- Email: [email protected]
Safety and efficacy of ketorolac tromethamine 0.4% ophthalmic solution in post-photorefractive keratectomy patients
Safety and efficacy of ketorolac tromethamine 0.4% ophthalmic solution in post-photorefractive keratectomy patients Kerry D. Solomon, MD, Eric D. Donnenfeld, MD, Michael Raizman, MD, Helga P. Sandoval, MD, Katherine Stern, MS, Amanda VanDenburgh, PhD, Janet K. Cheetham, PharmD, Rhett Schiffman, MD, for the Ketorolac Reformulation Study Groups 1 and 2 Purpose: To evaluate the safety and analgesic efficacy of ketorolac tromethamine 0.4% ophthalmic solution in postoperative photorefractive keratectomy (PRK) patients. Setting: Fifteen clinical sites in the eastern and southern United States. Methods: This pooled analysis of 2 multicenter, randomized, double-masked, vehicle-controlled, parallel-group studies comprised 313 patients having unilateral PRK. After surgery, patients were treated with 1 drop of ketorolac tromethamine 0.4% ophthalmic solution (Acular威 LS) (n ⫽ 156) or vehicle (n ⫽ 157) 4 times daily for up to 4 days. Pain intensity, pain relief, use of escape medication, and severity of ocular symptoms were assessed. Adverse events, epithelial healing, and visual acuity were recorded. Results: There was significantly less pain intensity experienced by patients in the ketorolac group (P⬍.001). During the first 12 hours post PRK, 50% fewer patients in the ketorolac group than in the vehicle group had severe to intolerable pain (41.6% [64/154] and 84.5% [131/155], respectively). The median time to no pain was 30 hours in the ketorolac group and 54 hours in the vehicle group (P⬍.001, survival analysis). Ketorolac patients reported significantly greater pain relief than vehicle patients throughout the study (P⬍.001). Ketorolac patients used significantly less escape medication than vehicle patients for 48 hours post PRK (Pⱕ.008). Treatment-related adverse events occurred in 2.6% (4/156) of ketorolac patients and 6.4% (10/157) of vehicle patients. Conclusion: Ketorolac 0.4% ophthalmic solution is safe and effective in reducing ocular pain when used 4 times daily for up to 4 days post PRK. J Cataract Refract Surg 2004; 30:1653–1660 2004 ASCRS and ESCRS
T
opical nonsteroidal antiinflammatory drugs (NSAIDs) are regularly used to control pain and photophobia associated with corneal incisions and abrasions and to control postoperative inflammation. They are also used to inhibit intraoperative miosis. Nonsteroidal antiinflammatory drugs act peripherally rather than Accepted for publication March 30, 2004. Reprint requests to Kerry D. Solomon, MD, Magill Research Center for Vision Correction, Medical University of South Carolina, Storm Eye Institute, 167 Ashley Avenue, Charleston, South Carolina 29425, USA. 2004 ASCRS and ESCRS Published by Elsevier Inc.
centrally, thereby eliminating central nervous system side effects. A topical NSAID offers several potential benefits over traditional systemic agents for ocular pain relief. These benefits include decreased systemic absorption and increased ocular absorption. Nonsteroidal antiinflammatory drugs are also a safe alternative to topical corticosteroids, which can cause elevated intraocular pressure, cataracts, and increased risk for infections.1 Typical of all drugs, topical NSAIDs have been associated with some adverse events. Surface toxicity is a concern, and incidents of keratitis, corneal subepithelial 0886-3350/04/$–see front matter doi:10.1016/j.jcrs.2004.05.019
KETOROLAC TROMETHAMINE AFTER PRK
infiltrates, ulceration, and corneal melts have been reported.2–4 Patients have often reported burning or stinging on instillation.1 This ocular irritation could be dose related.1 Ketorolac tromethamine ophthalmic solution 0.5% is an NSAID with proven analgesic and antiinflammatory activity and safety.2,5–7 The most common adverse event associated with the use of the ketorolac 0.5% formulation is ocular irritation, primarily burning and stinging on instillation. Ocular irritation has the potential to mask some of the drug’s efficacy. We evaluated the efficacy and safety of ketorolac tromethamine 0.4% ophthalmic solution for the treatment of ocular pain in postoperative photorefractive keratectomy (PRK) surgery patients. Photorefractive keratectomy was chosen as the model for this study because it often causes severe postoperative ocular pain. Since ketorolac 0.5% has been shown to be effective in reducing pain after PRK,7 we hypothesized that the lower concentration of the active and inactive agents would reduce the incidence of adverse events and enhance comfort while maintaining clinical efficacy.
Patients and Methods Study Design Ketorolac tromethamine 0.4% ophthalmic solution (Acular威 LS) was evaluated in 2 identical multicenter, randomized, double-masked, vehicle-controlled, parallel-group, phase 3 clinical studies in patients having unilateral PRK. One drop of ketorolac or its vehicle was administered 4 times daily in the study eye for up to 4 days. Patients exited the study when reepithelialization had occurred and they had no pain. Patients used electronic diaries (In Vivo Data) to record pain assessments and the use of masked study medications and escape medications. The electronic diary was programmed to prompt the patient to complete an assessment of pain relief and pain intensity. The electronic diary continued to prompt up to 4 times per day at 4-hour intervals. The diary entries were time and date stamped. The trial was conducted in compliance with good clinical practices, institutional review board regulations, sponsor and investigator obligations, informed consent regulations, and the Declaration of Helsinki.
Eligible patients were at least 18 years of age and were candidates for unilateral PRK within 7 days of the screening visit. Selected patients had a best corrected Early Treatment of Diabetic Retinopathy Study (ETDRS) visual acuity score equivalent to a Snellen score of 20/100 or better in each eye and willingness and demonstrated ability to use an electronic diary. Patients were excluded if they had a history of refractive ocular surgery in either eye; had poor tolerance to any component of the masked study treatments: NSAIDs, fluoroquinolones including ofloxacin ophthalmic solution (Ocuflox威), acetaminophen, codeine, or any component of Tylenol威 with Codeine #3; had used analgesic agents (including acetaminophen or NSAIDs), corticosteroids, or anticoagulants within the 7 days prior to surgery; and had a clinically significant operative complication from the PRK surgery.
Surgical Technique Unilateral PRK was performed according to each investigator’s usual standard of practice. All surgery was performed under topical anesthesia. The epithelium was removed using each surgeon’s preferred technique. The deepithelialization methods included laser, mechanical scrapping, alcohol removal, and other.
Intervention and Timing On the day of the PRK surgery, 1 dose of the masked study treatment was administered immediately postoperatively, at 3 hours, and then every 4 hours while the patient was awake for a maximum of 4 doses. During the next 3 days, 1 dose of the masked study treatment was applied 4 times daily. Throughout the study, 1 drop of ofloxacin was administered approximately 5 minutes before each dose of masked study treatment. Patients were given an escape medication, acetaminophen 300 mg with codeine 30 mg (Tylenol with Codeine #3), to be taken as 1 tablet up to every 4 hours as needed for intolerable pain. Patients were randomized in a 1:1 ratio in each study to receive ketorolac tromethamine 0.4% ophthalmic solution or its vehicle. Patients received study treatment for up to 4 days. They exited the study when reepithelialization had occurred and they were pain free. If reepithelialization did not occur by 4 days, patients immediately entered the posttreatment phase and continued daily visits until reepithelialization was completed. During the posttreatment phase, masked study treatment was discontinued and patients used ofloxacin and escape medications at the investigator’s discretion.
