Corresponding Author: Kyoung Woo Kim, MD, PhD. Department of Ophthalmology, Chung-Ang University Hospital, Chung-Ang University College of Medicine, 102 Heukseok-ro, Dongjak-gu, Seoul 06973, Korea. Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN), a more severe SJS variant, are acute inflammatory disorders that can affect multiple systems, including the skin and mucous membranes. They are one of the most severe ocular surface diseases, causing significant corneal damage and a substantial threat to vision. During the acute phase, approximately 50% of patients with SJS/TEN experience severe ocular complications including pseudomembranous conjunctivitis and corneal epithelial defects [1]. Early ocular manifestations of the diseases can be diverse, ranging from conjunctival hyperemia to nearly complete sloughing of the entire ocular surface and eyelid margin epithelium [2]. The end-stage status of SJS/TEN is often partial conjunctivalization, progressing toward complete conjunctivalization with accompanying vision loss [3].
Due to the small difference in refractive indexes between posterior cornea (1.367) and the aqueous humor (1.336), the evaluation of posterior cornea curvature has been challenging [4]. However, the recent development of anterior segment optical coherence tomography (AS-OCT) has enabled precise measurements of anterior and posterior corneal parameters in abnormal eyes [5]. With the help of AS-OCT, we can now gain a comprehensive understanding of the alterations in corneal optics caused by SJS/TEN, for both academic evaluation of their progression and appropriate clinical guidance.
Anterior corneal curvature alterations brought on by SJS/TEN have been the main subject of previous research. Due to the usual observances of the signs of SJS/TEN, alterations to the posterior cornea have been largely neglected from research; however, in more severe cases, distortion of the entire cornea, including the posterior surface, may occur. To precisely quantify the total optical changes in the cornea, including modifications in posterior curvature, we have used AS swept-source OCT (SS-OCT). Through this comprehensive analysis, we ultimately intend to deepen our knowledge on ocular involvement of SJS/TEN and contribute to the development of more effective treatment strategies.
Materials and Methods
Ethics statement
This study was approved by the Institutional Review Board of Chung-Ang University Hospital (No. 2408-006-19534). The requirement for informed consent was waived due to the retrospective nature of the study. The entire research process was conducted in accordance with the principles of the Declaration of Helsinki.
Subjects
This study was a retrospective, single center, controlled comparative analysis. This study included 23 eyes from 14 patients with SJS/TEN, assessed using AS SS-OCT between August 2021 and July 2024. The control group, which also included 23 eyes from 14 subjects in a pre-cataract surgery cohort, was matched for age and sex. The SJS/TEN group was further divided into two subgroups depending on the corneal opacity presence: with (subgroup A) and without (subgroup B).
Study design
First, we selected the SJS/TEN group and the control group according to the inclusion criteria. Using AS SS-OCT, we retrospectively evaluated corneal refractive values, including anterior, posterior, and total corneal curvature, as well as corneal pachymetry values in patients with SJS/TEN and controls. Statistical comparisons of these values were conducted between the SJS/TEN group and the control group, as well as among the control group, subgroup A, and subgroup B.
Measurement of corneal refractive and thickness profiles based on AS SS-OCT
Both the SJS/TEN group and the control group underwent measurements using AS SS-OCT (Anterion). On the anterior axial curvature map, the following parameters were obtained: simulated keratometry (Sim K) average (diopters, D), Sim K steep (D), Sim K flat (D), anterior corneal astigmatism (D) at the 3-mm ring, best-fit sphere (BFS; D), maximal keratometry (Kmax; D), and K max distance (mm) from the corneal vertex within the 8-mm zone. On the posterior axial curvature map, the parameters obtained included K average (D), steep K (D), flat K (D), posterior corneal astigmatism (D), posterior to anterior curvature ratio (P/A ratio) at the 3-mm ring, BFS (D), Kmax (D), Kmax distance (mm), and P/A ratio within the 8-mm zone. On the total corneal power map, total corneal power (TCP) profiles were assessed, including TCP average (D), TCP steep (D), TCP flat (D), and total corneal astigmatism (D) at the 3-mm ring. On the corneal pachymetry map, central corneal thickness (μm) and the thinnest point thickness (μm) were measured, and the thinnest point distance (mm) from the corneal vertex was estimated based on the thinnest point x/y (mm) values as part of the corneal thickness profile.
Automated keratometry
Anterior flat K, steep K, average K (Kave), and corneal astigmatism (Kastig) values were obtained using automated keratometry (Topcon) in both groups.
