Pulmonary Recruitment Maneuver Reduces Shoulder Pain and Nausea After Laparoscopic Cholecystectomy: A Randomized Controlled Trial

Background Pain and nausea are common after laparoscopic surgery. This prospective, randomized, controlled trial aimed to investigate postoperative pain and as a secondary endpoint nausea, when performing a ventilator-piloted Pulmonary Recruitment Maneuvre (PRM) at the end of laparoscopic cholecystectomy. Method Patients having elective laparoscopic cholecystectomy were randomized to either ordinary exsufflation or ventilator-piloted PRM, to evacuate intra-abdominal carbon dioxide (CO2) before abdominal closure. A questionnaire with numeric rating scales (NRS) was utilized to evaluate pain and nausea at five occasions during 48 h following surgery. Analgesic and antiemetic treatment was also analyzed. Results 147 patients were analyzed, 76 receiving PRM and 71 controls. Overall pain was well controlled, with no significant difference between the groups regarding incidence (P=0.149) nor intensity (P=0.739). Incidence of shoulder pain was lower in the PRM group during the 48 postoperative hours, 44.7% versus 63.4% (P=0.023). The number needed to treat (NNT) to reduce shoulder pain was 6 (95% Confidence Interval, CI, 2.9–35.5) for the 48-h period. Incidence of nausea was lower in the PRM group during the 48-h period, 51.3% versus 70.4% (P=0.018). NNT was 6 (95% CI 2.9–27.4) for the 48-h period. Nausea intensity was lower in the PRM group during the 48 h (P=0.025). Fewer in the PRM population required antiemetics, 25.0% versus 42.3% (P=0.027). Conclusion A ventilator-piloted PRM at the end of laparoscopic cholecystectomy reduced incidence of shoulder pain, and incidence and intensity of nausea. Clinical trial registration www.clinicaltrials.gov. Identifier: NCT03026543. Supplementary Information The online version contains supplementary material available at 10.1007/s00268-021-06262-6.


Introduction
Cholelithiasis affects up to 20% of the population in developed countries. The surgical treatment of choice worldwide is laparoscopic cholecystectomy [1]. Pain and nausea following the procedure are important reasons for prolonged hospital stay and readmissions [2]. In a previous study, we found that a pulmonary recruitment maneuvre (PRM) reduced pain after bariatric surgery [3], consistent with results from other studies, mainly gynecological [4][5][6][7][8][9][10]. This trial aimed to investigate postoperative pain when performing a ventilator-piloted PRM at the end of laparoscopic cholecystectomy. Nausea was analyzed as a secondary endpoint.
The theory behind the benefit of PRM is that part of the post-laparoscopic pain is due to trapped intra-abdominal CO 2 [11][12][13][14][15]. The entrapped gas may irritate or stretch the diaphragm, cause local acidosis and supposedly initiates referred shoulder pain via its effect on the phrenic nerve. [4,16] PRM lowers the diaphragm, which in turn raises the intra-abdominal pressure, hence mechanically facilitates evacuation of CO 2 [4][5][6].

