Einarsson, Jon I. MD, MPH*; Sun, Judy MD*; Orav, John PhD†; Young, Amy E. MD*
From the *Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, Texas; and †Department of Biostatistics, Harvard School of Public Health, Boston, Massachusetts.
Address reprint requests to: Jon Ivar Einarsson, MD, Baylor College of Medicine, Department of Obstetrics and Gynecology, 6550 Fannin, Suite 901, Houston, TX 77030; e-mail: email@example.com.
Received May 6, 2004. Received in revised form July 27, 2004. Accepted August 5, 2004.
OBJECTIVE: To investigate whether local infiltration of bupivacaine reduces postoperative pain at trocar sites during gynecologic laparoscopy.
METHODS: This was a randomized, placebo-controlled, double-blind clinical trial, using patients as their own controls. For each patient, 2 opposite trocar sites were infiltrated. One site was randomly chosen to receive 0.5% bupivacaine, and the other received 0.9% saline. In addition, patients were randomized into 2 cohorts to receive either preincision or postsurgical infiltration. Surgeons, patients, and interviewers were blinded toward the exposure. Postoperative pain was evaluated at 1 hour, 4 hours, and 24 hours after surgery using a 100-mm visual analog scale. Patients rated their pain at each of the infiltrated trocar sites. A 20-mm difference between pain scores was considered clinically significant. A paired t test was used for analysis.
RESULTS: Infiltration of bupivacaine at completion of surgery resulted in significantly decreased pain at 1 hour postoperatively (mean pain score 25.8 versus 48.6, P = .02). Mean pain scores at 4 hours and 24 hours were decreased, but not statistically different. Patients receiving bupivacaine before surgery did not have a statistically significant decrease in pain scores.
CONCLUSION: Infiltration of bupivacaine at completion of gynecologic laparoscopic surgery decreases pain at trocar sites in the immediate postoperative period.
LEVEL OF EVIDENCE: I
Laparoscopic techniques have evolved rapidly over the last 2 decades.1 This mode of access offers patients the benefits of quick recovery and early return to normal activities, compared with the traditional laparotomy approach. Pain is often the main factor that hinders return to normal activities. Preemptive analgesia has been advocated as an important factor in preventing postoperative pain.2 In theory, using a local anesthetic to block peripheral nerve fibers before trauma prevents hyper-excitability in the dorsal horn of the spinal cord. This could prevent prolonged sensation of pain in the periphery.
Currently no standard of care exists regarding the use of local analgesia in gynecologic laparoscopy. Practices vary widely among physicians, even within the same institution.3 Although it is important to reduce postoperative pain, it is also important to determine whether this practice is truly effective. The specific aim of our study was to estimate whether local infiltration of bupivacaine, a long-acting anesthetic, reduces the amount of pain experienced at trocar sites in the immediate postoperative period. In addition, we planned to investigate whether this effect is dependent on the timing of administration of analgesia, ie, before skin incision versus at completion of the surgical procedure.
MATERIALS AND METHODS
Our patient population consisted of women undergoing gynecologic laparoscopic procedures at a local community hospital located within a large academic center in Houston, Texas. Patients were recruited before surgery and an informed consent obtained. The research protocol was reviewed and approved by the Baylor College of Medicine Institutional Review Board.
To minimize confounding, we used patients as their own controls. Patients were eligible to participate if they were scheduled to undergo gynecologic laparoscopic procedures that required 2 peripheral trocars and/or 2 midline trocars. Exclusion criteria were defined as allergy to amide local anesthetics or a scheduled procedure that involved a one-puncture laparoscopy or the use of another trocar on the ipsilateral side. We also decided to exclude from the data analysis cases with conversion to exploratory laparotomy. The sites of bupivacaine and placebo injection were randomly assigned to either right versus left trocar site or upper versus lower trocar site. We used trocars of equal sizes (either 5 mm or 10 mm) and design. We employed a simple computer-generated randomization scheme for this purpose. Sequentially numbered, opaque, sealed envelopes containing the assignments were made available in the operating room. After patient consent and transfer to the operating room, a surgical assistant opened an envelope and prepared two 10-mL syringes according to specified instructions inside the envelope. One of the syringes contained 10 mL of 0.5% bupivacaine (50 mg), and the other syringe contained 10 mL of 0.9% normal saline. Bupivacaine is toxic in doses exceeding 2.5–3 mg/kg, making this dose well within the margin of safety.4 We chose to use 0.9% normal saline as our control to maintain blinding, assuming that there is no discernible analgesic effect arising from saline only.5 Surgeons, patients, and patient interviewers were blinded to patient exposure.
In addition, patients were randomized to participate in either of 2 experimental cohorts: preincision or postsurgical infiltration of bupivacaine/placebo. Through this randomization, we were not only able to assess the efficacy of bupivacaine versus placebo in reducing pain within a cohort, but we were also able to assess the effects of preincision bupivacaine compared with postsurgical bupivacaine by comparing the actively treated trocar sites between the 2 cohorts.
