A Comparison of Multimodal Perioperative Analgesia to Epidural Pain Management After Gastric Bypass Surgery

Abstract

We compared pain intensity, analgesic consumption, patient satisfaction, and length of stay in 114 patients undergoing gastric bypass surgery under general anesthesia. Patients were randomized to incisional local anesthetic infiltration plus postoperative patient-controlled analgesia (Group A), epidural anesthesia and analgesia (Group B), or postoperative patient-controlled analgesia (Group C). All received perioperative nonsteroidal antiinflammatory drugs. Age, sex, body mass index, length of stay, and patient satisfaction were equivalent in all groups. Pain at time 0 and 36 h was the smallest in Group B, greater in Group A, and greatest in Group C. Pain scores in a subset of Group A were lower at all times than in Groups B and C, but this difference was significant only at 0, 12, and 36 h. In responders, infiltration analgesia as part of a multimodal regimen offers a simple, safe, and inexpensive alternative to epidural pain control.

The recent adoption of pain assessment and management standards by the Joint Commission on Accreditation of Healthcare Organizations ⁽¹⁾ follows at least a decade of growing recognition of the benefits of acute pain control ⁽²⁾ on short-term outcomes, patient satisfaction, quality of life, and, in some studies, the development of chronic pain syndromes ⁽³⁾. A particular challenge in implementing such standards and to collaborative practice among surgeons, anesthesiologists, nurses, pharmacists, and others, is the care of obese patients. These patients are at increased risk of developing postoperative thromboembolic, pulmonary, and other complications, and the risk for such impediments can be reduced by regional anesthetic techniques, such as epidural analgesia or intercostal nerve block ⁽⁴⁾. Hence, epidural analgesia has become a common practice for patients undergoing gastric bypass. However, placement of the epidural catheter in obese patients can be time-consuming and technically difficult. There is growing interest for less demanding modes of perioperative analgesia such as wound infiltration with local anesthetic combined with systemic medications to achieve multimodal analgesia.

The term multimodal analgesia refers to the simultaneous use of multiple analgesic methods or drugs. Because acute pain is an integrated process that is mediated by activation of numerous biochemical and anatomical pathways ⁽⁵⁾, a multimodal approach typically deploys interventions such as local anesthesia, a nonsteroidal antiinflammatory drug, or an opioid so as to achieve combination analgesic chemotherapy ⁽⁶⁾. Most published trials dealing with surgical outcome in relation to postoperative pain control use unimodal treatment such as an opioid. Using a unimodal method, it may be difficult to attain adequate analgesia without inhibiting normal functions (e.g., movement or gastrointestinal [GI] function) or producing other side effects (e.g., sedation or nausea) ⁽⁷⁾ related to large-dose requirements. Preemptive analgesia (or preemptive antinociception)—a term that refers to the concept that blocking neuronal pathways before or during injury (surgery)—can reduce or eliminate the hyperexcitability of these pathways and pain memory during recovery. Preemptive analgesia can be achieved by infiltration of local anesthetic into surgical wounds. This technique improves postoperative pain relief and decreases the total opioid requirement, thus reducing postoperative nausea and vomiting ⁽⁸⁾.

In one study ⁽⁹⁾, it was indicated that multimodal analgesia in patients undergoing laparoscopic cholecystectomy is highly effective, resulting in faster recovery and discharge. Because pain from surgery has three major components ⁽¹⁰⁾ (tissue injury, nociceptor stimulation, and activation of central pathways), it is rational to think that a multimodal approach to analgesia based upon simple, low-tech interventions, such as preincisional infiltration with local anesthetic, supplemented by repeat incisional infiltration at the end of the operation, may be effective. The development of local anesthetics like levobupivacaine that permit larger total doses to be given safely extends opportunities for infiltration analgesia and suggests that such approaches could play an increasing role in perioperative regional anesthesia.

Therefore, the primary objective of this study was to incorporate multimodal analgesia including preemptive antinociception into the analgesic regimen of patients undergoing open gastric bypass surgery. We assessed several outcomes to determine whether current techniques for analgesia can be superseded by a simple and inexpensive approach.

Methods

After institutional review and approval of the protocol, all patients provided written informed consent. Patients aged 18–80 yr, scheduled for gastric bypass surgery as a treatment for obesity, and who were judged by their surgeon and anesthesiologist to be appropriate candidates for randomization to any of the analgesia regimens outlined were included. Obesity was defined as a body mass index (kg/m²) more than or equal to 30. Patients with significant cardiovascular, hepatic, pulmonary, renal, hematological, neurologic, or psychiatric disease were excluded from participation in the study. Patients with a known hypersensitivity to any of the study drugs, a history of drug or alcohol abuse within the previous year, preexisting chronic or acute pain, or with previous abdominal surgery or any surgery in the previous 3 mo were also excluded from the study.

