Advertisement

Predicting the Prognosis of Fuchs Endothelial Corneal Dystrophy by Using Scheimpflug Tomography

Published:September 27, 2019DOI:https://doi.org/10.1016/j.ophtha.2019.09.033

      Purpose

      To determine if Scheimpflug tomography pachymetry map and posterior elevation map patterns, central corneal thickness (CCT), and corneal backscatter can predict the prognosis of Fuchs endothelial corneal dystrophy (FECD).

      Design

      Cross-sectional study with follow-up of outcomes.

      Participants

      Ninety-six eyes (56 subjects) with a range of severity of FECD.

      Methods

      Corneas were graded by cornea specialists according to the area and confluence of guttae and the presence of clinically definite edema. Masked and randomized Scheimpflug imaging pachymetry map and posterior elevation map patterns were assessed by 1 observer for loss of regular isopachs, displacement of the thinnest point of the cornea, and the presence of posterior surface depression. The prognosis of eyes over a 5-year (median) follow-up period was determined based on FECD progression (new onset of clinically definite edema or ≥5% increase in CCT) or intervention by endothelial keratoplasty. Cumulative probabilities of progression and intervention were estimated from survival analyses, with risk factors determined by using Cox proportional hazards models.

      Main Outcome Measures

      Pachymetry map and posterior elevation map patterns, corneal backscatter, and CCT (ultrasonic pachymetry).

      Results

      In univariate analyses, loss of regular isopachs (hazard ratio [HR], 18.00) displacement of the thinnest point (HR, 11.53), focal posterior surface depression (HR, 10.21), and anterior corneal backscatter (HR, 1.22, per 1-grayscale unit increment), were risk factors for progression or intervention (P < 0.001), whereas CCT (HR, 1.30, per 25-μm increment) was not (P = 0.15). In multivariate analyses, loss of regular isopachs (HR, 11.57; P < 0.001) and displacement of the thinnest point (HR, 5.61; P = 0.02) were independent and clinically important risk factors for progression and intervention. The 5-year cumulative risk of disease progression and intervention was 7%, 48%, and 89% when none, 1 or 2, and all 3 pachymetry map and posterior elevation map parameters were present, respectively (P <0.001). The 4-year cumulative risk of disease progression and intervention after uncomplicated cataract surgery was 0%, 50%, and 75% when none, 1 or 2, and all 3 pachymetry map and posterior elevation map parameters were present, respectively (P < 0.001).

      Conclusions

      Three Scheimpflug tomography pachymetry map and posterior elevation map patterns can predict FECD prognosis independent of CCT. The risk of FECD progression and intervention, including after uncomplicated cataract surgery, increases according to the number of parameters present.

      Abbreviations and Acronyms:

