| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
*Department of Anesthesiology and Pain Medicine, University of Alberta Hospital, University of Alberta, Edmonton, Alberta, Canada; Department of Anesthesia, London Health Sciences Centre –University Campus Site, University of Western Ontario, London, Ontario, Canada; Department of Anesthesia, Montreal General Hospital, McGill University, Montreal, Quebec, Canada; Department of Anesthesia, Grace General Hospital, Memorial University, St. John’s, Newfoundland; ||Department of Anesthesia, University of Manitoba, Health Sciences Centre, Winnipeg, Manitoba, Canada; ¶Department of Anesthesia, New Halifax Infirmary, Queen Elizabeth II Health Sciences Centre, Halifax, Nova Scotia; #Astra Pain Control, Södertälje, Sweden
Address correspondence to B. T. Finucane, MD, BCh, FRCA, FRCPC, Department of Anesthesiology and Pain Medicine, 3B2.32 Walter C Mackenzie Health Sciences Centre, University of Alberta Hospital, 8440-112 St., Edmonton, Alberta, Canada, T6G 2B7. Address e-mail to bfinucan{at}ualberta.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Implications: Epidural infusions of ropivacaine with and without fentanyl were administered to patients to control pain after colonic surgery. Patients who received ropivacaine with fentanyl had better pain control, increased side effects, and delayed readiness to discharge. This study questions the value of adding opioids to epidural infusions of local anesthetics.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The main purpose of this study was to evaluate the safety and efficacy of a 72-h epidural infusion of ropivacaine and to measure the impact of adding fentanyl, 2 µg/mL, to ropivacaine on required infusion rates, pain scores, side effects, length of stay, and resource utilization in patients after colonic surgery.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Healthy patients, (ASA physical status I–III) scheduled for colonic surgery, aged between 18 and 79 yr, weighing between 50 and 110 kg, who were at least 152 cm in height, were considered eligible for the study. Patients with a history of peptic ulcer disease or major organ failure and those dependent on drugs or alcohol were not considered eligible candidates for the study. Patients who were not considered suitable candidates for epidural anesthesia were also excluded.
Patients were randomly assigned, in balanced blocks (1:1) of 10 patients in each center, in a double-blinded manner, to receive a continuous epidural infusion of ropivacaine (2 mg/mL) with (R) or without fentanyl (2 µg/mL) (R + F).
The infusion bags were identical, and the randomization procedure and distribution of infusion bags was handled by the pharmacy department in each medical center participating. The solutions were prepared in double-blinded fashion by Astra Pain Control, Södertälje, Sweden.
Fentanyl (up to 100 µg) and midazolam (up to 2 mg), were administered IV as needed before the procedure. An epidural catheter was inserted between T8 and T11, with the patient in the sitting position, by using a standard protocol. Catheters were advanced 4–6 cm into the epidural space and taped securely. A 3-mL test dose, consisting of lidocaine 1.5% with epinephrine 1:200,000, was administered to eliminate intravascular or intrathecal epidural catheter placement. If there were no signs of intravascular on intrathecal injection, 5 mL of ropivacaine (7.5 mg/mL) was administered epidurally for 1 min. The spread of sensory analgesia was assessed by using the skin prick method, and a minimum sensory spread from T6 to L1 was required bilaterally for the patient to remain in the study. Two additional 5-mL increments were injected at 15-min intervals, if an adequate sensory block was not achieved. If an adequate sensory block was not achieved within 45 min, the patient was withdrawn from the study and an alternate method of pain control was chosen at the discretion of the anesthesiologist.
A standard general anesthetic was administered consisting of thiopental, fentanyl, nitrous oxide, oxygen, isoflurane, and vecuronium or pancuronium for endotracheal intubation. Fentanyl (up to 200 µg) was given intraoperatively at the discretion of the anesthesiologist. Systemic opioids were not administered postoperatively while patients remained in the study.
The epidural infusion (consisting of ropivacaine 2 mg/mL with or without fentanyl 2 µg/mL) was started at an initial rate of 8 mL/h within 1 h of the induction of general anesthesia and was continued for 72 h after arrival in the postanesthesia care unit (PACU), which was considered "time zero." The initial infusion rate was changed according to the following protocol: if a patient complained of inadequate pain relief at any time, the infusion rate was increased, provided it remained constant for 30 min. Each dose adjustment consisted of a 4-mL top up followed by a 2-mL/h increase in the infusion rate, up to a maximum of 14 mL/h. If the maximum infusion rate of 14 mL/h was reached and pain relief was still inadequate, the infusion was terminated and an alternative treatment plan was implemented at the discretion of the anesthesiologist The infusion rate was decreased or discontinued temporarily if there was evidence of excessive blockade (above T2) or unacceptable side effects (hypotension on bradycardia requiring treatment). The infusion rate was recommenced at a rate of 2 mL/h less than the previous rate, to a minimum of 4 mL/h. All patients received IM injections of ketorolac 30 mg on arrival in the PACU and thereafter 15 mg every 8 h for 72 h. Prophylactic antiemetic treatments were not permitted.
