Visual outcomes of scleral lenses in different stages and cone locations of keratoconus: A retrospective observational study

Abstract

This study aims to evaluate the visual performance of scleral contact lenses (SCLs) in patients with keratoconus based on disease stage and cone localization and provide data to inform clinical practice. This retrospective study included 95 eyes from 61 keratoconus patients (34 females and 27 males) who underwent corneal cross-linking and were regularly followed at a single center between December 2019 and December 2023. Patients were categorized according to keratoconus stage (mild, moderate, or advanced) and cone localization (central, paracentral, or inferior) based on corneal topography. Visual acuity was recorded before and after mini SCL fitting, as were corneal topography and anterior segment parameters. Statistical analyses included the Wilcoxon signed-rank test, Mann–Whitney U test, and Kruskal–Wallis test for group comparisons, with Spearman correlation analysis used to assess associations between parameters. The mean patient age was 23.7 ± 7.1 years. Among the eyes evaluated, 44% had mild, 36% moderate, and 20% advanced keratoconus; cone localization was inferior in 56%, central in 34%, and paracentral in 10%. Mean logMAR visual acuity improved significantly from 0.95 ± 0.54 before SCL fitting to 0.14 ± 0.13 after application (P <.001), corresponding to a mean Snellen line gain of 0.59 ± 0.20 (95% CI: 0.54–0.64). Visual improvement was significantly greater in eyes with moderate-stage keratoconus and paracentral cone localization compared with other groups (P <.05). In advanced keratoconus, the gain in corrected distance visual acuity was negatively correlated with keratometry values and anterior chamber depth. SCLs significantly enhance visual acuity in keratoconus patients, with the most substantial benefit observed in those with moderate-stage disease and paracentral cone localization. However, the small sample size in the paracentral group limits the generalizability of subgroup findings. These findings highlight the importance of considering both disease stage and cone location when planning SCL fitting, and suggest that paracentral cones may be predictive of superior visual outcomes with mini SCLs.

1. Introduction

Scleral contact lenses (SCLs) provide excellent visual acuity in patients with keratoconus and can significantly reduce the need for corneal transplantation.^[1] Although spectacles and soft contact lenses may offer adequate visual rehabilitation in the early stages of the disease, they often become insufficient as keratoconus progresses.^[2] In more advanced stages, complex contact lens modalities such as rigid gas permeable (RGP) lenses, hybrid lenses, and SCLs are commonly employed.^[3] Over the past 15 years, the role of SCLs has evolved considerably from a niche solution primarily reserved for advanced keratoconus to a first-line therapeutic option in the management of the disease.^[4–6]

Scleral geometry and the stage of keratoconus are key factors that can influence visual outcomes following SCL fitting.^[7–9] Cone localization is a determinant of scleral geometry and disease stage may affect the effectiveness of lens application, so it is essential to assess these parameters during the fitting process. However, few studies have investigated the effects of cone localization on visual acuity outcomes with SCL use in keratoconus patients.^[10] Data comparing visual rehabilitation outcomes across different stages of keratoconus are also insufficient.

Therefore, this study evaluated the visual performance of SCLs in patients with keratoconus in relation to disease stage and cone localization; the findings may inform and guide clinical SCL applications in this context.

2. Materials and methods

This retrospective observational study evaluated the visual acuity outcomes of SCLs in patients with keratoconus based on disease stage and cone localization. Analyses included data from patients treated and followed at the cornea and contact lens department of a single center between December 2019 and December 2023. In total, 95 eyes from 61 patients (34 females and 27 males) were included. All patients had undergone corneal cross-linking at least 3 months prior, were under regular follow-up for keratoconus, had a confirmed diagnosis based on ophthalmologic examination and corneal topography, demonstrated compliance with contact lens assessment, and had complete medical documentation.

Patients were excluded from analyses if they had any additional ocular conditions that could affect visual acuity, a history of ocular surgery other than corneal cross-linking (e.g., keratoplasty or intrastromal ring implantation), or systemic diseases. Prior to SCL fitting, refractive errors were assessed using an autorefractometer (Topcon KR-1®, Japan), and best-corrected distance visual acuity (CDVA) as well as corneal topography were evaluated using Scheimpflug imaging (Pentacam HR; Oculus GmbH, Wetzlar, Germany). Following SCL application, refraction and CDVA were remeasured. Anterior segment imaging was performed using optical coherence tomography (Heidelberg SL-OCT, Heidelberg Engineering, Germany), and slit-lamp biomicroscopic images were obtained.

