Evaluation of Dynamic Corneal Response Parameters and the Biomechanical E-Staging After Accelerated Corneal Cross-Linking in Keratoconus : The Asia-Pacific Journal of Ophthalmology

Secondary Logo

Journal Logo

Original Study

Evaluation of Dynamic Corneal Response Parameters and the Biomechanical E-Staging After Accelerated Corneal Cross-Linking in Keratoconus

Flockerzi, Elias MD*; Xanthopoulou, Kassandra MD*; Daas, Loay MD*; Feld, Simon MD*; Langenbucher, Achim PhD; Seitz, Berthold MD*

Author Information
Asia-Pacific Journal of Ophthalmology: November/December 2022 - Volume 11 - Issue 6 - p 514-520
doi: 10.1097/APO.0000000000000580
  • Open

Abstract

Keratoconus (KC) is an ectatic corneal disease that is characterized by a progressive corneal thinning and irregular astigmatism resulting in visual impairment.1 Several indices and parameters have been proposed to assess and determine KC progression. A recent study reported that an initially advanced KC stage is a risk factor for KC progression.2 The parameters, (1) increase of the maximal anterior keratometry (Kmax) by >1 diopter in 12 months or (2) decrease of the thinnest corneal thickness (TCT) by >20 µm in 12 months, are still up-to-date and used in daily practice.3 Another recent study found an increase of the Pentacam-derived (Oculus, Wetzlar, Germany) Belin-Ambrosio deviation index (BAD-D) of >0.42 as indicative of KC progression.4 The Belin ABCD KC classification was introduced in 20165–8 and includes anterior (“A”) and posterior (“B”) corneal curvature analysis, a pachymetry-based (“C”) and spectacle-corrected visual acuity–based staging (“D”). The parameters are individually staged into 5 possible stages (0–4 for A, B, C, and D, respectively). KC progression may also be assessed based on this classification. On the basis of the definition of CIs for healthy and KC corneas, deviations in each parameter can be interpreted as progression7 requiring stabilization by corneal cross-linking (CXL).

From a biomechanical point of view, KC has a decreased resistance to deformation when compared with healthy corneas.9 The corneal deformation can be measured by the Corvis ST (CST; Oculus, Wetzlar, Germany) after the application of a standardized air puff.10,11 Many studies have already investigated whether the CXL effect can be measured based on CST parameters in different study populations, in some cases with contradictory results—whereas 1 study found an increase of stiffness parameter A1 (SP-A1) in 67 mild KC cases after 6 months,12 another study found no change for this parameter during 4 years of follow-up.13 The standard CST parameters Corvis Biomechanical Index (CBI) and Tomographic Biomechanical Index cannot be used for KC severity or progression assessment because of their nonlinear and quasi-binary behavior.14 The Corvis Biomechanical Factor (CBiF) is the linearized term of the nonlinear CBI that provides biomechanical information about KC severity.14 In a previous study, the CBiF was applied on 448 KC corneas of different severity stages derived from the Homburg Keratoconus Center (HKC) and the resulting CBiF value range was divided into 5 stages (E0 to E4) to create a grading system according to the ABCD KC classification.15 Stage E0 was characterized by values below the 2.5 percentile and the other thresholds were created by dividing the CBiF value range between the 2.5 and 97.5 percentiles into 4 groups of equal values (E1 to E4).15 The resulting biomechanical E-staging for KC and other ectatic corneal diseases as an addition to the ABCD KC classification was thereafter validated based on another independent 860 KC corneas data set from Milano and Rio de Janeiro.15

The purpose of this study was to assess whether the biomechanical E-staging for KC and other ectatic corneal diseases based on the CBiF as new standard parameters of the CST can document biomechanically the stabilizing effect of CXL and, thus, also KC progression in follow-up examinations in daily practice.

METHODS

This retrospective longitudinal study was conducted in the HKC, which is a clinical observational trial (trial number NCT03923101, US National Institutes of Health, https://ClinicalTrials.gov) that was approved by the regulatory body, the local ethics committee of Saarland (Ethikkommission bei der Ärztekammer des Saarlandes, reference number 121/20),8 and adheres to the tenets of the Declaration of Helsinki. Each patient within the HKC provided written consent for data analysis.

