Tsai H, Chen Y, Ho C, Hseu S, Chao K, Tsai S, Wang P. Maneuvers to Decrease Laparoscopy-Induced Shoulder and Upper Abdominal PainA Randomized Controlled Study. Arch Surg. 2011;146(12):1360-1366. doi:10.1001/archsurg.2011.597
Author Affiliations: Departments of Obstetrics and Gynecology (Drs H.-W. Tsai, Chen, Chao, and Wang) and Anesthesia (Drs Ho and Hseu), Taipei Veterans General Hospital, National Yang-Ming University, and Department of Anesthesia, Chen-Hsin General Hospital (Dr S.-K. Tsai), Taipei, Taiwan.
Objective To evaluate the effectiveness of the pulmonary recruitment maneuver (PRM) and intraperitoneal normal saline infusion (INSI) in removing postlaparoscopic carbon dioxide from the abdominal cavity to decrease laparoscopy-induced abdominal or shoulder pain after surgery.
Design, Setting, and Patients A prospective, randomized, controlled trial was conducted at Taipei Veterans General Hospital, Taipei, Taiwan, from August 1, 2009, through June 30, 2010. One hundred fifty-eight women undergoing laparoscopic surgery for benign gynecologic lesions were randomly assigned to 3 groups: the PRM group (n = 53), the INSI group (n = 54), and the control group (n = 51).
Interventions Postoperative maneuvers included PRM and INSI.
Main Outcome Measures Evaluation of pain, including abdominal pain and shoulder pain, was performed at 12, 24, and 48 hours postoperatively.
Results The frequency of postoperative shoulder pain at 24 and 48 hours was significantly decreased in the INSI group compared with that of either the PRM or control group (40.7% and 24.1% in the INSI group vs 66.0% and 50.9% in the PRM group [P = .009 and .004, respectively] or vs 72.5% and 54.9% in the control group [both P < .001]). Both methods significantly reduced the frequency of upper abdominal pain compared with the control condition (73.6% in the PRM group at 24 hours [P = .03] or 72.2% at 24 hours [P .02] and 44.4% at 48 hours [P = .01] in the INSI group vs 90.2% at 24 hours and 68.6% at 48 hours in the control group).
Conclusions Both PRM and INSI could effectively reduce pain after laparoscopic surgery, but INSI might be better for both upper abdominal and shoulder pain.
Trial Registration clinicaltrials.gov Identifier: NCT01135836
The advantages of laparoscopic surgery, including smaller incisions, shorter hospitalizations, less postoperative pain, and earlier return to normal activity, are attractive and promising.1 However, laparoscopy-induced shoulder pain, which rarely occurs in exploratory laparotomy, has gained importance2- 4 because the occurrence varies from 35% to 80% and it may be severe.2,5- 7 Many strategies, including treatment with nonsteroidal anti-inflammatory drugs, have been used to try to reduce laparoscopy-induced shoulder pain; however, no sufficiently reliable methods have been reported yet.8,9
The mechanism of laparoscopy-induced shoulder pain is mainly derived from carbon dioxide retention within the abdomen, subsequently irritating the phrenic nerve and causing referred pain in the C4 dermatome.10- 13 Moreover, carbon dioxide trapped between the liver and the right diaphragm, irritating the diaphragm, also causes upper abdominal pain.14- 16 All make the patient uncomfortable after laparoscopic surgery. Although the incidence and intensity of laparoscopy-induced upper abdominal pain are not clear, the problem is worthy of intervention.
Because carbon dioxide retention is a key factor in laparoscopy-induced upper abdominal pain, removing or washing out residual carbon dioxide might help reduce the occurrence or severity of this pain in both the shoulder and upper abdomen. Recently, 2 methods, the pulmonary recruitment maneuver (PRM) and intraperitoneal normal saline infusion (INSI), have been proposed to reduce shoulder pain.15- 19 Both maneuvers assist the expulsion of abdominal residual carbon dioxide, although they may be mediated through different mechanisms. The pulmonary recruitment maneuver works by mechanically increasing intraperitoneal pressure to facilitate the removal of residual carbon dioxide immediately after operation16,17; INSI is also mediated through increased intraperitoneal pressure but offers a physiologic buffer system to dissolve excess carbon dioxide.18- 20
Although both maneuvers are reported to be effective, no head-to-head comparison of the 2 methods was found in a MEDLINE search (English language; 1966-2010; search terms “laparoscopic” and “shoulder pain”). Therefore, a randomized controlled trial was conducted to compare the 2 methods and to test the efficacy of both; a control group using no maneuver was also included in the study.
