve pulmonary parameters.

Registration

The trial is retrospectively registered at ClinicalTrials.gov (NCT04670952) on December 17, 2020.

Background

Laparoscopic cholecystectomy (LC) has been considered the standard procedure for symptomatic gallstones or high-risk gallbladder polyps. It is performed by inflating the abdomen with carbon dioxide (CO₂) to create a surgical working space. A pneumoperitoneum of 12–16 mmHg is considered the standard value with proven safety [1]. However, CO₂ inflation causes surgical stress, changes in blood circulation, and pulmonary compliance, and it may even increase intracranial pressures [2,3,4]. Approximately 21.5%, 21.2%, and 30.3% of patients may have decreased forced vital capacity (FVC), forced expiratory volume in one second, and forced expiratory flow at 25–75% of FVC, respectively after LC [5,6,7]. To minimize the effect of pneumoperitoneum, Hohenberger et al. introduced the concept of low-pressure pneumoperitoneum (7–10 mmHg) in 2009 [8].

Previous studies have reported that low-pressure laparoscopic cholecystectomy (LPLC) might decrease postoperative pain [8,9,10]. Recent systematic reviews concluded that the quality of available evidence was sufficient to support the use of LPLC [11, 12]. However, existing studies mainly focused on low-risk population and postoperative pain as the primary outcome. Rare attention was paid on surgical comfort during LPLC. Extremely low pressure may limit the surgical working space, possibly causing adjacent organ impairment. Gin et al. compared 8 mmHg and 12 mmHg pressures for pneumoperitoneum in LC and reported that intraoperative visibility was significantly reduced in LPLC [13]. Thus, it is important to identify an optimal maintenance pressure to provide comfortable surgical working space while minimize the potential side effects caused by pneumoperitoneum. Therefore, high-quality evidence demonstrating the role of LPLC in patients with cardiopulmonary comorbidities is needed [1].

The present randomized controlled trial (RCT), launched on January 1, 2019, aimed to assess the safety and efficacy of LPLC (10 mmHg) when compared with those of standard-pressure laparoscopic cholecystectomy (SPLC, 14 mmHg) in patients with cardiopulmonary comorbidities.

Methods

Trial design

This multicenter, parallel, double-blind, randomized controlled trial was conducted at 3 centers (Peking Union Medical College Hospital, Baoquanling Hospital of Beidahuang Group, and Baotou Central Hospital) in 3 provinces of China. The enrollment period was from June 10, 2019, to October 31, 2021. The same team of surgeons from the 3 hospitals performed the LPLCs and SPLCs. Ethical approval for this study (approval number: ZS-1837) was provided by the Ethics Committee of the Chinese Academy of Medical Sciences Peking Union Medical College Hospital, Beijing China. The trial is registered at ClinicalTrials.gov (NCT04670952). The work has been reported in line with Consolidated Standards of Reporting Trials (CONSORT) Guidelines and included a completed CONSORT checklist as an additional file.

Participants and eligibility criteria

Patients scheduled for elective LC due to gallstones or polypoid lesions in the gallbladder were recruited. The inclusion criteria were as follows: adult patients aged from 18 to 85 years; presence of chronic cardiopulmonary diseases, including hypertension, diabetes mellitus, coronary heart disease, arrhythmias, chronic bronchitis, and emphysema; a history of cardiac, lung, or mediastinal surgery; or a history of asthma. Participants with contraindications to laparoscopic surgery, those with a history of major upper abdominal surgeries, those requiring surgery more extensive than LC, and those who refused to participate were excluded. Written informed consent was obtained from all participants. Detailed criteria for centers and surgeons were listed in the study protocol.

Randomization, concealment, and masking

Eligible participants were enrolled at each center and assigned (1:1) by simple randomization to undergo either LPLC (10 mmHg) or SPLC (14 mmHg). Participants were randomized before general anesthesia induction. Random numbers were pre-generated using SPSS^® version 23.0 (IBM, Armonk, New York, USA), and grouping information was kept in opaque, consecutively numbered envelopes. At the beginning of surgery, a co-investigator not involved in the surgery or data collection and analysis set the pneumoperitoneum pressure as assigned.

The participants and surgical team (including operators, assistants, and care providers in the surgical ward) were blinded to the intervention during the entire perioperative care process.

Interventions and surgical procedures

Prophylactic antibiotics were not administered routinely before surgery.

A standard anesthesia protocol was used for all participants (study protocol). LC in both groups was performed using a three-port technique (study protocol).

Each center followed its usual surgical procedure based on the study protocol. However, every center had to ensure that arterial blood gas (ABG) data were collected when CO₂ inflation was initiated (preintervention ABG) and right after CO₂ was released from the abdomen (postintervention ABG). Intraoperative respiratory physiology (such as respiratory pressures) was not recorded.