Corneal topography parameters assessed included flat keratometry (K1), steep keratometry (K2), mean keratometry (K_m), maximum keratometry (K_max), thinnest corneal thickness, white-to-white distance, anterior chamber depth, and pupil diameter. Based on the Km values, eyes were categorized into 3 stages of keratoconus: mild (K_m <48 D), moderate (48 D ≤K_m ≤53 D), and advanced (K_m >53 D). With regard to cone localization, eyes were classified into 3 groups: central, paracentral, and inferior. Cone location was defined using axial curvature and elevation maps as follows:

• Central: apex ≤ 1.5 mm from corneal vertex

• Paracentral: 1.5 to 3.0 mm from vertex

• Inferior: >3 mm inferior to vertex

2.1. SCL application

Mini Misa scleral lenses with diameters of 16.5 and 17.0 mm (Microlens Contact Lens Technology, Arnhem, The Netherlands) were used in this study. The most suitable lens diameter and base curve were selected individually for each patient. Following lens insertion, the distribution of fluorescein-stained saline between the posterior lens surface and the cornea was assessed under cobalt blue illumination to evaluate the fit. The lens edge was examined biomicroscopically to assess for any pressure on the conjunctiva or sclera. Lens vault and clearance were evaluated using anterior segment optical coherence tomography within the first hour of wear. Once the optimal lens parameters were determined, patients were provided with detailed instructions on lens hygiene and proper handling.

2.2. Ethical approval

The study protocol was approved by the local ethics committee (HRÜ 25.02.30; January 27, 2025). All procedures were conducted in accordance with the principles of the Declaration of Helsinki, and patient confidentiality was maintained throughout the study. Informed consent was obtained from all patients. This study involved retrospective review only, and no procedures were performed for research purposes.

2.3. Statistical analysis

Data analysis was performed using IBM SPSS 26 (Chicago). Descriptive statistics including means, standard deviations, frequencies, and percentages were calculated for variables including age, gender, and visual acuity. The Mann–Whitney U test was used for comparisons between 2 groups, and the Kruskal–Wallis test was used for comparisons involving 3 or more groups. Pre- and post-intervention comparisons within the same group were evaluated using the Wilcoxon test. Correlations between keratometric parameters and visual acuity were assessed via Spearman correlation analysis. A P-value of <.05 was deemed statistically significant. A post hoc power analysis indicated >90% power to detect a 0.20 logMAR change at α = 0.05. A priori calculation was not feasible due to the retrospective design.

3. Results

Patients had a mean age of 23.71 ± 7.06 years (range: 12–44). With regard to age distribution, 20% of patients were under 18 years, 65% were between 18 and 30 years, and 15% were over 30 years. Baseline mean values prior to SCL fitting were as follows: K1: 47.94 ± 7.31 D, K2: 52.52 ± 7.52 D, K_m: 50.13 ± 7.17 D, K_max: 59.12 ± 9.33 D, thinnest corneal thickness: 428.27 ± 62.54 µm, white-to-white: 11.92 ± 0.46 mm, anterior chamber depth: 3.45 ± 0.37 mm, and pupil diameter: 3.71 ± 0.64 mm (Table 1). Regarding disease severity, 44% of eyes had mild keratoconus, 36% moderate, and 20% advanced. In terms of cone localization, 56% of eyes had inferior cones, 34% central, and 10% paracentral.

Table 1.

Corneal topography data before scleral contact lens application. Distribution of patients according to stage and cone location.

	Mean	SD±	IQR	Range
Age (yr)	23.71	7.44	7.06	12–44
CDVA (logMAR)	0.95	0.54	0.78	0.22–3.00
K1 (D)	47.94	7.31	6.20	30.90–72.50
K2 (D)	52.52	7.22	5.30	39.10–79.30
Kmax (D)	59.12	9.33	8.90	41.60–86.30
TCT (µm)	428.27	62.54	67.0	201–572
ACD (mm)	3.45	0.37	0.51	2.64–4.46
WTW (mm)	11.92	0.46	0.60	9.90–12.80
PD (mm)	3.71	0.64	1.01	2.15–5.12
Age groups	<18 yr	18–30 yr	≥30 yr	–
	20% (n = 12)	65% (n = 40)	15% (n = 9)	–
Stage groups	Mild	Moderate	Advanced	–
	44% (n = 42)	36% (n = 34)	20% (n = 19)	–
Cone location	Inferior	Central	Paracentral	–
	56% (n = 53)	34% (n = 32)	10% (n = 10)	–

ACD = anterior chamber depth, CDVA = corrected distance visual acuity, IQR = interquartile range, K1 = flat keratometry, K2 = steep keratometry, KC = keratoconus, Kmax = maximum keratometry, PD = pupil diameter, TCT = thinnest corneal thickness, WTW = white-to-white.

