To manage pain and discomfort following photorefractive keratectomy (PRK), noninflammatory corticosteroids (NSAIDs) such as 0.5% ketorolac tromethamine (Acular) and 0.1% diclofenac sodium (Voltaren) have been widely used.1"4 Clinically, both eye drops have reduced pain or discomfort in some patients following PRK, but the mechanism by which these drugs induced analgesia remains unanswered. Previous studies have shown conflicting evidence regarding reduction of corneal sensation following administration of topical 0.1% diclofenac sodium compared to a control vehicle in normal human subjects.4 This evidence seems to support the mechanism by which the corneal pain or discomfort was reduced. More recently, several studies addressed this issue, but since the methodology and instrumentation were different, the results have been controversial.6"9
PATIENTS AND METHODS
Thirty normal volunteers participated in this study. They had no history of eye disease, corneal pathology, or previous ocular surgery. Those who wore contact lenses, used steroids or non-steroid agents, or other medication, were excluded as well as those with supersensitive eyes.
The corneal sensation of each patient was measured with a Cochet-Bonnet esthesiometer (Luneau, Chartres, France). This instrument is used routinely in the clinic, and consists of a 0.12 mm-diameter nylon monofilament of variable length (0 to 60 mm), and applies pressure on the cornea between 11 and 200 mg/0.0113 mmp 2. Patients were measured at a fixed distance under room light conditions. The measurement was started at the maximum length of the filament (60 mm) which applies the lowest pressure on the cornea. The filament was advanced perpendicular to the eye and the cornea gently touched. The central cornea as well as the four points on the peripheral cornea (12, 3, 6, and 9 o'clock) were measured and the results averaged for statistical analysis.
Figure 1: Comparison of the average change in corneal sensation in 15 normal volunteers with one eye treated with two drops of 0.5% ketorolac tromethamine, and the fellow eye treated with two drops of 0.1% diclofenac sodium. (Bars indicate standard error.)
Figure 2: Comparison of the average change in corneal sensation in another 15 normal volunteers. One drop of either 0.5% ketorolac tromethamine or 0.1% diclofenac sodium was administered, and then repeated every 5 minutes for 20 minutes. (Bars indicate standard error.)
A positive response was recorded when the patient indicated verbally that they felt the filament touch their cornea. If the patient did not feel the filament, the investigator pressed the instrument until the filament bent slightly. If the patient still had no sensation, the filament was shortened by 5 mm and the test was repeated until the patient reported sensation. A blinking reflex was not considered to be a positive response.
Fifteen volunteers (average age 37.8 years; 8 males and 7 females) randomly received two drops of 0.5% ketorolac tromethamine on one eye and two drops of 0.1% diclofenac sodium on the other eye, from masked, opaque bottles. Corneal sensation was measured with the Cochet-Bonnet esthesiometer by a masked investigator, before applying the drops, and at 15, 30, and 60-minute intervals after using the eye drops.
Another 15 volunteers (average age 36.4 years; 8 males and 7 females) were randomly administered only one drop in each eye of either 0.5% ketorolac tromethamine or 0.1% diclofenac sodium, which was then repeated every 5 minutes for 20 minutes. Corneal sensation was measured before application, and 0, 15, 30, and 60 minutes after applying the drops.
Changes in corneal sensation after applying the eye drops at each time interval were compared statistically with baseline (zero change) in each group, using one-tailed t-test (SYSTAT). A ? value less than 0.05 was considered statistically significant.
The baseline corneal sensation (before applying the eye drops) in all 30 volunteers was between 55 to 60 mm. In the two-drop 0.1% diclofenac sodium group, the mean reduction of corneal sensation was 0.53, 0.73, and 0.73 mm at 15, 30, and 60 minutes, respectively. The mean reduction for the 0.5% ketorolac tromethamine multiple-drop group was 1.2, 0.73, and 0.4 mm at the 15, 30, and 60-minute intervals, respectively (Fig 1). No statistically significant reduction of corneal sensation was found in either group.
