Under the current FDA regulations, once a drug or device is approved by the agency for a specific indication, physicians may choose to prescribe and use the drug or device as they see fit. Prescribing a pharmaceutical agent for indications other than those approved is considered an off-label use of the product.
A COMMON OCCURRENCE
The off-label use of drugs is a common practice in medicine, including ophthalmology. Due to the lengthy FDA approval process, pharmaceutical companies often investigate only a specific effect of a new drug. Manufacturers know that once the drug enters the marketplace, its usage can widen significantly. Many of the ophthalmic fluoroquinolones, for example, were initially approved only for the treatment of bacterial conjunctivitis. Just recently, after years of use for corneal ulcers, was that indication added. At this time, one of the most widespread uses for fluoroquinolones is for surgical prophylaxis before and after anterior segment and refractive surgery. As profitable as this portion of the prescription market is, pharmaceutical companies cannot directly advertise or encourage this off-label use. Consulting physicians and physician expert panels are often employed to disseminate the message instead.
As stated by the FDA, “If physicians use a product for an indication not in the approved labeling, they have the responsibility to be well informed about the product, to base its use on firm, scientific rationale and on sound medical evidence, and to maintain records of the product's use and effects.”1
With the rapid growth of US refractive surgery during the mid-1990s, new indications have emerged for older drugs. Initially, LASIK was considered an off-label use of the PRK-approved excimer laser. Today, as physicians try to optimize refractive results and improve surgical safety, a few pharmaceutical agents have become increasingly popular for uses never envisioned by their creators.
ACULAR AND ALPHAGAN
Two drugs, specifically ketorolac tromethamine (Acula; Allergan, Irvine, CA) and brimonidine tartrate (Alphagan, Allergan), are used off-label for various purposes in LASIK patients. Pharmaceutical companies have not directly promoted this practice, word of which has been disseminated through peer-to-peer exchanges at meetings, at conferences, and in various journals. Because the widespread off-label use of drugs typically precedes formal studies of their new indications, surgeons may be acting without certitude of the practice's efficacy or safety. Such a situation may have occurred with the use of brimonidine during LASIK. In this article, I examine what we currently know about the off-label use of ketorolac and brimonidine after LASIK surgery.
Promoting Refractive Regression
The FDA initially approved ketorolac, a topical NSAID, for the temporary relief of ocular itching. Its indications were later expanded to include the reduction of pain and photophobia following incisional refractive surgery. Today, it is used in ophthalmology for several different conditions and under many circumstances. In addition to its known analgesic and anti-inflammatory properties, ketorolac has been touted as a modulator useful in treating overcorrection following myopic LASIK. Through discussion in conferences and Internet users groups, many surgeons began to prescribe ketorolac after LASIK for their overcorrected, unhappy patients. Physicians hoped to medically promote refractive regression and avoid further enhancement surgery. Other surgeons remained skeptical, believing that regression occurred naturally in most patients regardless of ketorolac use.
Although no mechanism of action was ever explained, the off-label use of ketorolac still made sense to a large group of physicians. Many refractive surgeons were familiar with the ability of topical corticosteroids to modulate wound healing and the refractive result after PRK, so using ketorolac for this purpose after LASIK did not seem to be unfounded.
Unpublished Studies
To my knowledge, there have been two unpublished studies that attempted to evaluate ketorolac's effect on regression. I presented a paper assessing the efficacy of ketorolac in treating consecutive hyperopia following myopic LASIK.2 My colleagues and I examined 14 eyes of 14 patients who had hyperopia due to overcorrection in one of their eyes after bilateral myopic LASIK. The affected eye was treated with ketorolac q.i.d. beginning 1 week postoperatively and continuing for an average of 9 weeks. We compared the amount of refractive regression in the treated eye with that of the fellow eye. With a mean follow-up of 22 weeks, the ketorolac-treated eyes regressed from an initial mean spherical equivalent of +1.52 D to a final +0.39 D—a net regression of 1.10 D. The fellow control eyes, in the same time period, regressed from +0.10 D to -0.15 D, which was a net change of 0.25 D. The difference in regression was statistically significant between the two eyes (Figure 1).
Another analysis studied 51 patients. Investigators divided 72 overcorrected eyes into two groups.3 They treated one group with ketorolac for an average of 3 weeks and observed the other group. The investigators noted regression in some eyes, but there was no statistically significant difference between the two groups.
The efficacy of ketorolac is still in question, as both studies have inherent weaknesses. At the very least, the studies established the safety of this off-label use of ketorolac and did not report any adverse events. Unless a larger, placebo-controlled, randomized study is performed, surgeons will have to decide what they feel is best for their patients. With almost no potential downside, I continue to use ketorolac in select patients in hopes of avoiding more invasive enhancement surgeries.
