Effect of adding ketorolac to intravenous morphine
patient-controlled analgesia on bowel function in
colorectal surgery patients
— a prospective, randomized,
double-blind study
J-Y. CHEN1,2, G-J. WU1, M. S. MOK3, Y-H. CHOU4, W-Z. SUN5, P-L. CHEN6, W-S. CHAN1, H-W. YIEN2and Y-R. WEN1 1Department of Anesthesiology, Shin Kong Wu Ho-Su Memorial Hospital,2Institute of Emergency and Critical Care Medicine, National
Yang-Ming University,3Department of Anesthesiology, Taipei Medical University Hospital; Taipei,4Department of Surgery, Shin Kong Wu Ho-Su
Memorial Hospital,5Department of Anesthesia, National Taiwan University Hospital,6Taiwan College of Nursing, Taipei Medical University,
Taipei, Taiwan
Background: Postoperative ileus (PI) is the transient impair-ment of bowel motility due to surgical trauma and the asso-ciated physiological responses. Postoperative ileus results in patient discomfort, increases gastrointestinal risks, prolongs hospital stay and increases medical expenses. In this study, we investigated the effect of patient-controlled analgesia (PCA) morphine with or without ketorolac on bowel functions in patients after colorectal surgeries.
Methods: A total of 79 patients who received elective color-ectal resection were randomly allocated into two groups receiving either intravenous PCA morphine (M group) or intra-venous PCA morphine plus ketorolac (K group). Recovery of bowel functions (bowel movement, passage of flatus, and soft diet intake), pain scores, morphine consumption, time for first ambulation, and opioid-related side-effects were recorded. Results: Patients in the K group received 29% less morphine than patients in the M group with comparable pain scores. The first bowel movement (1.5 [0.7—1.9] vs. 1.7 [1.0—2.8] days, P <0.05) and the first ambulation (2.2 1.0 vs. 2.8 1.2 days,
P <0.05) were significantly earlier in the K group than in the M group. The time of the first flatus passing, the first intake of soft diet, and duration of hospital stay were not significantly different between the two groups.
Conclusions: The results of this study suggest that addition of ketorolac to intravenous morphine PCA provides an opioid-sparing effect but has limited benefit in shortening the duration of bowel immobility and time to first ambulation. These findings imply that postoperative ileus is attributable to multiple factors in addition to morphine consumption.
Accepted for publication 21 September 2004
Key words: Bowel movement; ketorolac; opioids; patient-controlled analgesia; postoperative ileus; postoperative pain. #Acta Anaesthesiologica Scandinavica 49 (2005)
P
OSTOPERATIVEileus (PI) is one of the commoncom-plications after abdominal surgery. It is usually defined as persistent bowel immobility for more than 3 days after surgery (1). Postoperative ileus not only results in abdominal discomfort, nausea, vomiting, delayed oral intake, and increase postoperative morbidity (2), but also prolongs the hospital stay and increases medical costs (3, 4). Prolonged recovery of postoperative gastrointestinal functions remains a significant healthcare problem (5).
Opioids are effective for the management of mod-erate to severe perioperative pain in major abdominal surgery (1, 6). However, adverse effects can compro-mise the usefulness of these agents for analgesia.
Bowel dysfunction is among the most important and distressing adverse effects associated with opioid administration. The use of opioids in the postopera-tive period after bowel surgery has been viewed as a double-edged sword with the benefit of pain relief on one hand and the impairment of bowel function on the other. Thus surgeons often limit the usage of post-operative opioids for fear of prolonged recovery of bowel function. However, somewhat surprisingly, only a few studies have investigated the correlation between postoperative opioids and PI (7—11), and most of them involved epidural opioids and non-gastrointestinal surgeries. Only one previous study has investigated the effect of morphine dose on PI in
colorectal surgery (12), but it did not investigate the influence of changing morphine formulation on PI. This is the first prospective, randomized, double-blind study to test the hypothesis that the opioid-sparing effect of the addition of ketorolac in patient-controlled analgesia (PCA) morphine can shorten the duration of PI and hospitalization after major colorectal surgery.
Methods
This prospective, randomized double-blind study was conducted from January 2003 to December 2003 after approval was obtained from the Shin Kong Wu Ho-Su Memorial Hospital Ethics Committee. Seventy-nine consecutive patients with American Society of Anesthesiologists (ASA) physical status I or II, with age ranging from 35 to 75 years, participated in this study after giving written informed consent. We calculated a sample size so that a between-group mean difference in cumulative morphine consump-tion to the endpoint of the first bowel movement would permit a type I error rate of 0.05 in a one-tailed test of significance with a power of 80%. This analysis indicated that a sample size of at least 32 patients per group was needed.