Statistical analysis
Statistical analyses were performed using Prism ver. (GraphPad). The D’Agostino-Pearson test was used to evaluate whether the data followed a normal distribution. To compare data between group 1 and group 2, the chisquare test, Fisher exact test, Student t-test, or the nonparametric Mann-Whitney U-test were applied as appropriate. For comparisons among three groups, one-way analysis of variance followed by Bonferroni post hoc test was performed. Values are presented as mean ± standard deviation, with statistical significance set at a p-value of <0.05. Anterior and total corneal refractive values, as well as corneal pachymetry values for six eyes that had previously undergone refractive surgery, were excluded from the statistical analyses.
Results
Demographics
This study included 23 eyes from 14 patients with SJS/TEN, as well as an equal number of eyes and patients from the control group in a pre-cataract cohort. Table 1 presents the demographic information of both groups. The control group comprised six women (42.9%) and eight men (57.1%), with an average age of 50.9 ± 13.3 years. The SJS/TEN group consisted of seven women (50.0%) and seven men (50.0%), with an average age of 50.1 ± 11.8 years. The age and sex distributions between the two groups did not differ significantly (p = 0.858 and p > 0.999, respectively).
Differences in AS SS-OCT-based anterior corneal refractive profiles between the control group and the SJS/TEN group
In the control group, the anterior astigmatism at the 3-mm ring, Kmax within the 8-mm zone, and Kmax distance within the 8-mm zone were measured at 0.96 ± 0.63 D, 43.82 ± 1.35 D, and 1.98 ± 0.82 mm, respectively. These measurements showed significant differences compared to those in the SJS/TEN group, where the corresponding values were 4.88 ± 4.33 D, 52.36 ± 10.94 D, and 1.42 ± 1.09 mm (p < 0.001, p < 0.001, and p = 0.018, respectively). However, no significant differences were observed between the two groups in terms of Sim K average, Sim K steep, Sim K flat, and BFS in the anterior corneal refractive profiles (Table 2).
Differences in AS SS-OCT-based posterior corneal refractive profiles between the control group and the SJS/TEN group
The differences in posterior corneal refractive profiles between the two groups are shown in Table 3. In the control group, the values of posterior astigmatism at the 3-mm ring and Kmax within the 8-mm zone were −0.34 ± 0.11 D and −6.39 ± 0.30 D, respectively. In comparison, the corresponding measurements in the SJS/TEN group were −0.60 ± 0.47 D and −6.92 ± 0.98 D, respectively, showing significant differences (p = 0.036 and p = 0.003, respectively). However, no significant differences were observed between the two groups in other posterior corneal refractive profile parameters, including K average, steep K, flat K, P/A ratio, BFS, and Kmax distance.
Differences in AS SS-OCT-based total corneal refractive profiles between the control group and the SJS/TEN group
Table 4 illustrates the differences in total corneal profiles between the groups. In the control group, the measurements of TCP steep and total corneal astigmatism at the 3-mm ring were 42.52 ± 1.57 D and 0.85 ± 0.63 D, respectively. These values were significantly different from those in the SJS/TEN group, where the corresponding measurements were 47.06 ± 7.18 D and 5.08 ± 5.10 D, respectively (p = 0.006 and p < 0.001, respectively). However, no significant differences were observed between the TCP average and TCP flat values of the two groups.
Differences in corneal thickness profiles between the control group and the SJS/TEN group
In the control group, the thinnest point thickness was measured at 543.80 ± 32.79 μm, and the thinnest point distance was 0.65 ± 0.27 mm. In contrast, in the SJS/TEN group, the thinnest corneal thickness was lower at 477.50 ± 124.10 μm (p = 0.019), and the thinnest point distance was greater at 1.82 ± 1.26 mm (p < 0.001) (Table 5). However, no significant difference was observed in the central corneal thickness between the two groups.
Differences in automated keratometry-based anterior corneal refractive profiles between the control group and the SJS/TEN group
Considering that automated keratometry is more commonly used than AS SS-OCT in routine ophthalmic surgeries, including cataract surgery, an additional comparative analysis of anterior K values obtained through automated keratometry was conducted between the two groups. The values of all parameters including flat K, steep K, Kave, and Kastig were significantly higher ( p = 0.029, p = 0.003, p = 0.003, and p < 0.001, respectively) in the SJS/TEN group compared to the control group (Table 6).
Differences in AS SS-OCT-based anterior, posterior, and total corneal refractive profiles and corneal thickness profiles among the control group and the SJS/TEN group with and without corneal opacity
Next, to verify whether SJS/TEN subjects with corneal opacity (subgroup A) exhibit different AS SS-OCT features compared to those without corneal opacity (subgroup B), we compared the values of subgroup A and subgroup B with those of the control group.
