Intraocular pressure trend following myopic photorefractive keratectomy

To evaluate the intraocular pressure (IOP) trend and risk factors for IOP rise after myopic photorefractive keratectomy (PRK). One eye of each patient undergone PRK for myopia was randomly assigned to this study. All eyes underwent tonometry by CorVis Scheimpflug Technology (CST) tonometer (Oculus Optikgeräte GmbH, Wetzlar, Germany) 1 week, 2 weeks, 1 month, 2 months, 3 months and 4 months after surgery. The eyes with IOP rise more than 5 mmHg and the risk factors were evaluated by Kaplan–Meier graph and multiple Cox regression analysis. A total of 348 eyes of 348 patients were enrolled in this study. Forty-three eyes (12.35%) experienced a steroid-induced IOP rise of more than 5 mmHg. Eyes with IOP rise had higher baseline IOP (Median 19 mmHg (IQR 18–22) versus Median 15 mmHg (IQR 14–16); p < 0.001). Baseline central corneal thickness (CCT) was higher in eyes without IOP rise (Median 520 µm (IQR 509–541) versus Median 535 µm (IQR 518–547); p = 0.009). In multivariate Cox regression analysis, higher baseline IOP was a risk factor for IOP rise (Hazard Ratio (HR) 1.59 (95% CI 1.43–1.77); p < 0.001) while higher baseline CCT was protective (HR 0.97 (95% CI 0.95–0.98); p < 0.001). Eyes with higher baseline IOP and lower baseline CCT are at increased risk of IOP rise after PRK and should be monitored more frequently.


Introduction
Myopia is the most common form of refractive error [1] and its prevalence is increasing in our modern era [2]. It is estimated that myopia and high myopia will affect 5 and 1 billion people by 2050, respectively [3]. Parallel to the increasing prevalence of myopia, the demand for refractive surgery has grown. Photorefractive keratectomy (PRK) is one of these procedures that has gained popularity for the correction of refractive error in myopia, particularly in thin corneas, because of its improved safety [4]. It seems that PRK is a more common procedure for correction of myopia in comparison to the laser in situ keratomileusis and small incision lenticule extraction in Iran. One reason is the more prevalence of thin cornea and keratoconus in the middle east [5].
Topical steroids are routinely prescribed after myopic PRK for the reduction of corneal opacity and myopic regression [6]. Intraocular pressure (IOP) elevation following steroid use is well-documented and long-term use of these agents can result in ocular hypertension (OHTN) and glaucoma. On the other hand, myopia itself is a risk factor for open-angle glaucoma (OAG) and steroid-induced OHTN [7]. So, understanding the incidence and magnitude of IOP elevation following PRK and also its associated risk factors may help clinicians detect, prevent, monitor, and treat any clinically significant IOP rise.
Goldmann applanation tonometry (GAT) is the gold standard for the measurement of IOP and has the greatest accuracy in certain limits of central corneal thickness (CCT) and corneal curvature and biomechanics. These factors are all altered after myopic PRK [4,8]. Therefore, relying on GAT in these patients may lead to underestimation of IOP and OHTN [9][10][11]. Also, different correction formulas have been proposed in different studies, but none of them is sufficiently reliable due to several factors [12,13].
Given the inaccuracy of tonometry with Goldmann applanation tonometer in patients undergone PRK, we used CorVis Scheimpflug Technology (CST) tonometer (Oculus Optikgeräte GmbH, Wetzlar, Germany) for IOP measurement pre-and postoperatively. It has already been documented that biomechanical IOP (bIOP) provided by CST tonometer does not seem to be affected by corneal thickness and corneal biomechanics, and in the case of keratorefractive surgery, it is similar to pre-surgery measurements [14,15].
This study aimed to evaluate IOP changes following PRK. To the best of our knowledge, the IOP trend after myopic refractive surgery has not been reported previously.

Patients and methods
In this prospective study, patients that underwent PRK for correction of nearsightedness between July 2016 and April 2019 were recruited. All surgeries were done by the same attending ophthalmologists (G.F.). Ethics committee approval of Tehran University of Medical Sciences was granted, and tenets of the declaration of Helsinki have been adhered to.

Inclusion criteria
One eye of each patient was randomly assigned for this study, and inclusion criteria were as follows: age older than 18 years old, spherical refraction between -1.00 and -10.00 diopters, stable refraction for at least 12 months, and cessation of wearing soft and rigid contact lenses for at least 2 and 3 weeks, respectively. All patients that were included in this study underwent PRK in both eyes and were followed for at least 4 months.

