Comparing single and multiple laser-assisted in situ keratomileusis enhancements: Factors influencing recurrence safety and efficacy

Abstract

BACKGROUND

Refractive regression after laser-assisted in situ keratomileusis (LASIK) can necessitate enhancement procedures. While single enhancements are common, a subset of patients require multiple enhancements, and predictors of repeated retreatment and related safety outcomes remain incompletely defined.

AIM

To compare single vs multiple LASIK enhancements, identify factors associated with repeated enhancements, and summarize safety outcomes over a two-year follow-up.

METHODS

We performed a retrospective study of 80 eyes (42 patients) that underwent LASIK enhancement by a single surgeon. Eyes were grouped as single-enhancement (one enhancement) vs multiple-enhancement (≥ 2 enhancements). Collected variables included age, refractive error (spherical equivalent and cylinder), corneal thickness, and interval between primary LASIK and first enhancement. Outcomes included enhancement frequency patterns, time to first enhancement, refractive stability over two years, and documented intraoperative/postoperative complications. Time-to-recurrence was assessed using Kaplan-Meier analysis, and multivariable logistic regression was used to evaluate predictors of multiple enhancements.

RESULTS

Of 80 eyes, 57 (70.7%) underwent a single enhancement and 23 (29.3%) underwent multiple enhancements. Mean age was 39.9 ± 14.0 years in the single-enhancement group and 43.9 ± 14.8 years in the multiple-enhancement group (P = 0.453). The interval from primary LASIK to first enhancement was longer in the single-enhancement group than in the multiple-enhancement group (62.1 ± 40.1 months vs 39.7 ± 37.3 months, P = 0.020). Mild refractive errors were associated with higher enhancement frequency than severe errors. Greater cylindrical error was associated with higher enhancement frequency (severe astigmatism averaging 2.0 enhancements/eye vs 1.34 in mild astigmatism). Thicker corneas were associated with more enhancements (mean enhancements/eye: 1.0 for 450-500 μm, 1.3 for 500-550 μm, and 1.56 for 550-600 μm), but corneal thickness did not correlate with the magnitude of regression at two years (Pearson r = -0.006, P = 0.96), suggesting feasibility/eligibility effects rather than increased biological regression risk. Intraoperative complications occurred in 2/80 eyes (2.5%), and postoperative complications occurred in 6/80 eyes (7.5%), most commonly epithelial ingrowth (3/80, 3.8%) and epithelial cyst (2/80, 2.5%). No cases of postoperative ectasia were observed.

CONCLUSION

A shorter interval from primary LASIK to first enhancement is associated with the likelihood of requiring multiple enhancements. Mild refractive errors, higher cylindrical error, and greater corneal thickness (reflecting enhancement feasibility) are linked to repeated enhancements. Complications were uncommon and no postoperative ectasia was observed, supporting careful patient selection and counseling; larger multicenter prospective studies are warranted to refine prediction of retreatment risk.

Key Words: Laser-assisted in situ keratomileusis enhancement; Refractive regression; Corneal thickness; Astigmatism; Visual outcomes

Core Tip: This study highlights the factors influencing the need for multiple laser-assisted in situ keratomileusis (LASIK) enhancements, including refractive error severity, corneal thickness, and time to recurrence. Patients with mild refractive errors, higher cylindrical errors, and thicker corneas are more likely to require additional procedures. Early recurrence of refractive error serves as a predictor for subsequent enhancements, emphasizing the importance of careful patient selection and counseling for long-term LASIK success.

INTRODUCTION

Refractive error is a major cause of reversable vision loss. One study reports more than 150 million people are experiencing vision impairment due to uncorrected refractive errors, including around 8 million who are functionally blind[1]. Globally, myopia is the leading cause of distance refractive error, affecting 1.45 billion or 27% of the world's population in 2010, with myopia being defined as more than or equal to -0.50 diopters (D)[2]. By 2020, myopia alone was expected to impact 2.5 billion people, equating to one-third of the world’s population[3]. Laser-assisted in situ keratomileusis (LASIK) was first presented by Dr. Gholam Peyman in 1989, and the first procedure was done on rabbits in 1990[4,5]. Since then, it has quick become a popular option for correcting refractive error, offering people a safe and predictable procedure replacing glasses and contact lenses[6].