Methods
This prospective, randomized, controlled trial was conducted in a secondary level hospital, engaging both experienced staff and staff under education, aspiring results of clinical significance. Participants, postoperative personnel and the investigator registering data were blinded to group affiliation. All study procedures were approved by the Regional Ethical Review Board in Linköping, Sweden (2014/120-31). The trial was registered at www.clin icaltrials.gov; registration number: NCT03026543.
Inclusion criteria: adults ([18 years); American Society of Anesthesiologists (ASA) physical status classification I-II (not including body mass index as a health variable); patients scheduled for elective laparoscopic cholecystectomy. Written consent was obtained. Exclusion criteria: acute cholecystitis, pancreatitis or cholangitis; conversion to open surgery; endoscopic retrograde cholangiopancreatography (ERCP) during or after surgery; extra surgical procedure; complications classified as Clavien-Dindo grade (CD)CII [17].
Patients were consecutively enrolled and randomized to one of two equally sized groups; intervention with PRM or control. Block randomization was computer-generated by an independent statistician, and the allocation decision was placed in sealed, opaque, sequentially numbered envelopes. The envelope in turn followed the participant to theater, where it was opened by the anesthetic staff. Preoperatively, participants filled in a health declaration. Six weeks of nonsmoking was a prerequisite for surgery.
Four-port laparoscopic cholecystectomy in anti-Trendelenburg position was performed. Gas pressure was maximum 12 mm Hg. General anesthesia was monitored by an anesthesiologist and nurse. Induction was achieved with either remifentanil and propofol or alfentanil and thiopental, the latter for patients with gastroesophageal reflux. Maintenance anesthesia was with remifentanil as analgesic, and propofol or sevoflurane in oxygen-enriched air as hypnotic. Pressure regulated volume control was standard ventilator mode. The surgeon was either an attending or a resident assisted by a senior colleague. Peroperative cholangiography was performed, according to Swedish routine. If a stone were obstructing the bile duct, an ERCP was performed during surgery or scheduled the following day, and the patient would then be excluded from the study.
In both groups, intra-abdominal CO 2 was evacuated passively through the open sleeve valve of the epigastria port, while the surgeon applied gentle abdominal pressure. In the intervention group, a PRM was performed according to a specific protocol [3] before removal of the port, using the ventilator (GE Datex-Ohmeda Aisys, Madison, WI, United States). During one-minute of pressure-controlled ventilation, the patient received 6 breaths with a total pressure of 40 cm H 2 O. CO 2 is heavier than air [18], wherefore PRM was performed with the patient in supine position to avoid retained CO 2 in the abdominal cavity. After gas exsufflation, the subumbilical fascia was sutured. Incisions were sutured intracutaneously. Local anesthesia (20 ml bupivacaine hydrochloride 5 mg/ml with epinephrine 5 ug/ml) was infiltrated subcutaneously around trocar incisions.
Participants received oral paracetamol preoperatively. 30 mg ketorolac and/or morphine was administered intravenously (IV) near the end of surgery. The morphine dose was documented. Droperidol (0.5 mg IV) and betamethasone (4 mg IV) were given as antiemetic prophylaxis. Those with a history of postoperative nausea and vomiting (PONV) also received 4 mg ondansetron IV. The analgesic regimen postoperatively consisted of 1 g paracetamol four times daily. In the post-anesthesia care unit IV opioids were given as needed, while a 5 mg immediate-release oxycodone capsule was offered as rescue analgesia in the surgical ward. While hospitalized, patients experiencing nausea were treated with ondansetron IV. The doses administrated were recorded for follow-up.
Participants completed a 48-h postoperative questionnaire, with the same questions asked at five different occasions, and the replies given by ticking boxes. Intensity of pain and nausea, respectively, was evaluated with a numeric rating scale (NRS), ranging from 0 (no pain/nausea) to 10 (worst imaginable pain/nausea) [19].
Assuming a clinically relevant difference in mean pain intensity score of 2 points between intervention and control groups, with a standard deviation of 3.5 points (based on results from Tsai et al. [6,7]), the sample size required for 90% power, and a 0.05 was 69 participants per group, using a two-tailed Wilcoxon-Mann-Whitney U test (90% power and a 0.05). Including 83 patients in each group allowed for 20% loss to follow-up.
For statistical analysis, v 2 test was used for comparison of binominal variables and Mann-Whitney U test for continuous variables. Evolution of NRS scores over time was compared using analysis of variance (ANOVA) for repeated measures. Univariate analysis of variance was used to control for possible confounding factors. Results are presented as median (inter quartile range) or numbers with percentages. A P value\0.050 was considered statistically significant. Statistical analyses were carried out with SPSS 26.0 (SPSS Inc., Chicago, IL), except Number Needed to Treat (NNT) which was calculated with QuickCalcs (GraphPad Software, San Diego, CA).

Results
Participants were recruited as of December 2014 until December 2018. Loss to follow-up, due to exclusion or missing questionnaires, was greater than expected. The Regional Ethical Review Board granted our request to randomize another 47 patients to each group (2018/2-32). Consequently, 260 participants were included and randomized, 130 to PRM and 130 to control (Fig. 1).
Of 39 excluded participants, 21 had an ERCP, 14 had an extra surgical intervention (5 closures of mesenteric defects after previous bariatric surgery, 2 umbilical hernia repairs, 3 had extra ports because of intra-abdominal adhesions obstructing the view, and 4 received tube drainage (3 because of subtotal cholecystectomy)). Two had a Baseline characteristics were similar between the groups (Table 1). Clinical data are presented in Table 2. The only parameter with a significant difference between the groups was a somewhat longer duration of surgery in the PRM group. No pulmonary complication was observed.
Incidence of overall pain did not differ significantly between the groups on any occasion. For the postoperative 48-h period, the incidence of overall pain (NRS 1-10) was 98.7% in the PRM group and 94.4% in the control group (P=0.149). Overall pain intensity did not differ over time (P=0.739), nor at a certain hour (Fig. 2).  . The NNT to benefit from reduced nausea incidence was 6 (95% CI 2.9-27.4) during the 48-h period, 5 (95% CI 2.6-16.5) 4 h after surgery, and 6 (95% CI 2.9-28.5) 12 h after surgery. Nausea intensity was low in both groups, though differed significantly between them during the 48 postoperative hours (P=0.025) (Fig. 4) Opioid consumption during hospital stay postoperatively did not differ significantly ( Table 3).
The percentage of patients receiving IV opioids postoperatively was somewhat larger in the intervention group.
Although not part of the study protocol, nonsteroidal antiinflammatory drugs (NSAID) were sometimes administered. In the PRM group 1.3% of the participants received NSAID, versus 9.9% in the control group (P=0.023). After discharge from hospital, there was no difference in analgesic consumption between the groups.
Antiemetics were more frequently needed in the control group postoperatively (P=0.027). The dose required was also higher in the control group (P=0.015). When adjusted for ondansetron given preoperatively, the difference between the groups was reinforced (P=0.003).