All patients received elective colorectal resections (Table 1) with simple anastomosis via midline incision by the same surgeon (Y.H. Chou). Exclusion criteria included active peptic ulcer, poor renal function (serum creatinine over 1.5 mg dl1), preoperative coagulopathy (abnormal platelet count, prolonged bleeding time, prothrombin time or activated partial thromboplastin time), previous history of allergy to morphine or ketorolac, opioid addiction, and inability to properly use the PCA device. All subjects received a written instruction and verbal training on the use of the PCA device, and then were randomly allocated to their treatment group with the aid of computer-derived random number. These preoperative prepara-tions were completed on the day before the surgery by one author (W.S. Chan) who did not personally participate in the postoperative evaluation. Patients
received PCA (Abbot aim1plus, Abbot Laboratories, Norton Chicago, IL) with either intravenous mor-phine (M group) or intravenous mormor-phine combined with ketorolac (K group), without knowledge of their group allocation.
No opioids or non-steroidal anti-inflammatory drugs (NSAIDs) were given within 1 week before surgery and no premedication drugs were used preoperatively. All patients received prophylactic antibiotics at 30 min before surgery. General anes-thesia was induced with 3—5 mg kg1thiamylal and 2—4 mg kg1fentanyl intravenously (i.v.). Endotracheal intubation was facilitated by 0.15 mg kg1 cis-atracurium i.v., and patients were mechanically ventilated. Anesthesia was maintained with sevo-flurane (at 1.5—2 MAC), and muscle relaxation was maintained with additional doses of cis-atracurium. A nasogastric tube was inserted after anesthesia had been established. Perioperative cardiovascular stabi-lity was maintained and blood transfusion was given when indicated. Every patient was sent to the post-anesthetic care unit postoperatively where they immediately received a bolus of intravenous meperi-dine 50—100 mg for pain control. Then, the use of the PCA device was initiated. All the study drugs were prepared by a pharmacist and no label was put on the PCA reservoir bag. The 100-ml solution in the PCA reservoir bag contained 100 mg of morphine in normal saline (1 mg ml1) in the M group or 100 mg morphine plus 120 mg of ketorolac in normal saline (morphine 1 mg ml1; ketorolac 1.2 mg ml1) in the K group. The PCA setting was set at a bolus of 2 ml with a 10-min lockout interval without a continuous infusion for all patients. After discharge from the PACU, all patients were visited twice per day by acute pain service staff who were not aware of the patients’ grouping. The pain ser-vice staff were trained to adjust the bolus dose by a 20—50% increase or a 20—25% decrease according to the patient’s pain intensity at the time of the daily visit. No other NSAIDs were given on the ward postoperatively. The use of PCA was discontinued when VAS on movement was less than 3 on two consecutive evaluations.
The pain scores at rest and on movement were assessed with the visual analog scale (VAS, 0 ¼ no pain, 10 ¼ the worst intolerant pain). Morphine consumption, doses of rescue analgesic (meperidine), nausea, vomiting, skin itching, and sedation at the time of each visit were documented. The time inter-vals from the end of surgery to the first bowel move-ment and to the first flatus were recorded by the patients. Data on the time to the first ambulation
Table 1
Distribution of types of elective colorectal surgical procedures in the two groups.
Procedure M Group K Group
Hemicolectomy 5 4
Abdominoperineal resection 8 7
Total colectomy 7 9
Sigmoid resection 15 19
(left bed and walked for at least 5 min) and to the first intake of soft diet (standard hospital formula permitted by the surgeon) were specifically asked and recorded by the pain service staff. Data on the duration of hospitalization after surgery were col-lected from the patient’s medical record after the patient was discharged.
Statistics
Continuous variables are expressed as the mean standard deviation or median (25th to 75th percen-tile) when data were not normally distributed, and categorical variables are presented as frequencies (percentage of patients). Data were analyzed by chi-square test for categorical variables and two-tailed Student’s t-test or Mann—Whitney U-test for continu-ous variables. A P-value < 0.05 was considered to indicate a significant difference.
Results
Among the 79 participants, 74 patients completed the study course and were included in the analysis. Of the five excluded patients, two in the K group and one in the M group received second surgeries due to leakage from bowel anastomosis, one patient in the M group was transferred to the intensive care unit due to respiratory failure unrelated to morphine use, and one patient in the M group had postoperative wound infection. There were no differences in pre-operative data between the two groups (Table 2).
