Comparison of Continuous Wound Infusion versus Continuous Epidural Infusion in Upper Abdominal Surgery: Noninferiority Randomized Controlled Trial.

CONTEXT: Wound catheter offers a less invasive alternative for postoperative analgesia in the abdominal surgery.
METHODS: We conducted a single-center, prospective, open-label noninferiority randomized controlled trial. A total of 40 patients who consented to this trial, undergoing upper abdominal surgery via an upper midline incision, were randomized into two groups. In the continuous wound infusion (CWI) group, the wound catheter was placed in the subcutaneous plane of the surgical incision; the continuous epidural infusion (CEI) group received thoracic epidural with a catheter placed. After the surgery, both the groups received 0.2% ropivacaine infusion at 10 mL/h following a 10 mL bolus for 48 h postsurgery. Postoperatively, the pain scores were noted at multiple time points, along with a record of morphine consumption and adverse effects.
RESULTS: There was no significant difference in pain scores both at rest and on movement between the two groups at all the time points assessed. The mean difference in numerical rating scale score 24 h postsurgery at rest (0.1, 95% confidence interval [CI] = -0.45, 0.65) and on movement (0.05, 95% CI = -0.73, 0.83), with 95% CI in both the groups, was within the noninferiority limit. Morphine consumption was less in the CEI group, though not significant. The time to appearance of bowel movement, time to ambulate, and length of hospital stay were significantly lower in the CWI group. The incidence of hypotension requiring intervention was higher in the CEI group.
CONCLUSION: Hence, analgesia provided by continuous wound catheter infusion is not inferior to CEI with better preservation of hemodynamics and faster recovery. Copyright:

INTRODUCTION

Wound infiltration is a simple, safe, and effective method of decreasing the postoperative pain and opioid consumption for several surgeries,[1] its limitation being the short duration of action. To overcome the limitation, the technique of placing a multi-holed catheter is used to provide continuous infusion of local anesthetic. This technique of continuous wound infusion has been gaining popularity and has proven efficacious in many surgeries including cardiothoracic, pelvic, breast, and spine surgeries,[2345] with contradictory results in abdominal surgeries.[678] This can be explained by the complex mechanisms of pain in abdominal pathology and surgery. In this study, we focused on upper abdominal surgeries done via an upper midline incision, as a measure to decrease the heterogeneity among different surgeries, as suggested in the literature.[8] The aim was to compare the efficacy of continuous wound infusion (CWI) with continuous epidural infusion (CEI) in patients undergoing upper abdominal surgery, done through an upper midline incision.

The primary outcome measure was the pain at 24 h after surgery using the numerical rating scale (NRS) score. Other outcomes were pain scores at rest and on movement during other time point postoperatively, the requirement of morphine, length of hospital stay, side effect profile, wound-related complications, and bowel recovery. The side effect profile included the incidence of postoperative hemodynamic instability, postoperative nausea and vomiting (PONV), pruritus, respiratory depression, wound infection, wound dehiscence, postoperative pulmonary complications, and other miscellaneous complications.

METHODS

It is a single-center, prospective, open-label noninferiority randomized controlled trial conducted over 1 year. The study was cleared by the Institute Ethical Committee. Informed consent was obtained from all the study participants.

All the patients included in this study belonged to the American Society of Anesthesiologists physical Status I to III, aged 18–65 years, who underwent elective upper gastrointestinal surgery strictly through an upper midline incision. The exclusion criteria were refusal of consent, contraindications to epidural analgesia (e.g., localized sepsis, coagulopathy, and peripheral neuropathy), planned incision length more than 20 cm, procedure involving retroperitoneal dissection, preoperative abdominal pain with NRS score more than 3, history of any substance abuse, allergy to local anesthetic drugs, pregnancy, inability to communicate, and lack of numeracy skills.

All the patients were divided into two groups, as follows:

CEI group: Included 20 patients, who received thoracic epidural catheter placed before anesthesia induction

CWI group: Included 20 patients who received a multi-holed wound catheter placed at the incision site by the surgeon before the closure of the wound.

All participants underwent a routine preanesthetic evaluation. On the evening, before the procedure, all patients were allocated to either CWI or CEI group using computerized randomization. The information sheet of the study was made available to the patient, and consent was obtained. The ability to count from 0 to 10 in the vernacular language was considered as adequate numeracy skill. Patients were sensitized to the 11 points NRS and were given hands-on training in using patient-controlled analgesia (PCA) pump (Perfusor Space, B. Braun, Germany).

