Refractive results of photorefractive keratectomy comparing trans-PRK and PTK−PRK for correction of myopia and myopic astigmatism

Refractive surgery is one of the most commonly performed procedures in medicine. The field of refractive surgery has evolved rapidly in recent years. Several surgical treatment options are now available, including photorefractive keratectomy (PRK), laser in situ keratomileusis (LASIK), and small incision lenticule extraction (SMILE). Nevertheless, PRK remains a popular method for correcting refractive errors, especially in eyes with thin corneas, high refractive errors, or recurrent corneal erosions [1, 2].

PRK can be performed in different ways. Either as a single-step procedure, in which the corneal epithelium and stroma are ablated simultaneously with an excimer laser, called transepithelial PRK (tPRK), or as a two-stage procedure, in which it is possible to distinguish between the mechanical, alcohol-assisted, or laser-assisted method of epithelial ablation. The introduction of tPRK in the 1990s aimed to avoid complications associated with epithelial debridement [3] and showed at least non-inferior results compared to two-step PRK [4]. However, there are only a few devices currently available in the market for the single-step PRK (e.g., SCHWIND Amaris and Alcon Wave Light). Representative of this procedure is the Amaris750 excimer laser. Therefore, the two-step method is still used by surgeons who do not have access to these laser platforms.

Basically, the two-stage procedure, consisting of mechanical epithelial and subsequent stromal ablation, is used in PRK treatment with the MEL90 excimer laser platform (Carl Zeiss Meditec). Excimer lasers like this are also used for PTK in other indications (e.g., degenerative diseases of the cornea) [5, 6]. Therefore, the sequential combination of PTK followed by PRK represents an option for non-contact laser correction with the MEL90.

On the one hand, the tPRK method has been proven many times as a safe and effective way to treat myopia [7]. On the other hand, the PTK−PRK procedure seems to show stable results, as demonstrated previously by two groups [8, 9] with the EX500 laser (Alcon Laboratories). Also, older publications indicate that the PTK−PRK combination is an effective treatment [10, 11]. However, only one of the previous works compared the PTK−PRK method with the tPRK and none of them evaluated the two-stage PTK−PRK method using the MEL90 excimer laser. This retrospective data analysis aims to compare these two approaches (tPRK vs. PTK−PRK) in terms of postoperative outcomes. Here, the standard target parameters of visual acuity, efficacy, and safety indices are considered and analyzed [12]. Hereby, we want to make an important contribution to this novel modified PRK application with the MEL90 and offer users of this excimer laser more evidence for individual therapy decisions. It is interesting to ask whether these two methods differ significantly. To the best of our knowledge, there is very little work on PTK−PRK procedures using MEL90, especially in the field of refractive surgery. Thus, further research on this issue is necessary to provide proper evidence for this alternative method.

All patients were examined before surgery according to a standard protocol. Visual acuity in terms of subjective refraction, including uncorrected distance visual acuity (UDVA) and corrected distance visual acuity (CDVA), was collected within 2 weeks before and 6 months after surgery. We also obtained topographic data using Schwind Anterior Segment Analyzers (Peramis and Sirius, Schwind eye−tech solutions) for tPRK patients and the TMS−5 Scheimpflug tomograph (Tomey) and WASCA aberrometer (Carl Zeiss) for PTK−PRK patients all at an optical zone (OZ) of 6.5 mm. For treatment planning, we used the K values of the respective topography device as well as those of an auto-refractometer/keratometer (Nidek).