Total morphine consumption was significantly lower in the K group (71.4 [55.0—96.1] mg) than in the M group (93.3 [70.4—127.1] mg) (Fig. 1). The respective cumulative morphine doses from the end of surgery to the first bowel movement and to the first passage of flatus were also significantly lower in the K group than in the M group (28.0 [18.0—44.0] mg in the K group vs. 57 [34.5—69.8] mg in the M group for bowel movement; 51.5 [38.0—71.2] mg in the K group
vs. 75.5 [60.8—93.3] mg in the M group for flatus). There were significant differences between the two groups in the time to the first bowel movement (1.5 [0.7—1.9] days in the K group vs. 1.7 [1.0—2.8] days in the M group) and the time to the first ambulation (2.2 1.0 days in the K group vs. 2.8 1.2 days in the M group) (Fig. 2). However, no significant differences were found in the time to the first flatus, the time to the first soft diet, and the number of postoperative hospitalization days between the two groups (Fig. 2). Duration of surgery, amount of bleeding, the duration of PCA use, postoperative pain scores, rescue meperidine dose and opioid-induced adverse effects were also similar between the two groups (Tables 3 and 4).
Discussion
This study suggests that the addition of ketorolac in intravenous PCA provides benefits in patients after colorectal surgery in reducing the total amount of morphine consumption, shortening the duration of
Table 2
Demographic characteristics and operative data of the patients.
Group Mean SD or median (25th75th percentile) P
M K Male/Female 22/13 19/20 0.22 Age 68 (47.8—74.0) 64.5 (48.5—71.0) 0.51 Weight (kg) 61 13.4 61.1 10.9 0.91 Height (cm) 161.1 8.8 161.7 8.1 0.78 Operation time (h) 4.0 1.1 3.7 1.2 0.38 Amount of bleeding (ml) 111 85 88 63 0.26
Rescue meperidine dose (mg) 60 41 67 52 0.47
0 20 40 60 80 100 120 140 160
Total morphine consumption Morphine
(mg)
*
K M
Fig. 1. Ketorolac addition was associated with a 29% reduction in total morphine consumption. (M: morphine group; K: morphine plus ketorolac group; error bar ¼ standard deviation; *P < 0.05).
recovery of postoperative bowel movement and ambulation. The duration to the first flatus, to the first soft diet, and postoperative hospital stay in the morphine-ketorolac group was decreased but were not significantly different from those in the morphine group. The opioid-sparing effect of ketorolac was pro-vided without loss of analgesia and did not increase NSAID-related side-effects.
Postoperative ileus is of particular concerned to surgeons. Ileus can cause abdominal discomfort, nausea and vomiting. More importantly, it delays return of bowel function and the resumption of oral intake, resulting in prolonged hospitalization (3). Use of opioids for postoperative pain could act at the spinal and cerebral opioid receptors and suppress pain transmission, but it also inhibits gut motility and decreases intestinal secretory function at the level of myenteric plexus (6). Increasing the morphine dose significantly delays transit time in the small bowel and colon (13), the time to first audible bowel sound, to the passage of flatus, and to the first bowel
movement (12). However, morphine did not delay the recovery of bowel function after abdominal surgery if given in small doses by intravenous PCA (14). In our study, morphine consumption had limited impact on PI recovery, which suggests there may be a certain range of morphine escalation within which there would be no significant inhibition of bowel motility. Among patients in the M group, the average time to flatus passage was not prolonged despite a 29% higher morphine dose than patients in the K group. This suggests that the existence and severity of surgical pain are the important indicators of the need for increasing the morphine dose to relieve pain or decreasing the dose to avoid PI.
In addition to the opioid effect, multiple factors like surgical pain, bowel inflammation and sympathetic hyperactivity are involved in bowel movement (7). A multimodal postcolorectal surgical program to reduce the PI has been advocated which included minimally invasive laparoscopic surgery, epidural anesthesia, reduction of opioid use, early mobilization, enteral feeding, prokinetic agents, and nasal-gastric suction (2, 5, 15, 16). Recent studies demonstrated that under the program, PI and hospital stay were substantially shortened (17). We did not adopt such a rehabilitation program, but continued with a conventional perioperative care technique because the National Health Insurance system in Taiwan lacks a compulsory clinical pathway for major surgeries and allows substantial variability in the surgical program. Nevertheless, this study suggests that a simple combination analgesic regimen can provide advan-tages over single opioid treatment in PI under such conditions, which we believe is of reference value to our colleagues in other countries in the world.
