Abbott, Jason MRANZCOG, MRCOG; Hawe, Jeremy MRCOG; Srivastava, Pankaj MB, BS, MS; Hunter, David MRCOG; Garry, Ray MD, FRCOG
OBJECTIVE: To determine whether a drain placed in the peritoneal cavity during laparoscopy is both a clinical and cost‐effective method of reducing postoperative pain.
METHODS: Two hundred twenty‐five women undergoing diagnostic or minor operative laparoscopic procedures were recruited. Women were assigned to receive either an intraperitoneal gas drain or a dummy drain during surgery. Sample size to detect a two‐point difference in visual analogue score was estimated at 158 subjects, with 79 in each arm. The patients and nursing staff were unaware of the position of the drain. A visual analogue score was used to assess pain preoperatively and at 4, 24, and 48 hours postoperatively. Data on the experience of nausea, frequency of vomiting, and site of pain were collected. The analgesic and antiemetic use was recorded. An economic evaluation of the analgesic use and the material costs for the two groups was performed.
RESULTS: One hundred sixty‐one complete sets of data (72%) were available for analysis. The two groups were well matched for age, parity, previous surgery, body mass index, volume of carbon dioxide used, and operative time. No significant differences were found between the two groups with regard to the overall pain scores preoperatively (8 versus 7) or at 4 (30 versus 34), 24 (40 versus 44), and 48 (26 versus 26) hours postoperatively, after adjusting for multiple point testing. On assessment at different sites, the dummy drain group experienced shoulder pain more frequently at 4 (19 of 79 versus 10 of 82, P = .05) and 48 (16 of 79 versus 7 of 82, P = .03) hours postoperatively compared with the drain group. The placebo group had a 33% greater usage of oral analgesia after discharge, but this was $2.50 cheaper than the use of an intraperitoneal drain. No statistically significant differences were found between the groups with regard to nausea and vomiting postoperatively.
CONCLUSION: An intraperitoneal drain after minor gynecologic laparoscopy decreases the frequency of shoulder pain and reduces postoperative analgesia requirements. However, it is less cost‐effective to reduce pain using an intraperitoneal gas drain than simple oral analgesia after minor gynecologic laparoscopy.
It has been suggested that the retention of gas in the peritoneal cavity plays a role in the production of postoperative pain after laparoscopy. Some authors have reported a reduction in pain after gas removal by drainage in the postoperative period in both gynecologic1 and general2 surgical settings; others have not confirmed this finding.3 A number of studies have compared traditional carbon dioxide laparoscopy with gasless laparoscopy using mechanical devices to elevate the abdominal wall; in none was a significant difference in pain scores demonstrated.4–6 One study compared the volume of residual gas (calculated from radiologic imaging) with the severity of pain that women experienced after gynecologic laparoscopic procedures. The amount of residual gas volume and amount of pain correlated: the greater the volume of residual gas, the greater the amount of shoulder pain.7 It was also suggested that the gas insufflation rate and pressures might also influence postoperative pain.
The objective of the current study was to determine whether placing a drain into the peritoneal cavity after minor gynecologic laparoscopic surgery reduced pain during the postoperative period. Secondary outcome measures included a functional comparison between the groups and an assessment of the site of pain, the degree of nausea and vomiting, the amount of analgesia required in the postoperative period, and the cost‐effectiveness compared with alternate pain‐relieving strategies.
MATERIALS AND METHODS
Approval for the study was obtained from the ethical committee. Two hundred twenty‐five women were enrolled; complete data sets were available for 161. Women undergoing diagnostic laparoscopy, laparoscopic sterilization, laparoscopic hydrotubation, and minor laparoscopic procedures (minor adhesiolysis, excision, or ablation of stage I or II endometriosis, or aspiration of ovarian cysts) were eligible for this trial. Participants consented to the specific laparoscopic procedure and participation in the randomized, double‐masked trial. Demographic data were collected, and the women were asked to complete a preoperative symptom assessment questionnaire.