Adult patients (≥18 y old) with progressive KC and treatment with accelerated CXL (9 mW/cm2, 10 min, 5.4 J/cm2) using the Avedro kxl system (Avedro, Waltham, MA) from January 2017 to December 2021 were assessed for inclusion. Criteria of KC progression were the following: (1) an increase of corneal astigmatism ≥1 diopter, (2) an increase of Kmax ≥1 diopter, or (3) a decrease of TCT ≥30 µm within 12 months.

CXL was performed as accelerated epithelium-off CXL. After corneal abrasion, riboflavin solution (VibeX Rapid 0.1% solution; Avedro, Waltham, MA) was instilled every 2 minutes for 20 minutes. Irradiation was performed with the Avedro kxl system for 10 minutes with an irradiance of 9 mW/cm2 and a fluence of 5.4 J/cm2. During irradiation, riboflavin solution was instilled every 2 minutes. Postoperatively, the patients received a 17-mm bandage contact lens, artificial tears, and topical antibiotic eyedrops (Floxal EDO; Bausch & Lomb, New York) for the first week. From the second week on, they received nonpreservative dexamethasone eyedrops (Dexa Sine; Novartis, Basel, Switzerland) 6 times daily tapering off weekly by 1 while continuing topical treatment with artificial tears.

Patients were included if a Pentacam and a CST examination with a quality score “ok” after at least 3 days without wearing contact lenses were available preoperatively and postoperatively. Patients were excluded (1) if they underwent CXL because of other diagnoses than KC, (2) if they had previous eye surgery in their medical history, (3) if there was no follow-up examination after about 12 months, or (4) if the preoperative corneal thickness was lower than 400 µm.

Preoperative and postoperative Pentacam and CST examinations were exported. Main outcome parameters included the Pentacam-derived Kmax, the anterior radius of curvature (ARC, measured over a 3.0-mm zone centered on the thinnest point—the “A” parameter of the ABCD KC classification), TCT, BAD-D, and the CST-derived CBiF, the dynamic corneal response (DCR) parameters included therein and the resulting biomechanical E-staging. The CBiF is composed of (1) the Ambrósio relational thickness to the horizontal profile (ARTh), (2) the SP-A1, (3) the integrated radius, (4) the deformation amplitude ratio at 2 mm, and (5) the velocity of the corneal apex at inward applanation (A1 velocity).14

The values of the main outcome measure parameters were checked for normal distribution using the Shapiro-Wilk test assuming normal distribution with P>0.05. Preoperative and postoperative findings were subsequently compared using the 2-tailed paired t test (if normally distributed) or the Wilcoxon matched-pairs test (if not normally distributed) assuming significant differences with P<0.05.

In addition, (1) a Pearson correlation was performed between the mean corneal densitometry through the cornea measured in the central annulus of 0–2 mm and TCT, and (2) the biomechanical E-staging was analyzed in dependence of TCT at the different follow-up examinations: 2 groups were formed for each follow-up time point. One group comprised the eyes with postoperatively decreased TCT and the other included the eyes with an equal or increased TCT measurement. Then, the biomechanical E-staging was assessed separately for both groups to determine, whether postoperative TCT measurement might influence the biomechanical E-staging. Calculations were done using GraphPad Prism software (version 5.0, GraphPad Software; San Diego, CA).

RESULTS

This study is based upon 49 corneas of 41 patients older than 18 years of age (mean age: 31±11 y, 34 male patients, 7 female patients) who underwent accelerated epithelium-off CXL because of progressive KC. Preoperative measurements and postoperative measurements after 11.3±1.8 months were available for these patients. For a part of these patients, postoperative measurements were also available after other follow-up times, which were evaluated separately: 22 corneas of 21 patients were examined after 5.4±1.4 months and 21 corneas of 18 patients were additionally examined after 23.4±1.6 months.

The analysis revealed that the CBiF was significantly lower 5 months after CXL than preoperatively (P=0.0338, Table 1A) and comparable to the preoperative findings after 11 (Table 1B) and 23 months (Table 1C). The biomechanical E-staging that results out of the CBiF was significantly higher 5 months after CXL than preoperatively (P=0.035, Table 1A), differed not significantly after 11 months (Table 1B) and was comparable to preoperatively after 23 months (Table 1C). Parallel to this, the BAD-D as a tomographic measure of KC severity increased significantly 5 months after CXL (Table 1A) but was comparable to preoperative findings in further follow-up examinations (Table 1B, C).

