To compare the effects of timolol on refractive outcomes in eyes with myopic regression after laser in situ keratomileusis (LASIK) with a control-matched group.
Prospective, randomized, parallel-controlled, double-masked clinical trial. A computer-generated randomization list based on random block permutation (length 4 to 8) was used for treatment allocation.
setting: Basir Eye Center, Tehran, Iran. Patient population: Of 124 eyes with myopic regression after LASIK using Technolas 217-Z, 45 eyes in each group were analyzed. Intervention: Patients were randomly assigned into either Group 1, who received timolol 0.5% eye drops, or Group 2, who received artificial tears for 6 months. Main outcome measure: Spherical equivalent (SE) at 6 months posttreatment.
In Group 1, SE improved from −1.48 ± 0.99 diopter (D) before treatment to −0.88 ± 0.91 D and −0.86 ± 0.93 D 6 months after treatment and 6 months after timolol discontinuation, respectively ( P < .001). In Group 2, it was −1.57 ± 0.67 D, −1.83 ± 0.76 D, and −1.91 ± 0.70 D, respectively ( P < .001). SE was significantly better in Group 1 6 months after treatment and 6 months after discontinuation of treatment ( P < .001 for both comparisons). There was a 0.26 D decrease in SE improvement every 4 months after the surgery in the Group 1 ( P < .001).
Timolol application is effective for the treatment of myopic regression after LASIK compared with control group. Its effects last for at least 6 months after its discontinuation.
Despite the improvements in the nomograms, ablation parameters, and advances in excimer laser technology, up to 28% of refractive surgery patients still continue to experience myopic regression. In previous studies, “regression” was defined as a 0.25 diopter (D) or greater myopic shift occurring between follow-up visits. The main possible explanations for regression are focused on the increases in corneal thickness and the postoperative forward shift of the cornea.
In the forward shift theory, thinner cornea, higher intraocular pressure, and higher myopia requiring greater laser ablation have been reported to increase the risk of myopic regression. It has been suggested that intraocular pressure (IOP)-lowering agents can decrease or even improve myopic regression after laser in situ keratomileusis (LASIK). However, in previous studies, the findings were not compared with a control-matched group. Moreover, it is still not clear what happens to the refractive error when the drops are stopped and whether the refractive error returns to the baseline after the discontinuation of medication. In this prospective, randomized, double-masked, parallel-group, placebo-controlled clinical trial, we compare the effects of timolol vs placebo for the treatment of myopic regression after LASIK; and afterwards, we evaluate what may come about 6 months after the discontinuation of treatment.
Materials and Methods
Study Design and Setting
This prospective, randomized, placebo-controlled, parallel-group, double-masked clinical trial was performed at Basir Eye Center between March 2009 and January 2011.
In this study, patients with myopic regression who were at least 20 years old with the cylinder ≤−1.00 D, preoperative corneal thickness of at least 500 μm, and a postoperative residual stromal bed thickness of more than 250 μm were included. Myopia regression was defined as a 0.25 D or greater myopic shift between the follow-up visits after month 1 postoperatively. Undercorrection is defined as failure to achieve within 1.00 D or greater of the intended correction by 1 week postoperatively; patients meeting this criterion were excluded from the study. Patients with a history of refractive surgery retreatment, previous ocular surgery other than previous LASIK, keratoconus or any ectatic corneal disorder, keratoconus suspect by topography, preoperative corneal opacity, any corneal dystrophies, presence of pterygium, retinal disorders, collagen vascular disorders, diabetes mellitus, glaucoma, cataract, pregnancy, breastfeeding, and systemic corticosteroid therapy were excluded. In order to obviate inter-eye correlation, only 1 eye of a patient was included, in the case that both eyes of a patient had myopic regression.