All procedures were performed under general anesthesia. Anesthesia was induced with propofol, fentanyl, and vecuronium or rocuronium as a muscle relaxant; inhalation anesthesia using a volatile agent was used for maintenance. All intraoperative analgesia was recorded and consisted of a narcotic (fentanyl, morphine, or pethidine), and all patients were given a nonsteroidal anti‐inflammatory suppository (100 mg diclofenac) if it was not otherwise contraindicated.
Laparoscopy was performed using carbon dioxide gas as the distension medium 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. Additional ports were placed as deemed necessary. The use of skin infiltration with local anesthetic was also recorded
Women suitable for inclusion in the trial were randomly assigned to receive either intraperitoneal gas drainage or placebo, using computer‐generated randomization blocks, stratified for procedure. Concealment was achieved by the allocation being placed in opaque envelopes, which were stored sequentially in operating rooms and opened by the nursing staff, when the entry criteria were satisfied. Subjects in the drain group had a single‐bore, non‐suction Yeates drain inserted into either the umbilical port or accessory port sites. When possible, intraperitoneal placement was confirmed by direct visualization. At the end of the procedure, gas was released from the abdomen by opening the gas tap in the umbilical port site or removal of the cannula cap with the patient still in the Trendelenburg position. A safety pin was then inserted through the drain to prevent its falling into the abdominal cavity, and an opaque dressing covered the drain and pin. Wounds were closed in a standard fashion. Women allocated to the placebo arm had the wounds closed in a standard manner and a single‐bore Yeates drain was then coiled over one of the wound sites (umbilical or accessory). A safety pin was placed through the drain and both were covered with an opaque dressing, so that no external difference was discernible between the two groups.
The recovery staff members were unaware of the position of the drain, as were the nursing personnel on the wards who removed the drain at 4 hours postoperatively. Before drain removal, patients were asked to complete a pain and nausea score using a visual analogue scale. All postoperative analgesia and antiemetics used were recorded by the nursing staff. One of the research team confirmed the accuracy of the recording by cross‐referencing the drug charts and data sheets. After the 4‐hour assessment was completed, the drain was removed by the nursing personnel who were instructed to remove it and the dressing in a standardized fashion. This was achieved by grasping the dressing and drain caudally and removing both together in a cephalad direction, to reduce the possibility of the patient becoming aware of its location. Patients were then discharged home after undergoing assessment by their consulting surgeon. Patients were asked to complete a follow‐up questionnaire 24 and 48 hours postoperatively that recorded their pain scores both at rest and during activity (eg, coughing), as well as nausea scores, the frequency of vomiting, and the location of their pain—lower abdomen, whole abdomen, or shoulder tip.
An assessment of function was performed using the EuroQOL thermometer, a validated instrument that measures self‐rated function.8 Patients were also asked to complete a postoperative analgesia chart, including recording “nil” if no drugs were used. The average analgesic intake was based on the dose of drug used per patient. The questionnaire was returned by prepaid envelope provided before discharge.
On the basis of a previous study by Alexander and Hull,1 the sample size for this study was calculated by assuming that a clinically significant difference would be determined if a two‐point difference in visual analogue scores was found. For a type I error of 5% and with 80% power to detect a statistically significant difference, a total sample size of 158, with 79 in each arm, was required. Dichotomous data were analyzed by χ2 test and continuous data by the Mann‐Whitney U test, with significance set at P ≤ .05.
From the 225 patients entered into the study, 161 complete sets of data were available for analysis, 79 in the placebo group and 82 in the drain group. Of those excluded from the study, 14 (6%) were deemed unsuitable because more major surgery than planned was required, 32 (14%) had incomplete data sets, and 5 (2%) were randomized but no drain was placed. The response rate for the returned questionnaires was 82%, with 41 patients (18%) not returning their questionnaire. There may have been more than one reason for patient exclusion. No differences in the demographic data were found for those patients excluded because of incomplete data sets.