      CCT (central corneal thickness), CI (confidence interval), EK (endothelial keratoplasty), FECD (Fuchs endothelial corneal dystrophy), HR (hazard ratio)
      Fuchs endothelial corneal dystrophy (FECD) affects approximately 5% of the United States population and can lead to corneal edema and poor vision. Progressive corneal thickening develops gradually with a period of subclinical edema (which we define as edema that is not clinically obvious on slit-lamp biomicroscopy) that can cause symptoms of glare and subjective lack of clarity in vision (i.e., subclinical edema is not necessarily asymptomatic).
      • Sun S.Y.
      • Wacker K.
      • Baratz K.H.
      • Patel S.V.
      Determining subclinical edema in Fuchs endothelial corneal dystrophy. Revised classification using Scheimpflug tomography for preoperative assessment.
      • Repp D.J.
      • Hodge D.O.
      • Baratz K.H.
      • et al.
      Fuchs’ endothelial corneal dystrophy. Subjective grading versus objective grading based on the central-to-peripheral thickness ratio.
      • van der Meulen I.J.
      • Patel S.V.
      • Lapid-Gortzak R.
      • et al.
      Quality of vision in patients with Fuchs endothelial dystrophy and after Descemet stripping endothelial keratoplasty.
      Although the diagnosis of FECD is easy to make clinically by detecting the presence of guttae by slit-lamp examination, detecting the presence of subclinical corneal edema is more challenging. The latter is important in general ophthalmology because these patients can demonstrate symptoms related to subclinical corneal edema, and they may need counseling regarding the risk of FECD progression, including in the setting of cataract surgery.
      Currently, there are no simple indicators of disease prognosis for FECD in clinical practice, and prognosis is frequently based on subjective clinical judgment. Methods have been proposed for determining whether to perform cataract surgery alone or in combination with keratoplasty in the setting of FECD, including assessing absolute central corneal thickness (CCT)
      • Seitzman G.D.
      • Gottsch J.D.
      • Stark W.J.
      Cataract surgery in patients with Fuchs’ corneal dystrophy: expanding recommendations for cataract surgery without simultaneous keratoplasty.
      ,
      • van Cleynenbreugel H.
      • Remeijer L.
      • Hillenaar T.
      Cataract surgery in patients with Fuchs’ endothelial corneal dystrophy: when to consider a triple procedure.
      and endothelial cell density,
      American Academy of Ophthalmology
      Cataract in the adult eye Preferred Practice Pattern.
      but these measurements are misleading for clinical decision making.
      • Sun S.Y.
      • Wacker K.
      • Baratz K.H.
      • Patel S.V.
      Determining subclinical edema in Fuchs endothelial corneal dystrophy. Revised classification using Scheimpflug tomography for preoperative assessment.
      ,
      • Patel S.V.
      Towards clinical trials in Fuchs endothelial corneal dystrophy: classification and outcome measures—the Bowman Club Lecture 2019.
      An increase in CCT over time can indicate disease progression but needs to be considered in the context of corneal thickness before disease onset. A cutoff of corneal backscatter measured by confocal microscopy has also been suggested to help with surgical decision making,
      • van Cleynenbreugel H.
      • Remeijer L.
      • Hillenaar T.
      Cataract surgery in patients with Fuchs’ endothelial corneal dystrophy: when to consider a triple procedure.
      although this method is difficult to perform in clinical practice.
      Recently, we described a Scheimpflug imaging classification of FECD that categorizes the disease into having clinically obvious edema (edema visible by slit-lamp examination with thickening of the stroma, Descemet or deep stromal folds, stromal clouding determined by sclerotic scatter, microcystic epithelial edema or bedewing, or subepithelial bullae), subclinical corneal edema (based on the presence of specific tomographic features; see below), or no edema (absence of the specific tomographic features of edema), independent of CCT.
      • Sun S.Y.
      • Wacker K.
      • Baratz K.H.
      • Patel S.V.
      Determining subclinical edema in Fuchs endothelial corneal dystrophy. Revised classification using Scheimpflug tomography for preoperative assessment.
      We found that 3 tomographic pachymetry map and posterior elevation map patterns—specifically the presence of irregular isopachs, displacement of the thinnest point of the cornea, and focal posterior surface depression—were present in FECD with subclinical corneal edema. The classification recommended that the diagnosis of FECD be made clinically with slit-lamp biomicroscopy and categorized further by using Scheimpflug tomography when corneal edema was not clinically obvious. In the present study, we determined Scheimpflug tomographic pachymetry and posterior elevation map patterns in a cohort of patients with FECD and reviewed their outcomes over a median follow-up period of 5 years to determine their prognosis. We included patients with a range of severity of FECD from having few scattered guttae to clinically obvious corneal edema. In addition, we assessed the usefulness of CCT measured by ultrasound and nonstandardized corneal backscatter measurements for determining the prognosis of the disease.