Exclusion criteria
Exclusion criteria included: a history of dry eye or any autoimmune disease that involves ocular surface, diabetes, previous ocular surgery, herpetic keratitis, OHTN, and glaucoma.

Data collection
All patients underwent thorough and comprehensive ophthalmic examination including measurement of corrected distance visual acuity (CDVA), slit-lamp biomicroscopy, tonometry by CST, gonioscopy, dilated funduscopy, measurement of the CCT by pachymetry (Tomey Corporation, Nagoya, Japan), and corneal imaging by Pentacam (OCULUS Optikgerate GmbH, Wetzlar, Germany). Measurement of the IOP at baseline and different time points was performed using CST which can compensate for corneal thickness and biomechanics [14,15]. Demographic data like age and gender were collected as well.

Surgical technique
All PRK procedures were done in the same manner and by the same surgeon (G. F.). Briefly, after instilling a drop of generic tetracaine in lower conjunctival fornices, a lid speculum was inserted. Epithelial debridement was done after applying ethyl alcohol 20% on the cornea for 10-15 s. Next, a 500 Hz Wavelight Allegretto Laser excimer laser (Wavelight Laser Technologie AG) was used for laser delivery. After ablation, a piece of sponge soaked in mitomycin 0.02% was applied to the stroma accordingly. Then, vigorous rinsing was done by balanced salt solution. At the end of the procedure, a bandage contact lens (BCL) was placed on the cornea and a drop of levofloxacin 0.5% and diclofenac 0.1% was instilled on the ocular surface.

Postoperative treatment
Topical levofloxacin 0.5% and betamethasone 0.1% were prescribed for the patients during 1 and 2 weeks after the procedure, respectively. Both drops were ordered 4 times a day. BCL was removed after healing the epithelial defect, during the first week. After two weeks, betamethasone 0.1% was replaced by fluorometholone 0.1% and the latter continued for 3 months in a tapered fashion. Meanwhile, the patient was advised to use preservative-free artificial tears regularly and ''pro re nata''.5

Follow-up
Patients were examined 1 day, 1 week, 2 weeks, 1 month, 2 months, 3 months, and 4 months after surgery. Tonometry by CST was done in each visit except postoperative day 1. Postoperative CCT was measured at the 1-month visit.

Statistical analysis
By considering that one-third of the normal population would reveal a steroid response after administration of topical betamethasone and setting type 1 error and margin of error to 0.05, we needed a minimum of 340 eyes [16]. To present data, we used mean, standard deviation, median, and interquartile range. To evaluate the normal distribution of data, we used the Kolmogorov-Smirnov test. To assess the IOP changes, we used a linear mixed model. Multiple comparisons were considered in this model by the Sidak method. The velocity of the patients to have an IOP rise of more than 5 mmHg compared to the baseline IOP was evaluated by the Kaplan-Meier graph. Also, the comparison of the groups regarding the time to this IOP rise was evaluated by univariate and multiple Cox regression. In the last step to obtain the most important risk factors, we used the Backward Likelihood Ratio (LR) method in our multiple cox regression models. All the analyses were performed by SPSS software (IBM Corp. Released 2017. IBM SPSS Statistics for Windows, Version 25.0. Armonk, NY: IBM Corp.). A P value of less than 0.05 is considered statistically significant.