Table 1 shows the demographic and operative data available for the two groups. In the drain group, 25% of patients underwent laparoscopic sterilization, 43% diagnostic laparoscopy, 9% laparoscopic tubal dye studies, and 2% a minor laparoscopic procedure. In the placebo group, these figures were 21%, 53%, 4%, and 4% for the respective procedures. No statistically significant difference was found in age, parity, body mass index, or history of previous surgery. Also, no statistically significant difference was found between the two groups in regard to the volume of carbon dioxide used to insufflate the peritoneal cavity, the length of the procedure, or the percentage of patients who had wounds infiltrated with local anesthetic during the procedure.
No statistically significant differences were noted between the groups for pain scores at the preoperative (8 versus 7) assessment or the 4‐hour (30 versus 34) assessment. No statistically significant differences were found in pain scores measured at 24 (40 versus 44) and 48 (26 versus 26) hours. When adjusted for multiple point testing, using the Bonferroni adjustment, no significant difference was found in any of these scores. Nor were there any significant differences in the return to normal function between the two groups as assessed by the EuroQOL thermometer.
Table 2 shows the assessment of pain by site. A significant difference in the number of patients who experienced shoulder pain at 4 hours and 48 hours was observed. Fewer women experienced shoulder pain at these times in the drain group. A trend toward a decreased frequency of shoulder pain at 24 hours occurred in women in the drain group, although this did not reach statistical significance. No difference in the frequency of vomiting occurred between the groups.
We calculated the average number of doses of analgesia (equivalent to 10 mg morphine or 100 mg diclofenac) and antiemetics used by patients at each of the assessment times and compared the two groups. No differences were found in the amount of intraoperative or 4‐hour analgesic or antiemetic use in the two groups. The difference in the amount of analgesia used after discharge from the hospital was significant, with patients in the placebo group using 33% more analgesia than those in the drain group at the 24‐ and 48‐hour assessment times.
Pain after laparoscopy is common, and the ability to decrease it further by simple, inexpensive measures would be advantageous. The site of pain is variable and is most commonly felt in the abdomen, shoulders, or back. Shoulder pain may occur in up to two thirds of patients,9,10 although it is most commonly transient and by 24 hours has decreased, rarely lasting more than 72 hours.11,12 Distension of the abdomen alone causes pain, and mechanical elevation of the abdominal wall is as painful as traditional gaseous laparoscopy.4–6 One study has suggested that a lower intra‐abdominal pressure during laparoscopy was associated with less pain, although 25% of cases were converted to high‐pressure laparoscopy because the vision was inadequate.13 One study reported a reduced frequency of shoulder pain with a low gas insufflation rate.14 The conversion of carbon dioxide to carbonic acid on the moist peritoneal surfaces may cause pain, although trials comparing the use of nitrous oxide with carbon dioxide in an attempt to reduce postoperative pain have shown no difference between the two gases.15,16
One study reported a 50% reduction in pain when a gas drain was used in the postoperative period to remove residual gas, although this did not reach statistical significance.1 Our study did not find a statistically significant difference between a drain and placebo for the overall reduction of pain, the experience of nausea, or the frequency of vomiting. The frequency of shoulder pain was reduced in the group using a drain. This suggests that a relationship may exist between shoulder pain and the amount of residual gas in the peritoneal cavity, as has been previously reported.9 Shoulder pain is reported after gasless laparoscopy and is thought to be related to a combination of stretch on the peritoneum and phrenic areas and retained room air within the peritoneal cavity.4–6,15,16 The amount of simple analgesia consumed after discharge from the hospital was, on average, 33% greater in the placebo group than in the drain group. The experience of shoulder pain, although not changing the overall pain score, did have an impact on analgesic consumption. Once discharged from the hospital, women had more control over their own analgesia intake. In this masked study, therefore, the consumption of increased analgesics in the placebo group indicates a greater experience of pain.
Since the use of a gas drain or simple analgesics achieved the same pain scores and had no impact on the other outcome measures, the question remains as to which intervention should be used. A drain costs $2.50 and an additional paracetamol or nonsteroidal tablet costs less than a penny. Although no complications were associated with the placement of a gas drain in this study, such simple problems as the loss of a small bore drain back into the abdomen and its necessary surgical retrieval carries with it significant potential morbidity. The additional use of a single analgesic tablet carries with it minimal morbidity.
