Intravenous Ketorolac as an Adjunct to Patient-Controlled Analgesia (PCA) for Management of Postgynecologic Surgical Pain

*.kociate Professor

+Research Nurse Specialist in Anesthesiol- W

IProfessor

Address reprint requests to Dr. Sevarino at the Department of Anesthesiology, Yale University School of Medicine, 333 Cedar Street, Box 3333, New Haven, CT 06510, USA.

Received for publication April 30, 1993; re- vised manuscript accepted for publication .July 23, 1993.

0 1994 Butterworth-Heinemann

1. Clin. Anesth. f&2.3-27. 1994.

Ferne B. Sevarino, MD,* Raymond S. Sinatra, MD, PhD,* Darcy Paige, RN, MBA,? David G. Silverman, MD:

Department of Anesthesiology, Yale Universiq School of Medicine, New Haven, CT.

Study Objective: To determine whether intravenous (IV) doses ofketorolac tromethamine provide safe and effective augmentation of postsurgical analgesia for patients using N patient-controlled analgesia (PCA) with morphine. Design: Randomized, double-blind, placebo--controlled, dose-response evaluation. Setting: Patient care unit at a university wledical center Patients: 62 ASA physical status I-III f ema,le.y recovering from intra-abdominal gneco- logic surgery with general anesthesia who requested posto@rative PCA. Interventions: Following initial pain assessment in the recovmy room, patients were randomized to receive either Nsaline (placebo) followed b n/saline every 6 houn (Group 1); N ketorolac 30 mg loading dose followed 4 N ketorolar 15 mg every 6 hours (Group 2); or N ketorolac 60 mg loading dose followed by N ketorolac 30 mg every 6 hours (Group 3). All patients were provided N PCA, which was programmed to gottide 1.2 mg of mmphine with a 6%minute lockout interval. Measurements and Main Results: Visual analogscale (LAS) restingpain and satisfac- tion scores were measured eoery 2 to 12 hours. Cumulative PCA with moqhine and the frequency and severity of side effects also were assessed. N ketorolac showed no clinically signi,ficant side effects. Group 2patients experienced significant reductions in VAS resting pain scores (p < O.OS), and a trend toward decreased morphine self-administration in both active groups was noted. Group 2 and Group 3 patients reported ,greater sat&faction with postsurgical analgesia than Group 1 patients. (p < 0.05). Conclusions: N ketorolac used as an analgesic adjunct provided safe and effective augmentation of PCA with morphine in patients recove?ingfrom intra-abdominal gyneco- logic surgery.

Keywords: Analgesia, postoperative; analgesia; intravenous delivery; ketorolac; morphine; patient-controlled analgesia.

Introduction

Ketorolac tromethamine is an injectable nonsteroidal anti-inflammatory drug (NSAID) that has been used as a primary analgesicIs for control of moderate to severe postsurgical pain. Ketorolac is recommended for use as an analgesic adjunct for intravenous (IV)“.; and epidural” patient-controlled analgesia

,J. Clin. Anesth., vol. 6, .Januarv/Februarp 1994 23

Originnl Contributions

Table 1. Demographic Data

Age (vr)

Height Weight

(cm) (kg)

Intraoperative

MSO, (mg)

Intraoperative

Fentanyl (mg)

PACU MSO,

(mg)

Group 1: placebo 39.1 162.8 (n = 21) (10.1) (8.9)

Group 2: ketorolac 15 mg 42.2 162.6 (n = 20) (8.4) (11.2)

Group 3: ketorolac 30 mg 41.2 163.3 (n = 21) (7.9) (8.6)

*p < 0.05 between this group and both other groups.

Note: Data are means (SEM).

MSO, = morphine; PACU = postanesthesia care unit.

64.9 3.9 213.2 3.5 (9.4) (4.5) (143.2) (3.7) 72.9* 6.5 179.5 3.6

(13.3) (6.8) (219.4) (3.5) 65.6 6.5 165.8 3.7

(12.0) (6.5) ( 160.8) (3.6)

(PCA). In a prior investigation,” patients treated with intramuscular (IM) doses of ketorolac self-administered significantly less morphine that did patients treated with a placebo. No improvement in satisfaction was observed, however, as patients were generally dissatisfied with the need for frequent and painful IM injections.