      Methods

       Participants

      Participants with a range of severity of FECD were recruited from the cornea service at Mayo Clinic (Rochester, MN) between July 2012 and December 2014. The diagnosis of FECD was made by cornea specialists (S.V.P., K.H.B.) based on the presence of guttae, and disease severity was graded according to the distribution of guttae and the presence of clinically obvious edema. Fuchs endothelial corneal dystrophy was graded by using a modified Krachmer scale.
      • Repp D.J.
      • Hodge D.O.
      • Baratz K.H.
      • et al.
      Fuchs’ endothelial corneal dystrophy. Subjective grading versus objective grading based on the central-to-peripheral thickness ratio.
      ,
      • Louttit M.D.
      • Kopplin L.J.
      • Igo Jr., R.P.
      • et al.
      A multicenter study to map genes for Fuchs endothelial corneal dystrophy: baseline characteristics and heritability.
      Exclusion criteria were the presence of previous or current corneal pathologic features except FECD, prior corneal surgery, and prior intraocular surgery with the exception of uncomplicated phacoemulsification with endocapsular intraocular lens placement (eyes undergoing cataract surgery within 1 month of enrollment were excluded). Although we did not exclude eyes with clinically obvious corneal edema to determine if our imaging classification could predict the outcome in these eyes (i.e., positive controls), we analyzed the results with and without including this group of eyes. If eligible, both eyes of each participant were included. The study was approved by the Mayo Clinic Institutional Review Board and adhered to the tenets of the Declaration of Helsinki; informed consent was obtained from all participants. Some data from this cohort, related to corneal backscatter measured by using confocal microscopy, have been reported previously.
      • Amin S.R.
      • Baratz K.H.
      • McLaren J.W.
      • Patel S.V.
      Corneal abnormalities early in the course of Fuchs’ endothelial dystrophy.

       Scheimpflug Imaging and Tomography Evaluation

      All participants underwent Scheimpflug imaging (Pentacam HR; Oculus, Lynnwood, WA) as described previously.
      • Wacker K.
      • McLaren J.W.
      • Amin S.R.
      • et al.
      Corneal high-order aberrations and backscatter in Fuchs’ endothelial corneal dystrophy.
      Images were acquired at any time during routine clinic hours (7:30am to 4:30pm), with 75% of all images acquired before 1:00pm. The “4 Maps Refractive” display of enrolled eyes derived from the instrument’s software (Pentacam version 1.21r43) was exported as a high-resolution image, as described previously
      • Sun S.Y.
      • Wacker K.
      • Baratz K.H.
      • Patel S.V.
      Determining subclinical edema in Fuchs endothelial corneal dystrophy. Revised classification using Scheimpflug tomography for preoperative assessment.
      ; only images meeting the instrument’s software quality criteria were included. De-identified images were presented in random order to 1 masked observer (S.V.P.) for assessment of the pachymetry map, for loss of regular isopachs and displacement of the thinnest point of the cornea, and the posterior elevation map for posterior surface depression (the elevation map indicates elevation or depression, i.e., negative elevation, of the corneal surface relative to a best-fit sphere; Fig 1).
      • Sun S.Y.
      • Wacker K.
      • Baratz K.H.
      • Patel S.V.
      Determining subclinical edema in Fuchs endothelial corneal dystrophy. Revised classification using Scheimpflug tomography for preoperative assessment.
      Loss of regular isopachs was defined as any single line of equal thickness not being almost circular or oval or not being parallel to adjacent isopachs within the central 4 mm of the cornea. Displacement of the thinnest point of the cornea was defined as being located outside of the inferotemporal quadrant,
      • Liu Z.
      • Huang A.J.
      • Pflugfelder S.C.
      Evaluation of corneal thickness and topography in normal eyes using the Orbscan corneal topography system.
      centered at the pupil center, or more than 1 mm from the pupil center in any quadrant. Focal posterior depression, that is, protrusion of the posterior corneal surface toward the anterior chamber, was defined as any isolated area of negative elevation relative to the best fit sphere within the central 4 mm of the cornea. Each feature was determined to be either present or absent.
      Figure thumbnail gr1
      Figure 1Scheimpflug tomography pachymetry (left column) and posterior elevation (right column) maps of the same eye with Fuchs endothelial corneal dystrophy in 2012 (top row) and 2019 (bottom row). Pachymetry maps were evaluated for loss of parallel or almost circular isopachs and for displacement of the thinnest point (small circle) from the inferotemporal quadrant relative to the pupil center (+). Posterior elevation maps were evaluated for negative elevation, that is, depression, relative to the best-fit sphere. In 2012, tomography showed subtle evidence of subclinical edema with gradual progression of tomographic findings through 2019. Note the same location of irregular isopachs and posterior depression at both examinations. Clinically obvious corneal edema was not present by slit-lamp examination at any stage, and central corneal thickness (CCT) increased only slightly between examinations (from 545 to 560 μm). After Descemet membrane endothelial keratoplasty (not shown) in 2019, CCT decreased to 476 μm and visual acuity improved to 20/20 from 20/40. OD = right eye.
      Nonstandardized (i.e., unadjusted) corneal backscatter measurements (termed densitometry by the manufacturer of the instrument) from the central 2-mm diameter area of the cornea were exported directly by using the instrument’s software. Backscatter measurements were available for the anterior 120 μm of the cornea, mid cornea, and posterior 60 μm of the cornea. Although we have previously advocated for standardizing corneal backscatter measurements to account for any variation in the instrument’s light source intensity or camera sensitivity between examinations,
      • McLaren J.W.
      • Bourne W.M.
      • Patel S.V.
      Standardization of corneal haze measurement in confocal microscopy.
      • McLaren J.W.
      • Wacker K.
      • Kane K.M.
      • Patel S.V.
      Measuring corneal haze by using Scheimpflug photography and confocal microscopy.
      • Patel S.V.
      • Winter E.J.
      • McLaren J.W.
      • Bourne W.M.
      Objective measurement of backscattered light from the anterior and posterior cornea in vivo.
      we deliberately did not adjust backscatter measurements in this study to mimic a simplified assessment of backscatter (i.e., one that can be made relatively easily in clinical practice) that could be implemented if found to be of value.