Previous studies have left a drain in situ for a longer time,1 with the reporting of reduced pain, although this is not possible with day case surgery. We chose to leave the drain in for 4 hours because this was the maximum length of observation for women undergoing the procedures in our study. This also reflects usual length of stay after simple gynecologic surgery. We used a single channel of a Yeates drain because it is universally available, has a diameter of only 3 mm, is easily passed down a laparoscopy cannula, and is inexpensive, with each unit costing less than $3. Some studies have used closed suction systems, which may be a more effective way of removing gas from the abdominal cavity, although these systems cost $28 each and are not a cost‐effective option for day case gynecologic surgery.
In conclusion, the use of an intraperitoneal gas drain reduces the frequency of shoulder pain and postoperative analgesia requirements. However, the use of simple analgesics is a more cost‐effective treatment of pain with potentially less morbidity.
1. Alexander JI, Hull MG. Abdominal pain after laparoscopy: The value of a gas drain. Br J Obstet Gynaecol 1987;94:267–9.
2. Jorgensen JO, Gillies RB, Hunt DR, Caplehorn JR, Lumley T. A simple and effective way to reduce postoperative pain after laparoscopic cholecystectomy. Aust N Z J Surg 1995;65:466–9.
3. Hawasli A, Brown E. The effects of drains in laparoscopic cholecystectomy. J Laproendosc Surg 1994;4:393–8.
4. Guido RS, Brooks K, McKenzie R, Gruss J, Krohn MA. A randomised, prospective comparison of pain after gasless laparoscopy and traditional laparoscopy. J Am Assoc Gynecol Laparosc 1998;5:149–53.
5. Johnson PL, Sibert KS. Laparoscopy: Gasless vs. CO2 pneumoperitoneum. J Reprod Med 1997;42:255–9.
6. Goldberg JM, Maurer WG. A randomised comparison of gasless laparoscopy and CO2 pneumoperitoneum. Obstet Gynecol 1997;90:416–20.
7. Jackson SA, Laurence AS, Hill JC. Does Post-laparoscopy pain relate to residual carbon dioxide? Anesthesia 1996; 51:485–7.
8. Kind P, Dolan P, Gudex C, Williams A. Variations in population health status: results from a United Kingdom national questionnaire survey. BMJ 1998;316:736–41.
9. Reidel HH, Semm K. Das postpelvikopiische schmerzsyndrom. Geburhilfe Frauenheilkunde 1980;40:635–43.
10. Alexander JI. Pain after laparoscopy. Br J Anaesth 1997; 79:369–78.
11. Dobbs FF, Kumar V, Alexander JI, Hull MGR. Pain after laparoscopy related to posture and ring versus clip sterilisation. Br J Obstet Gynaecol 1987;94:262–6.
12. Rosenbaum M, Weller RS, Conard P, Falvey EA, Gross JB. Ibuprofen provides longer lasting analgesia than fentanyl after laparoscopic surgery. Anaesth Anal 1991;73:255–9.
13. Wallace DH, Serpell MG, Baxter JN, O'Dwyer PJ. Randomised trial of different insufflation pressures for laparoscopic cholecystectomy. Br J Surg 1997;84:455–8.
14. Berberglu M, Dilek ON, Ercan F, Kati I, Ozmen M. The effect of CO2 insufflation rate on the postlaparoscopic shoulder pain. J Laparoendosc Adv Surg Tech A 1998;8:273–7.
15. Lipscomb GH, Summitt RL, McCord ML, Ling FW. The effect of nitrous oxide and carbon dioxide pneumoperitoneum on operative and postoperative pain during laparoscopic sterilization under local anaesthesia. J Am Assoc Gynecol Laparosc 1994;2:57–60.
16. Jensen AG, Prevedoros H, Kullman E, Andergerg B, Lennmarken C. Perioperative nitrous oxide does not influence recovery after laparoscopic cholecystectomy. Acta Anaesth Scand 1993;37:683–6.