Spread of the sensory blockade was assessed by using the pin prick method, and the degree of motor blockade was assessed by using a modified Bromage scale (0 = no motor block and 3 = maximum block) at regular intervals. Assessments of pain were performed at 1, 2, 4, 6, and 8 h from time zero and thereafter every 4 h up to 72 h (no assessments were performed during the night between 22:00 and 08:00 h to allow patients to sleep). Pain at rest, on coughing, and on mobilization was assessed by the patients using a 100-mm visual analog scale (VAS) ruler consisting of a 100-mm ungraduated scale with the ends marked "no pain" (0 mm) and "worst pain" (100 mm).
Patients were encouraged to ambulate twice daily (once AM and once PM) beginning the day after surgery, provided the Bromage score was 0. The degree of mobility was defined as follows: able to rise from a lying to sitting position, able to sit with legs outside the bed, able to walk with assistance from bed to chair, able to walk with assistance for 5 m within the ward. The first time all mobilization steps were completed was recorded.
The quality of pain management was rated by the patients as excellent, good, fair, or poor. When patients complained of inadequate pain relief, an adjustment was made to the infusion by the nurse observer following a standard protocol.
Discomfort during postoperative recovery was evaluated by using a self-assessment questionnaire analyzing 16 symptoms associated with postoperative recovery, including itching, difficulty concentrating, poor appetite, difficulty urinating, and pain when moving around (see Appendix 1). This graded discomfort ranging from "no discomfort at all" to "very severe discomfort," and was based on the Gastrointestinal Symptom Scale (4). Readiness to discharge from the hospital was checked twice daily beginning on the fourth postoperative day by using the following criteria:
30 mm), Safety variables such as respiratory rate, degree of sedation, pulse rate, blood pressure, peripheral oxygen saturation, and body temperature were closely monitored during the study. Routine laboratory tests were performed throughout the study. All adverse events were recorded during hospitalization and included a telephone follow up 3–4 wk after surgery.
All adverse and severe adverse events (SAEs) were recorded and compared in the two groups. An adverse event was defined as any unintended or unfavorable clinical symptom or sign reported during the patient’s hospital admission whether it was related to the treatment or not. A SAE was by definition any complication which resulted in death, permanent disability, prolongation of hospitalization, or required medical or surgical intervention to prevent permanent damage to an organ system, whether it was treatment related or not. Investigators were instructed to report all SAEs to a central office in Canada within 96 h.
A comparison of hospital costs for the two treatment regimens was made based on a patient’s length of stay in the following care units: intensive care unit (ICU) and step-down, and hospital ward. The cost of PACU was valued per case. The per diem hospital cost is based on data from the Ontario Case Cost Project (5). Direct costs included all those directly related to patient care such as nursing time, drugs, and medical supplies, but not physician fees. Indirect costs were based on a portion of the hospital’s overhead, such as maintenance, laundry and linen costs, patient meals, and administration. Length of stay for the acute inpatient hospital visit costs were calculated in Canadian dollars.
The primary efficacy variable was defined as the mean epidural infusion rate per hour over the 72-h postoperative period. Secondary efficacy variables included assessments of pain, degree of mobility, safety variables, adverse events, readiness to discharge, and cost analysis.
The sample size of at least 150 patients with equal numbers in each group, was not based on statistical considerations because no variability information was available about the primary efficacy variable. However, with 75 patients in each group, the size of a 95% confidence interval for the mean difference between two groups, is at most, 1.4 with a power of 90%, assuming that there was a standard deviation of ±2 in the primary efficacy variable.
Data were presented as medians followed by lower (Q1) and upper (Q3) quartiles in brackets or as mean ± SD. If applicable, comparisons of groups were performed by using a stratified Wilcoxon’s (mid) ranked sum test adjusting for centers (6). All P values were two-tailed and based on a normal approximation. The two smallest centers were pooled in these analyses. The SAS system, version 6.12 from SAS Institute Inc. (Cary, NC), was used for the statistical analysis.