Anesthesia induction was done with propofol (1.5–2.5 mg.kg-1), morphine (0.1 mg.kg-1), and vecuronium (0.1 mg.kg-1) as the relaxant to facilitate tracheal intubation and was maintained with isoflurane, oxygen, nitrous oxide, and intermittent boluses of muscle relaxant. Intravenous (i.v.) paracetamol 1 g and ondansetron 0.1 mg.kg-1 were administered toward the end of surgery. Reversal of neuromuscular blockade was done with neostigmine and glycopyrrolate.

In the CEI group, an epidural catheter was placed before the induction of general anesthesia by an experienced anesthesiologist in the T7–T10 region with the loss of resistance technique using normal saline. An 18 G Tuohy needle with 20 G catheter was used. A standard 3 mL test dose of lignocaine-epinephrine (2% lignocaine containing 5 mg/mL of epinephrine) was administered to rule out the intrathecal or intravascular placement of the catheter. The proper position of the catheter was further confirmed by testing for a band of anesthesia with the pinprick method. Intraoperatively, the epidural catheter was not used for analgesia. Local anesthetic infusion (0.2% ropivacaine at a rate of 10 mL/h) was started through the epidural catheter using an elastomeric pump (Dosi-Fuser Multi-flow, Admedus, Australia) at the end of the procedure following a bolus of 10 mL before the reversal of neuromuscular blockade.

In the CWI group, a single 19-gauge multi-holed wound infusion catheter (Wound catheter Infiltralong set, Pajunk®, Germany) of 15 cm active length was inserted into the wound by the surgeon, through the skin, 3–5 cm away from the caudal end of the surgical incision using the introducer (split cannula with puncture cannula). The catheter has multiple holes throughout the active length spread spatially around 360°. The catheter was placed with aseptic precautions in the deep subcutaneous plane after the closure of peritoneum, rectus sheath, and muscle as a single layer. In the event of incision length being lesser than the catheter length, the excess catheter was folded such that all the holes of the catheter were within the wound. If the incision length was more than 15 cm, the catheter was placed in the cephalad 15 cm; the extra few centimeters will be covered by the drug dripping down as all patients were nursed in the head-up position postoperatively. After the skin closure, the catheter was secured to the skin with a transparent film dressing. Similarly, local anesthetic infusion (0.2% ropivacaine at a rate of 10 mL/h) was given through the wound catheter following a bolus of 10 mL of 0.2% ropivacaine. In both the groups, the catheters were removed at the end of 48 h.

After the surgery, patients were received in the postanesthesia care unit before getting shifted to the high dependency unit (HDU). As rescue analgesia, all patients received IV morphine for 48 h with a PCA device set to deliver 1 mg per dose with a lockout interval of 10 min, without background infusion. IV paracetamol 1 g 4 times a day was used to complement analgesia in all patients.

In both the groups, the NRS score (0–10) was recorded at each of the study points (4, 8, 12, 24, 36, and 48 h postoperatively) verbally with 0 signifying no pain at all and 10 the worst imaginable pain. The NRS score and patient satisfaction scale were recorded by the nursing staff trained in pain service. The total dose of morphine and the number of requests made for morphine in both groups were recorded from the PCA pump at specific intervals. Hemodynamic parameters were noted hourly by nursing staff in the HDU. Side effects and complications were recorded until the patient got discharged home. The decision to discharge was made based on standard discharge criteria. The length of hospital stay was defined from the day of surgery till discharge. Respiratory rate lower than 12 breath/min was defined as respiratory depression, and a 20% fall in blood pressure of the baseline or systolic blood pressure lower than 90 mmHg was taken as hypotension, based on the data recorded by the nursing staff. All data were entered daily into a Microsoft Excel Sheet by the primary investigator. In both the groups, if hypotension occurred, efforts to correct reversible causes were made. If hypotension persisted, one-step reduction of infusion to 5 mL/h was made. Withdrawal criteria included an infraumbilical extension of incision, failure of epidural analgesia, premature removal of the catheter, and undue interruption in the local anesthetic infusion. Postoperatively, patients were motivated to perform incentive spirometry and ambulation as early as possible.

Randomization was done using a computer-based schedule into CWI or CEI. Enrolled patients were given a serial number and were allocated into either of the groups using the generated randomization schedule. Using a proposed noninferiority design, for sample size calculation, the noninferiority limit (the largest difference that is clinically acceptable) was chosen as 1.0 on the NRS (0–10). With the noninferiority limit of 1.0, a pooled standard deviation of 1.0,[910111213] a significance of 5%, and power of 90% on the NRS (0–10), a total of 18 patients in each group were required.[14] Allowing for a dropout rate of 10%, a total of 20 patients were studied in each group.