We enrolled 114 patients, randomized into one of three treatment groups discussed below. Random allocation of patients to different treatment groups was performed according to methods previously described in detail ^(11,12). Successive envelopes containing allocation codes were opened by one investigator (J.M.W.) when patients were in the preoperative holding area, after consent had been obtained, but before any anesthetic or surgical interventions. Participants in this study were provided with standard hospital care during their stay. Participation in this study did not change their unit/floor location or any other aspect of their care.

This was a randomized, comparative, open-label, single-site study. After screening, patients were randomized to one of three groups:

Group A (multimodal group) received preincision infiltration and postsurgical, preextubation supplemental field infiltration with bupivacaine. A total of 40–45 mL of 0.25% bupivacaine with epinephrine 1:200,000 was infiltrated by the surgeon before the incision, supplemented by an equal amount of the same solution before extubation. Postoperative analgesia was provided using a standard order sheet (systemic morphine via patient-controlled analgesia, PCA). In the postanesthesia care unit (PACU), patients received initial loading doses of 2.5 mg of morphine IV every 5 min to a maximum of 10 mg as required for initial pain control. Subsequent PCA orders allowed for 2.5 mg IV of morphine per PCA dose up to every 8 min with no basal (continuous) infusion and no 4-h limit. In addition to the PCA, up to 3 rescue doses of IV morphine 2.5 mg could be given as required within a 15-min interval every 8 h.

Group B received a thoracic epidural catheter placed according to our standard practice. Local anesthetic was administered via the epidural catheter during surgery. Postoperatively, epidural infusions were started with a combined mixture of meperidine 1 mg/mL and bupivacaine 0.1% and titrated to a maximal infusion rate of 14 mL/h.

Group C received systemic analgesia by morphine PCA. The rescue dose range of opioid was the same as in Group A. No local anesthetic wound infiltration was given. Because saline or water given into soft tissue may alter nociception, as may dry needling, we did not use placebo injections in this group.

All groups underwent a standardized general anesthetic in conjunction with the therapies outlined above. This consisted of premedication with midazolam 1 mg every 5 min up to a maximum of 5 mg before the induction, propofol 2 mg/kg for the induction, and fentanyl up to 10 μg/kg during the entire operation. The latter drug was given to maintain stability of vital signs according to the usual clinical practice of the attending staff anesthesiologist. Anesthesia was maintained during surgery with isoflurane 0.8%–2% in N₂O/O₂. Nonsteroidal antiinflammatory drug treatment consisted of ketorolac 30 mg four times a day every 6 h (first dose given before extubation) until the patient was able to take oral medications and then ibuprofen 600 mg four times a day every 6 h orally through postoperative Day 5. On discharge, patients received oxycodone 5 mg/acetaminophen 325 mg (Percocet) one to two tablets four times a day as required.

All patients underwent the same surgical procedure performed by a single surgeon (S.S.). A vertical upper midline incision was made with a scalpel through the skin. The subcutaneous fat was then pulled apart and the fascia divided down the midline with electrocautery. A standard Roux-en-Y gastric bypass was created. After completion of the procedure, the fascia was closed with a heavy (#2) Nylon full thickness running suture. The skin was reapproximated with skin clips. The duration of surgery was approximately 2 h.

Efficacy and safety assessments were made throughout the postoperative study period and adverse events recorded. After hospital discharge, all patients were contacted within a week to assess overall status.

The primary end-point of this study was to assess the analgesic efficacy of a multimodal approach compared with the standard unimodal approaches in patients undergoing gastric bypass surgery. To do so, we assessed visual analog scale (VAS; 0–10) pain scores every 6 h at rest, postoperative morphine consumption in the two groups (A and C) treated with PCA, patient satisfaction, and length of stay. We also monitored intraoperative fentanyl requirements in the three groups.

We evaluated the normality of the distribution of each variable by means of histograms and probability plots. Statistical analysis of the data was undertaken with both parametric and nonparametric methods. For those tests in which the comparison groups were normally distributed, we applied analysis of variance and χ² analysis. In comparisons where normality could not be assumed, the Kruskal-Wallis and Mann-Whitney tests were used.