For some repeated assessments, a summary measure was calculated as follows. First, the area under the curve (AUC) based on the repeated measurements was calculated by using the trapezoidal rule (7). The summary measure considered was defined as the AUC divided by the length of the time period on which it was based, so that it had the same scale as the underlying repeated measurements. This summary measure was denoted AUCM.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Therefore, 151 patients (79 in the R and 72 in the R + F group) were eligible for efficacy and safety evaluation. The study was terminated after arrival at the PACU in four additional patients, three in the R group and one in the R + F group, because the surgical procedure was changed (one patient in each group), a history of peptic ulcer disease was revealed in one patient, and one patient was not monitored after the test dose. The remaining 147 patients (76 in the R group and 71 in the R + F group) received the continuous postoperative epidural infusion of the study medication and were included in the efficacy analysis. The number of evaluable patients per center ranged between 11 and 35.
Demographic and procedural data were comparable in both groups ( Table 1). The mean duration of surgery and blood loss were similar in both groups.
|
|
|
|
|
|
Median times to first micturition were similar in both groups (90.0 vs 89.7 h). Median time to first flatus was shorter in the R group (33.8 vs 44 h). Neither of these differences was significant.
The median time to full mobilization, i.e., the time when the patient was able to walk 5 m with assistance, was similar in both treatment groups (23 and 22 h, respectively). The number of patients who reported moderate to severe discomfort on ambulation was larger in the R group during postoperative Days 1, 2, and 3 (34% vs 27%, 37% vs 17%, and 25% vs 18%).
Hypotension (defined as low blood pressure requiring vasopressor therapy) occurred more frequently in the R + F group (P < 0.01). Pruritus (defined as itchiness reported by the patient) occurring at any time during the 72-h infusion period, was reported more frequently in the opioid-containing group (20 patients in the R + F group vs 5 in the R group), and this difference was significant (P < 0.0001).
There were more SAEs in the R + F group. SAEs from both groups are presented in Table 4. Eighteen patients in the R group experienced SAEs and 33 patients in the R + F group experienced 90 SAEs. There was one convulsion in a patient on the 17th postoperative day.
|
|
One patient in the R + F group spent 20.9 days in the ICU and 47.9 days in the hospital. Excluding this outlier from the cost analysis, the difference in mean total cost per patient was $567 in favor of the R group.
Laboratory variables did not reveal any relevant differences between the treatment groups.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
There is little dispute among anesthesiologists about the value of adding opioids to epidural infusions other than the question of which opioid to use and how much.
Ropivacaine 2 mg/mL has been infused epidurally in a number of previous studies for shorter intervals (12–15), and this is one of the first studies to evaluate ropivacaine for prolonged infusions.
The results in this study confirmed the salutary effects of adding fentanyl to ropivacaine on the quality of analgesia. However, VAS scores in the nonopioid-containing group were satisfactory both at rest and also on coughing.
One may question the logic of using the infusion rate per hour as the primary efficacy variable, especially with the knowledge that there was a limit on the maximum allowable infusion rate (14 mL/h). Furthermore, the ability to reduce the infusion was determined only by side effects (excessive spread or hypotension). The starting rate of the infusion was 8 mL/h. A previous study showed that the minimum effective infusion rate/hour of ropivacaine 2 mg/mL was 10 mL/h for upper abdominal surgery (15). Ideally, the patient-controlled epidural anesthesia modality should have been used to determine the primary efficacy variable. However, the patient-controlled epidural anesthesia modality was relatively new when this study was designed and is still not widely used. It would have been difficult to educate anesthesiologists and nurses to make this change in six major medical centers and complete the study within a reasonable time frame.
Hypotension occurred more frequently in the R + F group, even though infusion rates were less in that group. The mechanism of epidural opioid-induced hypotension is not known but may be the result of an intrinsic effect on spinal cord receptors. Pruritus occurred more often in the R + F group, as anticipated. Readiness to discharge was significantly delayed in the R + F group and it would be difficult not to link this delay in satisfying discharge criteria, to the increased occurrence of serious adverse events reported in the R + F group.