The study was a prospective, randomized, controlled clinical trial. Participants were recruited at Taipei Veterans General Hospital from August 1, 2009, through June 30, 2010. Approval for the study was obtained from the local ethics committee, and informed consent was obtained from all patients.
A total of 180 patients were screened. The inclusion criteria were being female, being aged 24 to 65 years, and having an American Society of Anesthesiologists physical status classification of 1 or 2. Randomization (n = 177) was done via a computerized, balanced (1:1:1) method. Random numbers were computer generated. The randomization code was inserted into numbered, sealed, opaque envelopes. A single envelope was opened by the surgeon when the patient was hospitalized. Study coordinators, patients, gynecologists, anesthesiologists, and members of the panel were not masked to intervention after assignment. Finally, data for 158 patients were analyzed.
The subjects were anesthetized by 1 of 2 anesthesiologists (C.M.H. and S.S.H.), who followed a specified protocol. The patients' vital signs were monitored according to clinical standards. All procedures were performed by 2 experienced surgeons (H.W.T. and Y.J.C.) during the study period. The postoperative assessment was performed by 2 independent investigators (K.C.C. and P.H.W.).
We used 4 ports to perform all laparoscopic operations. One 12-mm port was inserted through the umbilicus, and the other 5-mm ports were inserted through the lateral lower abdominal wall and suprapubic area.21 Laparoscopy was performed using carbon dioxide gas as the distention medium, which was introduced through a Veress needle placed intraumbilically. The intra-abdominal gas pressure was monitored, as was the total volume of gas delivered during the procedure. The gas pressure was set at 15 mm Hg during the procedure. The flow of carbon dioxide did not exceed 2 L/min when creating the capnoperitoneum.
At the end of surgery, the routine method of removing carbon dioxide by passive exsufflation through the port site was used in the control group. Gentle abdominal pressure was applied to evacuate as much gas as possible. In the PRM group, in addition to the routine method, the patients were placed in the Trendelenburg position (30°) and a pulmonary recruitment maneuver consisting of 5 manual pulmonary inflations was performed with a maximal pressure of 60 cm H2O.16 The anesthesiologists held the fifth positive-pressure inflation for 5 seconds. During these maneuvers, the surgeon was instructed to ensure that the trocar sleeve valve was fully open to allow the carbon dioxide to escape the abdominal cavity. The patients were then placed in the level position, the trocar was removed, and the abdominal incisions were closed. In the INSI group, the upper part of the abdominal cavity was evenly and bilaterally filled with isotonic normal saline (25-30 mL/kg of body weight), which was then left in the abdominal cavity.19,20 During this procedure, the surgeon was instructed to ensure that the trocar sleeve valve was fully open to allow the carbon dioxide to escape the abdominal cavity. The patients were then placed in the level position, the trocar was removed, and the abdominal incisions were closed.
Postoperative pain control was provided with meperidine hydrochloride intravenously as needed within 48 hours after operation. Nonsteroidal anti-inflammatory drugs were not used within 48 hours after operation. Postoperative blood pressure, heart rate, and respiratory rate were recorded at 4-hour intervals. Postoperative hospital stay (in days), duration of surgery, estimated blood loss (in milliliters), and cumulative dose of meperidine were recorded. Flatulence-relief time, which indicates resumption of normal bowel function as expressed by the presence of bowel sounds and the passage of flatus, was also recorded.22 According to the guidelines of the national health insurance system in Taiwan, patients who receive laparoscopic surgery can stay in the hospital for 4 to 6 days.