Outcome Measures Patients Patients were recruited from 15 sites in the United States. All patients gave written informed consent at the time of enrollment in the study, which was conducted from August 2001 to December 2001. 1654
Primary Efficacy Variable. The primary efficacy variable was pain intensity during the first 12 hours post PRK. Pain intensity was scored by the patients on a 5-point scale (0 ⫽ no pain and 4 ⫽ intolerable pain). Patients recorded the pain intensity in their electronic diaries immediately before
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
KETOROLAC TROMETHAMINE AFTER PRK
each ofloxacin dose and immediately before use of escape medication. Pain intensity assessment was not recorded if escape medication was used more than 10 minutes before the assessment. Secondary Efficacy Variable. Pain intensity ratings at all time points except the first 12 hours post PRK were defined as secondary efficacy variables. Patients rated and recorded pain relief approximately 2 hours after each instillation of masked study treatment (except during the immediate postoperative period). Pain relief was scored on a 5-point scale (1 ⫽ complete pain relief and 5 ⫽ no pain relief). Patients recorded escape medication use in their electronic diaries. Escape medication could be self-administered by patients to treat intolerable pain. Site personnel recorded the severity and frequency of ocular symptoms associated with the surgery, including foreign-body sensation, photophobia, burning/stinging, tearing, and itching, at all visits. Severity of ocular symptoms was classified as none, trace, mild, moderate, or severe. Efficacy Analysis and Statistical Methods. The results are based on the total number of patients who recorded the information in electronic diaries. Pain intensity and pain relief were analyzed in 12-hour periods post PRK. If a patient had more than 1 pain intensity rating during any post-PRK analysis period, the maximum (ie, worst pain intensity) rating was used. Pain intensity recorded immediately after surgery and before the first dose of study medication was not included in the first 12-hour postPRK period. Statistical analyses were performed for the efficacy variables using the intent-to-treat population. For differences in distribution of pain intensity, pain relief, escape medication used, and symptoms of discomfort, differences between treatment groups were tested using the Cochran-Mantel-Haenszel (CMH) test stratified by investigator using modified ridit scores and testing for row mean score differences. Binomial response categories of severe or intolerable pain intensity/ less than severe, of escape medication taken/not taken, or of symptoms of discomfort mild or better/worse were compared between treatment groups using the CMH test stratified by the investigator for general association. Time to first zero pain and time to first use of escape medication were examined using the generalized Wilcoxon rank sum test from a survival analysis. Safety Analysis and Statistical Methods. Assessment of adverse events began after instillation of study treatment and continued throughout the study, including the posttreatment period. An adverse event whose onset or initial detection occurred within 14 days after the patient exited the study was also recorded. A reported adverse event was graded by the investigator for severity (mild, moderate, or severe) and assessed for its relationship to the study treatment (unrelated, possible, probable, definite). Patients had slitlamp biomicroscopy and visual acuity testing at the screening visit; in addition, biomicroscopy was
Table 1. Patient demographics. Characteristic
Number
Percentage
153
48.9
160
51.1
Sex Female Male Mean age (y) ⫾ SD
39 ⫾ 10
Race White
279
89.1
Black
16
5.1
Hispanic
10
3.2
Asian
7
2.2
Other
1
0.3
performed daily and visual acuity was tested prior to exit. Slitlamp biomicroscopy included examination of the lids, conjunctiva, cornea, lens, and anterior chamber. Visual acuity was measured using the ETDRS chart. Differences in the proportion of patients reporting an adverse event were analyzed using the Pearson chi-square test or the Fisher exact test. Corneal Reepithelialization. Corneal reepithelialization was evaluated (complete versus incomplete) for the study eye at each office visit until the patient exited the study.
Results Patient Flow and Follow-up Three hundred thirteen patients enrolled in the study; 156 received ketorolac and 157, vehicle. Of the 313 who entered the treatment period, 49 continued in the posttreatment period: 33 ketorolac patients and 16 vehicle patients. Two hundred ninety-four patients (93.9%) completed the study (ie, completed 4 days of treatment or exited early with no pain and complete reepithelialization). Six ketorolac patients (3.8%) and 13 vehicle patients (8.3%) discontinued the study during the treatment phase. Patient Demographics Demographic data between the 2 treatment groups did not differ significantly. The mean age in each group was 40 years (range 18 to 66 years). The population was primarily white (89.1%) (Table 1). Tables 2 and 3 summarize ablation depth and width, respectively, by epithelial removal technique. There were no significant between-group differences.
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
1655
KETOROLAC TROMETHAMINE AFTER PRK
Figure 1. (Solomon) Distribution of ketorolac and vehicle patients
Figure 2. (Solomon) Cumulative incidence rates of time to first no
by pain intensity during the first 12 hours post PRK (overall P⬍.001 versus vehicle patients).
pain post PRK in ketorolac and vehicle patients (overall survival analysis P⬍.001 versus vehicle patients).
Efficacy Effect of Ketorolac on Pain Intensity. During the first 12 hours, there was a significant difference in the distribution of patients by pain intensity, with ketorolac patients experiencing significantly less intense pain than did vehicle patients (P⬍.001) (Figure 1). There were significantly fewer patients in the severe and intolerable pain categories in the ketorolac group (41.6% [64/154]) than in the vehicle group (84.5% [131/155]) (P⬍.001). During the next 12 to 48 hours, there were significant differences in the distribution of patients by pain intensity, with ketorolac patients experiencing less pain intensity (Pⱕ.001). There were fewer ketorolac patients than vehicle patients in the severe and intolerable pain categories 12 to 48 hours post PRK. There were no significant differences between the treatment groups in pain inten-
sity during the remaining 12-hour periods (from 48 to 108 hours post PRK). There was a significant difference in the cumulative incidence rates of time to first no-pain report, with ketorolac patients experiencing no pain more quickly (P⬍.001) (Figure 2). The median time to the first nopain report was 30 hours in the ketorolac group and 54 hours in the vehicle group. Forty-two percent (65/156) of the ketorolac patients and 8% (13/157) of the vehicle patients experienced the first no pain during the first 12 hours post PRK. Effect of Ketorolac on Pain Relief. Ketorolac patients reported greater pain relief than the vehicle patients throughout the study (Figure 3). The percentage of patients experiencing pain relief was significantly higher in the ketorolac group for all 12-hour periods (P⬍.001).
Table 2. Mean ablation depth by method of epithelium removal.
Table 3. Mean ablation width by method of epithelium removal.