Discussion
The present study was designed for the evaluation of the IOP trend after myopic PRK. Given the CCT and corneal biomechanics changes after myopic PRK, we used CST for the measurement of bIOP. Our study showed that higher baseline IOP and lower baseline CCT are significant risk factors for the IOP rise after myopic PRK. All IOP rises occurred during the first month (within the first 4 weeks) after the procedure. In our study, there was no significant difference between the IOP at the first week and baseline IOP but mean IOP was higher in other follow-up visits. Previous studies considering this issue have mixed results. Lee et al. applied CST and a non-contact tonometer (NTC) for measurement of IOP in patients   [17]. These studies show that the measurement of IOP by a tonometer that compensates for CCT and corneal biomechanics is more reliable in patients that have undergone surface ablation. Previous studies have shown that continuous administration of topical steroids in patients who are steroid responders will result in an IOP rise after 3-6 weeks [18]. Steroid-induced IOP elevation is related to increased outflow resistance at the level of the trabecular meshwork and extracellular matrix [16]. Particularly, topical administration of betamethasone can result in an IOP elevation in one-third of the normal population [19]. In our study, 12.35% of eyes experienced steroid-induced IOP rise of more than 5 mmHg. Notably, we replaced betamethasone with fluorometholone after two weeks. In comparison to other topical steroids, fluorometholone causes smaller amounts of IOP elevation [20]. Also, in a report by Javadi et al., the time needed for IOP rise in patients undergone PRK was the same. In their study, betamethasone eye drop was prescribed for the patients after the procedure and the first steroid responders were diagnosed after 2-3 weeks [21]. In our study, 3 eyes showed the rise of IOP more than 5 mmHg, one week after the procedure. This shortterm IOP elevation may be related to the high potency of betamethasone. The more potent steroids would lead to IOP elevation in a shorter period [22].
Generally, gender is not a risk factor for steroid response and IOP elevation; however, in the study run by Busool et al., the male gender was recognized as a risk factor for steroid response after PRK [23]. In another study, Lau et al. evaluated IOP after intravitreal triamcinolone injection. Odds ratio for IOP elevation greater than 6 mmHg in male patients was 3.17 (p = 0.006). They mentioned that there is no reasonable explanation for gender being a risk factor for steroid response [24]. In our study, gender was not a risk factor for IOP rise.
Age as a risk factor for steroid response has been explored in several studies. In a study by Lam et al., topical dexamethasone was prescribed for children after strabismus surgery. They found that children younger than 6 years old may be great steroid responders [25]. Studies done in the adults have mentioned older age, age under fifty or thirty as risk factors for IOP elevation [26][27][28]. In our study, the HR for age was not significant. This may be due to the Association between myopia and glaucoma is well documented [29][30][31][32], and some studies have illustrated high myopia and longer axial length as predictors of steroid response. A study by Chang et al. showed that a longer axial length, particularly longer than 29.0 mm will increase the risk of IOP elevation after uneventful cataract surgery [33]. In a study by Busool et al. that was designed for evaluation of predictors of IOP response after PRK, there was a greater proportion of eyes with high myopia in steroid responders [23]. In our study, the degree of myopia was not a risk factor for IOP elevation and there was not any difference between the proportion of eyes with high myopia among eyes with and without IOP elevation.
The thinner cornea has been known as a significant risk factor for the development and progression of glaucoma [34,35]; however, the role of CCT in IOP elevation after steroid administration is not well established in the literature. Angelopoulos et al. compared the IOP of normal and keratoconic eyes after PRK. The latter group underwent combined partial PRK and Collagen cross-linking (CCL) and topical dexamethasone 0.1% was prescribed for both groups in the postoperative period. In their study, keratoconic eyes had lower CCT, and a greater proportion of keratoconic eyes were steroid responders [36]. On the other hand, another study showed that thicker cornea is associated with increased steroid response [23]. In the current study, the thicker cornea was shown to be protective against IOP elevation.
Baseline IOP was another predictor of IOP elevation in the current study. The higher the baseline IOP the higher risk for IOP elevation. In Rhee's study, which was designed for evaluation of IOP after intravitreal triamcinolone acetonide injection (IVTA), higher baseline IOP (greater than 16 mmHg) was the only risk factor for IOP elevation [37]. In another study by Smithen et al., nonglaucomatous eyes with baseline IOP greater than 15 mmHg were at higher risk for steroid response after IVTA injection [38]. However, since the medication used and the route of administration was different in these studies, the comparison between the current study and the abovementioned reports has its own drawbacks.
Our study had several limitations. We only evaluated eyes that underwent myopic surface ablation. The intraocular pressure trend after lamellar and hyperopic refractive surgeries may be different. Also, the sample size of our study was limited. Future studies with a larger sample size and with attention to the other forms of laser keratorefractive surgeries are recommended.
In conclusion, our study showed that IOP elevation may occur in the first week after surface ablation. Eyes with higher baseline IOP and lower CCT are at greater risk of steroid response and should be monitored more frequently.
Funding The authors received no financial support for the research, authorship, and/or publication of this article.
Availability of data and material The data that support the findings of this study are available from the corresponding author, [SMT], upon reasonable request.

Declarations
Conflict of interest Ghasem Fakhraie declares that he has no conflict of interest. Zakieh Vahedian declares that she has no conflict of interest. Reza Zarei declares that he has no conflict of interest. Yadollah Eslami declares that he has no conflict of interest. Seyed Mehdi Tabatabaei declares that he has no conflict of interest. Abdollah Hadi declares that he has no conflict of interest. Sepideh Ghods declares that she has no conflict of interest. Ali Fakhraie declares that he has no conflict of interest.
Ethics approval All procedures performed in this study which involve human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Consent to participate Informed consent was obtained from all individual participants included in the study.
Consent for publication There is no identifying information about participants available in the article, so this issue is not applicable. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.