| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the University of Wisconsin Comprehensive Cancer Center, Madison, Wisconsin.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Methods: A prospective, randomized, double-blind, placebo-controlled trial was performed. After closure of the fascia, flexible soaker catheters were placed in the deep subcutaneous space. The infusion pump was filled with 290 mL of either 0.5% bupivacaine or normal saline, to infuse over 72 hours. Daily assessments of pain scores utilized the Wisconsin Brief Pain Inventory. All patients received intravenous narcotics via patient-controlled devices.
Results: Eighty surgeries were evaluated in a total of 79 women (40 per arm). Mean age was 56 years, with 79% having invasive gynecologic pathology. The two groups were not significantly different in terms of type of surgery, length of incision, estimated blood loss, operative time, or medical history. Postoperative outcomes, including wound toxicity, time to flatus, and hospital stay, did not differ. Study patients averaged 75 mg intravenously and 107 mg total narcotic use (morphine equivalent), whereas controls averaged 60 mg intravenously and 86 mg total (P = .40 intravenously; P = .25 total). Acetaminophen and intravenous ketorolac consumption were equal between groups. The Brief Pain Inventory score for "current pain" was 2.84 for bupivacaine patients and 3.14 for controls (P = .46; least = 0, most = 10). There was no individual postoperative day when "current pain" BPI scores differed. "Worst pain" and "least pain" Brief Pain Inventory scores showed similar results.
Conclusion: The results suggest that although the continuous infusion system seems safe, it is not efficacious in this patient population.
Level of Evidence: I
Recent studies have suggested that a multimodal approach to accelerate postoperative recovery may lead to a reduction in the undesirable sequelae of surgical injury with accelerated recovery and reduction in morbidity and overall costs. Current literature supports the use of 2 or more analgesic techniques for optimal control of perioperative pain, especially when different sites and/or mechanism of action are involved.6 These include regional anesthetic and analgesic techniques such as nerve blocks, wound infiltration, and epidural delivery of local anesthetics and/or opioids, as well as nonsteroidal antiinflammatory drugs and opioids.
Many studies show that wound infiltration with an intraoperative local anesthetic functions well as an adjunct form of pain control, but that it is limited by its short duration of action. Investigators have created and evaluated various techniques for delivering continuous infusions of bupivacaine to overcome this limitation. Such early trials used multiple different pump mechanisms, tubing styles, infusion concentrations, and rates. Surgery types included cesarean,11 open cholecystectomy,9,10 inguinal herniorrhaphy,12 and colorectal procedures.8 The techniques have varied in their effectiveness. Additional benefits of continuous local anesthetic may include impaired leukocyte action at the incision site and antimicrobial action.13,14 Overall, the techniques have been shown to be simple, low maintenance, and low risk to patients.
More recently, The ON-Q Pain Management System was developed as a commercially available, standardized infusion device. It is manufactured by I-Flow Corporation (Lake Forest, CA). It is a lightweight latex-free elastomeric pump designed to be filled with local anesthetic solution and to be placed into a surgical site. Delivery time ranges from 36 to 84 hours and is dependant on amount of solution contained within the pump. This device has not yet been evaluated in the setting of large wounds used for the removal of abdominal cancers. The objective of the current study is to evaluate the ON-Q system for postoperative pain control in women undergoing laparotomy for suspected gynecologic malignancy.
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Eligible subjects were those women at least 18 years of age who were undergoing laparotomy performed by 1 of 4 gynecologic oncologists. Women who were pregnant or incarcerated, had a known allergy to local anesthetic, or were on daily opioid therapy for chronic pain were excluded. To detect a 30% difference in average pain scores between groups with an alpha of .05, 80 patients were needed to reach a power of 90%. The primary outcome was mean current pain over the 5 days of hospitalization. Secondary efficacy outcomes included mean maximum pain and oral and intravenous pain medicine use. The main safety outcome was presence of wound toxicities, with other postoperative complications, duration of hospital stay, and time to first bowel movement also monitored.