Epithelium Removal Method
Mean Ablation Depth ⫾ SD (m) Ketorolac
Vehicle
P Value*
Laser
44.3 ⫾ 29.1 (n ⫽ 36)
45.9 ⫾ 32.7 (n ⫽ 40)
.80
Mechanical scraping
50.8 ⫾ 29.0 (n ⫽ 67)
55.0 ⫾ 32.5 (n ⫽ 63)
Alcohol removal
47.9 ⫾ 26.7 (n ⫽ 42)
Other Total *Wilcoxon rank sum test
1656
Epithelium Removal Method
Mean Ablation Width ⫾ SD (mm) Ketorolac
Vehicle
P Value*
Laser
6.3 ⫾ 0.6 (n ⫽ 36)
6.3 ⫾ 0.6 (n ⫽ 40)
.92
.55
Mechanical scraping
7.4 ⫾ 0.9 (n ⫽ 67)
7.4 ⫾ 1.0 (n ⫽ 63)
.70
47.2 ⫾ 24.3 (n ⫽ 42)
.86
Alcohol removal
6.5 ⫾ 0.9 (n ⫽ 42)
6.6 ⫾ 0.9 (n ⫽ 42)
.27
46.9 ⫾ 28.5 (n ⫽ 10)
38.4 ⫾ 21.0 (n ⫽ 12)
.50
Other
7.6 ⫾ 0.7 (n ⫽ 10)
7.3 ⫾ 2.2 (n ⫽ 12)
.34
48.3 ⫾ 28.2 (n ⫽ 155)
49.3 ⫾ 30.0 (n ⫽ 157)
.83
Total
6.9 ⫾ 1.0 (n ⫽ 155)
6.9 ⫾ 1.1 (n ⫽ 157)
.83
*Wilcoxon rank sum test
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
KETOROLAC TROMETHAMINE AFTER PRK
Figure 3. (Solomon) Percentage of ketorolac and vehicle patients
Figure 4. Percentage of ketorolac and vehicle patients requiring
experiencing complete or great pain relief for each 12-hour period post PRK. Pain relief was assessed 2 hours after drop instillation.
no escape medication (*P⬍.001 compared with vehicle patients; **P ⫽ .005 compared with vehicle patients).
Fewer patients in the ketorolac group than in the vehicle group reported little or no pain relief at all time points. Use of Escape Medication. During the first 12 hours post PRK, significantly fewer patients in the ketorolac group (45.5% [71/156]) than in the vehicle group (90.4% [141/156]) took escape medication (P⬍.001) (Figure 4). From 12 to 48 hours, use of escape medication was also significantly lower in the ketorolac group than in the vehicle group (Pⱕ.005). There were no significant differences in the use of escape medication for subsequent periods. During the first 12 hours post PRK, ketorolac patients took fewer tablets of escape medication than vehicle-treated patients (medians of 0.0 and 2.0 tablets, respectively). There was a significant difference in the distribution of the number of tablets of escape medication, with ketorolac patients using fewer tablets than vehicle patients (P⬍.001). From 12 to 48 hours, there was also a significant difference, with fewer patients in the ketorolac group than in the vehicle group using 1 tablet or more (Pⱕ.008). There was no significant difference in the use of escape medication in subsequent periods. Frequency and Severity of Ocular Symptoms. Vehicle patients reported significantly more ocular symptoms (associated with the surgery itself) during the first 24 hours after surgery. Ketorolac patients had significantly fewer symptoms of burning/stinging, foreign-body sensation, photophobia, and tearing (P⬍.001). Two-thirds (75.0%, 117/156) of ketorolac patients and 52.2% (82/157) of vehicle patients reported mild or less proce-
dure-related burning/stinging during the first 24 hours (Figure 5). Similarly, 63.5% (99/156) of ketorolac patients and 41.1% (65/157) of vehicle patients reported
Figure 5. (Solomon) Percentage of ketorolac and vehicle patients experiencing ocular symptoms (associated with the surgery) during the first 24 hours post PRK (P⬍.001 for all symptoms except itching compared with vehicle patients).
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
1657
KETOROLAC TROMETHAMINE AFTER PRK
Table 4. Treatment-related adverse events. Frequency (%) Ketorolac (n ⫽ 36)
Vehicle (n ⫽ 157)
Eye pain
2 (1.3)
4 (2.5)
Headache
1 (0.6)
3 (1.9)
Corneal infiltrates
1 (0.6)
1 (0.6)
Eye edema
1 (0.6)
0 (0.0)
Conjunctival hyperemia
1 (0.6)
0 (0.0)
Adverse Event
Ocular irritation
0 (0.0)
1 (0.6)
Nausea
0 (0.0)
1 (0.6)
Vomiting
0 (0.0)
1 (0.6)
mild or better foreign-body sensation during the first 24 hours. More ketorolac patients (50.6%, 79/156) than vehicle patients (26.1%, 41/157) had mild or better photophobia during the first 24 hours, and more ketarolac patients (50.6%, 79/156) than vehicle patients (28.7%, 45/157) had mild or better tearing. There were no significant between-group differences in ocular symptoms at other time points. Safety One or more treatment-related adverse events were reported in 2.6% (4/156) of patients in the ketorolac group and 6.4% (10/157) of patients in the vehicle group. There were no statistically significant differences between the treatment groups for any reported treatment-related adverse event (Table 4). Ocular irritation, which includes burning and stinging, was reported by 1 vehicle patient. There were no serious treatmentrelated adverse events or changes in visual acuity. Corneal infiltrates were reported in 3 patients (regardless of relationship to treatment): 2 ketorolac patients and 1 vehicle patient. In the ketorolac patients, 1 incident was considered unrelated to treatment and 1, treatment related. In the vehicle patient, the corneal infiltrate was considered treatment related. All 3 patients recovered without sequelae. Overall, 14 patients (6 ketorolac and 8 vehicle) discontinued because of adverse events. In most patients (9), the reason for discontinuation was inadequate pain relief from the escape medication. Other adverse events leading to discontinuation included eye pain not related to escape medication use (2 patients), corneal infiltrates (2 patients), and rhinitis (1 patient). 1658
Given the nature of the PRK procedure, most patients in both treatment groups experienced an expected short-term worsening of the best corrected visual acuity. There was no statistical difference between the treatment groups in changes in visual acuity from baseline. Changes from baseline in the severity of any biomicroscopic findings were reported for a similar proportion of patients in both groups. The changes were consistent with the expected postoperative findings of PRK. A greater number of ketorolac patients (33/156) than vehicle patients (16/157) entered the posttreatment period because of a delay in complete corneal reepithelialization. The median time to reepithelialization was after day 2 and by day 3 post PRK in both treatment groups. The mean time to corneal reepithelialization was 3.3 days in the ketorolac group and 2.7 days in the vehicle group. The half-day difference was not considered clinically significant.