TABLE 1 - Main Tomographic and Biomechanical Outcome Measures at Different Follow-Up Examinations
A Preoperatively Follow-up 5 mo P values
Corneas (patients), n (%) 22 (21) 22 (21)
Kmax 54.9±4.9 54.3±5.1 0.0831T
ARC 49.8±3.5 48.9±3.2 0.006 T *
TCT 468±31 454±36 0.0003 T *
Mean densitometry 0–2 mm 16.5±1.9 20.6±3.0 <0.0001 T *
BAD-D 7.8±3.1 8.3±3.3 0.0257 T *
ARTh 294±134 232±111 0.0003 W *
SP-A1 70.5±16 67.6±16 0.164T
IR 10.5±2.2 10.3±2.2 0.1112T
DA ratio 2 mm 5.3±0.8 5.2±0.8 0.1711W
A1 velocity 0.16±0.0 0.16±0.0 0.5205T
CBiF 5.1±0.5 5.0±0.5 0.0338 T *
E-staging 2.4±0.9 2.6±0.8 0.035 T *
B Preoperatively Follow-up 11 mo P values
Corneas (patients), n (%) 49 (41) 49 (41)
Kmax 56.9±6.3 56.2±6.6 0.0004 W *
ARC 51.4±4.9 50.8±5.6 0.0003 W *
TCT 470±34 459±35 <0.0001 T *
Mean densitometry 0–2 mm 16.3±1.9 18.6±2.5 <0.0001 T *
BAD-D 8.8±3.9 9.0±3.8 0.08W
ARTh 259±113 214±83 <0.0001 W *
SP-A1 67.6±16 69.1±19 0.3125T
IR 11.4±2.6 11.0±2.3 0.1415T
DA ratio 2 mm 5.5±1.1 5.6±1.2 0.656W
A1 velocity 0.17±0.0 0.17±0.0 0.6497W
CBiF 5.0±0.6 5.0±0.6 0.0683T
E-staging 2.6±0.9 2.7±1.0 0.0578T
C Preoperatively Follow-up 23 mo P values
Corneas (patients), n (%) 21 (18) 21 (18)
Kmax 55.4±5.2 54.0±5.2 0.001* T
ARC 50.0±3.8 49.0±3.7 0.0005* T
TCT 475±37 466±39 0.0285* T
Mean densitometry 0–2 mm 15.9±2.1 18.5±4.7 0.0083* W
BAD-D 7.8±3.3 7.9±3.0 0.4164T
ARTh 280±123 221±87 0.0002* W
SP-A1 71.1±15 75.5±17 0.0321* T
IR 10.9±2.2 10.3±2.1 0.2544T
DA ratio 2 mm 5.3±0.8 5.2±0.9 0.3737T
A1 velocity 0.15±0.2 0.15±0.2 0.9195T
CBiF 5.1±0.5 5.1±0.5 0.8828T
E-staging 2.4±0.9 2.4±0.8 0.9683T
A, 5 months postoperatively. B, 11 months postoperatively. C, 23 months postoperatively. ARC measured over a 3.0-mm zone centered on the thinnest point. Mean densitometry of 3 layers (anterior 120 µm, center layer, and posterior 60 µm of the cornea) measured in the central annulus of 0–2 mm. E-staging based on the CBiF. The DCR parameter SP-A1 was slightly higher than preoperatively after 11 months and significantly higher than preoperatively after 23 months (P=0.0321, C).
Bold and asterisks indicate significant difference between preoperative values and values at follow-up examination [calculated by (1) paired 2-tailed t testT if normally distributed, or by (2) Wilcoxon matched-pairs testW if not normally distributed—as determined by Shapiro-Wilk test].
A1 velocity indicates velocity at inward applanation; ARC, anterior radius of curvature; ARTh, Ambrósio relational thickness to the horizontal profile; BAD-D, Belin-Ambrosio deviation index; CBiF, Corvis Biomechanical Factor (linearized Corvis Biomechanical Index); DA ratio 2 mm, deformation amplitude ratio at 2 mm; IR, integrated radius; Kmax, maximal anterior keratometry; SP-A1, stiffness parameter A1; TCT, thinnest corneal thickness.