       Clinical Outcomes

      The prognosis of enrolled eyes was determined by review of medical records from the time of Scheimpflug image acquisition to last follow-up. The time between Scheimpflug image acquisition and the onset of any of 3 specific outcomes was used to assess prognosis. The first outcome was a decision or recommendation to intervene with endothelial keratoplasty (EK), either because of the presence of definite corneal edema or because of vision symptoms attributed to subclinical edema. The second outcome was progression of FECD determined by the new onset of clinically definite corneal edema. The third outcome was progression of FECD determined by an increase in CCT of 5% or more (sustained over at least 2 consecutive examinations on different days or subsequently associated with clinically definite edema) measured by using ultrasonic pachymetry (Pachette 2; DGH Technology, Exton, PA) compared with that obtained at the enrollment study visit.

       Statistical Analysis

      Demographic and enrollment data were summarized descriptively. Prognosis (progression or intervention) was estimated using the Kaplan-Meier (survival analysis) method, and prognostic risk factors for these end points were assessed by using univariate and multivariate Cox proportional hazards models. All estimates from the Cox models were evaluated using sandwich estimators of the standard errors to account for including fellow eyes of the same subject. Optimal cutoff values for CCT and anterior corneal backscatter were determined statistically based on the sample. Analyses were performed for all available eyes (n = 96) and were also repeated for eyes without clinically obvious corneal edema (n = 81), that is, FECD grade 6 eyes were excluded because the prognosis of these eyes may be deemed obvious from clinical examination. Analyses accounted for the timing of any cataract surgery before or after Scheimpflug imaging. A 2-tailed probability of less than 5% was considered statistically significant.