BRIMONIDINEPost-LASIK Cosmesis
The FDA approved brimonidine, a selective alpha-2 adrenergic agonist, for the lowering of IOP in patients with open-angle glaucoma and ocular hypertension. Since its introduction, surgeons have used brimonidine for other off-label indications within the field of glaucoma. For example, studies have shown that this drug effectively blunts IOP spikes following anterior segment laser procedures.4 Interestingly, there are two very different indications for brimonidine within refractive surgery. The first indication is the use of brimonidine preoperatively to reduce bleeding complications (eg, subconjuctival hemorrhage, conjuctival hyperemia, and bleeding pannus) of LASIK. The second, via brimonidine's effect on pupil size, is to diminish the occurrence of glare and halos in symptomatic post-LASIK patients.
Brimonidine has the properties of a vasoconstrictor, and some surgeons have noted that it decreases the postoperative occurrence of subconjuctival hemorrhage in LASIK patients. They believed that enabling patients to leave the laser suite with noninjected eyes improved patient satisfaction and diminished anxiety. Through physician word-of-mouth, many surgeons began adding brimonidine to their intraoperative LASIK regimen. I began to use it also, after I heard this tip mentioned in the back hallways of a few ophthalmology meetings. The drug appeared to work well, and the patients were happy.
Reports of Flap Dislocation
Unexpectedly, various reports began surfacing about an increased rate of flap slippage in eyes that intraoperatively received brimonidine (Figure 2). At the 2000 ASCRS conference in Boston, two presentations were made on the topic. A retrospective chart review found that six of 45 brimonidine-treated eyes (13.3%) had early flap dislocations as compared with none of 129 control eyes.5 Likewise, a second group of investigators also retrospectively showed the same association with nine of 118 brimonidine-treated eyes (7.6%) having the complication, compared with seven of 1,777 (0.39%) eyes that did not receive brimonidine.6 The results of the first study were later published, again finding a statistically significant higher rate of abnormal flap adherence with topical perioperative brimonidine.7 The investigators postulated a few mechanisms of action for these findings. They believed that perhaps brimonidine induced lagophthalmos, thereby decreasing the lids' natural flap-smoothing function. Other possibilities were that brimonidine increased flap desiccation, consequently increasing the friction between the upper lid and the flap.
Following the previously mentioned studies, many surgeons remained skeptical and continued to use brimonidine without any adverse effects. Recently, a small, prospective double-blind study quantified the beneficial effect of brimonidine in significantly decreasing bleeding complications after LASIK.8 More importantly, it found no increase in the incidence of flap dislocations with the drug. The researcher of this study felt that the lack of adverse effects was due to copious postoperative lubrication. My theory is that maybe patients who received brimonidine had “whiter,” less-injected eyes postoperatively, were less aware of them, and were more likely to rub their eyes inadvertently within the first 24 hours. In the retrospective chart review previously described, patients were instructed to wear eye shields. In this study, patients had the shields placed directly on their eyes to wear until the following visit.
The possibility of vision-threatening complications overshadowed brimonidine's positive cosmetic effect. I no longer use the drug but find that dilating the pupils with phenylephrine, as required for pupil tracking on the Alcon LADARVision system (Alcon Laboratories, Fort Worth, TX), probably achieves the same beneficial effects on bleeding complications without any known added flap adherence problems.
Decreasing Post-LASIK Nighttime Glare
Brimonidine has also gained use in modulating the pupil size postoperatively in patients with subjective symptoms of glare and halos. The relationship between pupil size, optical zone size, and subjective visual complaints has become a controversial topic. Recently, juries have awarded large malpractice settlements (as reported in the July/August 2002 issue of Cataract & Refractive Surgery Today) for complaints arising from alleged discrepancies between the size of the pupil and the treatment zone.
Glare, halos, starbursts, and other qualitative visual disturbances following PRK or LASIK are due to corneal higher-order aberrations. These aberrations may originate from the flap, the excimer laser ablation itself, haze, or the abrupt transition between the treated and untreated cornea (such as may occur with an optical zone smaller than the pupil size). Patients with large mesopic pupils are more susceptible to these types of complications, and they should be carefully counseled, as well as screened, preoperatively. For patients who have already had LASIK and are symptomatic, pupil size modulation may improve the quality of their vision.
Changing the diameter of the entrance pupil is known to dramatically affect the measured higher-order aberrations of the eye. A 1998 study (before the common use of large treatment zones) illustrates this concept.9 The investigators measured the spherical-like aberrations before and after PRK on 112 eyes. All the wavefront measurements were performed at 3 mm and (with dilation) 7 mm. The study found that enlarging the pupil from 3 to 7 mm before PRK increased the spherical aberrations sevenfold. More striking, following PRK, pupillary dilation increased the spherical aberration of the eye 300-fold.
