COMPARING KETOROLAC TO IBUPROFEN FOR POSTOPERATIVE PAIN: A RANDOMIZED CLINICAL TRIAL

Abstract

Objectives:

To identify differences in pain perception and satisfaction with pain control in women receiving nonsteroidal anti-inflammatory drugs postoperatively.

Methods:

This was a prospective, randomized controlled trial including urogynecology surgical patients. After surgery, all patients were randomized to receive either intravenous ketorolac or ibuprofen. Patients completed three visual analog scales assessing pain at rest, pain with ambulation, and satisfaction with pain control. Post-operative opioid use was also measured.

Results:

A total of 224 patients (112 in each arm) were included. Pain scores at rest in all patients who received ketorolac versus those who received ibuprofen was 2.30 (SD± 2.1) versus 2.68 (SD± 2.34), p = 0.20. Pain scores with ambulation was 3.94 (SD± 2.57) versus 4.16 (SD± 2.73), p = 0.57 in patients who received ketorolac and ibuprofen respectively. Patients who received ketorolac rated their satisfaction with their pain regimen similarly to those who received ibuprofen (p = 0.50). The average amount of hydromorphone used in the ketorolac and ibuprofen arm was 3.68 (SD± 4.58) mg and 4.04 (SD± 4.97) mg respectively, p = 0.58. A sub-group analysis based on type of surgery showed decreased pain at rest (VAS 2.77 versus 4.88, p = 0.04), and increased satisfaction (VAS 1.69 versus 4.67, p = 0.003) in patients who had laparotomy and received ketorolac.

Conclusions:

There was no difference in pain and satisfaction with intravenous ketorolac compared to intravenous ibuprofen in patients who underwent all modalities of urogynecologic surgery. A subgroup of patients who underwent laparotomy had less pain with ketorolac.

Introduction

The World Health Organization (WHO) has identified pain and proper treatment of pain as an important health concern.¹ Postoperative pain can have adverse clinical implications on time to recovery and length of hospital stay.² With increased focus on the opioid epidemic in the United States, utilization of non-opioid medications during the perioperative period needs to be prioritized. Nonsteroidal anti-inflammatory drugs (NSAIDs) are often used in conjunction with opioids to improve pain management. Such multimodal analgesia have been shown to reduce pain intensity within the first 24 hours after surgery.³

Ketorolac and ibuprofen are two options for therapy after surgery. The Food and Drug Administration (FDA) approved intravenous (IV) ibuprofen (Caldolor®, Cumberland Pharmaceuticals; Nashville, TN) in 2009, creating an additional treatment option for providers. IV ibuprofen is well tolerated by patients when administered before and after surgery and no dose adjustment is needed in higher risk patients such as the elderly,⁴ This makes it an attractive option for postoperative pain management after urogynecologic surgeries, since surgeries for pelvic floor disorders increase with age.⁵

There continues to be a paucity of literature regarding postoperative pain management for urogynecology surgeries.⁶ A recent randomized trial, found that IV ketorolac did not reduce pain scores or total opioid use in patients undergoing transvaginal urogynecologic surgery.⁷ IV ibuprofen has not proven to reduce postoperative pain in patients who had an elective hernia repair.⁸ It is not known how ibuprofen compares to IV ketorolac in reducing postoperative pain management after urogynecologic surgery. The primary aim of this study is to identify differences in pain perception and satisfaction with pain control in patients receiving IV ketorolac, IV ibuprofen, and oral ibuprofen after undergoing urogynecologic surgery.

Materials and Methods

This was a prospective, randomized controlled trial of patients undergoing surgery in the division of Female Pelvic Medicine and Reconstructive Surgery (FPMRS) at the University of Oklahoma Health Sciences Center (OUHSC) between September 1, 2013 and May 31, 2015. The trial was registered through the National Institutes of Health (NIH) U.S. National Library of Medicine clinical trials website, NCT01971372 (https://register.clinicaltrials.gov/), before enrollment and received approval by the OUHSC Institutional Review Board. Informed consent was obtained from all participants who meet the eligibility criteria prior to surgery.

Patients were included if they were female, planning on admission for at least one overnight stay, reliable to follow up, able to complete the study assessment, and scheduled for surgery with an FPMRS surgeon. Patients were excluded if they were less than 18 years of age, and had known renal impairment, congestive heart failure, dementia, allergy to NSAIDs, allergy to hydromorphone, history of gastrointestinal bleeding or ulceration, or inflammatory bowel disease.