      Results

      Ninety-six eyes of 56 participants were enrolled in the study (Table 1); age at enrollment was 68±15 years (mean ± standard deviation), and 37 participants (66%) were women. Fifteen eyes showed clinically obvious corneal edema at enrollment (and were designated as grade 6). Median follow-up was 60 months (interquartile range, 45–72 months). Of the 16 fellow eyes not enrolled in this study, 9 had undergone previous EK, 4 had undergone cataract surgery within the previous month, and 3 were not included by patient choice (because eyes underwent more testing than just Scheimpflug imaging
      • Amin S.R.
      • Baratz K.H.
      • McLaren J.W.
      • Patel S.V.
      Corneal abnormalities early in the course of Fuchs’ endothelial dystrophy.
      ).
      Table 1Characteristics of Eyes at Enrollment
      Fuchs Endothelial Corneal Dystrophy Grade
      1–23–456
      Eyes (no.)30341715
      Median age, yrs (range)73 (53–87)63 (42–89)63 (42–81)69 (45–89)
      Phakic, no. (%)22 (73)27 (79)13 (76)14 (93)
      Loss of parallel isopachs, no. (%)9 (30)12 (35)12 (71)15 (100)
      Displacement of the thinnest point, no. (%)6 (20)9 (26)14 (82)15 (100)
      Focal posterior depression, no. (%)10 (33)11 (32)13 (76)15 (100)
      No. of tomographic features
       019 (63)21 (62)2 (12)0 (0)
       1 or 27 (23)6 (18)5 (29)0 (0)
       34 (13)7 (21)10 (59)15 (100)
      Central corneal thickness (μm), mean ± SD (range)560±27 (522–615)573±35 (499–639)571±43 (492–636)618±33 (519–660)
      Corneal backscatter (grayscale), mean ± SD (range)
       Anterior 120 μm27.9±2.7 (23.2–34.2)28.2±2.6 (23.6–34.9)32.2±5.3 (23.7–42.5)35.6±5.7 (27.6–46.7)
       Mid cornea17.3±1.7 (14.4–21.1)16.9±1.5 (14.9–22.4)19.6±3.2 (15.4–27.0)21.5±2.0 (18.3–24.5)
       Posterior 60 μm12.0±2.3 (8.6–18.7)12.5±2.2 (9.1–17.8)16.2±4.1 (11.2–27.6)19.6±3.9 (11.9–25.6)
      SD = standard deviation.
      Fuchs endothelial corneal dystrophy grades: ≤12 scattered guttae, grade 1; >12 scattered guttae, grade 2; confluent guttae with widest diameter ≤2 mm, grade 3; confluent guttae with widest diameter 2–5 mm, grade 4; confluent guttae with widest diameter >5 mm, grade 5; presence of clinically definite corneal edema, grade 6.
      During the follow-up period, intervention by EK was not recommended and progression did not occur in 54 eyes. Of these 54 eyes, 12 were pseudophakic at enrollment (cataract surgery occurred at a median of 32 months before enrollment), 20 underwent uncomplicated cataract surgery (at a median of 13 months after enrollment), and 22 remained phakic throughout the follow-up period. For the 20 eyes that underwent cataract surgery after enrollment, the median follow-up after cataract surgery was 41 months.
      Intervention by EK was recommended or progression was noted in the remaining 42 eyes; 23 eyes showed existing or new onset of clinically obvious edema, 14 eyes showed a 5% or more increase in CCT, and 5 were recommended to undergo EK because of vision symptoms. The cumulative probability of disease progression or intervention at 5 years was 49% (95% confidence interval [CI], 36–59%) for the group overall. Of the 42 eyes that underwent intervention or progression, 8 eyes were pseudophakic at enrollment (cataract surgery occurred at a median of 29 months before enrollment), 7 eyes underwent uncomplicated cataract surgery without EK (at a median of 13 months after enrollment), 23 eyes underwent combined cataract surgery with EK (at a median of 6 months after enrollment), and 4 remained phakic throughout the follow-up period. For the 7 eyes that underwent cataract surgery without EK after enrollment, all subsequently underwent EK or progression with a median interval of 18 months after cataract surgery.