The primary outcome of pain was rated using a 10-cm visual analog scale,23 anchored at one end with no pain at all and at the other with the worst imaginable pain. Women rated the degree of pain by placing a mark on the line, and scale values were obtained by measuring the distance from 0 to that mark. This measurement was validated as a sensitive measure for large-group comparisons.24 The visual analog scale ratings were obtained at 12, 24, and 48 postoperative hours for each of the 3 types of pain: upper abdominal, shoulder, and lower abdominal. Lower abdominal pain was defined as a trocar wound and intra-abdominal pain. All pain assessments were made at rest. The patients were asked about any occurrence of nausea, vomiting, or abdominal distention. Abdominal distention was defined as the symptoms of the patients or postoperative gastrointestinal dysfunction and prolonged ileus based on the clinical judgment of the attending physician.25 The number needed to treat for benefit is the reciprocal of absolute risk reduction, which helps clinicians understand the impact of interventions. Therefore, we also reported outcome data as the number needed to treat for benefit for both interventions.
The 2 primary end points of our study were the incidence and visual analog scale rating of the overall upper abdominal and shoulder pain. Expecting that the incidence of shoulder pain may be reduced from 80% to 50% by using the methods, we determined a required sample size of 45 patients per group for a 2-tailed χ2 test with 80% power and P = .05 based on another study.16 Allowing for a 20% loss to follow-up, the recruitment target was inflated to 54. Considering that existing research into postoperative pain has shown substantial rates of loss to follow-up, recruitment continued until the end of the finding period when 60 women in each group were included.
Statistical analysis was performed with SPSS version 17.0.0 statistical software (SPSS Inc, Chicago, Illinois). Analysis by χ2 test or Fisher exact test was done to evaluate discrete variables. For continuous variables, we used t test, analysis of variance, and Scheffé post hoc test with adjustment for multiple comparisons. Statistical significance was defined as P < .05.
Between August 1, 2009, and June 30, 2010, 180 patients were admitted and participated in our study. Three refused to give consent and were excluded. Therefore, 177 patients were randomly assigned to the 3 groups, with 59 in each group. Of these participants, 13 did not complete the questionnaire, 5 were converted to laparotomy, and 1 cancelled the surgery for personal reasons, ultimately resulting in 53 patients in the PRM group, 54 in the INSI group, and 51 in the control group. The groups did not differ in age, weight, height, operative time, type of surgery, operative blood loss, postoperative hospital stay, and flatulence-relief time (Table 1).
According to the patient description, upper abdominal pain was dull pain and prick pain. The incidence of laparoscopy-induced upper abdominal pain in the PRM group was lower than that in the control group at 12 and 24 hours postoperatively (P = .047 and P = .03, respectively). The laparoscopy-induced upper abdominal pain in the INSI group was significantly lower than that in the control group at 24 and 48 hours postoperatively (P = .02 and P = .01, respectively). The effect of INSI persisted, lasting to 48 hours after surgery. The number needed to treat for benefit to reduce upper abdominal pain incidence at 48 hours was 5 (95% CI, 3-18) for INSI (Table 2).
The mean laparoscopy-induced upper abdominal pain scores were compatible with the incidence of pain in the 3 groups. The postoperative upper abdominal pain scores were significantly lower in the INSI group than in the control group at 12, 24, and 48 hours (P < .001, P < .001, and P = .001, respectively) (Table 2). However, the postoperative upper abdominal pain scores were significantly lower in the PRM group than in the control group at 12 and 24 hours only (P = .002 and P = .007, respectively) (Table 2) and were significantly lower in the INSI group than in the PRM group at 48 hours (P = .01).
Shoulder pain was described as a sharp pain and soreness. At 24 and 48 hours, the incidence of postoperative shoulder pain was significantly lower in the INSI group (40.7% and 24.1%, respectively) than in the control group (72.5% and 54.9%, respectively; P = .001 and P = .001, respectively) and the PRM group (66.0% and 50.9%, respectively; P = .009 and P = .004, respectively) (Table 3).
The postoperative shoulder pain scores were also significantly lower in the INSI group than in the control group at 12 hours (mean [SD], 1.80 [2.41] vs 3.57 [3.59], respectively; P = .02), 24 hours (mean [SD], 1.50 [2.20] vs 4.22 [3.32], respectively; P < .001), and 48 hours (mean [SD], 0.70 [1.55] vs 2.20 [2.42], respectively; P = .004) (Table 3). They were significantly lower in the INSI group than in the PRM group at 48 hours (mean [SD], 0.70 [1.55] vs 1.94 [2.73], respectively; P = .02).