All operations were performed by the same surgeon using two different laser platforms with a repetition rate of 500-Hz for the MEL90 and 750-Hz for the Amaris 750. For all treatments, topical anesthesia (Oxybuprocaine, Conjuncain EDO® 0.4 mg/mL, Bausch & Lomb) was used. Patients were asked to focus their gaze on the fixation light to center the ablation zone. Group A patients underwent a single stage tPRK procedure in which the excimer laser simultaneously reshapes the corneal epithelium and stroma to correct the refractive error. Patients in group B first received PTK treatment for epithelial ablation, with a depth of 50 μm and diameter of 8mm. Following this, refractive ablation of the stroma was performed using the PRK mode, without any further modifications. Our own ablation nomogram was utilized for this step, and no wavefront-guided mode was employed. It is important to acknowledge that the SCHWIND Amaris750 incorporates an aspherical epithelium ablation, taking into consideration the impact of keratometry and the associated higher energy loss in the periphery. In contrast, the MEL−90 does not account for peripheral energy loss. Regardless of the method employed, all refractive ablations were consistently performed with an OZ of 6.5 mm and a transition zone of 1.5 mm. Mitomycin C (MMC, 0.02%) was applied for 15–30 s (sec), depending on ablation depth. For corrections up to −4 D, MMC was applied for 20 s, while durations of 45 s were employed for corrections exceeding −4 D.

At the end of the procedure, all eyes got a therapeutic contact lens for 5 days, as well as preservative-free eyedrops of ofloxacin (Floxal EDO®, Bausch & Lomb) and dexamethasone (Dexa EDO®, Bausch & Lomb). Postoperative care included the application of Nepafenac eyedrops (0.1%, Nevanac®, Novartis) for 5 days, as well as hyaluronan (Hylo−Comod®, Ursapharm) eyedrops and dexamethasone (Dexa EDO®, Bausch & Lomb) eyedrops for 6 weeks.

Statistical analysis was performed using R Core Team software (R Foundation for Statistical Computing 2021). Refractive results are represented as standard graphs for reporting outcomes in refractive surgery [12], which show the efficacy, safety, predictability, and accuracy of each treatment. The efficacy index (EI) describes the ratio between postoperative UDVA and preoperative CDVA, whereas the safety index (SI) describes the ratio between postoperative CDVA and preoperative CDVA. Predictability is evaluated as the proportion of eyes achieving a postoperative SE within ± 0.50 D of targeted visual acuity and was analyzed using the least squares method. The differences in percent (%) of eyes within 0.50 D between the groups were tested with Fisher's exact test.

The differences in pre- and postoperative parameters were tested using either the independent t−test or Mann–Whitney test, based on whether the assumptions of parametric test were satisfied. Normality was tested with Shapiro–Wilk test, homogeneity of variances with Levene test, and outliers, if any, were identified using the box plot method. The changes within the groups were tested with either Wilcoxon signed rank test or paired t test. And the differences in changes were tested with either Mann–Whitney test or independent t test.

Table 1 compares the preoperative descriptive data of the two groups and the total cohort. In addition to SE values, coma, trefoil, and spherical aberration (SA) and higher order aberrations (HOA) are also listed. There is no significant difference between the two groups regarding all parameters (p > 0.05). Table 2 shows the same parameters 6 months postoperatively for the two groups and the complete cohort. Again, there is no significant difference between the two groups (p > 0.05). Table 3 summarizes the differences in pre- and postoperative parameters within the two groups and between the groups, respectively. The postoperative change in sphere, cylinder, and SE was highly significant in the two groups (p < 0.001). However, the differences in changes between the two groups were not statistically significant in all other parameters.

Figure 1 presents four standard graphs reporting the refractive outcome of the two surgical methods in comparison. Regarding efficacy, which is postoperative UDVA in relation to preoperative CDVA, both surgical methods showed equally high results with no loss in Snellen lines (Fig. 1A). Binocular visual acuity (VA) of all patients was 20/20 (Snellen acuity). Similar results can be shown in terms of safety, which describes the postoperative CDVA in relation to preoperative CDVA (Fig. 1B). Therefore, the safety and efficacy indices (SI and EI) were 1.0 for both treatment groups. It is important to note that these visual results were obtained binocular for a more realistic examination of the visual outcome. To calculate predictability, the attempted SE is set in relation to the achieved SE. Here, 100% of eyes in Group A and 95.7% in Group B achieved the attempted SE target within a range of ± 0.50 D (Fig. 1C). There were no differences in predictability between the groups as shown by the Fisher's exact test (p = 0.245). The accuracy, indicating the proportion of eyes with a SE within ± 0.50 D of the target, is presented in Fig. 1D. All eyes were within this target range, except 4% of eyes in the PTK−PRK group, which were between −1.00 and −0.50 D, postoperatively.