Discussion The new ketorolac ophthalmic solution, with a 20% lower concentration of active ingredient (ketorolac) and a lower concentration of certain inactive ingredients, maintained efficacy and provided excellent comfort after PRK. Ketorolac 0.4% (Acular LS) received approval from the United States Food and Drug Administration on May 30, 2003. Similar to ketorolac tromethamine 0.5% ophthalmic solution, ketorolac tromethamine 0.4% is efficacious against severe ocular pain caused by PRK. Ketorolac 0.4% was significantly better than vehicle in relieving pain, reducing pain intensity, reducing escape medication use, and limiting ocular discomfort. The pain induced by PRK is greater than that induced by radial keratotomy8 and may equal or surpass that induced by corneal abrasions and cataract surgery. Therefore, ketorolac 0.4% might be beneficial in alleviating ocular pain due to a variety of causes. Furthermore, the analgesic efficacy was evident even though the drug was administered only postoperatively. Preoperative or intraoperative use would probably have enhanced the analgesic effect by enabling the ketorolac to inhibit mediators of ocular pain before their synthesis.7 The cornea is 1 of the most densely innervated and therefore most pain-sensitive tissues of the eye.8 Since PRK involves removal of the central corneal epithelium and physical reshaping of the cornea, substantial postop-
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
KETOROLAC TROMETHAMINE AFTER PRK
erative pain is anticipated.9 While PRK and other surface procedures are undergoing a resurgence with the advent of wavefront-guided procedures, the downside of surface procedures (including PRK) is corneal pain, photophobia, burning/stinging, and foreign-body sensation. Although NSAIDs are generally effective in reducing discomfort associated with corneal surface procedures, they are often irritating and uncomfortable to instill, leading to potential noncompliance. The new formulation of ketorolac was developed with the specific goal of maintaining efficacy while minimizing the side effects associated with NSAIDs. While the patients were not asked specifically whether the drop caused burning and stinging, they were asked generally whether they experienced burning and stinging. Overall, ketorolac patients experienced significantly less procedure-related burning and stinging; ie, 75% of patients using the new formulation were comfortable. This study supports the efficacy and safety of the ketorolac reformulation for postPRK discomfort. Patients often complain about pain post PRK. Maximum pain intensity typically occurs during the first 12 hours.8 Figure 1 shows that the maximum pain intensity was statistically less in ketorolac patients than in vehicle patients (P⬍.001). Of the patients receiving ketorolac, 40% had none to mild pain compared with 4% of the patients receiving vehicle. It is particularly noteworthy that ketorolac 0.4% had greater analgesic efficacy than vehicle during the period of maximum pain intensity, despite the use of escape medication. Ketorolac patients also experienced significantly less intense pain during the first 48 hours post PRK (Figure 2). During the first 6 hours, 7 times more ketorolac patients than vehicle patients experienced no pain. It was not until 72 hours post PRK that similar numbers of vehicle and ketorolac patients experienced no pain. Ketorolac patients experienced no pain significantly more quickly than vehicle patients (P⬍.001, Figure 2). Not only did ketorolac patients have significantly less intense pain, they also had greater pain relief (P⬍.001, Figure 3). Figure 3 shows that during the first 24 hours post PRK, 4- to 15-fold more ketorolac patients than vehicle patients experienced complete to high relief in discomfort. Ketorolac patients experienced this relief despite requiring significantly less escape medication than vehicle patients (Pⱕ.005, Figure 4). During the first 24 hours, two-thirds of ketorolac patients
did not require escape medication. Vehicle patients reported significantly more symptoms of procedurerelated burning/stinging, foreign-body sensation, photophobia, and tearing during the first 24 hours than ketorolac patients (P⬍.001, Figure 5). The product labeling for ketorolac tromethamine 0.5% ophthalmic solution (Acular) lists transient stinging and burning on instillation as the most frequent adverse event, with up to 40% of patients reporting it. In contrast, no patient in the ketorolac 0.4% group in this study reported treatment-related incidents of stinging and burning as an adverse event. Therefore, ketorolac 0.4% appears to be more comfortable. This increase in comfort may lead to improved patient compliance. Topical NSAIDs have been associated with adverse events such as keratitis, corneal subepithelial infiltrates, and corneal melts.3,4,10 In this study, only 3 of 313 patients (2 ketorolac and 1 vehicle) developed corneal infiltrates. In only 1 of the 2 ketorolac cases was the corneal infiltrate considered treatment related (Table 2). All 3 patients recovered without sequelae, with follow-up uncorrected visual acuity of 20/20 in the study eye. Although time to reepithelialization was slightly longer (ie, less than a day) in ketorolac patients, the difference between groups was not clinically or statistically significant. There were no cases of corneal melts, keratitis, infection, ulceration, or perforation. In the vehicle group, time to no pain (Figure 2) was associated with complete reepithelialization. In ketorolac patients, time to no pain was not associated with reepithelialization, emphasizing the strong analgesic efficacy of the ketorolac reformulation. Despite the longer time to reepithelialization, ketorolac patients were more comfortable. They had less pain, less severe pain, no treatment-related burning and stinging, and less use of escape medications then vehicle patients, despite the slightly faster reepithelialization in the latter group. Overall, the results demonstrate that ketorolac patients experienced less pain more rapidly and greater pain relief than vehicle patients. The new formulation of ketorolac, which contains reduced concentrations of active and certain inactive ingredients, decreases the potential for irritation and toxicity. Further research on ketorolac tromethamine 0.4% ophthalmic solution will be worthwhile.
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
1659
KETOROLAC TROMETHAMINE AFTER PRK
Ketorolac Reformulation Study Groups Bruce Altman, Nicole Anderson, Amin Ashrafzadeh, Frank Bishop, Lawrence C. Chao, Roy S. Chuck, Elizabeth A. Davis, Eric Donnenfeld, Ralph Falkenstein, David Hardten, Barry Katzman, Matthew Kaufman, Jane Ahn Lee, Edgar Levine, Ning Lin, Richard L. Lindstrom, Anthony Lombardo, David Markoff, Stephen Mathias, Peter J. McDonnell, Joseph M. Miller, Shrabanee Mitra, Sylvia Norton, Matthew Paul, Jonathan R. Pirnazar, Francis Price, Michael Raizman, Turpin Rose, Paul Ruggier, Thomas Samuelson, Steve Schallhorn, Timothy Schneider, Robert W. Snyder, Kerry D. Solomon, David Tanzer, James Tidwell, Elizabeth Tonon, David T. Vroman, James Weisz, John Wittpenn, Richard Yee, William Zeh, Stephen Zuckerman
5.
References
9.
1. Jaanus SD, Cheetham JK, Lesher GA. Antiinflammatory drugs. In: Bartlett JD, Jaanus SD, eds, Clinical Ocular Pharmacology, 4th ed. Boston, MA, Butterworth-Heinemann, 2001; 265–298 2. Solomon KD, Vroman DT, Barker D, Gehlken J. Comparison of ketorolac tromethamine 0.5% and rimexolone 1% to control inflammation after cataract extraction; prospective randomized double-masked study. J Cataract Refract Surg 2001; 27:1232–1237 3. Teal P, Breslin C, Arshinoff S, Edmison D. Corneal subepithelial infiltrates following excimer laser photorefractive keratectomy. J Cataract Refract Surg 1995; 21:516–518 4. Guidera AC, Luchs JI, Udell IJ. Keratitis, ulceration, and perforation associated with topical nonsteroidal anti-
10.
1660
6.
7.