Post-operative abdominal drainage was implemented only in cases with intraoperative bile contamination or a high risk of hemorrhage or bile leakage, which was judged by the surgeon based on intraoperative factors. At the end of the surgery, port site infiltration for analgesia was not performed.

Follow-up

The follow-up period was 30 days after surgery, and follow-up was performed by the treating physician in the outpatient clinic or via phone calls. Evaluation of chief postoperative complaints focused on cardiopulmonary events, abdominal physical examinations focused on wound healing, and routine blood and liver function tests when required were a part of the follow-up protocol.

Outcomes and data collection

The primary outcome measure was the postoperative partial pressure of CO₂ based on ABG.

Secondary outcomes included other ABG parameters, surgeon comfort during surgery, postoperative 30-day morbidity and mortality, and postoperative recovery. Definitions of possible complications and perioperative variables were listed in the study protocol. Complications were graded according to the Clavien–Dindo classification [14].

Perioperative pain was reported by the participants, using a visual analog scale (VAS) from 0 (no pain) to 10 (worst pain ever felt). All surgery-related pain sources, including visceral, incisional, and shoulder-related, were assessed. The VAS score was recorded for the worst pain source before surgery, immediately after the patient returned to the ward, and 6, 9, 12, 18, and 24 h thereafter, excluding sleep times. Since the end time of surgery was not fixed and the discharge times also differed, each patient responded at different time points. To address this limitation, the mean postoperative VAS score was calculated.

Surgeon comfort (combined perception of intraoperative visibility and working space) was assessed by interview and scored after surgery using a Likert scale as follows: Level 1, indicates unsatisfied with intraoperative visibility or working space, need to increase pneumoperitoneum pressure; Level 2, nearly excellent, but did not affect operation and there was no need to increase pneumoperitoneum pressure; Level 3, excellent, adequate view and working space. In cases involving poor comfort, the surgeon could request an increase in pneumoperitoneal pressure.

Statistical analyses

A “noninferiority test” was used to estimate the sample size according to the reference end-tidal CO₂ reported in a previous study, where a mean reduction of 14.6% in end-tidal CO₂ was observed for patients undergoing LPLC (30.20 ± 1.75 mmHg) or SPLC (35.37 ± 2.25 mmHg).¹⁰ The type-I and type-II errors were set at 0.05 and 0.20 (statistical power, 0.8), respectively. A δ (equivalence margin) value of -3 mmHg was a conservative estimation of no significant between-group difference. Therefore, with 1:1 matching, there were 50 patients in each group.¹⁵ We estimated that 100 patients would be required to obtain a -3 mmHg improvement in PaCO₂ with LPLC relative to SPLC, with a 5% α error (two-sided test) and 80% power. Assuming a 10% dropout rate, we planned to enroll 110 patients.

All statistical analyses were performed using SPSS^® version 23.0 (IBM). Continuous variables are expressed as mean (SD) or median (range) and were analyzed using the Mann–Whitney U test or Student’s t-test after normality testing. Categorical variables are shown as frequencies and percentages and were analyzed using chi-squared and Fisher’s exact tests. The Newcombe method was used to calculate 95% confidence intervals (CIs) for between-group differences in perioperative variables. Statistical significance was set at p < 0.05.

Patients were defined as dropped cases after randomization if random allocation errors occurred, surgery was not performed, procedures besides LC were performed, conversion to open surgery, or ABG data were not usable.

Results

A total of 144 patients were randomly assigned to the LPLC and SPLC groups (72 per group). After randomization, 20 patients were dropped out for various reasons, as shown in Fig. 1. Finally, 124 participants were included in the analysis (62 in the LPLC group; 62 in the SPLC group). Conversion to open surgery due to severe adhesions was required in 2.8% (2/71) and 1.5% (1/68) of participants in the LPLC and SPLC groups, respectively (p = 0.759). All participants were followed up within 30 days after surgery.

Fig. 1

Consort flowchart. LPLC indicates low pressure laparoscopic cholecystectomy; SPLC, standard pressure laparoscopic cholecystectomy; ABG, arterial blood gas. *Failed reports of arterial blood gas referred to incomplete or missing data because of machine errors or venous blood samples. The 11 patients with failed ABG aged from 56–70 years

Table 1 describes the patients’ demographic characteristics, laboratory test results, and cardiopulmonary comorbidities, which were nearly balanced between the 2 groups. Notably, most participants were elderly, with mean ages of 61.9 and 64.9 years in the LPLC and SPLC groups, respectively. Meanwhile, 46% had hypertension, 21% had diabetes mellitus, 35% had coronary heart disease, 10% had arrhythmias, and 15% had pulmonary emphysema. A previous history of biliary inflammation or pancreatitis was noted in 35% of patients. More proportion of coronary heart disease was observed in the SPLC group (27.4% vs. 12.9% of LPLC group; p = 0.044). Meanwhile, more proportion of patients had histories of biliary inflammation/pancreatitis in the LPLC group (48.4% vs. 22.6% of SPLC group; p = 0.003).