       Pachymetry Map and Elevation Map Patterns

      In univariate analyses, all 3 tomographic pachymetry map and posterior elevation map patterns at enrollment were associated significantly with disease progression or intervention, regardless of whether grade 6 eyes were included (Table 2; Fig 2). Of the 15 eyes with grade 6 FECD at baseline, all showed loss of regular isopachs, displacement of the thinnest point of the cornea, and focal posterior depression present (Table 1) at enrollment. When excluding grade 6 eyes, the cumulative risk of disease progression or intervention over 5 years was 82% (95% CI, 50%–93%) for loss of regular isopachs, 78% (95% CI, 55%–89%) for displacement of the thinnest point, and 77% (95% CI, 51%–89%) for focal posterior depression. The cumulative risk of disease progression or intervention over 5 years increased according to the number of parameters present, from 7% (95% CI, 0%–16%), when none of the pachymetry map and posterior elevation map features were present, to 48% (95% CI, 9%–70%), when any 1 or 2 of the features were present, to 89% (95% CI, 60%–97%), when all 3 features were present (P < 0.001, grade 6 eyes excluded; Fig 3).
      Table 2Univariate Cox Proportional Hazard Ratios for Risk Factors for Fuchs Endothelial Corneal Dystrophy Progression or Intervention
      All Eyes (n = 96)Grade 6 Eyes Excluded (n = 81)
      Hazard Ratio (95% Confidence Interval)P ValueHazard Ratio (95% Confidence Interval)P Value
      Pachymetry map and posterior elevation map patterns
       Loss of regular isopachs23.07 (8.39–63.40)<0.00118.00 (6.32–51.26)<0.001
       Displacement of the thinnest point of the cornea16.10 (6.33–40.90)<0.00111.53 (4.35–30.53)<0.001
       Focal posterior depression14.07 (5.61–35.24)<0.00110.21 (3.85–27.08)<0.001
      Corneal backscatter (per 1 grayscale unit-increment)
       Anterior 120 μm of cornea1.22 (1.16–1.28)<0.0011.22 (1.13–1.31)<0.001
       Mid cornea1.30 (1.16–1.47)<0.0011.25 (1.12–1.39)<0.001
       Posterior 60 μm of cornea1.20 (1.09–1.33)<0.0011.16 (1.06–1.27)<0.001
      Central corneal thickness (per 25 μm-increment)1.69 (1.22–2.36)0.0021.30 (0.91–1.87)0.149
      By excluding grade 6 eyes with clinically definite corneal edema by slit-lamp examination, analysis was confined to Fuchs endothelial corneal dystrophy eyes with subclinical or no edema.
      Figure thumbnail gr2
      Figure 2Graphs showing cumulative probability of progression or intervention for Fuchs endothelial corneal dystrophy based on (A) loss of regular isopachs, (B) displacement of the thinnest point of the cornea, and (C) presence of focal posterior depression. Eyes with visible corneal edema (grade 6) were excluded (n = 81). Univariate hazard ratios (HRs) are shown and were determined from Cox models.
      Figure thumbnail gr3
      Figure 3Graphs showing cumulative probability of progression or intervention for Fuchs endothelial corneal dystrophy (based on the number of pachymetry map and posterior elevation map patterns present for (A) all eyes (n = 96) and (B) after excluding (grade 6) eyes with visible corneal edema (n = 81). The cumulative risk of disease progression or intervention over 5 years was 7%, 48%, and 94% (latter 89% if grade 6 excluded) when none, 1 or 2, and all 3 pachymetry map and posterior elevation map parameters were present, respectively (P < 0.001).
      In a multivariate analysis with grade 6 eyes excluded, irregular isopachs (hazard ratio [HR], 8.66; 95% CI, 2.65–28.32; P < 0.001) and displacement of the thinnest point (HR, 4.19; 95% CI, 1.13–15.48; P = 0.03) were independent risk factors for progression or intervention, whereas focal posterior depression was not (HR, 0.95; 95% CI, 0.37–2.45; P = 0.92). However, when excluding irregular isopachs from the multivariate model, focal posterior depression was an independent risk factor for progression or intervention (HR, 3.93; 95% CI, 1.20–12.85; P = 0.02).