In the multiple-drop 0.1% diclofenac sodium group, the mean reduction of corneal sensation was 3.60, 4.00, 3.27, and 0.67 mm at intervals of 0, 15, 30, and 60 minutes, respectively; and 5.53, 4.00, 3.53, and 3.60 mm at 0, 15, 30, and 60 minutes, respectively in the 0.5% ketorolac tromethamine two-drop group. When compared to baseline measurements, both eye drops significantly reduced corneal sensation at 0, 15, and 30 minutes after multiple drops (p<0.05), and the analgesia effect lasted about 1 hour. Hypesthesia induced by 0.5% ketorolac tromethamine still existed 60 minutes after the last application (Fig 2).
Both 0.5% ketorolac tromethamine and 0.1% diclofenac sodium are noncorticosteroid, antiinflammatory drugs that block the cyclooxygenase pathway of arachidonic acid metabolism, and reduce the release of prostaglandin E2 which may play a role in reducing ocular inflammation and pain.4,10,11 In addition to an anti -inflammatory effect, previous studies have suggested that both drugs may have an anesthetic effect. Szerenyi et al first reported that 0.1% diclofenac sodium significantly reduced corneal sensation in normal humans compared to a control vehicle when the eye drops were given every 5 minutes for 20 minutes, and the drug-induced hypesthesia was reversed to normal within an hour in all volunteers.5 These findings were supported by subsequent studies that showed both 0.5% ketorolac tromethamine and 0.1% diclofenac sodium could significantly reduce corneal sensation in normal eyes when one drop or multiple drops were introduced.6,9 This has also been contradicted by studies that showed no reduction of corneal sensation using one or two drops of either drug in normal or non-normal eyes.7,8 This disagreement in results may be due to different methodology. In three studies6"8, patients were given one or two eye drops of either drug, while another two studies treated patients with multiple eye drops.5·9 We believe that the significant difference in dose and frequency could contribute to the conflicting results, as well as different instruments having been used for measuring corneal sensitivity (micro-anesthesiometer versus esthesiometer).5"9
For clarity, we divided our volunteers into two groups. In the two-drop group, volunteers were administered two drops of either 0.5% ketorolac tromethamine or 0.1% diclofenac sodium in each eye, and in the multiple-drop group, multiple eye drops were introduced. We found that the reduction of corneal sensation was not statistically significant when only two drops of either drug was administered. However, there was a statistically significant reduction in sensation when multiple drops were introduced repeatedly.
Our study confirms that both 0.5% ketorolac tromethamine and 0.1% diclofenac sodium can reduce corneal sensation in normal human eyes when multiple drops are administered, and that only one or two drops has little or no effect.
Szerenyi et al reported that corneal hypesthesia (induced by multiple 0.1% diclofenac sodium eye drops) returned to baseline level about 1 hour after the last application, which was confirmed in our study (multiple-drop group).5 However, we found that 0.5% ketorolac tromethamine-induced hypesthesia still existed at 60 minutes. Since we did not conduct sensitivity measurements after this time period, we cannot conclude the exact duration of hypesthesia, but believe the effect gradually disappears after 1 hour.
To date, no study has discovered the exact mechanism by which corneal analgesia is induced by these two drugs. It is well known that NSAID eye drops can block the cyclooxygenase pathway, and reduce prostaglandin E2 release, which reduces inflammatory reaction and pain, but the mechanism which affects corneal sensation by either of these drugs may also be like topical anesthetics, which block corneal nerves. Corneal sensation is mediated by sensory nerves that originate from the ciliary nerves and are concentrated in the anterior stroma beneath Bowman's membrane, eventually penetrating Bowman's membrane and branching forward into the epithelium.12 These sensory nerves have been reported to have different fibers which respond separately to heat, cold, light touch, and pain.13"16 Clinically, the reduction of corneal sensation has been associated with herpes simplex keratitis, corneal ulceration, contact lens, topical adrenergic blocking medication, cataract surgery, aging eye, and diabetes mellitus.
NSAIDs have only recently been widely used for managing pain and discomfort following PRK. Published studies have shown that either 0.5% ketorolac tromethamine or 0.1%' diclofenac sodium applied four times a day is enough to manage pain and discomfort in PRK volunteers. However, to achieve effective analgesia in normal eyes, multiple eye drops must be given.3·5"9 These findings seem to support the theory that light touch and pain sensations may be carried by different fibers. In addition, an animal study has shown that there was no difference in corneal sensation with or without the administration of 0.1% diclofenac sodium following PRK in rabbits.17 These results indicated that a selective effect may exist on pain fibers rather than light touch fibers of the cornea.