Table 1 Demographic characteristics and cardiopulmonary comorbidities of enrolled participants

As demonstrated in Table 2, most surgeons reported having an excellent working space and surgical view in the LPLC (90.3%) and SPLC (98.4%) groups. Meanwhile, 5 (8.1%) surgeons in the LPLC group and 1 (1.6%) in the SPLC group reported nearly excellent working conditions, whereas 1 surgeon in the LPLC group reported an unsatisfactory working space and required increased pressure (the case is a 56-year-old man, with a BMI of 27.7 kg/m², suffered from acute cholecystitis). The level of surgeon comfort during surgery did not differ significantly between the groups (p = 0.114).

Table 2 Surgeon comfort level during operation

Table 3 displayed perioperative outcomes. The intra-abdominal operative time and estimated blood loss (EBL) showed no difference between the LPLC and SPLC groups. Bile spillage during dissection of the gallbladder bed was observed in 16 (16.1%) and 7 (11.3%) patients in the LPLC and SPLC groups, respectively (p = 0.433). Meanwhile, bile duct injuries did not occur in either group. A similar proportion of patients required a drainage tube in the 2 groups (p = 0.412). The postoperative mortality rate was 0% in both groups. No patients developed major complications such as postoperative bile leakage, intra-abdominal hemorrhage, cardiopulmonary events, or minor complications such as delayed wound healing or subcutaneous emphysema. The highest temperature on postoperative day 1 (p = 0.211), the median time to the first flatus (p = 0.853), and postoperative hospital stay (p = 0.378) showed no significant between-group differences. The difference between the mean baseline (preoperative) and postoperative VAS scores was similar in the LPLC and SPLC groups (p = 0.710).

Table 3 Perioperative outcomes and pathological results

As displayed in Table 4, for the entire cohort, the postintervention pH and PaCO₂ were significantly lower than the preintervention ones (p < 0.001). However, serum bicarbonate (HCO₃) and base excess (BE) levels were similar before and after the intervention.

Table 4 Main arterial blood gas parameters of the whole cohort before and after intervention

Table 5 presents the main ABG parameters before and after pneumoperitoneum in both groups. Preoperatively, pH, PaCO₂, HCO₃, lactate, and BE were balanced between the 2 groups. Postoperatively, the median PaCO₂ was similar in the LPLC (43.3 mmHg) and SPLC (43.0 mmHg) groups (p = 0.988), and the pH, HCO₃, and lactate levels were also similar. However, the BE was significantly higher in the LPLC group than in the SPLC group (-0.6 mmol/L [-6.9–7.5] vs. -1.9 mmol/L [-6.6–5.4]; p = 0.031). Derived ABG parameters were not different between groups (p > 0.05).

Table 5 Comparison of arterial blood gas parameters between the two groups

Discussion

To our knowledge, this is the first RCT to verify the feasibility of LPLC focusing on patients with cardiopulmonary comorbidities. We found that LPLC in this population did not compromise perioperative safety, nor surgeon comfort during surgery and recovery. However, compared with SPLC, LPLC did not improve parameters related to pulmonary homeostasis.

Theoretically, LPLC could reduce the impact of CO₂ pneumoperitoneum on homeostasis to a greater extent than SPLC, with potential benefits in terms of postoperative recovery and cardiopulmonary function. Several studies comparing LPLC and SPLC have demonstrated that LPLC significantly decreases postoperative pain during the early recovery phase [1, 8, 9]. However, some studies have reported inconsistent results [16,17,18]. In a Korean RCT, data for various pneumoperitoneum pressures in LC did not justify the routine use of LPLC [19]. The controversial findings in previous studies mainly because most of them enrolled mixed participants and most of them were low-risk who could well tolerate pneumoperitoneum. Since rare studies have focused on patients with cardiopulmonary comorbidities and poor cardiopulmonary reserves, the role of LPLC for patients with cardiopulmonary comorbidities remain unknown. The present study only included patients with cardiopulmonary comorbidities who underwent LPLC (10 mmHg) or SPLC (14 mmHg).