Results

All three treatment groups were equivalent with respect to the demographic characteristics of age, sex, and body mass index (P = 0.971, 0.750, and 0.806, respectively) (Table 1). Decreases in the number of patients contributing data points across time were not because of additional dropouts but instead reflected incomplete capture or documentation of VAS scores by ward nursing staff (80% of subjects at time 0 versus 51% at 48 h).

Table 1: Clinical Characteristics of the Three Treatment Groups (mean ± sd), with corresponding P values by ANOVA and χ² analysis

Significant adverse events were equivalent in all three groups, as were length of stay (P = 0.529) and patient satisfaction (P = 0.790). The epidural failure rate was 27.8% (10 of 36). Three of these failures were preoperative, and seven were postoperative. Three of the 39 patients randomized to the epidural group (Group B) were dropped because of the epidural catheter not being placed, an intraoperative significant adverse event, and one of the cases being a revision. Additionally, two patients from the multimodal group (Group A) and four patients from the PCA group (Group C) were dropped from the study because of unexpected intra- or perioperative complications not attributable to the analgesic regimen (e.g., postoperative GI hemorrhage and emergent surgical reexploration). Patients in Groups A and C were significantly more likely to experience nausea in the postoperative period than those in Group B (n = 11, 11, and 5, respectively;P < 0.05). However, there was no significant difference in the experience of pruritus between Groups A, B, and C (n = 3, 1, and 2, respectively;P > 0.1).

VAS pain intensity scores between groups differed in the PACU (time 0;P = 0.001) and at 36 h after surgery (P = 0.011) because of lower scores in Group B (Fig. 1). The VAS scores for Group A at time 0 are bimodally distributed (Fig. 2). The subgroup of Group A with apparently successful infiltration experienced lower VAS pain intensity scores than the other subgroup and both Groups B and C at times 0, 12, and 36 h (P < 0.001, P = 0.020, and P = 0.011, respectively) (Fig. 3). This finding suggests that one subset of patients in Group A responded favorably to local anesthetic field infiltration, whereas the other subset did not respond at all. This bimodal distribution was consistently found with respect to analgesic requirement (see below).

Figure 1: Visual analog scale (VAS) pain intensity scores (0–10 scale) at 0, 12, 24, 36, and 48 h for Groups A, B and C. Horizontal lines indicate median values, boxes indicate ranges of scores, and “whiskers” indicate sd. Sample sizes at each time point are shown along the bottom of the x-axis.

Figure 2: Distribution of visual analog scale (VAS) pain intensity scores in the recovery room (time 0) for Group A.

Figure 3: Visual analog scale (VAS) pain intensity scores at 0, 12, 24, 36, and 48 h for Groups A (subgroup with successful infiltration), B, and C. Same symbol conventions as in Figure 1.

Incision length and depth were measured during surgery. These two variables, as well as incision area (calculated as the product of length and depth), did not differ among the three groups (Table 1), nor between subsets of Group A with successful and unsuccessful field infiltration (P = 0.463).

Intraoperative systemic fentanyl administration was smallest in Group B (P < 0.001), which is consistent with the intraoperative use of epidural local anesthetic in this group. This group had a mean fentanyl consumption of 422 ± 283 μg, whereas Groups A and C had means of 746 ± 310 μg and 704 ± 262 μg, respectively. The difference of mean intraoperative fentanyl consumption in the subsets of Group A with successful and unsuccessful incisional infiltration did not reach statistical significance with 877 ± 334 μg versus 657 ± 265 μg, respectively (P = 0.097).

With regard to PCA morphine consumption during the first two full postoperative days, Groups A and C were equivalent (P = 0.811) with a mean of 131 ± 78 mg for Group A and 136 ± 69 mg for Group C. When comparing the two subgroups within Group A, they did not differ significantly from Group C with respect to PCA morphine consumption (P = 0.056), with the apparently successful infiltration subgroup having a mean of 97 ± 65 mg and the other subgroup having a mean of 155 ± 79 mg.

Data collection by ward nurses to characterize the number of hours to first ambulation was insufficient for statistical analysis. However, three patients in the epidural group were documented to have lower extremity weakness that interfered with their early ambulation.