PILOCARPINE VS BRIMONIDINE
For years, surgeons have recommended dilute pilocarpine for their post-RK patients who complained of poor nighttime vision. The same treatment has been advocated for LASIK patients with disabling glare during nighttime driving. Unfortunately, pilocarpine often has intolerable, undesirable side effects, such as ciliary spasm, headache, and a change in refraction. Surgeons have tried many alternatives to decrease the pupil size; they have even recommended that while driving, patients leave on their car's interior dome light. Recently, brimonidine has been touted as the medicine of choice for modulating pupil size by preventing full nighttime dilation and decreasing visual complaints.
Although no study has directly assessed brimonidine's effect on glare and halos, one evaluated its effect on pupil size under different lighting conditions.10 The investigators measured pupil size in 16 eyes of 16 patients under three different luminance levels before and after brimonidine application. The study found that under room or outside lighting conditions, brimonidine had no significant effect on the pupil size. However, under nighttime (mesopic) luminance, brimonidine caused miosis in 100% of the pupils within 30 minutes. This effect persisted for 6 hours in 81.3% of the eyes.
This alpha-2 adrenergic agonist seems to selectively inhibit sympathetically mediated nighttime pupil dilation without significantly affecting the photopic pupil size. In the low-luminance group, before brimonidine application, the mean baseline pupil size was 5.8 mm (range, 5.0 to 7.0 mm); 30 minutes after brimonidine application, the average pupil size decreased to 4.2 mm (range, 3.0 to 5.0 mm). All pupils that began as greater than 6.5 mm achieved a reduction of (at least) 2 mm in pupil size after the application of brimonidine, yet no large pupils decreased to below 3.5 mm in size. These findings indicate that brimonidine may provide the type of pupil modulation that could benefit unhappy postrefractive patients by preventing nighttime dilation without causing systemic side effects, inciting ciliary spasm, or affecting the photopic pupil size.
Overall, the off-label use of pharmaceuticals allows physicians to expand the field of medicine and better serve the patient. By using their collective past experience and clinical acumen, physicians can practice the art of medicine when making prescription decisions. Nevertheless, nothing should replace hard science when prescribing medicines in new ways.
Tal Raviv, MD, is in private practice in New York City and is an attending corneal and refractive surgeon at the New York Eye and Ear Infirmary. He does not hold a financial interest in any of the products mentioned herein. Dr. Raviv may be reached at (212) 717-4609; Tal.Raviv@NYLaserEye.com.1. Guidance for Institutional Review Boards and Clinical Investigators 1998 Update. US Food and Drug Administration Web site. Available at: www.fda.gov. Accessed Oct 2002.
2. Raviv T, Epstein RJ. Ketorolac in the treatment of consecutive hyperopia after myopic LASIK.” Paper presented at: American Society of Cataract and Refractive Surgeons Annual Conference; Boston, MA; May 2000.
3. Hannush S. What's New in Refractive Surgery? Review of Ophthalmology. 2002;9.
4. Chevrier RL, Assalian A, Duperre J, Lesk MR. Apraclonidine 0.5% versus brimonidine 0.2% for the control of intraocular pressure elevation after anterior segment laser procedures. Ophthalmic Surg Laser. 1999;30:199-204.
5. Walter KA. Adverse effect of Alphagan on LASIK flap adherence. Presented at the Symposium on Cataract, IOL, and Refractive Surgery, American Society of Cataract and Refractive Surgeons, Session 3-H: LASIK Complications; Boston, MA; May 22, 2000.
6. Talamo JH, deLuise VP. Increased incidence of flap slippage after LASIK associated with the use of preoperative Alphagan. Presented at the Symposium on Cataract, IOL, and Refractive Surgery, American Society of Cataract and Refractive Surgeons, Session 3-H: LASIK Complications; Boston, MA; May 22, 2000.
7. Walter KA, Gilbert DD. The adverse effect of peri-operative brimonidine tartrate 0.2% on flap adherence and enhancement rates in laser in situ keratomileusis patients. Ophthalmology. 2001;108:1434-1438.
8. Norden RA. Effect of prophylactic brimonidine on bleeding complications and flap adherence after laser in situ keratomileusis. J Refract Surg. 2002;18:468-471.
9. Martinez CE, Applegate RA, Klyce SK, et al. Effect of pupillary dilation on corneal optical aberrations after photorefractive keratectomy. Arch Ophthalmol. 1998;116:1104-1105.
10. McDonald JE II, El-Moatassem Kotb AM, Decker BB. Effect of brimonidine tartrate ophthalmic solution 0.2% on pupil size in normal eyes under different luminance conditions. J Cataract Refract Surg. 2001;27:560-564.