       Corneal Thickness and Backscatter

      In univariate analyses, anterior, mid, and posterior backscatter were all associated with disease progression or intervention (Table 2); CCT was associated with disease progression or intervention when grade 6 eyes were included, but not when they were excluded (Table 2). In a multivariate analysis of all eyes that included CCT and all backscatter parameters, only anterior corneal backscatter was an independent risk factor for disease progression or intervention when grade 6 eyes were included (HR, 1.18; 95% CI, 1.07–1.29; P < 0.001) or excluded (HR, 1.30; 95% CI, 1.09–1.54; P = 0.003).
      Fuchs endothelial corneal dystrophy progression or intervention was estimated according to sample-specific optimal cutoff values for CCT and anterior corneal backscatter (Fig S1, available at www.aaojournal.org). After excluding grade 6 eyes, the cumulative risk of disease progression or intervention at 5 years was 54% (95% CI, 32%–69%) for eyes with CCT equal to or more than the sample-specific optimal cutoff and 25% (95% CI, 9%–38%) for CCT less than the sample-specific optimal cutoff (HR, 2.36; 95% CI, 1.06–5.25; P = 0.036). After excluding grade 6 eyes, the cumulative risk of disease progression or intervention at 5 years was 85% (95% CI, 56%–95%) when anterior corneal backscatter was equal to or more than the sample-specific optimal cutoff and 20% (95% CI, 8%–31%) when less than the sample-specific optimal cutoff (HR, 7.33; 95% CI, 3.34–16.07; P < 0.001).

       Combined Analyses

      In multivariate analyses that combined pachymetry map and posterior elevation map patterns with backscatter or CCT, anterior corneal backscatter and CCT were independent, but weak, risk factors for disease progression or intervention compared with loss of regular isopachs and displacement of the thinnest point (Table 3). Similar results were found regardless of whether or not grade 6 eyes were included (Table 3).
      Table 3Multivariate Analyses Combining Pachymetry Map and Posterior Elevation Map Patterns with Central Corneal Thickness or Corneal Backscatter
      All Eyes (n = 96)Grade 6 Eyes Excluded (n = 81)
      Hazard Ratio (95% Confidence Interval)P ValueHazard Ratio (95% Confidence Interval)P Value
      Multivariate model with CCT
       Loss of regular isopachs12.49 (4.07–38.33)<0.00111.57 (3.44–38.88)<0.001
       Displacement of the thinnest point of the cornea6.60 (1.88–23.13)0.0035.61 (1.33–23.66)0.02
       Focal posterior depression
      Focal posterior depression was not an independent risk factor in these multivariate models because it was highly correlated with loss of regular isopachs; when considered in a model without loss of regular isopachs, focal posterior depression was a significant risk factor, and therefore should not be discounted. CCT = central corneal thickness.
      0.56 (0.20–1.60)0.280.53 (0.14–1.95)0.34
       CCT (per 25 μm-increment)1.62 (1.30–2.01)<0.0011.43 (1.13–1.80)0.003
      Multivariate model with corneal backscatter
       Loss of regular isopachs10.25 (3.98–26.40)<0.00111.31 (3.44–37.12)<0.001
       Displacement of the thinnest point of the cornea3.95 (1.24–12.58)0.023.40 (0.86–13.40)0.08
       Focal posterior depression
      Focal posterior depression was not an independent risk factor in these multivariate models because it was highly correlated with loss of regular isopachs; when considered in a model without loss of regular isopachs, focal posterior depression was a significant risk factor, and therefore should not be discounted. CCT = central corneal thickness.
      0.71 (0.32–1.55)0.390.53 (0.17–1.59)0.26
       Anterior corneal backscatter (per 1 grayscale unit-increment)1.14 (1.07–1.21)<0.0011.14 (1.04–1.25)0.004
      Focal posterior depression was not an independent risk factor in these multivariate models because it was highly correlated with loss of regular isopachs; when considered in a model without loss of regular isopachs, focal posterior depression was a significant risk factor, and therefore should not be discounted. CCT = central corneal thickness.