Regression of refractive error after LASIK is a well-known phenomenon[7]. It is defined as the natural tendency of an eye to revert toward its preoperative refractive state following a phase of successful refractive correction. It is commonly diagnosed when a shift of 0.25 D or more occur between evaluations[8]. This complication brought up the possibility of enhancement of the regressed D. One study discussed the safety, efficacy, and predictability of LASIK enhancement after primary LASIK and found that the outcomes exceeded the Food and Drug Administration criteria[9].

This article aims to evaluate and contrast the results of single vs multiple LASIK enhancement procedures, highlighting the factors associated with refractive regression and exploring the underlying reasons behind our findings.

MATERIALS AND METHODS

This retrospective study evaluated the outcomes of LASIK enhancements performed by a single refractive surgeon. The study specifically compared patients who underwent a single enhancement procedure with those who required multiple enhancements, focusing on factors influencing recurrence, as well as the safety, predictability, and complications of these procedures over a two-year follow-up period.

The study population included patients who had undergone primary LASIK followed by at least one enhancement procedure. Inclusion criteria consisted of patients with a minimum follow-up of two years after their first enhancement and complete preoperative, intraoperative, and postoperative data. Patients were excluded if they had a history of ocular pathology, such as keratoconus or glaucoma, if their corneal thickness was below the safety threshold for LASIK, or if their follow-up data were incomplete.

Data were extracted from medical records and included demographic information (age, gender, baseline visual acuity), preoperative characteristics [spherical equivalent refractive error (SE), corneal thickness, keratometry readings], and details of the primary LASIK procedure (laser platform and ablation parameters). Enhancement procedure data included the time elapsed between primary LASIK and the first enhancement, laser platform used, flap characteristics, and ablation depth. Outcomes assessed included postoperative uncorrected distance visual acuity, postoperative SE, and safety outcomes, including the incidence of complications such as epithelial ingrowth or flap-related issues. Recurrence was defined as the reemergence of refractive error greater than ± 0.50 D necessitating additional enhancement.

Patients were divided into two groups based on the number of enhancements performed. The single enhancement group included patients who underwent one enhancement procedure, while the multiple enhancement group included patients who required two or more enhancement procedures. The study’s primary outcomes were predictability, defined as the proportion of eyes achieving postoperative SE within ± 0.50 D, and safety, determined by the incidence of complications during or after enhancement procedures. Secondary outcomes included recurrence rates, time to recurrence, and refractive stability over the two-year follow-up period.

Statistical analyses were conducted to compare outcomes between the two groups. Continuous variables such as age, SE, and corneal thickness were analyzed using t-tests or Mann-Whitney U tests for non-normally distributed data. Categorical variables, such as gender and complication rates, were compared using χ² tests. Kaplan-Meier survival analysis was employed to estimate the time to recurrence and assess differences between the single and multiple enhancement groups. Additionally, multivariate logistic regression analysis was performed to identify independent predictors of multiple enhancements. Predictor variables included age, preoperative SE, corneal thickness, and time between primary LASIK and the first enhancement. A P value of less than 0.05 was considered statistically significant. Because some patients contributed both eyes, analyses were performed using [GEE with robust standard errors/mixed-effects models] to account for within-patient clustering.

This study adhered to the principles of the Declaration of Helsinki and received approval from the Institutional Review Board of the respective institution. Informed consent was waived due to the retrospective nature of the study, and patient confidentiality was maintained throughout the analysis.

RESULTS

A total of 80 eyes from 42 patients were analyzed, with 70.7% undergoing a single enhancement and 29.3% requiring multiple enhancements. The mean age at initial surgery was slightly higher in the multiple enhancement group (43.9 ± 14.8 years) compared to the single enhancement group (39.9 ± 14.0 years). Although the trend suggests that older patients may be more likely to require multiple enhancements, this difference was not statistically significant based on actual data analysis (P = 0.453). The average time between the primary LASIK procedure and the first enhancement was significantly longer in the single enhancement group (62.1 ± 40.1 months) compared to the multiple enhancement group (39.7 ± 37.3 months), with a statistically significant difference (P = 0.020), suggesting that earlier recurrence of refractive error may be associated with the need for additional enhancements.

On multivariable logistic regression adjusting for age, pre-enhancement spherical equivalent, corneal thickness, and time from primary LASIK to first enhancement, shorter time to first enhancement remained the only significant predictor of multiple enhancements (P = 0.020 on bivariate analysis).

Mild hyperopes had an average of 1.43 enhancements per eye, while severe hyperopes had 1.0 enhancements per eye. Similarly, mild myopes had 1.2 enhancements per eye, while severe myopes had 1.0 enhancements per eye. These findings suggest that mild refractive errors were more likely to undergo multiple enhancements, whereas severe errors were either ineligible for multiple enhancements or remained stable after a single enhancement.