8.
inflammatory drugs. Ophthalmology 2001; 108:936– 944 Heier J, Cheetham JK, Degryse R, et al. Ketorolac tromethamine 0.5% ophthalmic solution in the treatment of moderate to severe ocular inflammation after cataract surgery: a randomized, vehicle-controlled clinical trial. Am J Ophthalmol 1999; 127:253–259 Holzer MP, Solomon KD, Sandoval HP, Vroman DT. Comparison of ketorolac tromethamine 0.5% and loteprednol etabonate 0.5% for inflammation after phacoemulsification; prospective randomized double-masked study. J Cataract Refract Surg 2002; 28:93–99 Rajpal RK, Cooperman BB. Analgesic efficacy and safety of ketorolac after photorefractive keratectomy; the Ketorolac Study Group. J Refract Surg 1999; 15:661–667 Mu¨ller LJ, Marfurt CF, Kruse F, Tervo TMT. Corneal nerves: structure, contents and function. Exp Eye Res 2003; 76:521–542; erratum 2003; 77:253 Binder PS. Excimer laser photoablation: miracle or menace? J Ophthalmic Nurs Technol 1994; 13:61–63, 76 Congdon NG, Schein OD, von Kulajta P, et al. Corneal complications associated with topical ophthalmic use of nonsteroidal antiinflammatory drugs. J Cataract Refract Surg 2001; 27:622–631
From Storm Eye Institute, Medical University of South Carolina, Department of Ophthalmology (Solomon, Sandoval), Charleston, South Carolina, Ophthalmic Consultants of Long Island (Donnenfeld), Long Island, New York, New England Eye Center, Tufts University School of Medicine (Raizman), Boston, Massachusetts, and Allergan Inc. (Stern, VanDenburgh, Cheetham, Schiffman), Irvine, California, USA. Supported by Allergan Inc., Irvine, California, USA. Drs.VanDenburgh, Stern, Cheetham, and Schiffman are employees of Allergan, Inc. None of the other authors has a proprietary interest in the drug used in the study.
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
T
opical nonsteroidal antiinflammatory drugs (NSAIDs) are regularly used to control pain and photophobia associated with corneal incisions and abrasions and to control postoperative inflammation. They are also used to inhibit intraoperative miosis. Nonsteroidal antiinflammatory drugs act peripherally rather than Accepted for publication March 30, 2004. Reprint requests to Kerry D. Solomon, MD, Magill Research Center for Vision Correction, Medical University of South Carolina, Storm Eye Institute, 167 Ashley Avenue, Charleston, South Carolina 29425, USA. 2004 ASCRS and ESCRS Published by Elsevier Inc.
centrally, thereby eliminating central nervous system side effects. A topical NSAID offers several potential benefits over traditional systemic agents for ocular pain relief. These benefits include decreased systemic absorption and increased ocular absorption. Nonsteroidal antiinflammatory drugs are also a safe alternative to topical corticosteroids, which can cause elevated intraocular pressure, cataracts, and increased risk for infections.1 Typical of all drugs, topical NSAIDs have been associated with some adverse events. Surface toxicity is a concern, and incidents of keratitis, corneal subepithelial 0886-3350/04/$–see front matter doi:10.1016/j.jcrs.2004.05.019
KETOROLAC TROMETHAMINE AFTER PRK
infiltrates, ulceration, and corneal melts have been reported.2–4 Patients have often reported burning or stinging on instillation.1 This ocular irritation could be dose related.1 Ketorolac tromethamine ophthalmic solution 0.5% is an NSAID with proven analgesic and antiinflammatory activity and safety.2,5–7 The most common adverse event associated with the use of the ketorolac 0.5% formulation is ocular irritation, primarily burning and stinging on instillation. Ocular irritation has the potential to mask some of the drug’s efficacy. We evaluated the efficacy and safety of ketorolac tromethamine 0.4% ophthalmic solution for the treatment of ocular pain in postoperative photorefractive keratectomy (PRK) surgery patients. Photorefractive keratectomy was chosen as the model for this study because it often causes severe postoperative ocular pain. Since ketorolac 0.5% has been shown to be effective in reducing pain after PRK,7 we hypothesized that the lower concentration of the active and inactive agents would reduce the incidence of adverse events and enhance comfort while maintaining clinical efficacy.
Patients and Methods Study Design Ketorolac tromethamine 0.4% ophthalmic solution (Acular威 LS) was evaluated in 2 identical multicenter, randomized, double-masked, vehicle-controlled, parallel-group, phase 3 clinical studies in patients having unilateral PRK. One drop of ketorolac or its vehicle was administered 4 times daily in the study eye for up to 4 days. Patients exited the study when reepithelialization had occurred and they had no pain. Patients used electronic diaries (In Vivo Data) to record pain assessments and the use of masked study medications and escape medications. The electronic diary was programmed to prompt the patient to complete an assessment of pain relief and pain intensity. The electronic diary continued to prompt up to 4 times per day at 4-hour intervals. The diary entries were time and date stamped. The trial was conducted in compliance with good clinical practices, institutional review board regulations, sponsor and investigator obligations, informed consent regulations, and the Declaration of Helsinki.
Eligible patients were at least 18 years of age and were candidates for unilateral PRK within 7 days of the screening visit. Selected patients had a best corrected Early Treatment of Diabetic Retinopathy Study (ETDRS) visual acuity score equivalent to a Snellen score of 20/100 or better in each eye and willingness and demonstrated ability to use an electronic diary. Patients were excluded if they had a history of refractive ocular surgery in either eye; had poor tolerance to any component of the masked study treatments: NSAIDs, fluoroquinolones including ofloxacin ophthalmic solution (Ocuflox威), acetaminophen, codeine, or any component of Tylenol威 with Codeine #3; had used analgesic agents (including acetaminophen or NSAIDs), corticosteroids, or anticoagulants within the 7 days prior to surgery; and had a clinically significant operative complication from the PRK surgery.
Surgical Technique Unilateral PRK was performed according to each investigator’s usual standard of practice. All surgery was performed under topical anesthesia. The epithelium was removed using each surgeon’s preferred technique. The deepithelialization methods included laser, mechanical scrapping, alcohol removal, and other.
Intervention and Timing On the day of the PRK surgery, 1 dose of the masked study treatment was administered immediately postoperatively, at 3 hours, and then every 4 hours while the patient was awake for a maximum of 4 doses. During the next 3 days, 1 dose of the masked study treatment was applied 4 times daily. Throughout the study, 1 drop of ofloxacin was administered approximately 5 minutes before each dose of masked study treatment. Patients were given an escape medication, acetaminophen 300 mg with codeine 30 mg (Tylenol with Codeine #3), to be taken as 1 tablet up to every 4 hours as needed for intolerable pain. Patients were randomized in a 1:1 ratio in each study to receive ketorolac tromethamine 0.4% ophthalmic solution or its vehicle. Patients received study treatment for up to 4 days. They exited the study when reepithelialization had occurred and they were pain free. If reepithelialization did not occur by 4 days, patients immediately entered the posttreatment phase and continued daily visits until reepithelialization was completed. During the posttreatment phase, masked study treatment was discontinued and patients used ofloxacin and escape medications at the investigator’s discretion.