The Gynecologic Oncology Research Office performed subject enrollment. Patients were randomized by the Pharmaceutical Research Center, after induction of anesthesia, to either continuous infusion of local anesthetic (bupivacaine 0.5%) or placebo. A random-number generator was used to produce a blocked randomization code without stratification. Randomization codes were generated by University of Wisconsin Hospital Investigational Drug Service staff who then composed bags of bupivacaine or placebo analgesic solution, as appropriate (identical in appearance). To avoid misrandomizations, a bag would be sent to the operating room only after an eligible consenting patient was anesthetized for surgery. All clinical personnel and patients were blinded; only the Investigational Drug Service staff and study statistician knew the patients' true treatments until after the final analysis. The ON-Q Pain Management System was the infusion system used (ON-Q Postop Pain Relief System, I-Flow Corporation, Lake Forest, CA). It is a lightweight latex-free elastomeric pump attached to 2 delivery catheters (Fig. 1).
|
At the time of surgery, after closure of the fascia, the flexible soaker catheters were placed in the deep subcutaneous space. The pumps in the treatment group were filled with 290 mL 0.5% bupivacaine, which provided continuous infusion of 2 mL per hour, per catheter. The pumps in the control group were filled with 290 mL normal saline. The infusion system was removed after 72 hours.
All subjects also received a patient-controlled analgesic pump postoperatively. The patient-controlled analgesic was filled with either morphine or hydromorphone, which is our current standard following laparotomy. Pharmacy records and nursing notes were reviewed, and all narcotic use was converted into morphine equivalents using the Opioid Equivalency Table.15 Most subjects received oral oxycodone after discontinuation of the patient-controlled analgesic.
The primary outcomes of this study were pain and narcotic use. Pain scores were evaluated using Wisconsin Brief Pain Inventory,16 which was developed to obtain estimates of pain prevalence and severity. It is brief, self-administered, and easily understandable. It possesses adequate reliability and validity for use in research, and its format suits postoperative patients. The Brief Pain Inventory consists of 17 items. Subjects use a numerical rating scale ("0" is no pain and "10" is worst pain possible) to indicate current, worst, and least pain in the past 24 hours.
The General Clinical Research Center staff obtained all pain data. Subjects completed the Brief Pain Inventory once daily, assisted by a study nurse if necessary, for 5 days. A research nurse called the subject at home if the subject was discharged before the fifth postoperative day. Subjects were followed up for adverse events until the scheduled 4-week postoperative visit.
Significance of the association of treatment with binary or discrete outcomes was assessed using Fisher exact test. Significance for difference in a continuous outcome between treatment groups was assessed using the equal-variance t test. Pain outcomes were also compared using nonparametric tests of medians instead of means, but with identical results. One patient was randomized and treated twice; under the intent-to-treat principal, data from both surgeries are included in the analysis. The study's results are nearly identical when data from the second procedure were omitted. No ancillary or subgroup analyses were performed. All computations were performed with the Splus statistical package.17
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Subject and intraoperative characteristics are displayed in Table 1. There were no significant differences between groups in any of the demographic categories. Concurrent medical problems existed in large numbers in both groups. The most commonly reported coexisting morbidities included heart disease, hypertension, stroke or transient ischemic attach, diabetes, and renal disease.
|
The majority of subjects were found to have cancer (65% of the bupivacaine group and 82% of the placebo group). Ovarian cancer was the most frequently encountered cancer in the treatment group (40%). In the control group, ovarian and endometrial cancers occurred most frequently (35%). Stage I and III cancers were most frequently diagnosed in the bupivacaine subjects (38% each) and stage I cancer in the placebo group (45%). This difference was not statistically significant.