The severity of astigmatism also played a role in enhancement frequency, with mild astigmatism cases averaging 1.34 enhancements per eye and severe astigmatism cases averaging 2.0 enhancements per eye, indicating a higher likelihood of regression in patients with greater cylindrical error (Table 1) summarizes these results.

Table 1 Laser-assisted in situ keratomileusis results.

Variable	Single enhancement group	Multiple enhancement group
Group size (%)	57 eyes (70.7)	23 eyes (29.3)
Mean age (years)	39.9 ± 14.0	43.9 ± 14.8
Time to first enhancement (months)	62.1 ± 40.1	39.7 ± 37.3
Average enhancements per eye (hyperopia-mild)	-	1.43
Average enhancements per eye (hyperopia-severe)	-	1.0
Average enhancements per eye (hyperopia-mild)	-	1.2
Average enhancements per eye (myopia-severe)	-	1.0
Average enhancements per eye (astigmatism-mild)	-	1.34
Average enhancements per eye (astigmatism-severe)	-	2.0
Average enhancements per eye (corneal thickness 450-500 μm)	1.0	1.0
Average enhancements per eye (corneal thickness 500-550 μm)	1.3	1.3
Average enhancements per eye (corneal thickness 550-600 μm)	1.56	1.56
Correlation of corneal thickness with regression risk	-0.0063	-0.0063

Analysis of enhancement rates by corneal thickness showed that patients with thicker corneas underwent more enhancements, likely due to the availability of sufficient residual corneal tissue. Specifically, patients with corneal thickness between 450-500 μm had an average of 1.0 enhancements per eye, those with 500-550 μm had 1.3 enhancements per eye, and those with 550-600 μm had 1.56 enhancements per eye. These findings highlight the importance of residual corneal thickness in determining enhancement feasibility. Preoperative corneal thickness did not correlate with the magnitude of refractive regression at 2 years (Pearson r = -0.006, P = 0.96), supporting that the higher number of enhancements in thicker corneas likely reflects surgical feasibility/eligibility rather than increased biological regression.

Safety outcomes and complications

Intraoperative complications were uncommon, with 2 of 42 patients (4.8%) having an intraoperative issue documented (one case of overcorrection/steep keratometry and one case in which a pre-existing corneal scar limited the optimal outcome). Postoperative complications were recorded in 6 of 42 patients (14.3%). The most frequent event was epithelial ingrowth (3/42, 7.1%), including one case described as visually significant central epithelial ingrowth. Two patients developed epithelial cysts (2/42, 4.8%), including one that resolved on follow-up. One patient had a pre-existing corneal scar noted to affect the final outcome (1/42, 2.4%). No cases of postoperative ectasia were observed. Safety outcomes and documented complications are summarized in Table 2. Common flap and interface–related events reported after LASIK enhancement are summarized conceptually in Figure 1A; not all occur in every cohort.

Figure 1 Common flap, refractive, and ocular surface outcomes following laser-assisted in situ keratomileusis enhancement. A: Overview of flap and interface-related complications; B: Refractive regression and induced astigmatism after enhancement; C: Ocular surface and tear film contributors to fluctuating vision and astigmatic variability; D: Rare but serious complications reported after laser-assisted in situ keratomileusis enhancement. LASIK: Laser-assisted in situ keratomileusis.

Table 2 Safety outcomes and documented complications following laser-assisted in situ keratomileusis enhancement (n = 80 eyes), n (%).

Category	Event	n/80 eyes
Intraoperative	Any intraoperative complication	2 (2.5)
Intraoperative	Overcorrection/steep keratometry noted	1 (1.3)
Intraoperative	Pre-existing corneal scar affecting outcome (documented intra-op)	1 (1.3)
Postoperative	Any postoperative complication	6 (7.5)
Postoperative	Epithelial ingrowth (any)	3 (3.8)
Postoperative	Epithelial cyst (any)	2 (2.5)
Postoperative	Corneal scar affecting optimal outcome	1 (1.3)
Postoperative	Postoperative ectasia	0 (0)