The incidence and intensity of lower abdominal pain (wound pain) did not differ among the 3 groups at 12, 24, and 48 hours (Table 4). Analgesic requirements and the incidence of postoperative nausea and vomiting were also not significantly different (Table 5).
Because the abdominal cavity of the subjects in the INSI group was filled with isotonic normal saline at 25 to 30 mL/kg of body weight, we suspected that the incidence of flatulence-relief time or abdominal distention would be different between the groups. However, there was no difference among the 3 groups (Table 1 and Table 5). There were no cardiovascular or pulmonary complications among the 3 groups.
Upper abdominal pain and shoulder pain after laparoscopic surgery may be transient or may persist for about 3 days26; sometimes, it may cause more discomfort to the patient than the pain at the incision sites. In this study, both the INSI and PRM interventions reduced the incidence and intensity of upper abdominal pain and shoulder pain after laparoscopic surgery. The effect of INSI was continuous and persistent until the intraperitoneal warmed normal saline was absorbed. The INSI maneuver seemed to be much more effective at reducing laparoscopy-induced upper abdominal and shoulder pain than PRM, and the effect lasted longer.
Phelps et al16 reduced positional pain from 63% to 31% using PRM because PRM used positive-pressure ventilations at the end of surgery to inflate the lungs and lower the diaphragm (increase intraperitoneal pressure), which eliminates carbon dioxide from the peritoneal cavity and results in less intra-abdominal acidosis and consequent phrenic nerve or peritoneal irritation. In a recent randomized clinical trial, the frequency and intensity of shoulder pain were significantly decreased in the PRM group at 12, 24, and 48 hours after surgery.17 Although similar positive results were also found in our study, these effects were observed immediately after operation (at 12 hours, P = .002; and at 24 hours, P = .007) but disappeared later. This might be related to the different or complicated operations in our study compared with the majority of diagnostic procedures in the study by Phelps and colleagues. Different surgical types and operation times might interfere with the interpretation of the results.17
The INSI maneuver is also used to reduce laparoscopy-induced pain. Tsimoyiannis et al19 treated patients with 37°C normal saline (25-30 mL/kg of body weight) under the right hemidiaphragm after laparoscopic cholecystectomy to show the significant decrease in frequency and intensity of laparoscopy-induced shoulder pain. In other studies, these patients were treated with low-pressure pneumoperitoneum, and the data also showed that INSI significantly reduced the incidence and intensity of shoulder pain after laparoscopic cholecystectomy at 12 and 24 hours (pain scores: 3.0 for the control group vs 2.5 for the INSI group [P = .01] and 1.29 vs 0.75 [P = .04], respectively).19,27,28 In our study, convention-pressure pneumoperitoneum was used. Consistent with previous studies,19,27,28 a significant reduction in intensity and frequency of laparoscopy-induced pain was seen 24 hours postoperatively (P < .001 and P = .001, respectively). We found that the effectiveness of INSI seemed to be long acting, continuous, and physiological until the warmed normal saline was absorbed, because significantly reduced pain was still noted 48 hours postoperatively (P = .004 and P = .001, respectively).
The amount of normal saline used in INSI was 1000 mL, which has been reported to be safe.19,27,28 Consistent with previous studies,19,27,28 the patients in the INSI group in this study did not have complications. The absence of a statistical difference in flatulence-relief time among the 3 groups in this study suggested that this extra fluid (1000 mL) might not prolong ileus. In fact, studies have attempted to verify the kinetics of intraperitoneal fluid absorption for preventing postoperative adhesions.29,30 It is estimated that intraperitoneal absorption is approximately 30 to 60 mL/h; therefore, there would be fewer or no problems with fluid shifts.31 The INSI maneuver is different from hysteroscopic procedures with the potential risk of fluid overload. Increased intrauterine pressure seems to affect the onset of fluid overload syndrome.32 The relative risk of fluid overload increases significantly when the intrauterine pressure exceeds the mean arterial pressure.33
In our study, the effect of INSI was longer lasting and more persistent than that of PRM. Besides increased intraperitoneal pressure similar to PRM, INSI had an additional buffer system.18 Carbon dioxide readily dissolves in water and forms carbonic acid. Carbonic acid is then absorbed into the intravascular space and contacts red blood cells, which contain carbonic anhydrase that transforms carbonic acid into bicarbonate. This reaction is reversed in the lungs, where bicarbonate is converted back into carbon dioxide to be expelled. This equilibrium plays an important role as a buffer in mammalian blood. As a physiologic buffer system, INSI facilitates the dissipation of carbon dioxide in the abdominal cavity, thus preventing diaphragmatic irritation and lessening postoperative upper abdominal and shoulder pain. The effect of INSI is long lasting, continuous, and physiological until the normal saline is absorbed. In contrast, PRM eliminates the retained carbon dioxide gas mechanically with high pressure but is a relatively short-acting maneuver, which means that the effect might stop when the maneuver stops. That is, its effect is transient and not persistent. Therefore, INSI would be more effective and longer lasting than mechanical PRM to reduce upper abdominal and shoulder pain.