Length of hospital stay and time to first bowel movement were not different between the two groups (Table 2). Incision length was recorded for 63 patients (32 in the treatment group and 31 control subjects). The mean length of 19.9 cm (range 11.5–31 cm) was similar to that of the control group (20.3 cm; range 12–31 cm).
|
Wound complications were seen in 20% of the treatment arm and 18% of controls (Table 2). The difference in wound toxicity rates is thus 2%, favoring the control arm, with a 95% confidence interval (CI) of –15% to 20%. The majority of these complications were mild to moderate in severity (15% in both groups; CTC criteria version 2.0). Regarding nonwound toxicities, 50% of all subjects experienced some type of adverse event (53% in the bupivacaine group and 48% in the placebo group). Again, the majority of these toxicities were mild in severity (ie, fever, atelectasis). The only postoperative outcomes with significantly different incidence between groups were serious adverse events that required reporting to the Institutional Review Board. Eight serious adverse events occurred in the placebo group (myocardial infarction, allergic reaction to paclitaxel, shortness of breath, nausea, 2 strokes, 2 pulmonary emboli), whereas none occurred in the bupivacaine group (P = .004).
Pain scores, the study's primary outcome, were available for 78 of 80 subjects, for a follow-up rate of 97.5%. Nine subjects missed 1 or 2 days' pain scores. Averaging the adjacent days' scores imputed these scores. Overall, pain scores are available for 95% of the days that pain was evaluated.
Figure 2 graphs the Brief Pain Inventory "current pain" scores, averaged over the 5-day evaluation period. Although the mean was slightly lower in the bupivacaine group compared with the placebo group (2.84 versus 3.14), this difference was not statistically significant (P = .46; 95% CI = –1.08 to 0.49). An analysis of "worst pain" returned similar results (not shown). The "least pain" scores, as indicated in Figure 2, approached statistical significance (P = .056; 95% CI = –1.09 to 0.01). Further analysis revealed no difference between groups for daily values in any of the reported pain scores.
|
Narcotic use was evaluated extensively (Fig. 3). No patients were missing intravenous narcotic dose data. Two patients were missing oral narcotic doses on days 4 and 3 patients on day 5. These results were imputed with the last observation carried forward. Subjects in the treatment group averaged 107 mg intravenous morphine equivalent per day (range 3.7–420.1 mg) and control group subjects averaged 86 mg (range 3.067–403.9 mg). This difference was not statistically significant (P = .25; 95% CI = –14.9 to 56.4). An analysis of total oral narcotic and total intravenous narcotic usage showed identical results. There was no difference in the amount of nonnarcotic pain medications used, such as ketorolac and acetaminophen (results not shown).
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although not all of the patients included in the trial had a cancer diagnosis, this population was representative of a typical gynecologic oncology practice. All patients had a preoperative diagnosis of cancer, suspected cancer, preinvasive disease, or suspected difficult gynecologic surgery. Patients in both groups had the expected high level of comorbidities. The incision types and lengths were consistent with the extensive pelvic and abdominal surgical fields required for procedures such as debulking, bowel resection, and paraaortic lymphadenectomy. This population of patients and surgical procedures contrast the previous studies examining continuous bupivacaine systems.11,12 These studies evaluated less complex surgeries with smaller wounds and found the pump to be more effective.
The ON-Q system is simple to place and maintain. The pumps are kept in a small bag with a shoulder harness to allow ambulation, but the catheters can easily become dislodged if the pump is placed under tension (eg, if it rolls off of the hospital bed onto the floor while the patient sleeps). This occurred in 4 patients (5%). Leakage of fluid from the dependant portion of the wound was common, but the incidence of wound toxicity did not differ between the two groups. The majority of wound complications were minor cellulitis or superficial separation. The 4% incidence of major wound toxicity is similar or decreased in comparison to previous reported gynecologic surgeries in this patient population.18,19
Pain and narcotic use were the major outcomes evaluated in this trial. The Brief Pain Inventory tool allowed multiple assessments of pain on each of the 5 postoperative days. The 3 major measurement instruments (ie, current, least, or worst pain) detected no difference in pain scores between groups. These similarities were found both for overall mean postoperative scores and daily scores. This lack of effectiveness on pain levels is consistent with a recent evaluation of the ON-Q system by Givens et al.20 In their prospective evaluation of 36 patients undergoing cesarean delivery, they found no difference in visual analog scale scores; however, there was a significant decrease in narcotic use in the bupivacaine group.