The TCT and the thickness-dependent DCR parameter ARTh were significantly lower at any follow-up examination than preoperatively (Table 1A–C). The Kmax was comparable to preoperatively 5 months after CXL (Table 1A), but significantly lower than preoperatively after 11 (P=0.0004, Table 1B) and 23 months (P=0.001, Table 1C). The ARC was significantly lower at any follow-up examination than preoperatively (Table 1). The mean densitometry of 3 layers (anterior 120 µm, center layer, and posterior 60 µm of the cornea) measured in the central annulus of 0–2 mm as a measure for corneal haze was also significantly higher than preoperatively at any follow-up examination (Table 1).

As shown in Table 2, the Pearson-correlation of mean densitometry through the cornea in the central annulus of 0–2 mm with TCT showed a significant correlation for postoperative values 5 and 23 months after CXL. The mean differences between preoperative and postoperative densitometry and TCT values at 5 and 11 months were also significantly correlated. A mean increase in densitometry values of 4.1 (5 mo) and 2.3 (11 mo after CXL) thus went along with a mean decrease of the TCT measurement of 13.8 µm (5 mo) and 10.8 µm (11 mo after CXL).

TABLE 2 - Pearson-Correlation (R2 and P Values) of Mean Densitometry of 3 Layers (Anterior 120 µm, Center Layer, and Posterior 60 µm of the Cornea) Measured in the Central Annulus of 0–2 mm With Thinnest Corneal Thickness (TCT) for the Group With 5 months Follow-Up, the Group With 11 Months Follow-Up, and the Group With 23 months Follow-Up
5 mo 11 mo 23 mo
Preop. R 2 0.0147 0.0519 0.1496
Preop. P 0.5908 0.1151 0.0833
Postop. R 2 0.2062 0.0322 0.2271
Postop. P 0.0337* 0.2171 0.0289*
ΔR 2 0.2408 0.1157 0.0687
ΔP 0.0204* 0.0168* 0.2510
Mean±SD ΔDensito. 4.1±3.6 2.3±3.2 2.6±4.9
Mean±SD ΔTCT −13.8±14.9 −10.8±16.6 −8.8±17.2
Preoperative (Preop.) R2 and P values, postoperative (Postop.) R2 and P values. Δ, R2, and P values for the difference between mean preoperative and postoperative densitometry (Densito) and TCT values. Mean difference (Δ) between preoperative and postoperative densitometry (Densito) and TCT values shown as mean±SD for the different follow-up time points.
Bold and asterisks indicate significant correlations.

The TCT-dependent analysis revealed that the majority of corneas showed a significantly lower TCT independent of the follow-up interval (Table 3).

TABLE 3 - Analysis of the Biomechanical E-staging (“E”) in Dependence of Thinnest Corneal Thickness (TCT)
n (corneas) TCT preop.|postop. P-value “E” preop.| postop. P-value
A. Postoperative TCT <Preoperative TCT
 5 mo: 16 of 21 469±31|450±36 <0.0001* T 2.3±1.0|2.6±0.9 0.0112 T *
 11 mo: 37 of 49 476±33|458±37 <0.0001* T 2.5±0.9|2.7±0.9 0.0581T
 23 mo: 16 of 21 481±34|465±34 <0.0001* T 2.3±1.0|2.4±0.8 0.3309T
B. Postoperative TCT ≥preoperative TCT
 n (corneas)
 5 mo: 5 of 21 466±33|469±32 0.0054* T 2.5±0.6|2.4±0.6 0.298T
 11 mo: 12 of 49 450±31|460±29 0.0008* T 2.7±1.1|2.7±1.3 0.6319T
 23 mo: 5 of 21 457±45|469±59 0.1726T 2.8±0.8|2.4±0.7 0.0993T
A, Postoperative decrease of TCT resulting in an increase of the E-staging. B, equal or higher postoperative TCT measurements resulting in equal or lower E-staging.
Bold and asterisks indicate significant difference between preoperative values and values at follow-up examination (calculated by paired 2-tailed t testT).

Postoperatively decreasing TCT values went along with an increasing E-staging (significant difference for 6 months postoperative follow-up, Table 3, Fig. 1).