Although ketorolac has yet to receive U.S. Food and Drug Administration approval for IV use, the IM prepara- tion provides safe and effective pain relief when adminis- tered as a parenteral analgesic’,“,7-” and is used with increasing frequency in general surgical and pediatric patients.* Our rationale for IV dosing includes the follow- ing: (1) patients are generally fearful of IM injections; (2) pharmacokinetic studies have shown that a 4minute IV infusion of ketorolac is bioequivalent to that observed with IVdosing;‘” (3) ketorolac is physically and chemically stable when mixed with a variety of commonly used infu- sion solutions, and it is not absorbed by IV tubing or polyvinyl chloride syringes;” (4) the IM preparation is nonirritative and is not associated with significant phlebi- tis when administered as a slow IV infusion.“.‘.+ Further- more, IV administration of this preparation is not associated with cardiac or hemodynamic changes.’

The present double-blind, placebo-controlled evalua- tion examined the safety and efficacy of IV ketorolac as an analgesic adjunct for PCA with morphine in patients recovering from gynecologic surgery.

Materials and Methods

This study was approved by the Human Investigation Committee of Yale University School of Medicine. In- formed written consent was obtained from 62 healthy (ASA physical status l-111) females scheduled for elective intra-abdominal gynecologic surgery with general anes- thesia. Excluded were patients over age 65 or under age 18; those with known bleeding diatheses, renal dysfunc-

*Oberlander T, Berde C: Ketorolac: raising the roof on ceiling effects for NSAIDs. IASP Newsletter September/October 1991:2-3. tSimione D, Seashore J: Chart review of infants receiving narcotics postoperatively. Unpublished data, Yale-New Haven Hospital, 1991.

tion, a history of asthma, or a sensitivity to NSAIDs; and patients weighing less than 50 kg or more than 100 kg. All patients were instructed preoperatively in the use of the PCA device (Abbott PCA Plus II, Abbott Park, Chi- cago, IL) and oriented to the visual analog scale (VAS) scoring for pain and satisfaction. If indicated, premedica- tion was with IM midazolam 0.03 to 0.05 mg/kg 1 hour prior to surgery.

General anesthesia was provided with nitrous oxide in oxygen plus isoflurane and morphine or fentanyl as required. At the conclusion of surgery, patients were randomized in a double-blind manner to one of three groups: Group 1 patients (n = 21) received IV saline (placebo) 2 ml in the postanesthesia care unit (PACU), followed by four doses of IV placebo 1 ml every 6 hours. Group 2 patients (n = 20) received IV ketorolac 30 mg in the PACU, followed by four doses of IV ketorolac 15 mg every 6 hours. Group 3 patients (n = 21) received IV ketorolac 60 mg in the PACU, followed by four doses of IV ketorolac 30 mg every 6 hours. While in the PACU, patients received IV morphine 1 to 2 mg as needed to achieve acceptable analgesia. Upon transfer to the ward, patients received PCA with IV morphine, with the infuser programmed to provide an incremental dose of 1.2 mg and a 4hour maximum dose of 30 mg, with a lockout interval of 6 minutes.

The following data were obtained at time 0 (time of first dose of study medication) and at 2, 4, 6, 9, 12, 18, 24, and 36 hours: cumulative interval PCA dose; VAS score for pain (0 = no pain to 10 = worst pain) ; VAS score for satisfaction (0 = totally dissatisfied to 10 = completely satisfied); level of alertness as scored by the blinded observer (1 = alert, oriented, initiates conversa- tion; 2 = drowsy, oriented, initiates conversation: 3 = drowsy, oriented, does not initiate conversation; 4 = very drowsy, disoriented; 5 = stuporous); blood pressure, heart rate, and respiratory rate; frequency and severity of nausea, vomiting, and pruritus; dose and frequency of antiemetic therapy.

Data were analyzed using the Wilcoxon rank sum test (VAS pain and satisfaction, pain intensity differences), chi-square analysis (adverse effects), and analysis of vari- ance (morphine use). A value of p < 0.05 was considered statistically significant.

24 J. Clin. Anesth., vol. 6, January/February 1994

Table 2. Number (%I) of Patients with Side Effects

Nausea > 4 Hours Treatment for N/V Pruritus

Group 1: placebo (n = 21) 12 (57%) 8 (38%) 2 (10%) Group 2: ketorolac 15 mg (n = 20) 11 (55%) 10 (50%) 0 (0%) Group 3: ketorolac 30 mg (n = 21) 7 (33%) 7 (33%) 0 (0%)

N/V = nausea and vomiting.

Figure 1. Cumulative PCA morphine requirements. Data

are means ? SD and represent cumulative morphine dose

for each group at the designated time points. No significant

differences were found.