Outcome Measures Patients Patients were recruited from 15 sites in the United States. All patients gave written informed consent at the time of enrollment in the study, which was conducted from August 2001 to December 2001. 1654
Primary Efficacy Variable. The primary efficacy variable was pain intensity during the first 12 hours post PRK. Pain intensity was scored by the patients on a 5-point scale (0 ⫽ no pain and 4 ⫽ intolerable pain). Patients recorded the pain intensity in their electronic diaries immediately before
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
KETOROLAC TROMETHAMINE AFTER PRK
each ofloxacin dose and immediately before use of escape medication. Pain intensity assessment was not recorded if escape medication was used more than 10 minutes before the assessment. Secondary Efficacy Variable. Pain intensity ratings at all time points except the first 12 hours post PRK were defined as secondary efficacy variables. Patients rated and recorded pain relief approximately 2 hours after each instillation of masked study treatment (except during the immediate postoperative period). Pain relief was scored on a 5-point scale (1 ⫽ complete pain relief and 5 ⫽ no pain relief). Patients recorded escape medication use in their electronic diaries. Escape medication could be self-administered by patients to treat intolerable pain. Site personnel recorded the severity and frequency of ocular symptoms associated with the surgery, including foreign-body sensation, photophobia, burning/stinging, tearing, and itching, at all visits. Severity of ocular symptoms was classified as none, trace, mild, moderate, or severe. Efficacy Analysis and Statistical Methods. The results are based on the total number of patients who recorded the information in electronic diaries. Pain intensity and pain relief were analyzed in 12-hour periods post PRK. If a patient had more than 1 pain intensity rating during any post-PRK analysis period, the maximum (ie, worst pain intensity) rating was used. Pain intensity recorded immediately after surgery and before the first dose of study medication was not included in the first 12-hour postPRK period. Statistical analyses were performed for the efficacy variables using the intent-to-treat population. For differences in distribution of pain intensity, pain relief, escape medication used, and symptoms of discomfort, differences between treatment groups were tested using the Cochran-Mantel-Haenszel (CMH) test stratified by investigator using modified ridit scores and testing for row mean score differences. Binomial response categories of severe or intolerable pain intensity/ less than severe, of escape medication taken/not taken, or of symptoms of discomfort mild or better/worse were compared between treatment groups using the CMH test stratified by the investigator for general association. Time to first zero pain and time to first use of escape medication were examined using the generalized Wilcoxon rank sum test from a survival analysis. Safety Analysis and Statistical Methods. Assessment of adverse events began after instillation of study treatment and continued throughout the study, including the posttreatment period. An adverse event whose onset or initial detection occurred within 14 days after the patient exited the study was also recorded. A reported adverse event was graded by the investigator for severity (mild, moderate, or severe) and assessed for its relationship to the study treatment (unrelated, possible, probable, definite). Patients had slitlamp biomicroscopy and visual acuity testing at the screening visit; in addition, biomicroscopy was
Table 1. Patient demographics. Characteristic
Number
Percentage
153
48.9
160
51.1
Sex Female Male Mean age (y) ⫾ SD
39 ⫾ 10
Race White
279
89.1
Black
16
5.1
Hispanic
10
3.2
Asian
7
2.2
Other
1
0.3
performed daily and visual acuity was tested prior to exit. Slitlamp biomicroscopy included examination of the lids, conjunctiva, cornea, lens, and anterior chamber. Visual acuity was measured using the ETDRS chart. Differences in the proportion of patients reporting an adverse event were analyzed using the Pearson chi-square test or the Fisher exact test. Corneal Reepithelialization. Corneal reepithelialization was evaluated (complete versus incomplete) for the study eye at each office visit until the patient exited the study.
Results Patient Flow and Follow-up Three hundred thirteen patients enrolled in the study; 156 received ketorolac and 157, vehicle. Of the 313 who entered the treatment period, 49 continued in the posttreatment period: 33 ketorolac patients and 16 vehicle patients. Two hundred ninety-four patients (93.9%) completed the study (ie, completed 4 days of treatment or exited early with no pain and complete reepithelialization). Six ketorolac patients (3.8%) and 13 vehicle patients (8.3%) discontinued the study during the treatment phase. Patient Demographics Demographic data between the 2 treatment groups did not differ significantly. The mean age in each group was 40 years (range 18 to 66 years). The population was primarily white (89.1%) (Table 1). Tables 2 and 3 summarize ablation depth and width, respectively, by epithelial removal technique. There were no significant between-group differences.
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
1655
KETOROLAC TROMETHAMINE AFTER PRK
Figure 1. (Solomon) Distribution of ketorolac and vehicle patients
Figure 2. (Solomon) Cumulative incidence rates of time to first no
by pain intensity during the first 12 hours post PRK (overall P⬍.001 versus vehicle patients).
pain post PRK in ketorolac and vehicle patients (overall survival analysis P⬍.001 versus vehicle patients).
Efficacy Effect of Ketorolac on Pain Intensity. During the first 12 hours, there was a significant difference in the distribution of patients by pain intensity, with ketorolac patients experiencing significantly less intense pain than did vehicle patients (P⬍.001) (Figure 1). There were significantly fewer patients in the severe and intolerable pain categories in the ketorolac group (41.6% [64/154]) than in the vehicle group (84.5% [131/155]) (P⬍.001). During the next 12 to 48 hours, there were significant differences in the distribution of patients by pain intensity, with ketorolac patients experiencing less pain intensity (Pⱕ.001). There were fewer ketorolac patients than vehicle patients in the severe and intolerable pain categories 12 to 48 hours post PRK. There were no significant differences between the treatment groups in pain inten-
sity during the remaining 12-hour periods (from 48 to 108 hours post PRK). There was a significant difference in the cumulative incidence rates of time to first no-pain report, with ketorolac patients experiencing no pain more quickly (P⬍.001) (Figure 2). The median time to the first nopain report was 30 hours in the ketorolac group and 54 hours in the vehicle group. Forty-two percent (65/156) of the ketorolac patients and 8% (13/157) of the vehicle patients experienced the first no pain during the first 12 hours post PRK. Effect of Ketorolac on Pain Relief. Ketorolac patients reported greater pain relief than the vehicle patients throughout the study (Figure 3). The percentage of patients experiencing pain relief was significantly higher in the ketorolac group for all 12-hour periods (P⬍.001).
Table 2. Mean ablation depth by method of epithelium removal.
Table 3. Mean ablation width by method of epithelium removal.
Epithelium Removal Method
Mean Ablation Depth ⫾ SD (m) Ketorolac
Vehicle
P Value*
Laser
44.3 ⫾ 29.1 (n ⫽ 36)
45.9 ⫾ 32.7 (n ⫽ 40)
.80
Mechanical scraping
50.8 ⫾ 29.0 (n ⫽ 67)
55.0 ⫾ 32.5 (n ⫽ 63)
Alcohol removal
47.9 ⫾ 26.7 (n ⫽ 42)
Other Total *Wilcoxon rank sum test
1656
Epithelium Removal Method
Mean Ablation Width ⫾ SD (mm) Ketorolac
Vehicle
P Value*
Laser
6.3 ⫾ 0.6 (n ⫽ 36)
6.3 ⫾ 0.6 (n ⫽ 40)
.92
.55
Mechanical scraping
7.4 ⫾ 0.9 (n ⫽ 67)
7.4 ⫾ 1.0 (n ⫽ 63)
.70
47.2 ⫾ 24.3 (n ⫽ 42)
.86
Alcohol removal
6.5 ⫾ 0.9 (n ⫽ 42)
6.6 ⫾ 0.9 (n ⫽ 42)
.27
46.9 ⫾ 28.5 (n ⫽ 10)
38.4 ⫾ 21.0 (n ⫽ 12)
.50
Other
7.6 ⫾ 0.7 (n ⫽ 10)
7.3 ⫾ 2.2 (n ⫽ 12)
.34
48.3 ⫾ 28.2 (n ⫽ 155)
49.3 ⫾ 30.0 (n ⫽ 157)
.83
Total
6.9 ⫾ 1.0 (n ⫽ 155)
6.9 ⫾ 1.1 (n ⫽ 157)
.83
*Wilcoxon rank sum test
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
KETOROLAC TROMETHAMINE AFTER PRK
Figure 3. (Solomon) Percentage of ketorolac and vehicle patients
Figure 4. Percentage of ketorolac and vehicle patients requiring
experiencing complete or great pain relief for each 12-hour period post PRK. Pain relief was assessed 2 hours after drop instillation.
no escape medication (*P⬍.001 compared with vehicle patients; **P ⫽ .005 compared with vehicle patients).