F1
FIGURE 1:
Homburg biomechanical E-staging as provided by the Corvis ST (Oculus, Wetzlar, Germany). Red, baseline examination (October 22, 2019). Accelerated corneal cross-linking was performed on November 18, 2019. Green: first follow-up (June 15, 2020), yellow: second follow-up (September 14, 2020), each with increased E-stage compared with baseline. Blue: third follow-up (September 27, 2021) with decreased E-stage compared with baseline. Thinnest corneal thickness measurements: 477 µm|433 µm|447 µm|445 µm. Maximal anterior keratometry readings: 54.5D|52.5D|51.9D|51.5D. Decreased values for Ambrósio relational thickness to the horizontal profile after cross-linking (149|173|153) when compared with baseline (204). Increased value for stiffness parameter A1 at last follow-up (80) compared with baseline (70). ARTh indicates Ambrósio relational thickness to the horizontal profile; CBiF, Corvis Biomechanical Factor; DA, deformation amplitude; SP-A1, stiffness parameter A1; SSI, stress-strain-index; OD, oculus dexter (right eye).

An equal or higher postoperative TCT measurement, however, was associated with equal or lower results in the biomechanical E-staging (Table 3, Fig. 2).

F2
FIGURE 2:
Homburg biomechanical E-staging as provided by the Corvis ST (Oculus, Wetzlar, Germany). Red, baseline examination (February 19, 2019). Accelerated corneal cross-linking was performed on March 11, 2019. Green: first follow-up (January 16, 2020), yellow: second follow-up (July 28, 2021), each with decreased E-stage compared with baseline. Thinnest corneal thickness measurements: 444 µm|456 µm|444 µm. Maximal anterior keratometry readings: 59.1D|57.7D|56.3D. Slightly increased values for Ambrósio relational thickness to the horizontal profile after cross-linking (156|166) when compared with baseline (149). Increased value for stiffness parameter A1 (SP-A1) at last follow-up (65) compared with baseline (62). ARTh indicates Ambrósio relational thickness to the horizontal profile; CBiF, Corvis Biomechanical Factor; DA, deformation amplitude; SP-A1, stiffness parameter A1; SSI, stress-strain-index; OS, oculus sinister (left eye).

DISCUSSION

This study assessed the CBiF, the respective DCR parameters included therein and the resulting biomechanical E-staging for KC and other ectatic corneal diseases in KC corneas after accelerated epithelium-off CXL. It aimed to answer the question whether the biomechanical E-staging could be a suitable and intuitive standard CST parameter to biomechanically assess the stabilizing CXL effect in KC.

Many recent studies analyzed single CST parameters with the aim of biomechanically measuring the CXL effect on KC corneas. One study compared different CXL treatment protocols and found a significant increase of SP-A1 in 21 patients who underwent accelerated CXL (9 mW/cm2, as in this study).16 This finding was also observed in another study that assessed 67 corneas with mild KC 6 months after CXL.12 A third study, however, did not find a change of SP-A1 in 18 patients that were followed-up for up to 4 years after CXL according to the Dresden protocol.13 An increase in SP-A1 was also observed postoperatively in transepithelial accelerated CXL.17 An increasing corneal stiffness after CXL seems plausible, as CXL has been reported to result in an increase in collagen fiber diameter of about 12% in the anterior corneal stroma.18 The current study found comparable values preoperatively and postoperatively for SP-A1 within the first months after CXL, a slightly higher SP-A1 value after 11 months and a significantly higher SP-A1 value 23 months after CXL. The hypothesis of a biomechanical stabilization over several months after CXL could explain this result and also the seemingly contradictory results regarding an increase or decrease of SP-A1 after CXL in other studies.

Another study identified the difference between the first and second applanation length to discriminate cross-linked from non–cross-linked KC and healthy corneas.19 It was also hypothesized that parameters related to the second applanation might be more sensitive to detect biomechanical changes after CXL.18 These studies have in common that they are based on single CST parameters.