All patients underwent surgery by one of four FPMRS surgeons and received general anesthesia. After surgery, all patients in the study were assigned to receive either oral (PO) ibuprofen, IV ibuprofen, or IV ketorolac by a computer-generated randomization schedule. Randomization was performed in blocks of six in a 1:1 ratio regardless of route of surgery. After surgery, a sequentially numbered, opaque, sealed envelope was opened to reveal the randomly allocated medication. (Figure 1) Patients were not blinded to the medications administered.

Figure 1:

Consort flow chart

Based on their randomization arm, patients were given either ketorolac 30 mg IV (15 mg if >65 years of age or <50 kg), ibuprofen 800 mg IV, or ibuprofen 800 mg PO every eight hours for three doses. All patients were placed on patient-controlled analgesia (PCA) of hydromorphone with standardized settings, allowing the patient to dispense up to 0.2 mg of medication every six minutes (lockout rate) and an hourly maximum of 2.0 mg/hour. A one-time nursing bolus of 0.4 mg of hydromorphone was also ordered at initiation of the PCA. All patients were given acetaminophen 650 mg PO every six hours. All patients were able to receive additional opioid if needed for breakthrough pain by increasing the PCA demand dose. Narcotic medications and doses given intraoperatively were not recorded in the study.

The PCA was discontinued on the morning after surgery no later than 12:00 pm. The patients were then started on an oral pain medication with either hydrocodone/acetaminophen or oxycodone/acetaminophen. On the first postoperative day before noon, three visual analog scale (VAS) scores were obtained from the patients to assess pain at rest, ambulation, and satisfaction with pain control. (Figure 2)

Figure 2:

Visual analog scale

The primary outcome was pain and satisfaction with pain control as assessed by the VAS. The VAS allowed patients to rate their pain level at rest and with ambulation on a scale of 0 – 10.0 cm with 0 being no pain and 10.0 being severe pain. The VAS for satisfaction was reversed with 10.0 cm being unsatisfied. (Figure 2) Our secondary outcome was the amount of hydromorphone used. Demographic and surgical data were collected including the type of surgery (vaginal, laparoscopic, or open) as well as concomitant procedures. The amount of hydromorphone used while the patient had their PCA was recorded.

Power analysis was performed using a study from Crisp et al. 2012, comparing patient-controlled versus scheduled, nurse-administered analgesia, who reported 90% power to detect a mean difference of 2.68 cm on a 10 cm VAS for pain with an alpha of 0.05.⁹ Using this as the basis for our a priori power analysis, we planned for 117 patients (40 patients in each of the three planned treatment groups: IV ketorolac, IV ibuprofen, and PO ibuprofen) to be included to demonstrate a difference in mean VAS pain scores. An interim power analysis was planned after the first sixty patients to ensure that adequate power would be achieved at the conclusion of the study. The interim power analysis, however, showed that 112 patients would be needed in each arm to detect the same difference in VAS scores noted above, assuming a power of 80% and alpha of 0.05. Given this marked increase in estimated required sample size, the decision was made to discontinue the PO ibuprofen arm, and continue enrollment for IV ketorolac and IV ibuprofen only for feasibility purposes. IV ketorolac was the institutional standard so this arm remained in order to compare it to the IV ibuprofen.

Statistical analysis was performed using SAS v9.4 (Cary, NC). Continuous variables were summarized using mean, median, and standard deviations; categorical variables were summarized using counts and proportions. Statistical analysis was by intention to treat. The parametric p-value was calculated by ANOVA for numerical covariates and chi-square test for categorical covariates. A p value of <0.05 was considered significant.

Results

A total of 332 patients were approached for the study between September 1, 2013 and May 31, 2015. Of those approached, 11 patients did not meet eligibility criteria and 30 patients declined to participate. Twenty patients who were initially randomized to oral ibuprofen were removed from the study after the interim analysis and 29 patients were excluded for various reasons. (Figure 1) The remaining 242 patients were enrolled and randomized in either the IV ketorolac arm or the IV ibuprofen arm. In the ketorolac arm, there were 2 patients that declined a portion of the pain regimen and were removed from the study, and 5 similar patients in the ibuprofen arm. In follow up, there were 3 patients in the ketorolac arm, and 2 patients in the ibuprofen arm that did not complete the visual analogue scales. One patient in the ketorolac arm was removed from the study after they were found to have low urine output overnight and IV ketorolac was stopped. One patient in the ibuprofen arm had an early discontinuation of the medication after they complained of feeling “light-headed”. Four patients were removed from the final analysis due to missing medication, resulting in 224 patients (112 patients in each arm) included in the statistical analysis. (Figure 1)