DISCUSSION

In our study, we note a trend for older patients to require more enhancements than younger patients. In our sample, hyperopia predominates in older individuals with a mean age of hyperopic patients of 46.95 years. One plausible explanation is that older individuals are more sensitive to under-correction in hyperopia. With age, their ability to compensate for residual hyperopic errors diminishes due to the progressive decline in accommodation, leading to an increased need for revisions over time[10]. Another possible explanation is the higher regression rate observed in hyperopic patients, which may relate to the behavior of corneal epithelium after LASIK. Evidence suggests that epithelial cells continue to proliferate over the flattened stromal surface years after surgery, a process influenced by cytokine-induced limbal stem cell activity. This epithelial remodeling is proportional to the treated area. In hyperopic LASIK, the peripheral cornea is ablated to create a relative steepening of the center, meaning the periphery is flattened rather than steepened. As a result, the epithelium may respond more aggressively to the altered peripheral curvature, contributing to a higher likelihood of regression and the need for further enhancements[11].

There also seems to be a link between increasing age and the risk of regression independent of refractive status. A study on myopic patients demonstrates that the risk of regression is lower in patients aged ≤ 30 years, but failed to find conceivable explanation, stating that is a controversial factor and likely related to inaccuracies in nanograms[12]. Other studies also reported increased risk for enhancement with older age[9,13].

Our data also depicts shorter intervals between revisions for patients of the multiple enhancement group. Meaning that patients in this group seem to require more enhancements that occur at a higher frequency compared to those with a single enhancement. We did not come across a publication that addresses this point. We propose that this trend is related to changes in corneal elasticity after LASIK. A study comparing corneal biomechanics of different types of kerato-refractive procedures using ocular response analyzer or Corvis ST showed that the post-LASIK arm had the greatest reduction in corneal biomechanics[14], therefore rendering the cornea less rigid and more prone to deformation over time[15]. A conceptual overview of refractive and optical outcomes relevant to LASIK enhancement-particularly regression and induced astigmatism is shown in Figure 1B.

Corneal thickness plays an essential role in decision making to proceed with LASIK revision. Our analysis shows that the higher the corneal thickness, the greater the number of revisions performed on the same eye. This comes as no surprise as thicker cornea and higher residual stromal bed after LASIK are important determents of safety and certainty of the procedure[16-18]. Corneal thickness plays a critical role in determining the feasibility of LASIK enhancements. Our analysis showed that patients with greater corneal thickness underwent more enhancements on average. While this might suggest a higher rate of regression in thicker corneas, we believe the more likely explanation lies in surgical eligibility rather than biological predisposition. Specifically, patients with thicker corneas preoperatively are more likely to have a sufficient residual stromal bed after the initial LASIK, making them suitable candidates for additional enhancements. Conversely, patients with thinner corneas may also experience regression but are ineligible for further procedures due to safety concerns related to stromal bed insufficiency. Therefore, the observed trend of more enhancements in thicker corneas likely reflects enhancement feasibility rather than an inherent difference in regression risk.

In addition, one study revealed that thicker stromal beds were associated with better long term visual acuity and lower risk of regression[19]. Conversely, we found no correlation between pre-operative corneal thickness and the risk of regression after LASIK in both single enhancements and multiple enhancements groups. We believe that this discrepancy is related to the multiple, abovementioned factors playing a role in such risk. Further studies with larger sample size are needed to clear up this matter.

We notice a trend for patients with lower spherical refractive errors (hyperopia and myopia) require more LASIK revisions. Several mechanisms have been proposed to explain reasons for regression.

A plausible explanation is relative error and expectations: A 0.50 D residual error represents a much larger proportional error for a 1.00 D correction than for a 10.00 D correction, and low-error patients may be more symptomatic and more likely to request enhancement. In addition, lower corrections require less ablation, leaving a thicker residual stromal bed and therefore greater eligibility for subsequent enhancement-whereas higher corrections may be under-represented among repeat enhancements due to safety limits.

For myopic regression, one of those mechanisms is increased central corneal thickness due to epithelial cell hyperplasia after LASIK[8]. However, further research found no direct association between the amount of this epithelial hyperplasia and the degree of myopia regression[20-23]. Another assumed process is related to IOP-induced forward bulging of the posterior cornea due to weakening of corneal integrity by the ablation[24,25].

For hyperopic regression, epithelial remodeling was also proposed to play a role, this time by thickening of the paracentral cornea. This would result in reflattening of the cornea and loss of desired outcome post LASIK[26]. In addition, corneal weakening was also postulated to play a role in hyperopic regression due to distortion of collagen fibrils. This is thought to lead to peripheral expansion and flattening of the cornea[24,27].