The preponderance of SAEs reported in the opioid-containing group was an unexpected finding, and this is the first study to make such an observation. The majority of SAEs involved the gastrointestinal (GI) tract (>40%) and favored the opioid group by a factor of 5. Major differences were also observed in the cardiovascular, respiratory, and genitourinary systems. (Table 5) The mechanism of bowel dysfunction after abdominal surgery is not fully understood. Current theory suggests that nociceptive impulses from the bowel reach the spinal cord via afferent fibers. The efferent limb of this spinal reflex arc transmits inhibitory impulses to the bowel leading to ileus (16). Increased sympathetic activity from pain and surgical stress further impairs bowel motility, and these effects are most pronounced after colonic surgery (17,18). The use of systemic and epidural opioids to control pain also interferes with bowel motility (19). Patients in the opioid-containing group received approximately 500 µg of fentanyl every 24 hours for up to three days. Systemic absorption of fentanyl in that quantity (500 µg/24 h) would be sufficient to have a direct inhibitory effect on bowel motility. Epidural opioids also inhibit bowel motility via a spinal mechanism (20,21). Finally, animal studies have shown a supraspinal mechanism of inhibition of bowel motility which would have more relevance to hydrophilic opioids (22). However, prolonged epidural infusions of lipophilic opioids may also reach the supraspinal cerebrospinal fluid. The salutary effects of thoracic epidural local anesthetics on bowel function are well established (23,24). Epidural anesthesia blocks both limbs of the inhibitory spinal reflex arc, reduces sympathetic tone (leaving the parasympathetic fibers to the bowel to function unimpaired), and improves blood flow to the bowel. Finally, systemic absorption of local anesthetics also blocks the inhibitory reflex arc to the bowel by a direct effect.
The frequent incidence of SAEs in the respiratory, cardiovascular, and genitourinary systems in the R + F group were also of concern in this study. We did not have clear-cut definitions for all of the side effects reported in this study. However, the definitions of adverse events and SAEs were clearly defined and were reported in double-blinded fashion.
Ideally, we should have standardized perioperative surgical care (e.g., nasogastric tubes, feeding schedules, fluid regimes and definitions); however, it would have been very difficult to get compliance from several colorectal surgeons in six medical centers across Canada. We did establish clear-cut guidelines for pain control, mobilization, readiness to discharge, and adverse events and SAEs. Length of stay is not necessarily an accurate measure of a patient’s fitness for discharge. There are a number of reasons why patients remain in hospital even though the discharge criteria have been met (e.g., travel arrangements, social factors, home care, prescriptions, and general inefficiency). Readiness to discharge was delayed by 1.1 days in the opioid-containing group and that difference was significant. Time to actual discharge was also delayed by one day in this group, but the results were not statistically significant.
Readiness to discharge, actual length of hospital stay, and cost are very important issues in the managed care environment. This is one of the few studies to report an in-depth analysis and cost comparison of different treatment modalities in the Canadian health system. The cost analysis showed savings of $567.00 in the R group; however, this result did not reach statistical significance. Post hoc analysis revealed that the sample size was too small and, therefore, had inadequate power to demonstrate a difference. Twice as many patients would have been required to have sufficient power to demonstrate an economic benefit.
There are very few studies in the literature that compare outcomes after epidural infusions of local anesthetics alone and combinations of local anesthetics and opioids. One such study is that reported by Liu et al. (25) who compared four different postoperative pain regimens on the quality of analgesia, restoration of bowel function, and readiness to discharge after colonic surgery. To understand the relevance of that study, we must briefly describe some aspects of the methodology. The four groups were: epidural infusion of bupivacaine (B) and bupivacaine plus epidural morphine (MB), epidural infusion of morphine alone (M), and PCA morphine (P).
The groups that had epidural infusions of local anesthetics had the best analgesia, had a shorter time to first flatus, and satisfied the discharge criteria much sooner. There was no difference in outcomes when the epidural morphine group was compared with the PCA morphine group. The group that received epidural bupivacaine alone (B) had an unacceptably frequent incidence of orthostatic hypotension (57% vs 17%). The authors concluded that the best results were obtained in the MB group because they had the best balance between positive and negative outcomes. In contrast to Liu et al.’s (25), findings we observed that hypotension occurred more frequently in the opioid-containing group and readiness to discharge was significantly delayed in that group. We also showed an increased incidence of SAEs in many organ systems but particularly the GI tract. It is also worth noting that in Liu et al.’s (25) study, the mean infusion rate per hour in the bupivacaine group alone (B) was 21.6 mL, which was more than twice the infusion rate observed in the MB group. In the current study, the infusion rate per hour differed by only 2.2 mL. Perhaps these differences can be explained by the different drug combinations used in the two studies (ropivacaine ± fentanyl versus bupivacaine ± morphine).
A recent study by Berti et al. (26) also questioned the advantage of adding fentanyl to epidural infusions of for pain control after abdominal surgery. They also showed that the fentanyl-containing group had lower oxygen saturations at 12, 24, and 48 hours. There were no differences in GI side effects in that study.