The statistical analysis was done using Statistical Package for the Social Sciences (IBM SPSS version 22.0. IBM Corp., Armonk, NY, USA). All the data were tabulated and checked for normalcy. For normal data, Student's t-test (unpaired) was used to compare the mean difference between the two study groups i.e., CWI and CEI. All parametric data were expressed as the mean ± standard deviation. Postoperatively side effects such as hypotension (yes/no), PONV (yes/no), and pruritus (yes/no) in the two arms were compared by Chi-square test. P ≤ 0.05 was considered significant at 95% CIs with a two-sided tailed test. The noninferiority limit value was set as 1.0 on NRS score (0–10), hence to prove that the new intervention is not inferior to the standard technique of epidural analgesia, the mean difference and the 95% confidence limits of the mean difference in pain scores among the two groups should not be more than 1.0.

RESULTS

A total of 40 patients were studied: 20 in CEI group and 20 in CWI group [Figure 1]. As shown in Table 1, there was no significant difference in patient demographic data such as age, sex, height, weight, and ASA physical status between the two groups. Patients underwent various upper abdominal procedures such as palliative gastrojejunostomy, feeding jejunostomy, splenectomy, and cystogastrostomy. Table 1 shows that there was no significant difference in the type of surgery, duration of surgery, and the incision length between the two groups. No additional dose of morphine was required intraoperatively in both the groups. In the CEI group, all the epidurals were placed successfully; there were no procedure-related complications, and intrathecal or intravascular catheter placements were ruled out. By using the pinprick test, the anesthetic band was established in the corresponding dermatomes, confirming proper placement of the epidural catheter in all the patients in the CEI group. All the patients were extubated on the table by the end of the surgery, and no major intraoperative events were noted.

Figure 1

Consort diagram

Table 1

Baseline characteristics of patients, type, and surgery

Variable	CEI group (n=20), n (%)	CWI group (n=20), n (%)	P*
Age (years)	43.5±12.58	45.05±12.6	0.69
Gender (male/female)	16/4	14/6	0.46
Height (cm)	165.25±5.4	161.4±6.3	0.45
Weight (kg)	65.3±9.49	61.1±11.03	0.20
ASA PS
I and II	19 (95)	18 (90)	0.73
III	1 (5)	2 (10)
Type of surgery
Palliative gastrojejunostomy	10 (25)	7 (17.5)	0.78
Feeding Jejunostomy	3 (7.5)	5 (12.5)
Splenectomy	3 (7.5)	4 (10)
Cystogastrostomy	2 (5)	3 (7.5)
Others	2 (5)	1 (2.5)
Duration of surgery (min)	141.5±43.01	123.75±54.43	0.26
Incision length (cm)	16.43±1.27	16.225±1.49	0.65

*Unpaired Student’s t-test used for normal data, Chi-Square test used for categorical data. CEI=Continuous epidural infusion, CWI=Continuous wound infusion, ASA PS=American Society of Anesthesiologists Physical Status

Figure 2 shows the mean NRS scores in both the groups at rest and on movement, at different time points postsurgery. The NRS scores were not significantly different between the groups at any time point. As shown in Figure 3, the 95% confidence limits for the differences between the mean NRS scores in the CWI and CEI groups 24 h postsurgery at rest (0.1, 95% CI = −0.45, 0.65) and on movement (0.05, 95% CI = −0.73, 0.83) were within the noninferiority limit of 1.0 in the NRS. Morphine consumption and the number of requests made in the PCA pump were higher in the CWI group at all the time points, though none of them was statistically significant [Table 2]. Blood pressure was significantly lower in the CEI group in the first 24 h postsurgery as seen in Table 2. The incidence of hypotension in the CEI group was 25%, five patients requiring a dose reduction to 5 mL/h of 0.2% ropivacaine. The time to appearance of bowel movement, time to ambulate, and length of hospital stay were significantly lower in the CWI group [Table 3]. Some of the patients in the CWI group had excellent analgesia and were ready to perform incentive spirometry as early as 4 to 6 h after surgery. Wound soakage with serous fluid was noted in all the cases in the CWI group, but there was no incidence of wound dehiscence, wound infection, delayed healing, or fever in both the groups. There was no significant difference in the incidence of PONV, pruritus, or respiratory depression. There was no incidence of major pulmonary or vascular complications.