Demographic data are summarized in Table 1. The majority of patients were White (92%) with mean ages of 58.73 (SD± 12.34) years and 55.52 (SD± 14.30) years in the ketorolac versus IV ibuprofen arm respectively, p = 0.07. The average BMI was 29.56 (SD± 6.45) kg/m² and 28.94 (SD± 7.43) kg/m², p = 0.51. Opioid use prior to surgery was not statistically significant between groups with 14% of patients reporting opioid use in the ketorolac arm and 17% reporting opioid use in the IV ibuprofen arm. (Table 1)

Table 1.

Demographic data.

	Ketorolac (n=112)	IV Ibuprofen (n=112)	P value
Age, mean (SD)	58.73 (12.34)	55.52 (14.30)	0.07
Race, n (%)			0.85
Caucasian	103 (91.96)	102 (91.07)
African American	3 (2.68)	5 (4.46)
Hispanic	2 (1.79)	1 (0.89)
Asian	2 (1.79)	1 (0.89)
Native American	2 (1.79)	3 (2.68)
Height, inches (SD)	64.15 (2.69)	64.22 (3.69)	0.87
Weight, pounds (SD)	173.49 (39.95)	168.77(36.54)	0.36
BMI, kg/m2 (SD)	29.56 (6.45)	28.94 (7.43)	0.51
Prior opioid use, n (%)	16 (14.29)	19 (16.96)	0.58
Prior abdominal surgery, n (%)	89 (79.46)	79 (70.54)	0.15
Type of Surgery, n (%)			0.36
Vaginal	31 (27.68)	38 (33.93)
Laparoscopic/robotic	67 (59.82)	58 (51.79)
Laparotomy	13 (11.61)	12 (10.71)
Uncategorized¥	1 (0.89)	4 (3.57)
Hysterectomy, n (%)	29 (26.85)	29 (27.62)	0.90
Sacrocolpopexy, n (%)	63 (58.33)	47 (44.76)	0.048
Midurethral sling, n (%)	22 (20.37)	14 (13.33)	0.17
Local anesthetic, n (%)	80 (74.07)	95 (84.82)	0.048
Length of Surgery, minutes (SD)	165.73 (58.74)	160.21 (63.28)	0.51
Length of Stay, hours (SD)	32.98 (15.11)	30.29 (11.26)	0.14
Post-op hydromorphone (mg), mean (SD)	3.68 (4.58)	4.04 (4.97)	0.58

^¥- uncategorized surgery included excision of vaginal mesh (2), rectus fascia pubovaginal sling (2), and injection of onabotulinumtoxinA to pelvic floor (1)

During the study period, 26% versus 27%, p = 0.90, of patients had a hysterectomy performed and 20% versus 13%, p = 0.17, had a sling performed in the ketorolac arm versus the IV ibuprofen arm respectively. The mean length of surgery was 165.73 (SD± 58.74) and 160.21 (SD± 63.28) minutes, p = 0.51. The average length of stay was 32.98 (SD± 15.11) hours and 30.29 (SD± 11.26) hours in those who received ketorolac and IV ibuprofen respectively. (Table 1)

There were statistically significant differences in the two arms of the study with more patients in the ketorolac arm undergoing sacrocolpopexy (58.33% vs. 44.76%, p = 0.048), and more patients in the ibuprofen arm receiving local anesthetic (84.82% vs. 74.07%, p= 0.048). (Table 1)

Patients underwent different routes of surgery including vaginal, laparoscopic or robotic, and laparotomy (Table 1). Five patients could not be categorized within the previously listed categories. Within the ketorolac arm, 1 patient had a rectus fascia harvest and pubovaginal sling placement. Within the IV ibuprofen arm, 1 patient had onabotulinumtoxinA injections to the pelvic floor, 1 patient had a rectus fascia harvest and pubovaginal sling, and 2 patients underwent excision of mesh.