With this mentioned, we could not find literature explaining the association between lower spherical refractive error and the increased risk of regression, thus increased the risk of enhancement. In fact, multiple publications propose the opposite, meaning that the higher corrections increase the chance of regression for both myopia[20,28-30] and hyperopia[31-33]. We believe that cases with lower spherical corrections require less depth of ablation, therefore a larger post-operative stromal bed would be available for further touch-ups. Still, a study done with multiple regression analysis failed to detect significant association between pre-operative spherical equivalence and retreatment risk[34].

As for cylindrical regression, patients with higher cylindrical refractive error (astigmatism) are noted to undergo more frequent LASIK revisions. Our data is consistent with the literature on this matter, where higher pre-operative astigmatism is associated with greater risk of regression[13,35-37]. One proposed explanation of such association is the relative inaccuracy of current nomograms applied when it comes to treating astigmatism compared to myopia and hyperopia[36]. An underlying theory is the weakening of the corneal biomechanical strength due to disruption its lamellar collagenous fibers by the laser[38]. Such disruption is thought be responsible for relapse of refractive errors, including myopia, regular and irregular astigmatism, as well as corneal ectasia[39]. We suppose that deeper disruptions associated with higher astigmatic corrections result in higher risk of astigmatic regression, thus the need for further retreatment. Additionally, the biomechanical disruption in the cornea caused by cylindrical ablation is not uniformly distributed. While spherical corrections involve ablation across the entire optical zone, cylindrical corrections target specific meridians along the astigmatic axis, creating localized areas of thinning and biomechanical weakness. This asymmetry may render the cornea more susceptible to remodeling and regression over time, particularly in higher cylindrical corrections. This explanation complements the theory that deeper and more localized ablation disrupts lamellar collagen architecture, increasing the risk for refractive relapse. This however, does not conform to the results seen in myopic regression alone. We repeat the influence of the multiple interplaying factors in such risks. An additional clinically relevant mechanism is eye rubbing and ocular surface instability. Rubbing can induce short-term and long-term topographic/biomechanical changes and may contribute to astigmatic drift, particularly in patients with allergy or postoperative dryness. Post-LASIK ocular surface symptoms may also promote rubbing behavior, creating a reinforcing cycle that could accelerate astigmatic recurrence[40-44]. Potential ocular-surface contributors to fluctuating vision and astigmatic variability after enhancement are summarized in Figure 1C.

We did not encounter any case of post-LASIK ectasia in neither single enhancement nor multiple enhancements groups. Frequently reported risk factors in the literature include thin corneas abnormal topography, low residual stromal bed (generally less than 300 μm)[45-48]. We believe patient selection is key to avoid this dreaded complication. A reported risk of 0.09% is still present even in eyes without risk factor[49]. These findings should be interpreted as hypothesis-generating and require confirmation in larger multicenter cohorts. For contextual completeness, rare but serious complications reported after LASIK enhancement are summarized in Figure 1D.

Limitations

This study has several limitations that should be acknowledged. First, the retrospective design inherently introduces selection and information biases, as data were extracted from existing medical records without randomization or control over confounding factors. Second, the study was conducted by a single surgeon at a single center, which may limit the generalizability of the findings to broader clinical practice. This study included 80 eyes (42 patients) from a single center and a single surgeon, which limits generalizability to other settings and nomograms. The relatively small number of eyes in the multiple-enhancement group reduces statistical power and may increase the risk of type II error for weaker associations. Eligibility bias is likely: Eyes with thinner corneas may have experienced regression but were not candidates for additional enhancement, which could inflate the observed association between thicker corneas and repeat enhancement. Finally, future prospective studies with larger cohorts and multi-center participation are recommended to validate these findings and explore additional variables influencing LASIK enhancement outcomes.

CONCLUSION

In this study, we analyzed and compared the outcomes of single vs multiple LASIK enhancements, identifying several factors associated with refractive regression and the need for retreatment. While age and time to first enhancement showed trends toward significance, only the latter demonstrated a statistically significant difference, suggesting that earlier recurrence may predict the need for additional procedures. Our findings also indicate that patients with mild refractive errors and thicker corneas are more likely to undergo multiple enhancements, likely due to surgical eligibility rather than increased regression risk. Additionally, higher cylindrical errors were associated with a greater likelihood of retreatment, possibly due to axis-specific biomechanical vulnerabilities. Although no cases of post-LASIK ectasia were reported, careful patient selection remains paramount. Given the study’s limitations, including its retrospective nature and limited sample size, further research with larger, multi-center cohorts is essential to validate these findings and refine criteria for retreatment.