Discussion
This trial is the first that evaluated the effect of PRM on postoperative pain solely in patients treated with laparoscopic cholecystectomy, and the second to use the ventilator to perform PRM [3]. We found that a PRM significantly reduced the incidence of postoperative shoulder pain during the postoperative 48-h period. This is consistent with previous studies [4-7, 9, 15]. The NNT, to benefit from reduced shoulder pain incidence was 6, similar to NNT reported by Tsai et al. [7].
Incidence of wound pain differed at two occasions, 4 and 24 h after surgery, in favor of the control group.  [25,26]. Two studies by Tsai et al. and one study by Güngördük did not find any difference between groups regarding neither incidence nor intensity of wound pain [6,7,27]. The varying results, and lack of a connecting theory, leave us in doubt of a correlation. Ryu et al. evaluated the effect of PRM combined with saline instillation on postoperative pain in wounds and shoulders. Their results showed that the relative pain severity was differently perceived postoperatively. Wound pain was considered more intense than shoulder pain in the intervention group, while the control group perceived shoulder pain more intense than wound pain [14]. Possibly, not experiencing shoulder pain shifts the attention to other pain areas.
Postoperative pain was well controlled in both groups, the median NRS score being maximum 2 in the PRM group and 3 in the control group. No significant difference was found between the groups regarding intensity of overall pain, though there was a trend in favor of the PRM group. Two previous studies have analyzed overall pain, both reporting less pain in the PRM group [3,8]. Our study design, with many members of the hospital staff involved, intending to detect differences of generalizable clinical relevance, means minor differences in the care might appear. Potentially, this could obscure the results and explain why no significant difference in overall pain was detected.
Incidence of PONV after laparoscopic cholecystectomy ranges from 53 to 72% [28]. Despite routine PONV prophylaxis, PONV was frequent, affecting 70.4% in the control group, compared to 51.3% in the PRM group. The reduction in the PRM group was significant, and the NNT to benefit from this reduction was 6 patients overall. The two previous studies analyzing PRM in patients undergoing laparoscopic cholecystectomy (among others) did not comment on postoperative nausea [8,15]. Nausea intensity was generally low, though significantly lower in the PRM group. The PRM population was up on their feet earlier, and the requirement of antiemetics was lower, further supporting a lower intensity of nausea in this group.
Duration of surgery in the PRM group was a few minutes longer than in the control group. 10 of 11 prior studies indicate that PRM do not prolong surgery [3,4,6,7,9,13,14,[25][26][27]. A reason surgical duration was slightly longer in our PRM group could be that the operating team had varying experience of the maneuvre. That participants were excluded after randomization, due to for example ERCPs, could have entailed imbalances in baseline characteristics. Though the participants appear similar, we do recognize this as a limitation in the study protocol.
The anesthetic management was not completely standardized. It did not differ significantly between the groups though. Induction analgesia was achieved using remifentanil or alfentanil, both having rapid onset of action and short terminal half-lives [29]. Alfentanils duration of activity is \ 10-24 min after a single dose [29,30]. Remifentanil, also used to maintain anesthesia, has a terminal elimination half-life of 6-12 min, independent of renal and hepatic function [29]. Nausea and vomiting are common adverse effects of the selected analgesic and hypnotic drugs. Minor differences in anesthetic treatment ought therefore not to have biased our results.

Conclusions
Our study indicates that a one-minute, ventilator-piloted PRM reduces incidence of shoulder pain after laparoscopic cholecystectomy. It further suggests that the PRM result in reduced incidence and intensity of postoperative nausea, and decreased need for antiemetics. The maneuvre is safe and uncomplicated to perform.
Acknowledgements The authors express their gratitude to anesthesiologists L. Abildgaard and H. Krook for their support in designing a ventilator-piloted PRM protocol. We also extend our appreciation to statistician Mats Fredriksson for his assistance with randomization and statistical analysis.
Funding The study was supported by grants from Region Ö stergötland; RALF and Hälsofonden.

Declarations
Conflict of interest The authors have no potential conflict of interest to disclose.
Informed consent Informed consent was obtained from all individual participants included in the study.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/.