Bupivacaine infusion using the ON-Q pump has been shown to positively affect pain scores in patients undergoing shoulder surgery21 and inguinal hernia repair.22,23 Schurr et al23 examined patients with inguinal herniorrhaphy. They found no difference in narcotic consumption between the two groups at any point during the study. Our current trial also found no difference in narcotic use, with a surprising trend toward increased use in the bupivacaine group. Schurr et al evaluated pain with the same BPI tool used in this trial. Using this more detailed instrument, they deciphered an interesting pattern. On day 1, patients who received bupivacaine had lower rating for both "least pain" and "worst pain." However, no difference was discovered between groups on days 2 to 5. They conclude that this modest and short-lived effect limits the usefulness of this intervention.
Why are there such varying results regarding both the pain control and narcotic-limiting effects of bupivacaine infusion? One answer could be the anatomical placement of the catheters. In the current trial, the catheters were placed in the deep subcutaneous space overlying the fascia. In reviewing the literature of continuous bupivacaine infusion (including non–ON-Q trials), 2 studies place the pump in this same manner.8,20 Both of these trials show no difference in pain scores, but a significant difference in narcotic consumption. Of the 7 trials that place the pump in a deeper location (usually deep to the fascial layer),9–12,21–23 5 show improvement in pain scores with bupivacaine and 4 show a decrease in narcotic usage.
No safety concerns for ON-Q have arisen from this study, which admittedly does not have a sample size sufficient to prove clinical equivalence within a small margin. The main safety outcome, wound toxicity rate, was shown to differ by no more than 20%.
Because the pumps were implanted in both the saline and treatment groups, this trial does not provide evidence about the risks due to implantation of the pump itself. Although there was no difference in overall complications between groups, there was an unexpected finding of decreased serious adverse events in the bupivacaine subjects. This was not a main outcome variable for the trial, and is therefore subject to bias. Determining what adverse events to report as a serious adverse event is difficult. Also, the category is very heterogeneous, which makes it hard to draw firm conclusions. This finding has not been noted in previous studies and could be evaluated in future investigation.
The results of the current randomized trial do not support the use of the ON-Q bupivacaine infusion system for the management of postoperative pain in gynecologic oncology procedures. The large incisions and extensive dissections could potentially overwhelm the ability of the system to affect pain control, and its utility seems limited in this patient group.
|
Footnotes |
|---|
Funding Support: This work was supported by grant M01 RR003186 from the General Clinical Research Centers Program of the National Center for Research Resources, National Institutes of Health.
Presented at the Society of Gynecologic Oncologists' 36th Annual Meeting, March 20, 2005, at Miami Beach, Florida.
Corresponding author: David M. Kushner, MD, Department of Gynecologic Oncology, University of Wisconsin Hospital & Clinics, 600 Highland Avenue, H4–636 Madison, WI 53792; e-mail: dmkushner{at}wisc.edu.
doi:10.1097/01.AOG.0000171111.68015.af
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Practice guidelines for acute pain management in the perioperative setting: an updated report by the American Society of Anesthesiologists Task Force on Acute Pain Management. Anesthesiology 2004;100:1573–81.[Medline]
3. Doyle E, Bowler GM. Pre-emptive effect of multimodal analgesia in thoracic surgery. Br J Anaesth 1998;80:147–51.
4. Kehlet H. Multimodal approach to control postoperative pathophysiology and rehabilitation. Br J Anaesth 1997;78:606–17.
5. Kehlet H. Acute pain control and accelerated postoperative surgical recovery. Surg Clin North Am 1999;79:431–43.[Medline]
6. Agency for Health Care Policy and Research. Acute pain management: operative or medical procedures and trauma. Clinical practice guideline No. 1. AHCPR publication 92–0032. Rockville, MD; 1992.
7. Harrison CA, Morris S, Harvey JS. Effect of ilioinguinal and iliohypogastric nerve block and wound infiltration with 0.5% bupivacaine on postoperative pain after hernia repair. Br J Anaesth 1994;72:691–3.