In assessing the additive effect of ketorolac, our results showed a reduction of total morphine consumption by 29% which is comparable to other
BM Flatus Mobilization Soft diet
Days * * 12 10 8 6 4 2 0 K M
Fig. 2. Duration from the end of surgery to the first bowel movement, the first passage of flatus, the first mobilization, and the first soft diet feeding (M: morphine group; K: morphine plus ketorolac group; BM: bowel movement; error bar ¼ standard deviation; *P < 0.05).
Table 3
Duration of patient-controlled analgesia use, data on visual analog scale pain, and duration of hospital stay.
Group Median (25th75th percentile) P
M K
Duration of PCA-treatment (days) 5.7 (5.4-6.5) 5.8 (4.9-6.0) 0.57
VAS at rest 1st POD 2.5 (1.9-2.6) 2.2 (1.9-2.5) 0.65 2nd POD 2.0 (1.7-2.3) 1.9 (1.4-2.1) 0.10 3rd POD 1.4 (1.3-1.8) 1.3 (1.1-1.8) 0.39 VAS on movement 1st POD 4.4 (4.0-4.7) 4.5 (4.0-4.9) 0.91 2nd POD 4.1 (3.7-4.4) 4.0 (3.6-4.4) 0.90 3rd POD 3.6 (3.2-3.9) 3.4 (3.2-4.0) 0.91
Hospital stay (days) 15 (14.0-19.3) 14.5 (12.0-20.0) 0.73
studies reporting that ketorolac addition could reduce morphine consumption by 25% to 45% (18—20). Clinical assessment of the recovery from PI is difficult because there is a lack of consensus on a reliable end-point for use by the surgeon. The passage of flatus reported by patients is a good clinical sign and is frequently used (21), but the correlation between flatus and bowel movement is questionable (9). Others have advocated that time to first bowel movement is a more reliable and sensitive endpoint indicating the return of bowel motility (10). Studies have strongly suggested that early oral feeding is safe and tolerated well by the majority of patients (11, 22); however, we waited for the resolution of ileus to initiate the feeding. This probably resulted in delayed oral intake and hospital discharge, and obscured the difference between the groups with or without ketorolac.
Although many studies have shown that epidural analgesia with local anesthetics or combination of local anesthetics and opioids is superior to intra-venous PCA in the earlier recovery of postoperative bowel function (23, 24), epidural catheterization does carry certain risks due to its invasiveness, e.g. motor blockade, postdural puncture headache, nerve injury, and meningitis. The inhibitory effect on bowel moti-lity is not completely avoided with epidural analgesia (25). On the other hand, intravenous PCA has the benefits of convenience, less technical difficulties, and less potential risks. In Asia, especially for the Chinese, intravenous PCA is more acceptable because of the incorrigible fear of potential neurological sequelae following spinal interventions like spinal surgeries, spinal tapping, or injections. Besides, the addition of adjuncts in intravenous morphine could ameliorate some of the opioid-related adverse effects like somnolence, nausea and vomiting as well as enhance analgesia (8, 18).
In conclusion, this randomized, double-blind study reports that adding ketorolac in intravenous morphine PCA provides an opioid-sparing effect but only mildly shortens the duration of postoperative bowel immo-bility in colorectal surgery patients. Postoperative ileus appears to be a multifactorial problem, and
opioid-induced impairment of bowel function might be overemphasized. A multimodal postoperative rehabilitation program aiming at multiple mechan-isms affecting postoperative ileus would improve the early restoration of normal bowel functions.
References
1. Livingston EH, Passaro EP Jr. Postoperative ileus. Digest Dis Sci1990; 35: 121—32.
2. Spiller CM. Mechanisms of postoperative intestinal motor dysfunction. Curr Opin Gastroenterol 2003; 19: 103—5. 3. Moss G, Regal ME, Lichtig L. Reducing postoperative pain,
narcotics, and length of hospitalization. Surgery 1986; 99: 206—10. 4. Ferraz AA, Cowles VE, Condon RE et al. Nonopioid analge-sics shorten the duration of postoperative ileus. Am Surgeon 1995; 61: 1079—83.
5. Kehlet H, Holte K. Review of postoperative ileus. Am J Surg 2001; 182: S3—S10.
6. Friedman JD, Dello Buono FA. Opioid antagonists in the treatment of opioid-induced constipation and pruritus. Ann Pharmacother2001; 35: 85—91.