The pain scores peaked at 12 to 24 hours postoperatively in our study, which revealed that the pain was most frequent and severe after early mobilization of the patient.34 This result was similar to that of laparoscopic cholecystectomy and gynecologic laparoscopic surgery.35,36 It suggests that mobilization increases traction on the peritoneal reflections of the heavy viscera, which then lose suction support for their weight owing to the creation of peritoneal spaces by the carbon dioxide.34
The strengths of this study include the prospective and randomized design comparing INSI, PRM, and controls and the fact that surgery on all patients was conducted by the same surgeons using the same technique and instruments.
This study has certain limitations, however, including the relatively healthy women and the fixed national health insurance payment per case for these operations. Furthermore, we have focused only on laparoscopic surgery for benign gynecologic diseases, but other procedures (laparoscopic cholecystectomy, for example) need validation. Different surgical types and the length of surgery might interfere with pain evaluation, although there was no difference in the surgical types among the 3 groups. In addition, the effect of low-pressure pneumoperitoneum was not tested in this study because 2 studies showed the superiority of the use of low-pressure pneumoperitoneum in reducing laparoscopy-induced pain compared with conventional-pressure pneumoperitoneum.27,28 The mean (SD) postoperative pain score was lower by 2.64 (0.86) in the low-pressure pneumoperitoneum group and the incidence of shoulder pain was also lower compared with conventional-pressure pneumoperitoneum (1.6% vs 11.29%, respectively).27 Finally, the analgesic requirements did not differ among the 3 groups in this study, which suggested that these strategies may need further modification to improve clinical efficacy. Moreover, other causes might be overlooked. For example, the retention of some intraperitoneal blood or inflammatory cystic fluid or residual carbon dioxide might contribute to it.
In conclusion, both INSI and PRM could significantly reduce laparoscopy-induced pain, but the INSI method is recommended because it was more effective.
Correspondence: Peng-Hui Wang, MD, PhD, Department of Obstetrics and Gynecology, Taipei Veterans General Hospital, National Yang-Ming University, 201, Section 2, Shih-Pai Road, Taipei 112, Taiwan (email@example.com).
Accepted for Publication: May 26, 2011.
Author Contributions: Drs H.-W. Tsai and Chen contributed equally to this work. Study concept and design: H.-W. Tsai, Chen, and Wang. Acquisition of data: Chao and S.-K. Tsai. Analysis and interpretation of data: H.-W. Tsai, Chen, Ho, Hseu, and Wang. Drafting of the manuscript: H.-W. Tsai, Chen, Ho, Hseu, Chao, S.-K. Tsai, and Wang. Critical revision of the manuscript for important intellectual content: Wang. Statistical analysis: H.-W. Tsai, Chen, and Wang. Obtained funding: Chao and Wang. Administrative, technical, and material support: H.-W. Tsai, Chen, Ho, Hseu, S.-K. Tsai, and Wang. Study supervision: Wang.
Financial Disclosure: None reported.
Funding/Support: This work was supported in part by grants V99-C1-085, V99-C1-166, V99F-014, and V99-A-065 from the Taipei Veterans General Hospital and grants CI-97-6 and CI-99-7 from the Yen-Tjing-Ling Medical Foundation.
Previous Presentation: This paper was presented at the 39th Global Congress of Minimally Invasive Gynecology of the American Association of Gynecologic Laparoscopists; November 9, 2010; Las Vegas, Nevada; and the XXII Asian and Oceanic Congress of Obstetrics and Gynecology; September 26, 2011; Taipei, Taiwan.