Fewer patients in the ketorolac group than in the vehicle group reported little or no pain relief at all time points. Use of Escape Medication. During the first 12 hours post PRK, significantly fewer patients in the ketorolac group (45.5% [71/156]) than in the vehicle group (90.4% [141/156]) took escape medication (P⬍.001) (Figure 4). From 12 to 48 hours, use of escape medication was also significantly lower in the ketorolac group than in the vehicle group (Pⱕ.005). There were no significant differences in the use of escape medication for subsequent periods. During the first 12 hours post PRK, ketorolac patients took fewer tablets of escape medication than vehicle-treated patients (medians of 0.0 and 2.0 tablets, respectively). There was a significant difference in the distribution of the number of tablets of escape medication, with ketorolac patients using fewer tablets than vehicle patients (P⬍.001). From 12 to 48 hours, there was also a significant difference, with fewer patients in the ketorolac group than in the vehicle group using 1 tablet or more (Pⱕ.008). There was no significant difference in the use of escape medication in subsequent periods. Frequency and Severity of Ocular Symptoms. Vehicle patients reported significantly more ocular symptoms (associated with the surgery itself) during the first 24 hours after surgery. Ketorolac patients had significantly fewer symptoms of burning/stinging, foreign-body sensation, photophobia, and tearing (P⬍.001). Two-thirds (75.0%, 117/156) of ketorolac patients and 52.2% (82/157) of vehicle patients reported mild or less proce-
dure-related burning/stinging during the first 24 hours (Figure 5). Similarly, 63.5% (99/156) of ketorolac patients and 41.1% (65/157) of vehicle patients reported
Figure 5. (Solomon) Percentage of ketorolac and vehicle patients experiencing ocular symptoms (associated with the surgery) during the first 24 hours post PRK (P⬍.001 for all symptoms except itching compared with vehicle patients).
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
1657
KETOROLAC TROMETHAMINE AFTER PRK
Table 4. Treatment-related adverse events. Frequency (%) Ketorolac (n ⫽ 36)
Vehicle (n ⫽ 157)
Eye pain
2 (1.3)
4 (2.5)
Headache
1 (0.6)
3 (1.9)
Corneal infiltrates
1 (0.6)
1 (0.6)
Eye edema
1 (0.6)
0 (0.0)
Conjunctival hyperemia
1 (0.6)
0 (0.0)
Adverse Event
Ocular irritation
0 (0.0)
1 (0.6)
Nausea
0 (0.0)
1 (0.6)
Vomiting
0 (0.0)
1 (0.6)
mild or better foreign-body sensation during the first 24 hours. More ketorolac patients (50.6%, 79/156) than vehicle patients (26.1%, 41/157) had mild or better photophobia during the first 24 hours, and more ketarolac patients (50.6%, 79/156) than vehicle patients (28.7%, 45/157) had mild or better tearing. There were no significant between-group differences in ocular symptoms at other time points. Safety One or more treatment-related adverse events were reported in 2.6% (4/156) of patients in the ketorolac group and 6.4% (10/157) of patients in the vehicle group. There were no statistically significant differences between the treatment groups for any reported treatment-related adverse event (Table 4). Ocular irritation, which includes burning and stinging, was reported by 1 vehicle patient. There were no serious treatmentrelated adverse events or changes in visual acuity. Corneal infiltrates were reported in 3 patients (regardless of relationship to treatment): 2 ketorolac patients and 1 vehicle patient. In the ketorolac patients, 1 incident was considered unrelated to treatment and 1, treatment related. In the vehicle patient, the corneal infiltrate was considered treatment related. All 3 patients recovered without sequelae. Overall, 14 patients (6 ketorolac and 8 vehicle) discontinued because of adverse events. In most patients (9), the reason for discontinuation was inadequate pain relief from the escape medication. Other adverse events leading to discontinuation included eye pain not related to escape medication use (2 patients), corneal infiltrates (2 patients), and rhinitis (1 patient). 1658
Given the nature of the PRK procedure, most patients in both treatment groups experienced an expected short-term worsening of the best corrected visual acuity. There was no statistical difference between the treatment groups in changes in visual acuity from baseline. Changes from baseline in the severity of any biomicroscopic findings were reported for a similar proportion of patients in both groups. The changes were consistent with the expected postoperative findings of PRK. A greater number of ketorolac patients (33/156) than vehicle patients (16/157) entered the posttreatment period because of a delay in complete corneal reepithelialization. The median time to reepithelialization was after day 2 and by day 3 post PRK in both treatment groups. The mean time to corneal reepithelialization was 3.3 days in the ketorolac group and 2.7 days in the vehicle group. The half-day difference was not considered clinically significant.
Discussion The new ketorolac ophthalmic solution, with a 20% lower concentration of active ingredient (ketorolac) and a lower concentration of certain inactive ingredients, maintained efficacy and provided excellent comfort after PRK. Ketorolac 0.4% (Acular LS) received approval from the United States Food and Drug Administration on May 30, 2003. Similar to ketorolac tromethamine 0.5% ophthalmic solution, ketorolac tromethamine 0.4% is efficacious against severe ocular pain caused by PRK. Ketorolac 0.4% was significantly better than vehicle in relieving pain, reducing pain intensity, reducing escape medication use, and limiting ocular discomfort. The pain induced by PRK is greater than that induced by radial keratotomy8 and may equal or surpass that induced by corneal abrasions and cataract surgery. Therefore, ketorolac 0.4% might be beneficial in alleviating ocular pain due to a variety of causes. Furthermore, the analgesic efficacy was evident even though the drug was administered only postoperatively. Preoperative or intraoperative use would probably have enhanced the analgesic effect by enabling the ketorolac to inhibit mediators of ocular pain before their synthesis.7 The cornea is 1 of the most densely innervated and therefore most pain-sensitive tissues of the eye.8 Since PRK involves removal of the central corneal epithelium and physical reshaping of the cornea, substantial postop-
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
KETOROLAC TROMETHAMINE AFTER PRK