All patients underwent LASIK using Technolas 217-Z (Bausch and Lomb, Rochester, New York, USA) using the standard method. Lamellar keratotomy had been performed using the M2 microkeratome (Moria, Antony, France) to create an intended 160-μm flap thickness with a superior hinge. The optical zone was 5.5 to 6.0 mm based on the corneal thickness and curvature. In all eyes, attempted correction was aimed at emmetropia. All surgical procedures were performed by 1 of the authors (A.S.). Patients were examined on postoperative days 1, 3, 7, 14, and 28, and then on a monthly basis for a year, and once every 3 months thereafter until 2 years.
Patients were randomly assigned into 2 groups. Group 1 included the patients with myopic regression who received timolol 0.5% eye drops twice a day for 6 months; Group 2 included matched controls who received artificial tears in the same manner as did Group 1.
The main outcome measure was the mean spherical equivalent (SE) 6 months after treatment. Secondary outcome measures were the change in visual acuity and the central corneal thickness (CCT) 6 months after treatment.
Included patients had a complete eye examination including refraction, uncorrected distance visual acuity (UDVA), corrected distance visual acuity (CDVA), slit-lamp biomicroscopy, IOP measurement, corneal pachymetry (UP-800 ultrasound-4, NIDEK Inc, Gamagori, Japan), and dilated funduscopy in different follow-up visits. It has been shown that LASIK for myopia produces underestimation of IOP measured by Goldmann applanation tonometry. PASCAL dynamic contour tonometer (Ziemer Ophthalmic Systems AG, Port, Switzerland) was used for IOP measurement to obviate the effect of corneal morphologic changes after refractive surgery. Follow-up visits were scheduled for 3 and 6 months after the treatment. Patients were called for another follow-up visit 6 months after discontinuation of the treatment. Follow-up examinations consisted of slit-lamp microscopy, refraction, UDVA, CDVA, IOP measurement, corneal pachymetry, and dilated funduscopy. All measurements were carried out by 1 optometrist, who was masked to the randomization list, with the same devices throughout the study.
Randomization was performed using the random block permutation method according to a computer-generated randomization list. The block length varied randomly (4 to 6). Random allocation sequence was performed by a biostatistician. The investigators in the study were not informed of the details of the series.
Labels of both medication bottles were designed similarly with the same color and shape for masking. A number and a letter indicating in which eye the drop should be administered were stated on the labels. All bottles were prepared and dispensed by a drug store near the clinic based on an instruction given to them by 1 of the authors (Y.V.), who had access to the randomization list and was also responsible for data entry. All eligible patients were referred to the assigned drug store with a letter in hand showing that they were included in this study and in which eye the drop should be administered. The pharmacist gave them appropriate bottles according to the instruction and randomization list.
To have a 90% power for detection of 0.5 D difference in the mean SE between the groups as significant (at the 2-sided 5% level) with an assumed standard deviation of 0.48, considering 20% loss to follow-up, 51 eyes for each group were required.
Data were analyzed using SPSS version 17 statistical software (SPSS Inc, Chicago, Illinois, USA). In order to test the normality of the data, we used the Kolmogorov-Smirnov test. To compare the values within groups in different times, we used repeated-measures analysis of variance (ANOVA) test. We used Bonferroni method to adjust multiple comparisons. To compare the results between 2 groups, we used independent-samples t test or Mann-Whitney U test based on normality test results. To obviate the effect of preoperative refraction, attempted correction, degree of myopic regression after LASIK, and the time interval between LASIK and myopic regression as well as their 2 × 2 interactions (effect modification) on the SE and UDVA in each follow-up, we used analysis of covariance. We used repeated-measures ANOVA to evaluate any interaction between 2 groups in terms of trend of changes. To evaluate the influence of preoperative degree of myopia, all eyes were stratified into 3 groups: “low myopia,” with preoperative SE below −6.0 D; “moderate myopia,” with preoperative SE from −6.0 D to less than −10 D; and “high myopia,” with preoperative SE greater than or equal to −10 D. P value less than .05 was considered significant.