Figure 2

Postoperative numerical rating scale score at rest and on movement at different time points

Figure 3

Noninferiority graph – The mean difference in numerical rating scale score 24 h postsurgery at rest (0.1, 95% confidence interval =– 0.45, 0.65) and on movement (0.05, 95% confidence interval =–0.73, 0.83), with 95% confidence interval in both the groups. The dashed line is the inferiority limit

Table 2

Comparison of morphine consumption and number of requests

Variable	CEI group	CWI group	P*
Morphine consumption (mg), h
0-4 h	1.35±1.5	1.95±2.1	0.31
4-24 h	1.0±1.9	1.4±2.2	0.55
24-48 h	0.45±1.2	0.95±1.7	0.31
Total morphine consumption (mg)	3.05±4.5	4.25±5.7	0.46
Number of requests, h
0-4	1.95±2.3	2.7±3.2	0.40
4-24	1.1±2.2	1.6±2.6	0.52
24-48	0.45±1.2	1.05±1.7	0.22
Total number of requests	3.35±5.4	5.25±7.6	0.37
Mean blood pressure (mmHg), h
0-4	75.4±7.87	81.7±6.9	0.01
4-24	76.7±5.7	82.4±6.07	0.00
24-48	79.1±4.9	81.78±5.8	0.13

*Unpaired Student’s t-test. CEI=Continuous epidural infusion, CWI=Continuous wound infusion

Table 3

Recovery and adverse effects

Variable	CEI group (n=20)*	CWI group (n=20)*	P†
Appearance of bowel movement (h)	29.6 (26.36-32.83)	25 (22.01-27.98)	0.031
Time to ambulate (h)	31.85 (21.68-35.01)	21.4 (18.69-24.01)	0.001
Length of hospital stay (h)	70.75 (65.09-76.40)	57.75 (50.7-64.79)	0.024
Hypotension, n (%)	5 (25)	1 (5)	0.07
PONV, n (%)	1 (5)	1 (5)	1
Pruritis, n (%)	Nil	1 (5)	0.311
Respiratory depression, n (%)	Nil	1 (5)	0.311
Wound soakage, n (%)	Nil	20 (100)	<0.00
Wound infection, dehiscence	Nil	Nil	-

*Mean (95% confidence limits), †Log rank test and Chi-Square test used for categorical data. CEI=Continuous epidural infusion, CWI=Continuous wound infusion, PONV=Postoperative nausea and vomiting

DISCUSSION

In this study, the results show that the continuous wound infusion in upper midline laparotomy cases is not inferior to that of epidural infusion. Our results were comparable with that of the previous studies comparing epidural analgesia with wound catheter infusion in different major abdominal surgeries.[81315] A recent study in open gastrectomies, comparing epidural, continuous wound infusion, and patient-controlled opioid analgesia, produced a similar result.[16] A systematic review on this topic which analyzed nine good quality randomized controlled trials (RCT s) concluded that analgesia by wound catheter infusion is comparable to epidural analgesia.[8] The main limitation of this review was heterogeneity in methodology among the studies analyzed. The studies included were not procedure-specific, different incisions were used, site of catheter placement was variable, and several additional drugs were used for analgesia. To overcome this limitation, we chose upper gastrointestinal procedures done only through an upper midline incision, subcutaneous placement of the catheter with minimal additional analgesics, and PCA morphine as rescue analgesia.

Very few studies were performed in abdominal surgery through an upper midline laparotomy. Notably, two studies reported the efficacy of wound catheter infusion in major abdominal surgery done through a midline laparotomy,[1718] but the control group either used parenteral analgesics or placebos.

Effective pain management in abdominal surgery mandates a proper understanding of the underlying mechanisms of pain. The pain can originate from nociceptors in the anterior abdominal wall, peritoneum, and viscera. The peritoneum lines most of the abdominal viscera and has two layers: the parietal peritoneum and the visceral peritoneum. The parietal peritoneum is sensitive to touch, pressure, painful stimuli, and temperature changes and causes sharp, well-localized pain. The visceral peritoneum is sensitive to stretch, distension, tearing, but insensitive to touch, cutting, and burning sensations, and the resulting pain is dull aching and poorly localized.[19] The intraperitoneal organs are invested both anteriorly and posteriorly by the visceral peritoneum; in contrast, the retroperitoneal structures are not associated with visceral peritoneum, but their anterior surface is invested by parietal peritoneum, which carries somatic nociceptive receptors. The mechanism how the wound catheter infusion functions are hypothesized as follows: initially, the local anesthetic solution blocks the free nerve endings in the skin, subcutaneous tissue, and fascia followed by a deep somatic nociceptive block seeping through the musculofascial layer acting on the nociceptors in the parietal peritoneum as well as somatic nerves in the rectus sheath. Hence, in this study, we have tried to exclude surgeries that involve retroperitoneal structures.