Discussion

The present study demonstrated that incisional infiltration with local anesthetic can be effective as a component of a multimodal perioperative analgesic strategy for obese patients undergoing open gastric bypass surgery. However, this result only applied to a subgroup of those with local infiltration (Group A). Aggregate results for Group A did not seem to differ from those for the PCA group (Group C). The only group to show analgesic superiority was the epidural group (Group B) in the PACU and 36 hours after surgery. On examining the local infiltration group (Group A), it became clear that there was a bimodal distribution: approximately equal numbers of patients in Group A had either minimal or mild pain (3 or less on the 0–10 VAS score) or moderate to severe pain (4 or more on a 0–10 VAS score). When we separately examined those patients in Group A who had effective field infiltration resulting in absent or mild pain in the PACU, we found that their pain scores were significantly lower than those for either Groups B or C at 0, 12, and 36 hours.

Although statistical significance was not present for the subgroups within Group A, again possibly because of a small number, the consistently smaller consumption of morphine in the successful subgroup during the first 48 hours after surgery merits reexamination in a larger-scale replication study. There was a nonsignificant trend towards increased fentanyl use under general anesthesia for the successful infiltration subgroup of Group A. Interpretation of this finding is difficult in light of the small sample size.

These results suggest that for patients in Group A who experienced successful infiltration, this inexpensive, simple intervention augmented the multimodal approach of this subgroup to be equivalent if not superior to postoperative epidural analgesia. The reason for the separation of Group A into two subsets of patients with apparent responders to local infiltration and nonresponders remains unclear. The surgeon was the same for all cases, and the infiltration technique did not consciously change during the study period. However, further investigation examining possible differences in local anesthetic infiltration techniques may be warranted. Our results are limited by the relatively small number of patients studied and the even smaller number who remained in the working epidural or infiltration groups.

Morbid obesity increases the risks of anesthesia and surgery including, but not limited to, cardiovascular, thromboembolic, pulmonary, and GI impairment. Obesity also carries an increased risk of postoperative hypoxemia and complex psychosocial issues, which may contribute to management difficulties ⁽¹³⁾. It is important for this population to ambulate soon after their surgery. By encouraging this patient population to move their legs, sit up, and stand, we can reduce the risk of postoperative complications ⁽¹⁴⁾. An important factor influencing postoperative activity is the level of postoperative pain. Most gastric bypass surgeries currently use a traditional open approach via a midline incision, although there is an increasing trend towards laparoscopic surgical management.

Our current anesthetic standard for gastric bypass surgery is combined epidural analgesia and general anesthesia. This option provides good outcomes, as documented in several studies ^(13,15,16), including one in an obese patient population ⁽¹³⁾. The latter study found that patients with epidurals had smaller opioid requirements and therefore had fewer opioid-related side effects (respiratory depression, nausea, and ileus) than those who received IM opioids. These patients were also able to move, sit, and stand earlier with a shorter hospital stay.

The benefits of the epidural technique seem convincing, but other studies have failed to show improved effect when using perioperative epidural analgesia ^(14,17–19). These inconsistent results have created controversy regarding the role of epidurals in patient outcome, as they are not without risks such as epidural-related infection, potential neurological complications, placement failure, and other technical complications. In this obese population, epidural placement failure is relatively frequent because of difficulty in identifying anatomic landmarks and the need for longer needles (and hence to sacrifice tactile cues) ^(20,21). Even after successful placement, inadequate epidural function requiring catheter replacement is more common in obese than non-obese patients ⁽²¹⁾, possibly because of increased mobility of overlying excess soft tissue. Our results confirm these findings.

An exciting implication of the present findings is that newer drugs are now being introduced into clinical practice that permit larger doses of local anesthesia to be used, owing to the reduced cardiotoxicity of novel local anesthetics like levobupivacaine ⁽²²⁾. Hence, the safety and feasibility of infiltration techniques that use large volumes or large concentrations of local anesthetics are increasing. Further, novel methods of drug delivery such as through polymer or microsphere technology or continuous infusion through small catheters placed in the incision allow the provision of regional local anesthesia for some days after operation. In a previous study, we have found that the short-term (one to two days) benefits of preemptive local anesthetic interscalene block before overnight-stay shoulder surgery were no longer evident a week after discharge ⁽²³⁾. Thus, to reduce the likelihood of a postoperative analgesic gap and to prolong the benefits of preemptive analgesia in front-line clinical practice may require a combination of safer local anesthetics plus novel delivery systems for continuous, prolonged postoperative administration at the site of incision ⁽²⁴⁾. Otherwise, the benefits of this theoretically attractive concept may be elusive in practice ^(25–27).

We are optimistic that as these dual approaches become optimized, cost-effective solutions to a variety of in- and outpatient surgical pain problems will emerge. Infiltration analgesia, often thought of as suitable for superficial or small procedures only, may then become a standard method in the armamentarium for provision of analgesia in obese patients with large abdominal incisions.