Another possibility is that light touch and pain sensations may be caused by the same fibers. As we know, prostaglandin synthesis and release following a corneal wound plays a major role in causing pain. Nonsteroidal agents could reduce synthesis and release of prostaglandins, and consequently reduce corneal pain. In addition, corneal epithelial defect following PRK allows easier penetration of the drugs into the corneal tissue. Therefore, fewer eye drops may reduce pain in a wounded cornea compared to a normal cornea. The exact mechanism by which corneal sensation was reduced with either of the drugs remains unanswered.
We did not observe any clinical sign of corneal toxicity when multiple NSAID drops were used. However, we do not know whether corneal toxicity could result when multiple doses of the drugs are delivered through soft contact lenses in abnormal corneas, such as those with a corneal epithelial defect.
1. Salz JJ, Reader AL III, Schwartz IJ, Van Le K. Treatment of corneal abrasions with soft contact lenses and topical diclofenac. J Refract Corneal Surg 1994;10:640-646.
2. Arshinoff S, D'Addario D, Sadler C, Bilotta R, Johnson TM. Use of topical nonsteroidal anti-inflammatory drugs in excimer laser photorefractive keratectomy, J Cataract Refract Surg 1994; 20:(suppl):216-222.
3. Sher NA, Frantz JM, Talley A, Parker P, Lane SS, Ostrov C, Carpel E, Doughman D, DeMarchi J, Lindstrom R. Topical diclofenac in the treatment of ocular pain after excimer photorefractive keratectomy. Refract Corneal Surg 1993;9:425-436.
4. Phillips AF, Szerenyi K, Campos M, Krueger RR, McDonnell PJ. Arachidomic acid metabolites after excimer laser corneal surgery. Arch Ophthalmol 1993;111:1273-1278.
5. Szerenyi K, Sorken K, Garbus JJ, Lee M, McDonnell PJ. Decrease in normal human corneal sensitivity with topical diclofenac sodium. Am J Ophthalmol 1994;118:312-315.
6. Zaidman GW, Yoon S, Amesur K. The effect of nonsteroidal anti-inflammatory agents (NSAIDs) on corneal sensation. Invest Ophthalmol Vis Sci 1995;36(suppl):S576.
7. Tauber J, Garner L, Niknia GH, Crosser V. The effects of instillation of topical NSAIDs on corneal sensation. Invest Ophthalmol Vis Sci 1995;36(suppl):S576.
8. Narvaez J, Kroll P, Guzek JP. Effects of diclofenac and ketorolac on corneal sensitivity. Invest Ophthalmol Vis Sci 1995;36(suppl):S576.
9. Perry HD, Donnenfeld ED, Kanellopoulos AJ, Aslanides I. Decrease in normal human corneal sensitivity with topical ketorolac tromethamine 0.5%. Invest Ophthalmol Vis Sci 1995;36(suppl):S702.
10. Ho SM, Brennan NA. Corneal sensitivity and contact lens wear. Invest Ophthalmol Vis Sci 1993;34:1007.
11. Kantor, TG. Current modalities in arthritic diseases. Am J Med 1987; 83:2-5.
12. Newell FW. Anatomy and embryology. In: Newell FW, Ophthalmology: Principles and Concepts, 6th Edition. St. Louis, MO; The CV Mosby Company: 1986:6-13.
13. Schimmelpfennig B. Nerve structures in human central corneal epithelium. Graefes Arch CUn Exp Ophthalmol 1982;218:14-20.
14. Millodot M. A review of research on the sensitivity of the cornea. Ophthalmic Physiol Opt 1984; 4:305-318.
15. Rozsa AJ, Beuerman RW. Density and organization of free nerve endings in the corneal epithelium of the rabbit. Pain 1982;14:1105-120.
16. Dohlman CH. Conjunctiva, cornea, and sclera. In: Dohlman CH, et al (eds). Prinicples and Practice of Ophthalmology. Clinical Practice, 1st Edition. Philadelphia, PA; WB. Saunders Company; 1994:10-40.
17. Loya N, Bassage S, Vyas S, Cerro MD, Park SB, Aquavella JV. Topical diclofenac following excimer laser: effect on corneal sensitivity and wound healing in rabbits. J Refract Corneal Surg 1994;10:423-427.