Table 3. Percentage Reduction in Patient-Controlled Analgesia with Morphine Compared with Group 1 (Placebo)

Figure 2. Visual analog scale (VAS) pain scores (0 = no pain to 10 = worst pain imaginable). Graph represents median, 25th percentile, and 75th percentile for each group at each time point. *Significant differences (p < 0.05) in

VAS pain scores were seen at 4, 6, 9, and 12 hours between the placebo and ketorolac 15 mg groups.

12 Hours 24 Hours 36 Hours

Group 2: ketorolac 15 mg 25 24 14 Group 3: ketorofac 30 mg 22 30 23

Results

All patients completed the study protocol without com- plications. There was no difference among groups with respect to patient age, height, dose of premedicant, intra- operative opioid use, or PACU morphine dose; however, patients in Group 2 (ketorolac 15 mg) weighed more

There were no intergroup differences with respect to frequency of opioid-related side effects (Table 2)) PCA

than those in the other two groups (p < 0.05;

attempts, or hourly morphine requirements; however, the overall trend was toward reduced self-administration of morphine (F&WE 1). Patients in Groups 2 and 3 re-

Table 1).

quired an average of 20% less morphine during the study interval than patients treated with saline placebo (Group 1) ( Tible 3); however, this difference did not reach statisti- cal significance. In the placebo group, baseline pain scores increased to a median value of 5.3 cm at 2 hours

and remained essentially unchanged through the first 9 hours. Thereafter, pain scores progressively decreased

Patients in Group 3 (ketorolac 30 mgj began the study

over the next 24 hours. Patients in Group 2 showed a significant reduction in VAS pain scores at 4, 6, 9, and

with a median VAS pain score of 2.15, which increased

12 hours (p < 0.05, ketorolac 15 mg us. placebo). Pain intensity differences from baseline scores were superior

over the next 2 hours to 4.3, then gradually diminished

with ketorolac 15 mg uersus the placebo at 2, 4, 6, 9, and

to 1.6 during the next 24 hours. VAS pain scores and pain

12 hours (Fiere 2).

intensity differences in this group were not significantly different from either the placebo or the ketorolac 15 mg group (Figure 2).

VAS scores for patient satisfaction with therapy (Figurc~ 3) indicated that although all patients were satisfied with their therapy (VAS score greater than 5’)) satisfaction was greatest in those individuals receiving either IV dose of’ ketorolac (15 mg or 30 mg), with significant improve-- ment (p < 0.05) seen at 4, 6, and 9 hours for ketorolac 15 mg and at 9 and 18 hours for ketorolac 30 mg. No patient receiving ketorolac experienced clinically signifi- cant postsurgical bleeding, phlebitis, or gastric upset dur- ing the 36hour study period.

.J. Ghn. Anesth., vol. 6, January/February 1994 25

Original Contributions

0

2 4 6 9 12 18 24 36

HOW

Figure 3. Visual analog scale (VAS) satisfaction scores (0 = dissatisfied to 10 = completely satisfied). Graph represents median, 25th percentile, and 75th percentile for each group at each time point. *Significant differences (p < 0.05) in VAS satisfaction scores were seen at 4,6, and 9 hours between the placebo and ketorolac 15 mg groups. **Significant dif- ferences were seen at 9 and 18 hours between the placebo and ketorolac 30 mg groups.

Discussion

JIV ketorolac provided safe and effective augmentation of PC4 with morphine in patients recovering from gyneco- logic surgery. To our knowledge, this is the first double- blind, dose-response study of IV ketorolac in which pa- tients self-administered opioids benefited from reduced intensityofpostsurgical pain. Although the drug manufac- turer (Syntex Corp., Palo Alto, CA) is unable to recom- mend IV dosing of ketorolac until safety and efficacy trials are completed, we found that this form of administration was well accepted by patients and nursing staff.

Our observations of improved pain scores and a trend toward reduced requirement of PCA with opioids reflect the advantage of balanced or multimodal analgesic ther- apy in which two or more drugs mediating effects at different sites or via different mechanisms may improve postoperative analgesic efficacy and outcome.” Ketoro- lac and other NSAIDs reduce pain after surgery by inhib- iting synthesis and release of prostaglandins at the site of surgical trauma,‘” whereas morphine activates spinal and supraspinal opioid receptors that inhibit the release of nociceptive neurotransmitters.14 Our findings support those of Parker et uL.,~ who reported that IV ketorolac reduced PCA with morphine and meperidine require- ments in patients recovering from abdominal hysterec- tomy. The opioid-sparing effect of PCA noted in their single-dose (ketorolac 30 mg) trial was associated with a more rapid return of bowel function and a shortened hospital stay. Although the combination of PCA plus ad- junctive analgesic therapy is able to reduce self-adminis- tration attempts and total opioid dose,R-5.” improvements in analgesic efficacy are less commonly observed. Our findings are consistent with those of Grass et al.,” who

26 J. Clin. Anesth., vol. 6, January/February 1994

reported that the combination of epidural PCA with fen- tanyl plus ketorolac 30 mg reduced movement-associated pain scores to a greater degree than did fentanyl alone in patients recovering from radical prostatic surgery.