In subgroup analyses based on cone location (Table 2), distinct baseline demographic and keratometric differences were observed among the groups. The mean age was measured as 21.40 ± 8.14, 23.71 ± 5.44, 31.10 ± 8.23 in the central, inferior and parcentral groups, respectively. Eyes with paracentral cone localization had significantly higher baseline logMAR visual acuity (1.36 ± 0.82), indicating worse visual function initially, despite having comparatively lower Kmax values (54.02 ± 7.35 D) relative to the central cone group (61.70 ± 11.17 D).

Table 2.

Baseline demographic data of cone groups. (mean ± standard deviation).

Cone location	Inferior	Central	Paracentral
	(n = 53)	(n = 32)	(n = 10)
Age (yr)	23.71 ± 5.44	21.40 ± 8.14	31.10 ± 8.23
CDVA (logMAR)	0.82 ± 0.46	1.00 ± 0.51	1.36 ± 0.82
K1 (D)	46.77 ± 4.75	50.59 ± 8.37	41.51 ± 6.08
K2 (D)	50.65 ± 5.04	55.26 ± 8.30	48.46 ± 8.30
Kmax (D)	57.05 ± 6.89	61.70 ± 11.17	54.02 ± 7.35

CDVA = corrected distance visual acuity, K1 = flat keratometry, K2 = steep keratometry, KC = keratoconus, Kmax = maximum keratometry.

Mean visual acuity prior to lens fitting was 0.95 ± 0.54 logMAR, which significantly improved to 0.14 ± 0.13 logMAR following SCL application (P <.001). The mean Snellen line gain after SCL fitting, compared to pre-lens CDVA, was 0.59 ± 0.20 (95% CI: 0.54–0.64). Table 1 lists baseline corneal topography parameters and Table 3 summarizes the correlation analysis between changes in visual acuity and corneal topographic parameters across mild, moderate, and advanced stages of keratoconus following SCL application. In eyes with advanced keratoconus, a significant negative correlation was observed between CDVA improvement and both keratometry values and anterior chamber depth.

Table 3.

Correlation analysis of CDVA changes with corneal tomographic parameters in different keratoconus stages.

CDVA differences
	Mild KC	Modarate KC	Advanced KC	All groups
	n = 42	n = 34	n = 19	n = 95
Age*
r	0.211	−0.049	0.272	0.100
P	.180	.781	.274	.337
K1
r	−0.098	0.060	−0.580	−0.138
P	.538	.736	.009	.183
K2
r	0.059	0.043	−0.590	−0.113
P	.711	.810	.008	.276
Km
r	−0.034	0.071	−0.624	−0.132
P	.831	.688	.004	.201
Kmax
r	0.069	−0.100	−0.521	−0.099
P	.665	.574	.022	.338
TCT
r	0.041	−0.154	0.326	0.049
P	.799	.384	.174	.637
WTW
r	0.052	0.085	0.028	0.069
P	.756	.649	.913	.525
ACD*
r	0.095	−0.145	−0.698	−0.168
P	.564	.413	.001	.111
PD*
r	−0.194	−0.145	−0.052	−0.131
P	.236	.415	.836	.216

ACD = anterior chamber depth, CDVA = corrected distance visual acuity, K1 = flat keratometry, K2 = steep keratometry, KC = keratoconus, Km = mean keratometry, Kmax = maximum keratometry, PD = pupil diameter, TCT = thinnest corneal thickness, WTW = white to white.

^*Parametric Pearson.

In eyes with mild keratoconus, the mean CDVA without lens was 0.79 ± 0.45 logMAR, improving to 0.11 ± 0.10 logMAR with SCL. In eyes with moderate keratoconus, CDVA improved from 0.95 ± 0.52 logMAR (without lens) to 0.12 ± 0.09 logMAR (with SCL). In eyes with advanced keratoconus, CDVA improved from 1.29 ± 0.63 logMAR to 0.24 ± 0.29 logMAR after SCL fitting (Table 4). Figure 1 presents the Snellen line improvement stratified by keratoconus stage. The letter gain was significantly greater in cases of moderate and advanced keratoconus compared with mild keratoconus (P = .013). Following SCL application, the mean Snellen line gain was 0.64 ± 0.19 in the paracentral cone group and 0.59 ± 0.20 in the central cone group. Visual improvement was significantly greater in eyes with paracentral cone localization compared to those with central cones (P <.01). The subgroup with moderate-stage keratoconus and paracentral cone localization demonstrated the greatest Snellen line improvement compared to all other groups (Fig. 1).