erative pain is anticipated.9 While PRK and other surface procedures are undergoing a resurgence with the advent of wavefront-guided procedures, the downside of surface procedures (including PRK) is corneal pain, photophobia, burning/stinging, and foreign-body sensation. Although NSAIDs are generally effective in reducing discomfort associated with corneal surface procedures, they are often irritating and uncomfortable to instill, leading to potential noncompliance. The new formulation of ketorolac was developed with the specific goal of maintaining efficacy while minimizing the side effects associated with NSAIDs. While the patients were not asked specifically whether the drop caused burning and stinging, they were asked generally whether they experienced burning and stinging. Overall, ketorolac patients experienced significantly less procedure-related burning and stinging; ie, 75% of patients using the new formulation were comfortable. This study supports the efficacy and safety of the ketorolac reformulation for postPRK discomfort. Patients often complain about pain post PRK. Maximum pain intensity typically occurs during the first 12 hours.8 Figure 1 shows that the maximum pain intensity was statistically less in ketorolac patients than in vehicle patients (P⬍.001). Of the patients receiving ketorolac, 40% had none to mild pain compared with 4% of the patients receiving vehicle. It is particularly noteworthy that ketorolac 0.4% had greater analgesic efficacy than vehicle during the period of maximum pain intensity, despite the use of escape medication. Ketorolac patients also experienced significantly less intense pain during the first 48 hours post PRK (Figure 2). During the first 6 hours, 7 times more ketorolac patients than vehicle patients experienced no pain. It was not until 72 hours post PRK that similar numbers of vehicle and ketorolac patients experienced no pain. Ketorolac patients experienced no pain significantly more quickly than vehicle patients (P⬍.001, Figure 2). Not only did ketorolac patients have significantly less intense pain, they also had greater pain relief (P⬍.001, Figure 3). Figure 3 shows that during the first 24 hours post PRK, 4- to 15-fold more ketorolac patients than vehicle patients experienced complete to high relief in discomfort. Ketorolac patients experienced this relief despite requiring significantly less escape medication than vehicle patients (Pⱕ.005, Figure 4). During the first 24 hours, two-thirds of ketorolac patients
did not require escape medication. Vehicle patients reported significantly more symptoms of procedurerelated burning/stinging, foreign-body sensation, photophobia, and tearing during the first 24 hours than ketorolac patients (P⬍.001, Figure 5). The product labeling for ketorolac tromethamine 0.5% ophthalmic solution (Acular) lists transient stinging and burning on instillation as the most frequent adverse event, with up to 40% of patients reporting it. In contrast, no patient in the ketorolac 0.4% group in this study reported treatment-related incidents of stinging and burning as an adverse event. Therefore, ketorolac 0.4% appears to be more comfortable. This increase in comfort may lead to improved patient compliance. Topical NSAIDs have been associated with adverse events such as keratitis, corneal subepithelial infiltrates, and corneal melts.3,4,10 In this study, only 3 of 313 patients (2 ketorolac and 1 vehicle) developed corneal infiltrates. In only 1 of the 2 ketorolac cases was the corneal infiltrate considered treatment related (Table 2). All 3 patients recovered without sequelae, with follow-up uncorrected visual acuity of 20/20 in the study eye. Although time to reepithelialization was slightly longer (ie, less than a day) in ketorolac patients, the difference between groups was not clinically or statistically significant. There were no cases of corneal melts, keratitis, infection, ulceration, or perforation. In the vehicle group, time to no pain (Figure 2) was associated with complete reepithelialization. In ketorolac patients, time to no pain was not associated with reepithelialization, emphasizing the strong analgesic efficacy of the ketorolac reformulation. Despite the longer time to reepithelialization, ketorolac patients were more comfortable. They had less pain, less severe pain, no treatment-related burning and stinging, and less use of escape medications then vehicle patients, despite the slightly faster reepithelialization in the latter group. Overall, the results demonstrate that ketorolac patients experienced less pain more rapidly and greater pain relief than vehicle patients. The new formulation of ketorolac, which contains reduced concentrations of active and certain inactive ingredients, decreases the potential for irritation and toxicity. Further research on ketorolac tromethamine 0.4% ophthalmic solution will be worthwhile.
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004
1659
KETOROLAC TROMETHAMINE AFTER PRK
Ketorolac Reformulation Study Groups Bruce Altman, Nicole Anderson, Amin Ashrafzadeh, Frank Bishop, Lawrence C. Chao, Roy S. Chuck, Elizabeth A. Davis, Eric Donnenfeld, Ralph Falkenstein, David Hardten, Barry Katzman, Matthew Kaufman, Jane Ahn Lee, Edgar Levine, Ning Lin, Richard L. Lindstrom, Anthony Lombardo, David Markoff, Stephen Mathias, Peter J. McDonnell, Joseph M. Miller, Shrabanee Mitra, Sylvia Norton, Matthew Paul, Jonathan R. Pirnazar, Francis Price, Michael Raizman, Turpin Rose, Paul Ruggier, Thomas Samuelson, Steve Schallhorn, Timothy Schneider, Robert W. Snyder, Kerry D. Solomon, David Tanzer, James Tidwell, Elizabeth Tonon, David T. Vroman, James Weisz, John Wittpenn, Richard Yee, William Zeh, Stephen Zuckerman
5.
References
9.
1. Jaanus SD, Cheetham JK, Lesher GA. Antiinflammatory drugs. In: Bartlett JD, Jaanus SD, eds, Clinical Ocular Pharmacology, 4th ed. Boston, MA, Butterworth-Heinemann, 2001; 265–298 2. Solomon KD, Vroman DT, Barker D, Gehlken J. Comparison of ketorolac tromethamine 0.5% and rimexolone 1% to control inflammation after cataract extraction; prospective randomized double-masked study. J Cataract Refract Surg 2001; 27:1232–1237 3. Teal P, Breslin C, Arshinoff S, Edmison D. Corneal subepithelial infiltrates following excimer laser photorefractive keratectomy. J Cataract Refract Surg 1995; 21:516–518 4. Guidera AC, Luchs JI, Udell IJ. Keratitis, ulceration, and perforation associated with topical nonsteroidal anti-
10.
1660
6.
7.
8.
inflammatory drugs. Ophthalmology 2001; 108:936– 944 Heier J, Cheetham JK, Degryse R, et al. Ketorolac tromethamine 0.5% ophthalmic solution in the treatment of moderate to severe ocular inflammation after cataract surgery: a randomized, vehicle-controlled clinical trial. Am J Ophthalmol 1999; 127:253–259 Holzer MP, Solomon KD, Sandoval HP, Vroman DT. Comparison of ketorolac tromethamine 0.5% and loteprednol etabonate 0.5% for inflammation after phacoemulsification; prospective randomized double-masked study. J Cataract Refract Surg 2002; 28:93–99 Rajpal RK, Cooperman BB. Analgesic efficacy and safety of ketorolac after photorefractive keratectomy; the Ketorolac Study Group. J Refract Surg 1999; 15:661–667 Mu¨ller LJ, Marfurt CF, Kruse F, Tervo TMT. Corneal nerves: structure, contents and function. Exp Eye Res 2003; 76:521–542; erratum 2003; 77:253 Binder PS. Excimer laser photoablation: miracle or menace? J Ophthalmic Nurs Technol 1994; 13:61–63, 76 Congdon NG, Schein OD, von Kulajta P, et al. Corneal complications associated with topical ophthalmic use of nonsteroidal antiinflammatory drugs. J Cataract Refract Surg 2001; 27:622–631
From Storm Eye Institute, Medical University of South Carolina, Department of Ophthalmology (Solomon, Sandoval), Charleston, South Carolina, Ophthalmic Consultants of Long Island (Donnenfeld), Long Island, New York, New England Eye Center, Tufts University School of Medicine (Raizman), Boston, Massachusetts, and Allergan Inc. (Stern, VanDenburgh, Cheetham, Schiffman), Irvine, California, USA. Supported by Allergan Inc., Irvine, California, USA. Drs.VanDenburgh, Stern, Cheetham, and Schiffman are employees of Allergan, Inc. None of the other authors has a proprietary interest in the drug used in the study.
J CATARACT REFRACT SURG—VOL 30, AUGUST 2004