When used as an analgesic adjunct, ketorolac appears to have a very shallow dose response.‘,9,“,X In agreement with previous findings with IM and IV bolus doses or continuous infusions of ketorolac,7.9.“.X our study shows that increasing dose size is not associated with greater analgesic effect. In this regard, Kenny et al.’ and Gillies et al.” reported that a ketorolac 3 kg/hr infusion was no more efficacious than a 1.5 kg/hr infusion. In earlier evaluations of IM ketorola? and IV ketorolac,‘.7.X 10 to 15 mg administered every 6 hours was as effective as 30 mg in reducing 24hour requirements of PCA with opioids.

The possible additive (or even synergistic) interactions of morphine and ketorolac will require more extensive dose-response studies. However, based on our finding that 15 mg was just as effective as 30 mg, we again empha- size that overall dose administered should be reduced when ketorolac is coadministered with opioid analgesics. Since ketorolac-related side effects appear to be dose related,“‘,* the improved analgesic effectiveness noted i n

the low-dose group was gratifying. In this regard, even smaller doses (5 to 7.5 mg) should be examined, espe- cially in patients who are at risk for increased surgical and gastrointestinal bleeding.

Reductions in pain intensity noted in our lower-dose ketorolac group (Group 2) may have been responsible in part for increases in patient satisfaction with analgesic therapy. Overall, patient satisfaction with analgesic ther- apy reflects a composite of the effectiveness of pain relief balanced by the occurrence of troublesome side effects.“’ Despite requiring approximately 20% less PCA with mor- phine, patients treated with IV ketorolac did not benefit from a reduction in opioid-related side effects. This find- ing is similar to that reported by Parker et al.,’ who noted that patients receiving IV ketorolac experienced an almost identical rate of pruritus and nausea despite significant reductions in opioid self-administration. Ketorolac-in- duced side effects are essentially unrelated to, and nonad- ditive with, those associated with PCA with opioids. Thus, the combination of improved pain relief and a stable (ac- ceptable) level of opioid-induced side effects may have been responsible for the greater level of satisfaction ob- served. However, the fact that patients receiving the higher dose did not experience statistically significant improve- ment in pain relief but indicated greater satisfaction with their therapy suggests that other factors may be responsi- ble for patient satisfaction with PCA therapy.

We feel strongly that IM dosing of analgesics is a throw-

*Lanza FL, Karlin DA, Vee JP: A double-blind, placebo-controlled endoscopic study comparing the mucosa injury seen with an orally and parenterally administered new non-steroid analgesia ketorolac tromethamine at therapeutic and subtherapeutic doses [Abstract]. Am J Gastroenterol 1987;82:939.

back to older, less reliable forms of pain therapy and has no place in the setting of modern acute pain manage- ment. The ability to administer both NSAID and opioid analgesia via the IV route avoids absorption variability and pain associated with injection, and it represents an improvement in overall patient care. In this preliminary evaluation, IV ketorolac was withheld during the opera- tive period and administered postsurgically only to pa- tients having minimal intraoperative bleeding. It is conceivable that the effectiveness of ketorolac may be further improved by administering a loading dose prior to the surgical incision rather than waiting for the patient to experience postsurgical discomfort in the recovery room. This may explain why studies using preemptive administration of ketorolac reported significant reduc- tions in requirements of PCA with morphine, especially during the early postsurgical interval.‘~4~fi~x

Based on our experience with IV ketorolac, we recom- mend that a slow loading dose of 30 mg be given following anesthetic induction. Such timing may ensure that ketor- olac plasma levels are adequate to attenuate release of prostaglandin and other pain activators associated with incision and intraoperative dissection.‘l Maintenance doses of IV ketorolac (15 mg) should be administered as a dilute solution over a lo- to 1%minute period. This slow rate of administration may lower the risk of phlebitis, reduce peak plasma concentrations, and avoid potential toxicity. In this regard, a recent pharmacokinetic investi- gation I’) found that peak plasma levels of ketorolac oc- curred sooner and were of greater magnitude with IV administration than with IM bolus administration. How- ever, plasma levels declined rapidly in the IV group and after 10 minutes were identical with either mode of ad- ministration.‘”

To summarize, combination analgesic therapy con- sisting of PCA with opioids and the NSAID ketorolac tromethaminr provided improved postsurgical pain scores and greater patient satisfaction than did PCA alone. Further work is required to assess the optimal dose and timing of IV ketorolac administration when combined with PCA with opioids for postoperative anal- gesia.