Sim K steep (p < 0.001) and anterior corneal astigmatism (p < 0.001) at the anterior 3-mm ring, Kmax within the anterior 8-mm zone (p < 0.001), posterior corneal astigmatism at the posterior 3-mm ring ( p = 0.006), Kmax (p = 0.006) and P/A ratio within the posterior 8-mm zone ( p = 0.005), and TCP average ( p = 0.050), TCP steep ( p < 0.001), and total corneal a stigmatism at the TCP 3-mm ring (p < 0.001) were highest in subgroup A among three groups (Table 7).
Among these parameters, anterior corneal astigmatism at the anterior 3-mm ring, Kmax within the anterior 8-mm zone, and TCP steep and total corneal astigmatism at the TCP 3-mm ring were significantly higher in subgroup A compared to both the control group (p < 0.001 for all) and subgroup B (p < 0.001, p = 0.002, p = 0.026, and p < 0.001, respectively). Sim K steep at the anterior 3-mm ring, posterior corneal astigmatism at the posterior 3-mm ring, Kmax within the posterior 8-mm zone, and TCP average at the TCP 3-mm ring were higher in subgroup A compared only to the control group (p = 0.009, p = 0.005, p = 0.004, and p = 0.046, respectively). On the other hand, the P/A ratio within the posterior 8-mm zone was significantly higher in subgroup A compared only to subgroup B ( p = 0.004) (Table 7).
Among corneal thickness parameters, central corneal thickness values did not differ significantly among the groups (p = 0.937). However, the thinnest point thickness was lowest in subgroup B (p = 0.010), with significant differences compared to the control group (p = 0.009). The thinnest point distance was longest in subgroup A (p < 0.001), with significant differences observed compared to both the control group (p < 0.001) and subgroup B (p = 0.001) (Table 7).
AS SS-OCT findings in the control group and the SJS/TEN group
Representative AS SS-OCT findings from the anterior axial curvature map, anterior tangential curvature map, posterior axial curvature map, and corneal pachymetry map are presented for the control group (Fig. 1A-1D), the SJS/TEN group without corneal opacity (Fig. 1E-1H), and the SJS/TEN group with corneal opacity (Fig. 1I-1L). In SJS/TEN cases with corneal opacity, pronounced posterior curvature irregularity as well as severe anterior corneal irregular astigmatism is observed (Fig. 1J, 1L).
Discussion
The possibility of a long-term progression of ocular surface cicatrization is well-established from earlier research on SJS/TEN. There are two categories of ocular surface failures: conjunctival failure, which includes squamous metaplasia, symblepharon, and fornix shortening; and limbal stem cell failure, which is characterized by conjunctivalization, corneal neovascularization, and persistent epithelial defects [3]. The corneal damage through limbal stem cell failure or scarring is one of the most devastating ocular outcomes of SJS/TEN.
To accurately diagnose a corneal pathology and assess its structure, the most accurate corneal measurements are needed. Accurate measurements of both the anterior and posterior cornea are essential not only for the diagnosis but also for the treatment of different ocular diseases [6]. This requirement also applies to SJS/TEN, which affects the cornea. Despite this, research examining the quantitative effect of SJS/TEN on corneal optical characteristics has been lacking. Using the AS SS-OCT, we quantitatively assessed anterior, posterior, and total corneal curvature alterations, as well as corneal pachmetry features, in patients with SJS or TEN. The accurate corneal measurements are important because SJS/TEN-induced limbal stem cell deficit can alter corneal structure in a way that negatively impacts visual function and quality of life in general [7].
Since its introduction, AS-OCT has become crucial for assessing the cornea and anterior eye segment. AS-OCT has been discovered to exceed Placido-Scheimpflug imaging in reproducibility for central corneal thickness and keratometry measurements after laser in situ keratomileusis (LASIK) [8]. The high-quality, detailed anterior and posterior cornea maps are essential for diagnosing and managing conditions like corneal ectasia, assessing corneal astigmatism, or planning refractive surgery [6]. Notably, AS SS-OCT, a type of AS-OCT, uses a high-speed, narrow-bandwidth light source to obtain up to 2 million A-scans per second [9-11]. AS SS-OCT provides detailed maps such as anterior and posterior curvature, elevation, TCP, wavefront, and pachymetry with superior depth range, decreased sensitivity roll-off, and reduced susceptibility to motion artifacts [12]. Due to its excellent performance, AS SS-OCT is increasingly used for precise corneal measurement in many clinical practices: we have used AS SS-OCT to gather accurate measurements of the corneal parameters.