After introduction of general anesthesia, trocar sites were infiltrated with a 22-gauge needle 4 minutes before incision in the preincision group and immediately after closure of trocar sites in the postsurgical group. It takes about 4 minutes for Bupivacaine to take effect6; therefore, it is essential to wait at least that long before the incision is made. Infiltration of bupivacaine and placebo was distributed equally on both sides, incorporating all tissue layers from skin to peritoneum. Only 2 incisions were infiltrated in patients with more than 2 skin incisions; the remaining skin incisions were not infiltrated with a local anesthetic or placebo. Incisions were closed with a 3/0 polyglactic, interrupted, subcuticular suture, as is standard practice at our institution. Using a skin marker, we placed 3- to 4-cm diameter circles around the trocar sites that were infiltrated for easier identification. Postoperative pain management consisted of either intravenous morphine via patient-controlled analgesia and/or nonsteroidal anti-inflammatory oral pain medication per standard protocol.
Postoperative pain was evaluated by interview at 1 hour, 4 hours, and 24 hours after surgery using a 100-mm visual analog scale (VAS). The VAS scale has been used in prior studies evaluating similar outcomes.7,8 Patients were introduced to the VAS scale before surgery. Patients were asked to rate their level of pain from 0 (no pain) to 10 (worst pain ever) at each of the 2 infiltrated trocar sites. The first 2 evaluations were performed in the recovery room, and the third evaluation was performed on the ward if possible. A member of the surgical team that was not involved in the index surgical case administered the VAS scale. Patients discharged home before 24 hours received a copy of a VAS scale with instructions and a prestamped, self-addressed envelope to send the VAS scale to the investigators.
We defined 20 mm on the VAS scale as a clinically significant difference. We assumed the standard deviation to be 22 mm. Because there are no previous studies in the literature on this subject that use patients as their own controls, we had to estimate the potential standard deviation in our study. In studies employing paired design, such as ours, one can normally expect decreased variance of the data.9 With a level of significance of α = .05 and 90% power to detect a clinically significant difference of 20 mm on a 100-mm VAS scale, we needed to enroll 26 patients in the preincision and postsurgical infiltration groups, for a total of 52 patients. We further anticipated a 10% exclusion rate because of conversion to an open procedure. This increased the planned sample size to 58 patients.
The primary outcome measure was predefined as the difference in mean pain scores between treatment sites and placebo sites at 1, 4, and 24 hours after surgery. This was calculated separately in the preincision and postincision groups. Because we found the pain scores to follow an approximately normal distribution, we used the paired t test to compare mean pain scores between treatment sites and placebo sites. In addition, we compared mean pain scores at bupivacaine sites in the before and after cohorts using a 2-sample Student t test. Finally, we combined the pain scores at placebo sites in the before and after cohorts in an attempt to improve our statistical power. We then reran the data analyses with this larger comparison group against each bupivacaine group. However, these analyses did not yield any new results and are not reported in any further detail.
A flow diagram of study participants is displayed in Figure 1. Patient recruitment commenced in January 2003 and was completed in November 2003. The demographics of the participants are displayed in Table 1. Our patient population was mostly Hispanic, with a small proportion of other ethnic groups represented. Table 2 displays the difference in mean pain scores in the preincision and postsurgical infiltration groups at 1 and 4 hours postoperatively. The mean pain score was lower at the bupivacaine site in all groups at all time points, but this difference only reached statistical significance in the postsurgical infiltration group at 1 hour postoperatively. Table 3 displays the difference in mean pain scores in the preincision and postsurgical infiltration groups at 24 hours postoperatively. Because of the large number of patients who dropped out by 24 hours, the baseline data are restricted to the subset of patients who also provided 24-hour data. Again, no statistically significant differences were detected. In addition, there were no statistically significant differences in mean pain scores at bupivacaine sites between the before and after groups. Patients seemed to have significant pain at trocar sites immediately after surgery, with patients receiving placebo reporting a mean pain score just under 50 on a 100-point scale.
Our results indicate that trocar site infiltration of bupivacaine after gynecologic laparoscopic surgery significantly reduces trocar site pain at 1 hour postoperatively. We were unable to detect other significant differences between trocar sites infiltrated with bupivacaine and those infiltrated with placebo. This could be partially explained by noting that the analgesic effects of bupivacaine last approximately 3–6 hours, with peak effect at 30–45 minutes after infiltration.4 The pain reduction seen in our study seems to mirror these time periods. We were unable to demonstrate the effectiveness of preemptive analgesia for pain relief. It is possible that the analgesic effects were diminished in these patients by the time their surgery was completed and the 1-hour evaluation of pain took place.
A number of studies have examined, with inconsistent results, whether local infiltration of a long-acting anesthetic reduces pain in patients undergoing gynecologic laparoscopic surgery.7–8,10–12 In a study by Ke et al,10 preemptive analgesia using 0.5% bupivacaine was found to be superior to postsurgical analgesia at 24 hours postoperatively. There were no significant differences in pain scores between the groups at 1 and 4 hours postoperatively. The results of this study might have been biased because 16 of 75 randomized patients were excluded after randomization. A study by Saleh et al12 found significant differences in pain levels only at 30 minutes after surgery when comparing preemptive analgesia using 0.5% bupivacaine with placebo. This study did not evaluate use of postsurgical analgesia. A study by Fong et al7 found no difference in postoperative pain scores when either preemptive or postsurgical infiltration of 30 mL of 0.25% bupivacaine was used. However, a recent study by Lam et al8 found that postsurgical infiltration of 1% lidocaine was effective for postoperative pain control and that presurgical infiltration was not.