Table 4.

Visual acuity outcomes by keratoconus stage after SCL application.

	CDVA before (logMAR)	CDVA after (logMAR)	CDVA gain (main ±)	95% CI (CDVA gain)	P-value
Mild	0.79 ± 0.45	0.11 ± 0.10	0.68 ± 0.20	0.63–0.74	<.001
Moderate	0.95 ± 0.52	0.12 ± 0.09	0.83 ± 0.18	0.78–0.88	<.001
Advanced	1.29 ± 0.63	0.24 ± 0.29	1.05 ± 0.34	0.91–1.19	<.0001

CDVA = corrected distance visual acuity, CI = confidence interval, SCL = scleral contact lens.

Figure 1.

Distribution of Snellen line improvement after scleral contact lens in different keratoconus groups.

4. Discussion

We found that Snellen line improvement following mini SCL application was significantly greater in the group with moderate-stage keratoconus and paracentral cone localization compared to other groups. When both cone localization and keratoconus stage were considered together, greater visual improvement was observed in eyes with mild keratoconus and central cones, whereas the highest gain was noted in cases of moderate to advanced keratoconus with paracentral cone localization (Fig. 1).

A previous study compared different contact lens designs and reported that visual improvement was greater in eyes with central cones when using RGP lenses, but that cone localization did not significantly influence visual outcomes with mini SCLs.^[10] SCLs are generally more suitable for advanced and peripherally located keratoconus, as RGP and hybrid lenses often exhibit decentration and instability in these cases.^[11] Currently, SCLs are increasingly favored for enhancing visual quality in keratoconus patients. Several studies have reported significant improvements in both visual acuity and contrast sensitivity following SCL use, supporting their emergence as a first-line therapeutic option.^[6,12–14]

Scleral asymmetry is a critical factor in successful scleral lens fitting. Parameters including flat and steep keratometry values, along with cone localization, can serve as useful predictors for assessing scleral toricity.^[7] We found that paracentral cone localization was associated with greater visual improvement. Kowalski et al^[8] reported that the primary factor influencing scleral lens centration was the disparity between the horizontal and vertical meridians. These results align with our findings that paracentral cone localization yielded the most pronounced improvement in visual acuity following SCL application. This suggests that visual impairment in paracentral keratoconus may result more from complex irregularities and asymmetric corneal distortion than from absolute corneal steepening alone. Additionally, the older mean age (31.10 ± 8.23 years) observed in the paracentral group may reflect longer disease progression or later clinical diagnosis. These baseline differences likely influenced the superior visual acuity gains observed post-SCL fitting in the paracentral group, as scleral lenses effectively neutralize complex corneal irregularities. Thus, clinicians should consider cone location-specific corneal features in addition to keratometric severity when predicting scleral lens performance. In cases of inferior cone localization, visual acuity often remains relatively preserved if the central visual axis is unaffected. However, in advanced inferior cones, significant visual deterioration can occur, and it can be challenging to achieve stable lens centration on the corneal surface.

The uneven numerical distribution among groups may have influenced the outcomes of our study. Only 10% of patients had paracentral cones, and 36% were classified as having moderate keratoconus, whereas the majority were in the mild stage and had inferior cone localization. The relatively lower visual improvement observed in cases of advanced keratoconus compared with the moderate group may be explained by reduced visual potential due to corneal scarring. Nevertheless, scleral lenses significantly improved visual acuity across all stages of keratoconus, with particularly notable benefits in moderate and advanced cases. Paracentral cone localization was associated with greater visual gains, likely due to improved lens centration and enhanced peripheral optical quality. These findings suggest that even patients with advanced keratoconus can achieve meaningful visual rehabilitation with SCLs, supporting their use as a valuable nonsurgical treatment option. A limitation of this study is the uneven distribution among groups, particularly the small sample size of the paracentral cone group (10%). This may have affected the generalizability of the outcomes.

5. Conclusion

SCLs significantly improve visual acuity in patients with keratoconus, with the most pronounced benefits observed in moderate to advanced stages and in eyes with paracentral cone localization. These findings reinforce the role of SCLs as an effective therapeutic option for visual rehabilitation in keratoconus, particularly beyond the early stages of the disease.

Author contributions

Data curation: Sinem Kaya.

Formal analysis: Sinem Kaya, Ayhan Sağlik.

Methodology: Sinem Kaya, Ayhan Sağlik.

Project administration: Sinem Kaya, Ayhan Sağlik.

Writing – original draft: Sinem Kaya, Ayhan Sağlik.

Writing – review & editing: Ayhan Sağlik.