The most relevant study to compare with the current study is that reported by Scott et al. (27), the basic design of which was similar to the current study. Major differences were as follows: the current study was conducted on patients presenting for a single surgical procedure (colectomy). Scott et al.’s (27) study included patients presenting for a variety of abdominal procedures, and there were four treatment groups, with patients receiving a ropivacaine infusion alone or with 1, 2, or 4 µg/mL fentanyl. Finally, the nonsteroidal antiinflammatory drug used by Scott et al. (27) was acetaminophen. There were some notable differences in the results of the two studies. The attrition rate from Scott et al.’s (27) study, more than 72 hours in comparable groups, was remarkably larger than that observed in this study (51% vs 33% in the R group and 44% vs 24% in the R + F group). These differences may be a reflection of the differences in study design. Scott et al. (27) concluded that fentanyl 4 µg/mL, when added to ropivacaine 2 mg/mL, provided the best pain relief and the least attrition from the study. Details about serious adverse events were not included in Scott et al.’s (27) study. There were also some comparative results in these two studies. Fentanyl-containing groups reported the best pain control in both studies; however, pain control was acceptable when fentanyl was not added to ropivacaine. The occurrence of hypotension was linked with fentanyl in both studies. Median length of stay was extended in both studies in comparable groups (8 days in both R groups and 9 days vs 10 days in R + F group), and although these results were not statistically significant, there was a tendency toward increased length of stay in fentanyl-containing groups.
In summary, this is the first major, randomized, double-blinded study to question the routine use of epidural opioids in combination with local anesthetics to control postoperative pain. It would be difficult not to link the epidural opioid with the preponderance of SAEs reported in that group. What other factors could explain this disparity? There is a possibility that there is some confounding factor in the R + F group that we are not aware of. Perhaps there are insufficient data in this study to invoke a change in practice at this time; however, there is sufficient evidence to design additional double-blinded studies with this and other combinations of opioids and local anesthetics, to corroborate this unexpected finding. In the meantime, we must question the practice of routinely adding opioids to epidural infusions to control postoperative pain, especially when adequate pain relief can be achieved by using epidural infusions of local anesthetics without opioids, or we must at least minimize the amount of opioid added to epidural infusions of local anesthetics.
In conclusion, prolonged epidural infusions of ropivacaine appear to be safe and effective. The addition of fentanyl to ropivacaine enhanced the quality of pain relief, but side effects were increased and readiness to discharge was delayed. Further studies are required to evaluate the role of adding epidural opioids to local anesthetic infusions in major abdominal surgery.
|
|
Acknowledgments |
|---|
Canadian Ropivacaine Research Group: Mike Bautista, MD, FRCPC, Rob Forward, MD, FRCPC, Sunil Gupta, MD, FRCPC, Urban Gustafsson, PhD, Donald Jolly, MD, FRCPC, Holly Muir, MD, FRCPC, Steve Kowalski, MD, Jamie McKishnie, MD, FRCPC, Eric Patz, Ban Tsui, MD, FRCPC, Peter Wildgrube, MD, FRCPC, Tim Yeh, MD, FRCPC. We would also like to acknowledge the following individuals who coordinated research studies at the various medical centers: Deb Chalupa, RN, Angele Diepen, Joanne Morgan, RN, Maggie Rossiter, RN. We would like to acknowledge to cooperation of the surgeons and the operating room nursing staff at each of the medical centers involved. Finally, we would like to acknowledge Cynthia Lewis and Marilyn Blake who helped coordinate completion of the final manuscript.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  X. Hong, G. Mistraletti, S. Zandi, B. Stein, P. Charlebois, and F. Carli Laparoscopy for colectomy accelerates restoration of bowel function when using patient controlled analgesia: [La laparoscopie pour colectomie accelere la restauration de la fonction intestinale quand on utilise l'analgesie autocontrolee]. Can J Anesth, June 1, 2006; 53(6): 544 - 550. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Bonnet and E. Marret Influence of anaesthetic and analgesic techniques on outcome after surgery Br. J. Anaesth., July 1, 2005; 95(1): 52 - 58. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Niiyama, T. Kawamata, H. Shimizu, K. Omote, and A. Namiki The addition of epidural morphine to ropivacaine improves epidural analgesia after lower abdominal surgery: [L'addition de morphine peridurale a la ropivacaine ameliore l'analgesie peridurale apres une intervention chirurgicale abdominale basse] Can J Anesth, February 1, 2005; 52(2): 181 - 185. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Niemi and H. Breivik Epinephrine Markedly Improves Thoracic Epidural Analgesia Produced by a Small-Dose Infusion of Ropivacaine, Fentanyl, and Epinephrine After Major Thoracic or Abdominal Surgery: A Randomized, Double-Blinded Crossover Study With and Without Epinephrine Anesth. Analg., June 1, 2002; 94(6): 1598 - 1605. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