As discussed in the results of this study, corneal astigmatism and Kmax values were significantly higher in SJS/TEN group for both the anterior and posterior corneal curvatures. There have been prior studies investigating the changes in the anterior surface of patients with SJS/TEN [13]. Many previous studies have also discovered a positive correlation between the anterior and posterior corneal asphericities [14]. However, studies investigating the posterior cornea have been scarce. In this current study, the posterior astigmatisms of both the control group and the SJS/TEN group fall within the known average range of posterior corneal surface (−0.26 to −0.78 D) [15,16], and there is a significant difference between two groups. Although the posterior corneal surface only contributes to about 10% of the total refractive power of the eye, a precise assessment of its morphology is important because evaluations neglecting the posterior corneal surface measurements may cause significant deviations from the precise corneal astigmatism estimation [14]. We think that the possibility of extensive corneal impairment requires an accurate assessment of the posterior cornea in patients with SJS/TEN for better depiction of the damage.
The subgroup analysis comparing SJS/TEN cases with and without corneal opacity provided valuable insights into the severity of corneal involvement in these conditions. Patients with SJS/TEN with corneal opacity (subgroup A) demonstrated significantly higher values for several corneal parameters, including anterior and posterior corneal astigmatism, anterior and posterior Kmax, and total corneal astigmatism compared to the control group. Notably, patients with SJS/TEN without corneal opacity (subgroup B) did not show significant differences in all of parameters compared to the control group (Table 7). This finding suggests that the presence of corneal opacity may be a crucial indicator of more severe corneal involvement in SJS/TEN. The observation of higher values in anterior corneal astigmatism, Kmax within the anterior 8-mm zone, TCP steep, and total corneal astigmatism in subgroup A compared to both the control group and subgroup B underscores the impact of corneal opacity on corneal structure and function.
The identification of severe posterior corneal changes may influence the pre-cataract surgery assessments and the effectiveness of treatments designed to correct anterior curvature issues, such as scleral lenses. When conducting a cataract surgery, neglecting the posterior astigmatism can cause some critical errors in correcting the astigmatism. Because most of the eyes have a posterior astigmatism, which is against-the-rule, if this negligence occurs in toric intraocular lens use, it overcorrects with-the-rule astigmatism and undercorrects against-the-rule astigmatism [14]. Similar to the result of a miscalculation in cataract surgery, an inaccurate calculation of the amount of posterior corneal astigmatism with a steep meridian in prescription of scleral lenses may result in the inaccurate corrections of the astigmatism [14]. We consider that the AS SS-OCT can be beneficial for patients with SJS/TEN by objectively detecting the overall optical damages of the cornea that can impact the subsequent treatments.
Another important finding of our current study is that corneal ectasia is noted in the SJS/TEN group. Generally, the corneal ectasia progression shows a consistent change in at least two of the following parameters: (1) progressive steepening of the anterior corneal surface; (2) progressive steepening of the posterior corneal surface; or (3) progressive thinning and/or an increase in the rate of corneal thickness change from the periphery to the thinnest point [17]. From previous research, the prevalence of corneal ectasia among patients with SJS/TEN as compared with that of the general population was also statistically significant [18]. However, there has been a lack of reports on the prevalence of corneal ectasia and SJS/TEN as associated disorders. The results of our study quantitively show corneal ectasia in the SJS/TEN group.
Additional point to note is that the anterior and posterior steep, flat, and average corneal powers, as well as the BFS and P/A ratio values, did not differ significantly between two groups. This may be attributed to the fact that localized changes induced by SJS/TEN may not translate into global corneal curvature changes. SJS/TEN can cause more patchy, peripheral thinning and scarring rather than uniform thinning at the center. For example, severe inflammation and ulceration of the tarsal conjunctiva and eyelid margins in patients with acute SJS/TEN can cause tarsal scarring, eyelid margin keratinization, and lipid tear deficiency [18]. These adnexal changes can indirectly induce localized scars on cornea via microtraumas. Since this study is retrospective, there are certain limitation in observing the progression and mechanisms of corneal changes. It would be beneficial to investigate these alterations in future studies.
The limitations of this current study are as follows: first, it was a retrospective study, not a prospective study. Due to the rarity of SJS/TEN, it is hard to collect data prospectively from enough patients. Second, the SJS/TEN cases that had undergone ocular refractive surgeries were excluded for the analysis of anterior and total corneal profiles. This exclusion further decreased the number of patients from statistical analysis of those aspects; however, it was inevitable to increase the validity of the results. Moreover, future studies can aim to evaluate the correlations between the ocular surface severity grades and the degrees of corneal axial changes.
In conclusion, AS SS-OCT allows for the objective detection of corneal damage, providing patients with SJS/TEN with valuable long-term monitoring. Since AS SS-OCT shows significant alterations in the posterior corneal curvature, we need to consider posterior corneal curvature changes in pre-cataract surgery evaluations and contact lens prescriptions in patients with SJS/TEN. We also recommend cautious interpretation of these clinically relevant AS SS-OCT parameters when used in treatment decision-making.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