PRK is still highly valued today, either as a first treatment for eyes with thin corneas, high refractive error, or as a retreatment after a previous refractive intervention [13]. Therefore, studies investigating different PRK methods are still important. Our data show high efficacy and safety indices for both treatment modalities, tPRK and PTK−PRK, without significant differences.

Previous studies have already provided numerous proofs of the high efficacy and safety of the tPRK method. Alasmari et al. demonstrated a good refractive outcome with tPRK in mild myopia correction [14]. Furthermore, two meta-analyses underline these promising results [7, 15] in addition to several comparative studies [16‐18].

Regarding the comparison of the two laser ablation methods, two- and one-step PRK, there seems to be no significant difference in refractive outcomes expressed in the standard parameters of efficacy, safety, and predictability [19]. However, in highly myopic eyes (> −6.00 D), tPRK appears to be more effective than the conventional PRK methods [20]. One disadvantage of conventional PRK is the uneven dehydration of the stroma caused either by mechanical- or alcohol-assisted debridement of the corneal epithelium [15]. In contrast, tPRK allows for a more consistent corneal ablation, resulting not only in stable refractive outcomes, but also in less pain and haze formation over time [21]. Controversial results may be caused by the different surgical regimens in terms of MMC use or the follow-up period [3].

Laser-assisted epithelial removal, as used in PTK, is a less traumatic technique which provides a smoother ablation profile by reducing the postoperative fibroblast hyperplasia and increasing the epithelial adherence [22, 23]. Published data on the novel PRK method we studied, which consists of PTK treatment followed by PRK, are very scarce. The few available studies have been published on combined PTK−PRK treatment for myopic eyes with corneal scars or other opacities [24]. In addition, this combination is also used to correct complications from previous refractive treatments [25]. This is the first study comparing the tPRK technique using the Amaris750 with PTK−PRK performed with MEL90. Tangmonkongvoragul et al. demonstrated good refractive results of the PTK−PRK method for treating mild myopia with the EX500 excimer laser [8]. A recently published study also confirmed the efficacy of PTK−PRK surgery compared to tPRK treatment [9]. These results are consistent with our findings. This two-step procedure has some similarities with the tPRK, as it is a non-contact laser treatment. Nevertheless, Abdel−Radi et al. demonstrated a faster recovery time in tPRK eyes compared with PTK−PRK patients, which they attributed to a shorter surgical time and thus more precise alignment between the epithelial and stromal ablation with respect to their contraction. They concluded that tPRK surgery is superior to PTK−PRK regarding visual outcome. The faster recovery of this single-step technique is known from previous studies, which compared tPRK to mechanical two-step PRK [16]. Regarding HOA, we found no significant differences between the two groups. However, it is worth mentioning that an aspheric PTK theoretically produces different aberrations than a PTK-based epithelial ablation without aspheric correction, where less energy reaches the peripheral cornea. This may have clinical relevance under scotopic lighting conditions.

The use of MMC in refractive surgery is frequently discussed. While some authors promise stable results without MMC [26], others recommend its use to avoid postoperative complications such as haze and scarring [27]. Some also suggest the use of MMC after PTK treatment [28]. We have used MMC for all our treatments with no postoperative complications in the follow-up period, which is in line with previous works. Abdel−Radi et al. reported minimal haze formation (grade 1) even after MMC application [9]. However, they used a different laser platform, which makes comparison difficult.

A limitation of our study is the short follow-up period, which may have led us to overlook potential postoperative complications. However, from our experience, we know that the use of MMC prevents haze formation in almost all PRK cases, which can be one of the biggest issues after PRK. Further limitations may be the small group size and the lack of comparison data of other study groups using the same laser platforms. Additionally, due to differences in equipment availability at the respective practice locations, different devices were utilized for topographic measurements in the two groups, potentially introducing a bias. However, we made concerted efforts to alleviate this potential influence by ensuring consistent optical zone settings across all employed devices.