The advantage of the CBiF is that it is based on the standard CST parameter CBI and thus combines several single DCR parameters. It serves as a basis for the biomechanical E-staging. Increasing KC severity is characterized by an increasing E-staging and a decreasing CBiF value, which is because of the definition of the CBiF as described previously.15 The CBiF decreased significantly from preoperatively to 5 months postoperatively, and consequently, the biomechanical E-staging increased significantly during this period (Table 1). At later follow-ups, however, this difference first became smaller (11 mo, P=0.0683), and finally disappeared completely (23 mo, P=0.8828, Table 1). In parallel, there was an increase of tomographic severity (indicated by BAD-D, P=0.0257, Table 1) 5 months after CXL with subsequent stabilization and similar results compared with preoperatively at further follow-up examinations. One could hypothesize that this could indicate that the cornea seems to tomographically and biomechanically approximate the preoperative status 1–2 years after CXL.

The thickness-based parameters TCT and ARTh were the only parameters that were significantly smaller than preoperatively at any postoperative follow-up time (Table 1). Lower ARTh values indicate a centrally thinner cornea with a fast thickness increase toward the corneal periphery.10 Because it both has been reported that TCT may be measured lower after CXL than preoperatively20 and that there may be a real decrease in TCT after CXL,21 lower postoperative ARTh values could have been expected after CXL—resulting out of postoperatively decreasing corneal thickness and flattening of the corneal apex.

A significantly lower TCT measurement was observed for the majority of corneas at every follow-up time point when compared with preoperative values (Table 1). As the CBI and thus also the CBiF and E-staging depend on corneal thickness because they include ARTh, the dependence of the postoperative E-staging on postoperative TCT was investigated in a further step. Corneas with postoperatively lower TCT measurement were characterized by a higher E-staging than preoperatively, whereas corneas with postoperatively equal or higher TCT measurements showed an equal or lower E-staging postoperatively (Table 3). The postoperatively lower TCT measurement after CXL is in line with reports in literature. One study reported a decrease in corneal thickness measurements 1, 3, and 6 months after CXL that approximated baseline measurements after 12 months.22 Another study reported a pseudoprogression after CXL going along with lower TCT measurements 6 weeks and 6 months after CXL.23 It has been reported that this postoperative corneal thickness reduction is related to a stromal hyperdensity resulting in an underestimation of the real corneal thickness by Scheimpflug cameras20 within the first 6–12 months after CXL24 thus leading to a TCT-measurement artefact.

Although KC is histopathologically characterized by changes in epithelial maps,25 breaks in Bowman layer, and an overall corneal thinning,26 keratocytes have been reported to disappear from the anterior and intermediate stroma as a sequence of apoptosis and photonecrosis within the first months after CXL.27 In addition, transient collagen modifications can become visible in form of a hyperdensity (“haze”) of the anterior to midstroma after CXL, which has been reported to decrease from 6 to 12 months postoperatively and to generally disappear at 12 months after CXL.27 This may result in a postoperative pseudoprogression23 with a worsening of uncorrected and best-corrected visual acuity as reported also by the Siena CXL study20 and the Siena Eye-Cross Study 2.28 In line with these reports, our current study found a weak, yet significant correlation between corneal haze (as measured by corneal densitometry) and a reduced TCT measurement at 5 and 11 months after CXL, which disappeared at 23 months after CXL (Table 2). This indicates that the postoperative stromal hyperdensity after CXL did influence TCT measurements.

In turn, the E-staging after CXL may be influenced by TCT measurements with increasing postoperative E-values when TCT is measured lower than preoperatively (Fig. 1). Consequently, the postoperative increase of the E-staging in cases with lower postoperative TCT measurements can be interpreted as another artefact depending on the artefact of TCT underestimation by Scheimpflug cameras. This hypothesis is supported by the result of a stabilization or even decrease of the E-staging when TCT does not change, or is even measured higher postoperatively than preoperatively (Fig. 2). Thus, the biomechanical E-staging cannot be used as a valid parameter to assess stabilization after CXL unless corneal transparency, light scattering and, thus, TCT measurements return to baseline values.

The ARC and Kmax values help distinguish postoperatively increasing E-values from a real progression: as both were lower than preoperatively at every postoperative follow-up time, they indicate the stabilizing CXL effect early on, as evidenced later on by the postoperatively higher SP-A1 value after 11 months that reached statistical significance after 23 months in our study (Table 1). The clinician might prefer the ARC value for this purpose because it is not a single point parameter, but measured over a 3.0-mm zone centered at the thinnest corneal point.25

Finally, there was no significant difference between the preoperative and postoperative E-staging value for the entire study population at 11 and 23 months after CXL (Table 1B, C, Fig. 1). This indicates that the biomechanical E-staging can be used for post-CXL assessment of the biomechanics in KC in the long term, but not within the first year.