First of all, to objectively reflect CO₂ pneumoperitoneum-induced fluctuations in pulmonary parameters, we exactly collected ABG data at the start and end of the intervention. After CO₂ insufflation, pH and PaCO₂ significantly deviated (Table 4). However, there were no differences in the main ABG parameters between the LPLC and SPLC groups, and only BE showed a significant difference between the groups (Table 5). There are 2 main reasons for the similarity in ABG parameters. Firstly, a 4-mmHg difference in pneumoperitoneal pressure is insufficient to evoke evident fluctuations in homeostasis. Secondly, intraabdominal pressure might fluctuate to some degree due to suction, gas leak and so on, potentially affected between-group difference of the intraabdominal pressure. Third, the duration of LC is too short to determine the different effects in the 2 groups. Moreover, in cases of decreased pH and increased PaCO₂ after CO₂ insufflation, HCO₃ and BE remained stable; this indicated that participating patients can also withstand fluctuations in homeostasis caused by pneumoperitoneum in short surgeries such as LC. Thus, for patients with cardiopulmonary comorbidities, a 10-mmHg pneumoperitoneal pressure did not improve pulmonary parameters. Nevertheless, the beneficial effects of a 10-mmHg pneumoperitoneal pressure in longer procedures such as pancreatectomy and gastrectomy are worth discussing in the future.

Secondly, LPLC did not reduce surgeon comfort during surgery. With a higher pneumoperitoneal pressure, there is increased stretching of the diaphragm and decreased pulmonary volume and compliance [20]. With lower pneumoperitoneal pressure, the working space becomes limited, and surgeon comfort decreases. Gin et al. compared 8 and 12 mmHg pressures and reported that the surgical view was significantly reduced in LC with a pressure of 8 mmHg, with a higher proportion of surgeries requiring increased intra-abdominal pressure [13]. Neogi et al. reported significantly worse comfort in the 7 mmHg LPLC group than in the 14 mmHg group, with 19.5% (8/41) of patients requiring increased pneumoperitoneal pressure [18]. Thus, an appropriate pneumoperitoneal pressure should strike a balance between safety and homeostasis. In the present study, 10 mmHg was set for LPLC. Only 1 surgeon in the LPLC group required increased pressures, with no between-group differences in surgeon comfort. It should be noted that we estimated surgeon comfort using postoperative interview of the surgeon. There may be advanced software assisting measurements applied to the laparoscopic videos. However, due to the limited budget, we were unable to apply more measurable ways of assessing surgeon comfort at different pneumoperitoneum pressure.

Besides surgical comfort, perioperative safety was not compromised during LPLC. Possible adverse effects of lower pressures include prolonged surgery, intraoperative bleeding, and even a high risk of damage to adjacent structures because of the increased difficulty of the surgery. However, in this study, intraoperative complications (such as bile spillage and bile duct injury) and EBL were similar in both groups. Meanwhile, LPLC did not increase the intra-abdominal operative time relative to that with SPLC. In a recent meta-analysis based on 28 studies and 1729 patients, Ortenzi et al. concluded that the mean operative time was significantly higher in the low-pressure group (most pressures were 7–8 mmHg) than in the standard-pressure group [11]. Furthermore, LPLC did not increase the open surgery conversion rate, and there was no significant between-group difference in the proportion of patients requiring a drainage tube. These results demonstrated that a 10-mmHg pneumoperitoneal pressure did not compromise safety or efficacy and is reasonable. A low postoperative complication rate further validated this point.

Finally, postoperative recovery was another concerned aspect for LPLC. There were no between-group differences in the postoperative temperature, time to the first flatus, and postoperative hospital stay. Moro et al. also found that compared with SPLC (14 mmHg), LPLC (10 mmHg) did not improve the quality of recovery in a cohort of 80 elective LCs [16]. There was also no significant difference in the mean postoperative pain score between the LPLC and SPLC groups, similar to the results of Koc et al. and Perrakis et al. [21, 22]. There are various possible reasons for this finding. Firstly, LC is a short procedure, and the time is insufficient to evoke pain differences. Secondly, approximately one-fourth of the participants had a drainage tube. A high drainage rate might neutralize the effect of CO₂ pressure by reducing phrenic nerve irritation or peritoneal stretching caused by CO₂. Thirdly, our study focused on elderly individuals who have a higher pain threshold than younger individuals.

This study had several limitations. Firstly, all participating surgeons were experienced specialists. This might limit the generalizability of LC under a 10-mmHg pneumoperitoneal pressure. Secondly, we collected and calculated the average postoperative VAS score after considering patient convenience and real-world conditions; however, this may not reflect the VAS score at different time points.

Conclusion

For patients with cardiopulmonary comorbidities, LPLC with a 10-mmHg pneumoperitoneal pressure did not improve pulmonary parameters. Nevertheless, LPLC did not downregulate surgical comfort, nor increase the perioperative complications in experienced hands. For specialists, a 10mmHg pneumoperitoneum pressure was reasonable for LC, and might be considered as an alternative for patients with cardiopulmonary comorbidities. More pieces of evidence are needed to confirm our results. Furthermore, it is worthwhile to explore the role of lower pneumoperitoneum pressure in long-time major abdominal operations.