The preferred location of catheter placement is unclear from the previous studies. The catheters have been placed in preperitoneal,[913] suprafascial or subcutaneous,[1720] and transversus abdominis plane.[10] We chose to place the catheter in the subcutaneous plane as the institution protocol favors a mass closure technique (the suture bite includes all layers of abdomen except skin) for incision closure. Further, the closure of the peritoneum as a separate layer is not recommended by the European Hernia Society guidelines on the closure of abdominal incisions.[21] de Almeida et al. in their study placed the wound catheter in the subcutaneous plane and found it to be inferior to epidural analgesia, but there was neither mention of the type of surgeries nor its duration.[20] The safety of ropivacaine in the subcutaneous tissue has been well established in studies measuring serum concentrations.[22] The dosage of ropivacaine was chosen based on the study by Bertoglio et al., which proved that the analgesia by wound catheter infusion is as efficacious as thoracic epidural analgesia in colorectal surgery.[13] It was a large multi-centered RCT with a sample size of 106 patients. The study established both the safety and efficacy with a dosage of 0.2% ropivacaine at 10 mL/h in both the groups.

Morphine consumption was slightly higher in the CWI group, although not statistically significant. The opioid consumption was lower in both the groups, compared to primarily IV opioid-based analgesia in the previous studies, which indicates excellent analgesia provided by both the techniques.[616] The incidence of hypotension was higher in the CEI group, the mean blood pressure in the first 24 h being significantly lower than the CWI group. The hypotension was managed with IV fluids, boluses of vasopressors, and epidural dose reduction. Hypotension, the most common adverse effect with thoracic epidural analgesia, is well-documented in the literature.[23] No mention of wound soakage was found in the previous literature, which was significantly seen in the CWI group. It is probably because of the subcutaneous placement of the catheter. No incidence of wound infection or wound dehiscence was found in this study, in concordance with the literature, as opposed to the general apprehension.[678]

Good analgesia is especially needed for upper gastrointestinal surgery, as with good analgesia, the respiratory function is better and perioperative pulmonary complications are lesser, leading to faster wound healing, recovery, and discharge from the hospital. Thoracic epidural analgesia has been considered as the standard practice for pain management in such cases. Yet, it is a blind, technically challenging procedure and not free from causing discomfort to the patient, both on insertion and maintenance in the postoperative period. Further, local sepsis, deranged coagulogram, anticoagulant use, and lack of consent preclude its use. Our method of CWI offers a simple, quick, less invasive, and effective technique, which is devoid of complications such as hypotension, PONV, paralytic ileus, pruritis, and urinary retention. The procedure-specific postoperative pain management group, which provides procedure-specific recommendations, has already recommended wound catheter infusion for cesarean section, radical prostatectomy, and colonic resections. With more procedure-specific studies like ours, the utility of wound catheter infusion will further widen.[24]

In the CEI group, although intraoperative epidural analgesia was not provided, the test dose used could have added a preemptive analgesic effect. Although the current standard practice is to add opioids in epidural drug preparation, we avoided it so that postoperative morphine consumption can be compared between the groups. In the current era, laparoscopic procedures are replacing open abdominal procedures; nevertheless, the importance of the latter can never be underestimated. Moreover, the wound catheter infusion technique can be the analgesic technique of choice when a conversion is made from the laparoscopic approach to open procedure, wherein epidural analgesia is commonly avoided. Currently, the cost of the wound catheter infusion is more than that of the epidural; but the benefits far outweigh the cost difference, especially in high-risk terminally ill patients. Our study proves that understanding of the underlying pain mechanism and careful case selection can increase the efficacy of wound catheter infusion in abdominal surgeries. This novel technique of postoperative analgesia will play a huge role in palliative abdominal procedures but needs further studies focusing on such procedures.

CONCLUSION

We conclude, saying that the analgesia provided by continuous wound infusion is not inferior to CEI, with better preservation of hemodynamics and faster recovery. Wound catheter infusion is an effective alternative to the epidural technique in abdominal surgeries and will play a crucial role in the palliative procedures.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.