8. Cheong WK, Seow-Choen F, Eu KW, Tang CL, Heah SM. Randomized clinical trial of local bupivacaine perfusion versus parenteral morphine infusion for pain relief after laparotomy. Br J Surg 2001;88:357–9.[Medline]
9. Chester JF, Ravindranath K, White BD, Shanahan D, Taylor RS, Lloyd-Williams K. Wound perfusion with bupivacaine: objective evidence for efficacy in postoperative pain relief. Ann R Coll Surg Engl 1989;71:394–6.[Medline]
10. Levack ID, Holmes JD, Robertson GS. Abdominal wound perfusion for the relief of postoperative pain. Br J Anaesth 1986;58:615–9.
11. Mecklem DW, Humphrey MD, Hicks RW. Efficacy of bupivacaine delivered by wound catheter for post-caesarean section analgesia. Aust N Z J Obstet Gynaecol 1995;35:416–21.[Medline]
12. Oakley MJ, Smith JS, Anderson JR, Fenton-Lee D. Randomized placebo-controlled trial of local anaesthetic infusion in day-case inguinal hernia repair. Br J Surg 1998;85:797–9.[Medline]
13. Borris TJ, Scuba JR. Continuous postoperative pain control after bone graft harvest using a pump delivery system. J Oral Maxillofac Surg 2001;59:707–8.[Medline]
14. Eriksson AS, Sinclair R, Cassuto J, Thomsen P. Influence of lidocaine on leukocyte function in the surgical wound. Anesthesiology 1992;77:74–8.[Medline]
15. Principles of analgesic use in the treatment of acute pain and cancer pain. 5th ed. Glenview (IL): American Pain Society; 2003.
16. Daut RL, Cleeland CS, Flanery RC. Development of the Wisconsin Brief Pain Questionnaire to assess pain in cancer and other diseases. Pain 1983;17:197–210.[Medline]
17. Venables WN, Ripley BD. Modern applied statistics with S. 4th ed. New York (NY): Springer-Verlag; 2002.
18. Gemignani ML, Curtin JP, Zelmanovich J, Patel DA, Venkatraman E, Barakat RR. Laparoscopic-assisted vaginal hysterectomy for endometrial cancer: clinical outcomes and hospital charges. Gynecol Oncol 1999;73:5–11.[Medline]
19. Hemsell DL, Johnson ER, Hemsell PG, Nobles BJ, Little BB, Heard MC. Cefazolin is inferior to cefotetan as single-dose prophylaxis for women undergoing elective total abdominal hysterectomy. Clin Infect Dis 1995;20:677–84.[Medline]
20. Givens VA, Lipscomb GH, Meyer NL. A randomized trial of postoperative wound irrigation with local anesthetic for pain after cesarean delivery. Am J Obstet Gynecol 2002;186:1188–91.[Medline]
21. Savoie FH, Field LD, Jenkins RN, Mallon WJ, Phelps RA II. The pain control infusion pump for postoperative pain control in shoulder surgery. Arthroscopy 2000;16:339–42.[Medline]
22. Lau H, Patil NG, Lee F. Randomized clinical trial of postoperative subfascial infusion with bupivacaine following ambulatory open mesh repair of inguinal hernia. Dig Surg 2003;20:285–9.[Medline]
23. Schurr MJ, Gordon DB, Pellino TA, Scanlon TA. Continuous local anesthetic infusion for pain management after outpatient inguinal herniorrhaphy. Surgery 2004;136:761–9.[Medline]
This article has been cited by other articles:
|
|
 |
|
  K.-M. Chiu, C.-C. Wu, M.-J. Wang, C.-W. Lu, J.-S. Shieh, T.-Y. Lin, and S.-H. Chu Local infusion of bupivacaine combined with intravenous patient-controlled analgesia provides better pain relief than intravenous patient-controlled analgesia alone in patients undergoing minimally invasive cardiac surgery. J. Thorac. Cardiovasc. Surg., June 1, 2008; 135(6): 1348 - 1352. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Collins Which Randomized Controlled Trials Are Relevant to Clinical Practice? Obstet. Gynecol., August 1, 2005; 106(2): 216 - 218. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