Addendum

The average cost of 24 hours of PCA therapy with mor- phine for each placebo group patient (Group 1) was $10.84; for the group receiving IV ketorolac (Torodol) 30 mg loading dose followed by IV Torodol 15 mg every six hours (Group 2), cost per patient was $23.85; and for the group receiving IV Torodol 60 mg loading dose followed by IV Torodol 30 mg every six hours (Group 3), cost per patient was $24.20.

References

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14

15.

16

Pierce RJ, Frdgen RJ, Pemberton DM: Intravenous ketorolac tromethamine verslls morphine sulfate in the treatment of immediate postoperative pain. Phnrmncothrwpy 1990;10[6(Pt 2)]:lllS-5s. O’Hara DA, Frdgen ICJ, Kinzer M, Pemberton DM: Ketorolac tromethamine as compared with morphine sulfate for treat- ment of postoperative pain. Clilz Pharmncof Thor 1987;41:556-61. Sevarino FB, Sinatra RS, Paige D, Ning T, Brull SJ, Silverman DG: The efficacy of intramuscular ketorolac in combination with intravenous PCA morphine for postoperative pain relief: J Clin An&h 3992;4:285-8. Parker RK, Holtmann B, White PF: Effect of ketorolac on the postoperative opioid requirement and recovery profile [Ah- stracr] Anathaiolqq 1991;75:A758. Kenny GNC, McArdle CS, Aitken HH: Parenteral ketorolx: oplate-sparing effect and lack of cardiorespiratory depression in the perioperative patient. Phnrmacothrmapy 199O;lO: [6(Pt 2)]:127S-31s. Grass,JA, Sakima NT, Valley M. et al: \sscssment of ketorolac as an adjuvant to fentanyl patient controlled epidural analgesia after radical retropubic prostatectomv. .~nc:rthesioloff: 1993;78: fi42-8. Camu F, Van Overberge I., Bullingham K, Lloyd J: Hemody- namic effects of two intravenous doses of ketorolac trometha- mine compared with morphine. Phnrmarothrrapy 1990; 10[6(Pt 2)]:122S-6s. Brown CR, Moodie JE, Wild VM, Bynum I:J: Comparison of intravenous ketor-olac tromethamine and morphine sulfate in the treatment of’ postoperative pain. Phnwnnrothwap~ 1990; 10[6(Pt 2)]:116S-21s. Wheatley RG, Blackburn A, Stevens J, Madej TH, Hunter D: Analgesic effects of intravenous ketorolac in lower abdominal surgery [Abstract]. Anesthuzology 1992:77:4X18. Jung D, Mroszczak E, Bynum I.: Pharmacokinetics of ketorolar tromethamine in humans after intravenous, intramuscular and oral administration. Eur.] C&l Pharmnrol 3988;35:423-5. Flay BJ, Royko (X, Fleitman JS: (Compatibility of ketorolac tromethamine injection with common inf&ion fluids and ad- ministration sets. Am] Hasp Pharm 1990;47:1097-100. DahlJB, Rosenberg J, Dirkes WE, Mogensen T, Kehlet H: Pre- vention of postoperative pain by halanred analgesia. Rr,/Annesth 1990;64:518-20. Cousins MJ, Mathrr 1.E: Intrdthecdl and epidural administra- tion of opioids. Anathesiolo~ 1984;6 I:276-310. Ferrira SH: Prostaglandins: peripheral and central analgesia. In: Bonica JJ, Lmdblom U, Iggo A (eds): ilduances in Pain Research ond Therapy, vol. 5. New York: Raven Press, 19833627-34. Gillies GWA, Kenny GNC, Bullingham RE, McArdle CS: The morphine sparing effect of ketorolac tromethamine. A study of a new parenteral non-steroidal anti-inflammatory agent after abdominal surgery. Anoestheia 1987;42:727-31. Harrison DM, Sinatra RS, Morgese 1.. <:hung JH: Epidural narcotic and patient-controlled analgesia for post-cesarean sec- tion pain relief. lnesthe.siology 1988;68:454-‘7.

J. Clin. Anesth., vol. 6, ,January/February 1994 27