7. Graber JN, Schulte WJ, Condon RE, Cowles VE. Relationship of duration of postoperative ileus to extent and site of operative dissection. Surgery 1982; 92: 87—92.
8. Pavy TJ, Paech MJ, Evans SF. The effect of intravenous ketorolac on opioid requirement and pain after cesarean delivery. Anesth Analg 2001; 92: 1010—4.
9. Waldhausen JH, Shaffrey ME, Skenderis BS, Jones RS, Schirmer BD. Gastrointestinal myoelectric and clinical patterns of recovery after laparotomy. Ann Surg 1990; 211: 777—84. 10. Luckey A, Livingston E, Tache Y. Mechanisms and treatment
of postoperative ileus. Arch Surg 2003; 138: 206—14.
11. Reissman P, Teoh TA, Cohen SM, Weiss EG, Nogueras JJ, Wexner SD. Is early oral feeding safe after elective colorectal surgery? A prospective randomized trial. Ann Surg 1995; 222: 73—7.
12. Cali RL, Meade PG, Swanson MS, Freeman C. Effect of morphine and incision length on bowel function after colectomy. Dis Colon Rectum 2000; 43: 163—8.
13. Kaufman PN, Krevsky B, Malmud LS et al. Role of opiate receptors in the regulation of colonic transit. Gastroenterology 1988; 94: 1351—6.
14. Chan KC, Cheng YJ, Huang GT et al. The effect of IVPCA morphine on post-hysterectomy bowel function. Acta Anaesth Sinica2002; 40: 61—4.
15. Holte K, Kehlet H. Postoperative ileus: progress towards effective management. Drugs 2002; 62: 2603—15.
16. Pasero C. Epidural Analgesia for Postoperative Pain, Part 2: Multimodal recovery programs improve patient outcomes. Am J Nurs2003; 103: 43—5.
17. Basse L, Billesbolle P, Kehlet H. Early recovery after abdominal rectopexy with multimodal rehabilitation. Dis Colon Rectum2002; 45: 195—9.
18. Picard P, Bazin JE, Conio N, Ruiz F, Schoeffler P. Ketorolac potentiates morphine in postoperative patient-controlled analgesia. Pain 1997; 73: 401—6.
19. O’Hara DA, Fanciullo G, Hubbard L et al. Evaluation of the safety and efficacy of ketorolac versus morphine by patient-controlled analgesia for postoperative pain. Pharmacotherapy1997; 17: 891—9.
Table 4
Side-effects of patient-controlled analgesia in the two groups. Group Mild Moderate Severe Total (%)
Pruritus M 6 3 1 10 (28.5) K 8 2 0 10 (25.6) Nausea or vomiting M 7 1 1 9 (25.7) K 6 1 0 7 (17.9) Dizziness M 4 0 0 4 (11.4) K 4 0 0 4 (10.2)
20. Cataldo PA, Senagore AJ, Kilbride MJ. Ketorolac and patient controlled analgesia in the treatment of postoperative pain. Surg Gynecol Obstet1993; 176: 435—8.
21. Yukioka H, Bogod DG, Rosen M. Recovery of bowel motility after surgery. Detection of time of first flatus from carbon dioxide concentration and patient estimate after nalbuphine and placebo. Br J Anaesth 1987; 59: 581—4.
22. Di Fronzo LA, Cymerman J, O’Connell TX. Factors affecting early postoperative feeding following elective open colon resection. Arch Surg 1999; 134: 941—5.
23. Carli F, Trudel JL, Belliveau P. The effect of intra-operative thoracic epidural anesthesia and postintra-operative analgesia on bowel function after colorectal surgery: a prospective, randomized trial. Dis Colon Rectum 2001; 44: 1083—9.
24. Steinberg RB, Liu SS, Wu CL et al. Comparison of ropiva-caine-fentanyl patient-controlled epidural analgesia with
morphine intravenous patient-controlled analgesia for perioperative analgesia and recovery after open colon surgery. J Clin Anesth 2002; 14: 571—7.
25. Brix-Christensen V, Tonnesen E, Sanchez RG, Bilfinger TV, Stefano GB. Endogenous morphine levels increase following cardiac surgery as part of the antiinflammatory response? Int J Cardiol1997; 62: 191—7.
Address:
Yeong-Ray Wen, MD
Department of Anesthesiology
Shin Kong Wu Ho-Su Memorial Hospital 95, Wen-Chung Road
Shih-Lin, 111, Taipei Taiwan