The CST can display up to 3 follow-up examinations simultaneously for 1 eye based on a baseline examination within the Homburg biomechanical E-staging card (Figs. 1, 2). The CST provides an assessment of whether the cornea is stiffer, not significantly different, or softer compared with the baseline examination (check marks in Figs. 1, 2) for each of 6 biomechanical parameters [deformation amplitude ratio 2 mm, integrated radius, ARTh, SP-A1, stress-strain-index, and the biomechanical E-staging]. In addition to the tomographic assessment, this enables a biomechanical progression assessment, which can be helpful especially in tomographically borderline or unclear situations. However, this raises the question whether the tomographic or the biomechanical progression assessment can detect progression earlier—this has to be answered in larger studies.

Limitations of this study are (1) the moderate sample size of patients, (2) the dropout rate during the follow-up examinations at 5 months and 23 months postoperatively, and (3) the inclusion of both eyes per patient in 8 cases. Furthermore, the TCT-dependent analysis is based on a very limited number of subjects with stable or increased TCT measurements, so that these results have to be verified by larger-scale studies.

CONCLUSIONS

The biomechanical E-staging for KC and other ectatic corneal diseases, and thereby, the CBiF, are suitable and intuitive parameters that enable a biomechanical KC severity assessment. Although documenting the stabilizing CXL effect on the long term, the E-staging depends on TCT measurements what limits its use after CXL, because of a reported TCT underestimation by Scheimpflug cameras within the first postoperative months. After about 1 year, there was no significant difference between the E-staging preoperatively and postoperatively, which indicates that the human cornea seems to approximate the preoperative biomechanical status between 1 and 2 years after CXL.

Acknowledgments

The authors thank all colleagues within the Department of Ophthalmology at the Saarland University Medical Center in Homburg, who recruited Keratoconus patients to the Homburg Keratoconus Center.