The mean pain VAS score at rest on the first postoperative day was 2.30 (SD± 2.10) cm in the IV ketorolac arm and 2.68 (SD± 2.34) cm in the IV ibuprofen arm. The mean pain score with ambulation was 3.94 (SD± 2.57) cm in the IV ketorolac arm and 4.16 (SD± 2.73) cm in the IV ibuprofen arm. Overall, there was no statistically significant difference in the pain scores at rest or ambulation between the two study groups (p = 0.20 and p = 0.57, respectively). Additionally, patients were generally satisfied with their pain control. Subjects in the ketorolac arm reported a mean satisfaction score of 1.87 (SD± 2.74) cm which was similar to the ibuprofen arm score of 2.11 (SD± 2.65) cm (p = 0.50). (Table 2)

Table 2.

Comparisons of mean VAS scores by IV NSAID use

VAS	Ketorolac (n=112)	IV Ibuprofen (n=112)	P value
Pain
At rest	2.30 (2.10)	2.68 (2.34)	0.20
With ambulation	3.94 (2.57)	4.16 (2.73)	0.57
Satisfaction	1.87 (2.74)	2.11 (2.65)	0.50

Standard deviation is stated in parentheses.

Patients in both arms used similar amounts of narcotic. Subjects in the ketorolac arm required a mean of 3.68 (SD± 4.58) mg of hydromorphone and those in the ibuprofen arms used a mean of 4.04 (SD± 4.97) mg during the study period. The differences in the amount of hydromorphone between the two groups did not reach statistical significance (p = 0.58).

A sub analysis of the data was performed to determine VAS scores and narcotic use based on route of surgery. A significant difference in VAS scores was noted in those patients who underwent laparotomy. Patients who received IV ketorolac were found to have lower pain scores at rest when compared to those who received IV ibuprofen (2.77 (SD± 2.42) versus 4.88 (SD± 2.49), p = 0.04. In addition, pain satisfaction was found to be better in those who received IV ketorolac, with a VAS score of 1.69 (SD± 1.89) compared to those who received IV ibuprofen 4.67 (SD± 2.57), p = 0.003. There was no difference in the amount of hydromorphone used between groups. (Table 3)

Table 3.

Comparison of VAS scale scores by type of surgery and IV NSAID use

Surgery	VAS	Ketorolac (n=111)	IV Ibuprofen (n=108)	P value
Vaginal	Pain
	At rest	2.52 (2.45)	2.30 (2.22)	0.71
	With ambulation	3.64 (2.71)	3.26 (2.80)	0.59
	Satisfaction	2.68 (3.48)	1.59 (2.36)	0.13
Laparoscopic/Robotic	Pain
	At rest	2.11 (1.85)	2.38 (2.06)	0.44
	With ambulation	3.78 (2.49)	4.02 (2.29)	0.59
	Satisfaction	1.51 (2.45)	1.94 (2.58)	0.34
Laparotomy	Pain
	At rest	2.77 (2.42)	4.88 (2.49)	0.04
	With ambulation	5.38 (2.43)	6.83 (2.29)	0.14
	Satisfaction	1.69 (1.89)	4.67 (2.57)	0.003

Standard deviation is stated in parentheses.

Discussion

This was a randomized controlled trial comparing two IV NSAIDs, ketorolac and ibuprofen, to determine pain control and patient satisfaction on a standardized postoperative pain regimen after urogynecologic surgery. Overall, patients were generally very satisfied with their pain control while in the hospital and experienced low levels of pain. Patients who received IV ketorolac had similar pain control and satisfaction when compared to patients who received IV ibuprofen.

A sub analysis based on surgery route demonstrated that patients undergoing laparotomy had better pain control at rest and better satisfaction with their post-operative pain when they received IV ketorolac. For patients who underwent vaginal and laparoscopic surgery, both methods provided similar pain control with no statistically significant difference in VAS scores at rest and ambulation on the first postoperative day. While our study was not powered to detect these differences in surgery type, it does create an area of future study.

It should be noted that originally, our sample size was calculated based on a study focused on pain scores following vaginal reconstructive surgery which showed a mean difference of 2.68 in VAS scores.⁹ An interim analysis was planned in order to remain adaptive to the possible variations in study population. We concluded we could not replicate the effect size of the study by Crisp et al based on our original sample size given the overall low VAS scores found in our population.