Pain following laparoscopic surgery is multifactorial, arising from trocar sites, intra-abdominal trauma, and diaphragmatic irritation.13 Infiltration of trocar sites with a local analgesic probably does not reduce the severity of alternative sources of pain, and the perception of pain is very subjective and varies among patients. This could confound the results of any study on this subject. We used patients as their own controls to reduce potential confounding. By comparing pain scores at 2 surgical sites in the same patient, we believed we were better able to evaluate the pain emanating from the trocar sites, rather than the overall pain experienced by the patient. In addition, all previous studies have been hampered by the use of postoperative pain medications, which could potentially affect each patient differently, thereby providing an additional source of bias. This source of bias is avoided by using patients as their own controls. We hypothesized that if patients were unable to detect a difference between the trocar sites in question, either their pain was similar at both sites or other sources of pain were more prominent. In either case, this would make our estimations more conservative.
Based on available evidence, it seems that the extent and length of the surgical procedure would influence the efficacy of infiltration of trocar sites with a local analgesic. For example, in the study by Lam et al8 where preclosure infiltration was found to be effective, operating times were short (average operating time 21–25 minutes) and the procedures generally very simple (diagnostic laparoscopy with chromopertubation). However, in the study by Saleh et al,12 where there was only a statistically significant difference in pain scores at 30 minutes postprocedure, the average procedure time was 43–44 minutes, with a large proportion of patients undergoing invasive procedures such as excision of endometriosis, ovarian cystectomy, and ovarian drilling. This might indicate that the efficacy of the preemptive bupivacaine had waned during surgery and that alternative sources of postoperative pain were more prominent than pain from the trocar sites.
Given our predominantly Hispanic patient population, the results of our study would ideally need to be repeated in alternative ethnic groups for validation of our results. Another potential limitation is the small number of patients that returned their pain scales at 24 hours postoperatively. This predisposes our analysis to a type-II error at this time. Therefore, no firm conclusions can be drawn from our data about the analgesic efficacy of bupivacaine outside of the immediate postoperative period.
In conclusion, patients report experiencing considerable pain at trocar sites after laparoscopic surgery. Postoperative infiltration of bupivacaine significantly reduces postoperative pain in the immediate postoperative period, but preincision infiltration of bupivacaine does not seem to be effective.
1. Nezhat C, Nezhat F, Luciano A. Operative gynecologic laparoscopy: principles and techniques. 2nd ed. New York (NY): McGraw-Hill; 2000.
2. Woolf CJ. Evidence for a central component of post-injury pain hypersensitivity. Nature 1983;306:686–8.
3. Simpson RB, Russell D. Anaesthesia for daycase gynaecological laparoscopy: a survey of clinical practice in the United Kingdom. Anaesthesia 1999;54:72–6.
4. Tetzlaff JE. The pharmacology of local anesthetics. Anesthesiol Clin North America 2000;18:217–33.
5. Lepner U, Goroshina J, Samarutel J. Postoperative pain relief after laparoscopic cholecystectomy: a randomized prospective double-blind clinical trial. Scand J Surg 2003;92:121–4.
6. Moore DC. Bupivacaine: a review of 11,080 cases. Anesth Analg 1978;57:42–53.
7. Fong SY, Pavy TJ, Yeo ST, Paech MJ, Gurrin LC. Assessment of wound infiltration with bupivacaine in women undergoing day-case gynecologic laparoscopy. Reg Anesth Pain Med 2001;26:131–6.
8. Lam KW, Pun TC, Ng EH, Wong KS. Efficacy of preemptive analgesia for wound pain after laparoscopic operations in infertile women: a randomized, double blind and placebo control study. BJOG 2004;111:340–4.
9. Dawson B, Trapp RG. Basic and clinical biostatistics. 3rd ed. New York (NY): Lange Medical Books—McGraw-Hill;2001.
10. Ke RW, Portera SG, Bagous W, Lincoln SR. A randomized, double-blinded trial of preemptive analgesia in laparoscopy. Obstet Gynecol 1998;92:972–5.
11. Salman MA, Yucebaz MA, Cozkun F, Aypar U. Day case laparoscopy: a comparison of prophylactic opioid, NSAID or local anesthesia for postoperative analgesia. Acta Anaesthesiol Scand 2000;44:536–42.
12. Saleh A, Fox G, Felemban A, Guerra C, Tulandi T. Effects of local bupivacaine instillation on pain after laparoscopy. J Am Assoc Gynecol Laparosc 2001;8:203–6.
13. Mouton WG, Bessell JR, Otten KT, Maddern GJ. Pain after laparoscopy. Surg Endosc 1999;13:445–8.