REFERENCES

1. Seitz B, Daas L, Hamon L, et al. Stage-appropriate treatment of keratoconus. Ophthalmologe. 2021;118:1069–1088.
2. Meyer JJ, Gokul A, Vellara HR, et al. Progression of keratoconus in children and adolescents. Br J Ophthalmol. 2021. Online ahead of print. doi:10.1136/bjophthalmol-2020-316481
3. Ferdi A, Nguyen V, Kandel H, et al. Predictors of progression in untreated keratoconus: a Save Sight Keratoconus Registry study. Br J Ophthalmol. 2022;106:1206–1211.
4. Shajari M, Steinwender G, Herrmann K, et al. Evaluation of keratoconus progression. Br J Ophthalmol. 2019;103:551–557.
5. Belin MW, Duncan J, Ambrósio R, et al. A new tomographic method of staging/classifying keratoconus: the ABCD grading system. Int J Kerat Ect Cor Dis. 2015;4:85–93.
6. Belin MW, Duncan JK. Keratoconus: the ABCD Grading System. Klin Monbl Augenheilkd. 2016;233:701–707.
7. Belin MW, Meyer JJ, Duncan JK, et al. Assessing progression of keratoconus and cross-linking efficacy: the Belin ABCD Progression Display. Int J Kerat Ect Cor Dis. 2017;6:1–10.
8. Flockerzi E, Xanthopoulou K, Goebels SC, et al. Keratoconus staging by decades: a baseline ABCD classification of 1000 patients in the Homburg Keratoconus Center. Br J Ophthalmol. 2021;105:1069–1075.
9. Ambrósio R, Lopes BT, Faria-Correia F, et al. Integration of Scheimpflug-based corneal tomography and biomechanical assessments for enhancing ectasia detection. J Refract Surg. 2017;33:434–443.
10. Vinciguerra R, Ambrósio R, Elsheikh A, et al. Detection of keratoconus with a new biomechanical index. J Refract Surg. 2016;32:803–810.
11. Vinciguerra R, Ambrósio R, Roberts CJ, et al. Biomechanical characterization of subclinical keratoconus without topographic or tomographic abnormalities. J Refract Surg. 2017;33:399–407.
12. Jabbarvand M, Moravvej Z, Shahraki K, et al. Corneal biomechanical outcome of collagen cross-linking in keratoconic patients evaluated by Corvis ST. Eur J Ophthalmol. 2021;31:1577–1583.
13. Sedaghat M-R, Momeni-Moghaddam H, Ambrósio R, et al. Long-term evaluation of corneal biomechanical properties after corneal cross-linking for keratoconus: a 4-year longitudinal study. J Refract Surg. 2018;34:849–856.
14. Flockerzi E, Vinciguerra R, Belin MW, et al. Correlation of the corvis biomechanical factor CBiF with tomographic parameters in keratoconus. J Cataract Refract Surg. 2022;48:215–221.
15. Flockerzi E, Vinciguerra R, Belin MW, et al. Combined biomechanical and tomographic keratoconus staging: adding a biomechanical parameter to the ABCD keratoconus staging system. Acta Ophthalmol. 2022;100:e1135–e1142.
16. Hashemi H, Ambrósio R, Vinciguerra R, et al. Two-year changes in corneal stiffness parameters after accelerated corneal cross-linking. J Biomech. 2019;93:209–212.
17. Jian W, Tian M, Zhang X, et al. One-year follow-up of corneal biomechanical changes after accelerated transepithelial corneal cross-linking in pediatric patients with progressive keratoconus. Front Med. 2021;8:663494.
18. Salouti R, Khalili MR, Zamani M, et al. Assessment of the changes in corneal biomechanical properties after collagen cross-linking in patients with keratoconus. J Curr Ophthalmol. 2019;31:262–267.
19. Fuchsluger TA, Brettl S, Geerling G, et al. Biomechanical assessment of healthy and keratoconic corneas (with/without crosslinking) using dynamic ultrahigh-speed Scheimpflug technology and the relevance of the parameter (A1L−A2L. Br J Ophthalmol. 2019;103:558–564.
20. Caporossi A, Mazzotta C, Baiocchi S, et al. Long-term results of riboflavin ultraviolet a corneal collagen cross-linking for keratoconus in Italy: the Siena Eye Cross Study. Am J Ophthalmol. 2010;149:585–593.
21. Vounotrypidis E, Athanasiou A, Kortüm K, et al. Long-term database analysis of conventional and accelerated crosslinked keratoconic mid-European eyes. Graefes Arch Clin Exp Ophthalmol. 2018;256:1165–1172.
22. Greenstein SA, Shah VP, Fry KL, et al. Corneal thickness changes after corneal collagen crosslinking for keratoconus and corneal ectasia: one-year results. J Cataract Refract Surg. 2011;37:691–700.
23. Xanthopoulou K, Milioti G, Daas L, et al. Accelerated corneal crosslinking causes pseudoprogression in keratoconus within the first 6 weeks without affecting posterior corneal curvature. Eur J Ophthalmol. 2022;32:2565–2576.
24. Greenstein SA, Fry KL, Bhatt J, et al. Natural history of corneal haze after collagen crosslinking for keratoconus and corneal ectasia: Scheimpflug and biomicroscopic analysis. J Cataract Refract Surg. 2010;36:2105–2114.
25. Reinstein DZ, Archer TJ, Urs R, et al. Detection of keratoconus in clinically and algorithmically topographically normal fellow eyes using epithelial thickness analysis. J Refract Surg. 2015;31:736–744.
26. Müller PL, Loeffler KU, Messmer E, et al. Histological corneal alterations in keratoconus after crosslinking—expansion of findings. Cornea. 2020;39:333–341.
27. Mazzotta C, Hafezi F, Kymionis G, et al. In vivo confocal microscopy after corneal collagen crosslinking. Ocul Surf. 2015;13:298–314.
28. Mazzotta C, Raiskup F, Hafezi F, et al. Long term results of accelerated 9 mW corneal crosslinking for early progressive keratoconus: the Siena Eye-Cross Study 2. Eye and Vis. 2021;8:16.
Keywords:

biomechanical E-staging; biomechanics; CBI; CBiF; Corvis; cross-linking; keratoconus

Copyright © 2022 Asia-Pacific Academy of Ophthalmology. Published by Wolters Kluwer Health, Inc. on behalf of the Asia-Pacific Academy of Ophthalmology.