The current data encourages a multimodal approach to pain control rather than just opioids alone.³ At the crux of a multimodal pain regimen is the use of NSAIDs. Studies looking at multimodal analgesia have shown that there is a reduction in pain intensity even in the first 24 hours after surgery with just NSAIDs alone.³ Other studies have looked at postoperative pain after gynecologic laparoscopic surgery, and showed that both diclofenac patches and lidocaine patches lower overall pain intensity when compared with a placebo patch.^10,11 Since we did not observe a difference in pain or satisfaction between IV ketorolac compared to IV ibuprofen, we recommend the use of IV ketorolac in urogynecologic patients given its low cost compared to IV ibuprofen as part of a multimodal approach.

Determining the best postoperative pain regimen for patients admitted to the hospital has yet to be determined. PCAs have been used to manage postoperative pain. Crisp et al. 2012 looked specifically at the use of PCA versus nurse-administered analgesia after vaginal reconstructive surgery and found that patients with a PCA experienced less pain on postoperative day 1 without prolonging their hospital stay or increasing side effects.⁹ For this reason, we chose to include a PCA in our protocol. Additionally, the amount of narcotics used also allowed us to measure a surrogate marker for pain that was patient driven. In this study, there was no difference in the amount of hydromorphone used between the two study groups.

The strength of this study is that it is a randomized controlled trial in a large number of patients undergoing a variety of urogynecologic procedures. The pain management protocol was standardized in order to minimize confounders. The subjective measures of pain and satisfaction with pain control were both assessed in addition to the objective measure of overall narcotic use by standardizing all patients to receive a hydromorphone PCA. We found that 14% of patients who received IV ketorolac and 17% of patients who received IV ibuprofen had used an opioid prior to their surgery. This supports recently published data which demonstrate that approximately 14% of new urogynecologic patients have a history of opioid pain medication use.¹²

Some weaknesses of this study included that we did not look at long-term pain control after the patient was discharged from the hospital. Additionally, we did not use the maximum daily dosage of ketorolac or ibuprofen, which would have allowed for a dose every six hours. We chose to dose every eight hours as this was the typical dose regimen used in the hospital, and it minimized the risk of severe side effects of the medications. Another weakness was a lack of a formal cost analysis of each medication. In our hospital at the time of this study, a single dose of ketorolac cost $0.55 compared to a single dose of IV ibuprofen which cost approximately $7.60. These are direct pharmacy costs so they do not reflect the charges to patients. Due to surgical volume and feasibility, we did not power the study to detect differences in routes of surgery. Lastly, we were unable to analyze a third arm of patients receiving oral ibuprofen. This type of research carries other ongoing challenges. As shown in our study, urogynecologic surgeries are associated with low pain scores making it difficult to demonstrate an effect size of significant magnitude.

There were five patients included in the intention to treat analysis who we could not categorize as either vaginal, laparoscopic, or laparotomy. The surgeries that fell into this category were 1 onabotulinumtoxinA injection to the pelvic floor, 2 surgeries for excision of vaginal mesh, and 2 surgeries for rectus fascia harvest with pubovaginal sling. A sub-analysis of VAS scores and hydromorphone use was not performed on this group given the small numbers.

A comparison between oral ibuprofen and IV ketorolac remains an area of interest in urogynecology. One study looking at IV ibuprofen has shown that maximum plasma concentration of IV ibuprofen was approximately twice that of oral ibuprofen.¹³ In this same study, the time to maximum plasma concentration with IV ibuprofen was 6.5 minutes compared to 90 minutes for oral ibuprofen.¹³ The elimination half-life of the IV and oral ibuprofen is the same, with a rate of 120 minutes.⁶ This allows patients to experience the maximal effects of the medication faster when it is dispensed intravenously. Another study, which compared intramuscular (IM) ketorolac to oral ibuprofen in emergency room patients showed that levels of analgesia in patients with moderate to severe pain, was comparable.¹⁴ Similar studies comparing IV acetaminophen to oral acetaminophen in orthopedic patients demonstrated that IV acetaminophen did not provide significant pain relief or reduction in narcotic use when compared to oral therapy.¹⁵

Our study suggests that FPMRS patients who undergo laparotomy have better pain control with the use of IV ketorolac. While a majority of FPRMS surgery has moved to minimally invasive techniques, this finding creates an area for future research in optimizing pain control in FPMRS patients who do undergo laparotomies. Overall, when considering a multimodal approach to post-operative pain management in urogynecologic patients, FPMRS providers should consider use of ketorolac as a standardized option in their post-operative order set as it is likely the most cost effective.