       Prognosis after Cataract Surgery

      Twenty-seven eyes underwent uncomplicated cataract surgery after enrollment in the study. The 4-year cumulative risk of disease progression or intervention after uncomplicated cataract surgery was 0%, 50% (95% CI, 0%–78%), and 75% (95% CI, 17%–93%), when none, 1 or 2, and all 3 pachymetry map and posterior elevation map parameters were present, respectively (P < 0.001; Fig 4).
      Figure thumbnail gr4
      Figure 4Graph showing cumulative probability of progression or intervention for Fuchs endothelial corneal dystrophy after uncomplicated cataract surgery (n = 27). The cumulative risk of disease progression or intervention over 4 years was 0%, 50%, and 75% when none, 1 or 2, and all 3 pachymetry map and posterior elevation map parameters were present, respectively (P < 0.001).

      Discussion

      Assessment of 3 pachymetry map and posterior elevation map patterns derived from Scheimpflug tomography of corneas with FECD predicted disease prognosis independent of CCT. The cumulative probability of disease progression or intervention over 5 years increased from 7% when none of the patterns were present to 89% when all were present. The number of parameters present also predicted FECD prognosis after uncomplicated cataract surgery. The findings are important for counseling patients about their prognosis and can also help to identify patients at risk of progression for enrollment in clinical trials.
      We recently described a revised classification of FECD that combined clinical examination and Scheimpflug imaging.
      • Sun S.Y.
      • Wacker K.
      • Baratz K.H.
      • Patel S.V.
      Determining subclinical edema in Fuchs endothelial corneal dystrophy. Revised classification using Scheimpflug tomography for preoperative assessment.
      Fuchs endothelial corneal dystrophy can be classified as having clinically definite corneal edema, subclinical edema, or no edema based on the presence of 3 pachymetry map and posterior elevation map patterns. In this study, we determined the presence of these specific patterns that signify clinical or subclinical corneal edema in a cohort of eyes with a range of severity of FECD and determined their outcomes over a median follow-up period of 5 years. In univariate analyses, loss of regular isopachs, displacement of the thinnest point of the cornea, and focal posterior depression were all associated with a higher risk of disease progression or intervention regardless of whether eyes with clinically definite corneal edema were included. A multivariate analysis of all 3 parameters and cataract surgery (if performed) found that only loss of regular isopachs and displacement of the thinnest point were independent risk factors for progression, conferring an 8- and 4-fold increased risk of progression, respectively. Because the pachymetry map is derived from anterior and posterior elevation data, it is not surprising that focal posterior depression was not an independent risk factor; after excluding irregular isopachs from the multivariate model, focal posterior depression was in fact a significant risk factor, indicating the correlation between the 2 risk factors. Therefore, we still recommend interpreting pachymetry and posterior elevation map maps together because the patterns in each map should complement each other, that is, the region of posterior depression should correspond to the region of thickening of the cornea with irregular isopachs (Fig 1),
      • Sun S.Y.
      • Wacker K.
      • Baratz K.H.
      • Patel S.V.
      Determining subclinical edema in Fuchs endothelial corneal dystrophy. Revised classification using Scheimpflug tomography for preoperative assessment.
      and because the cumulative probability of disease progression or intervention over 5 years increased according to the number of these parameters that were present (7% when none were present, 48% when any 1 or 2 of the patterns were present, and 89% when all 3 were present).
      Different cutoff values for CCT have been proposed previously for when to consider cataract surgery combined with keratoplasty.
      • Seitzman G.D.
      • Gottsch J.D.
      • Stark W.J.
      Cataract surgery in patients with Fuchs’ corneal dystrophy: expanding recommendations for cataract surgery without simultaneous keratoplasty.
      ,
      • van Cleynenbreugel H.
      • Remeijer L.
      • Hillenaar T.
      Cataract surgery in patients with Fuchs’ endothelial corneal dystrophy: when to consider a triple procedure.
      We have recommended not basing such treatment decisions on CCT and to assess pachymetry map and posterior elevation map patterns instead.
      • Sun S.Y.
      • Wacker K.
      • Baratz K.H.
      • Patel S.V.
      Determining subclinical edema in Fuchs endothelial corneal dystrophy. Revised classification using Scheimpflug tomography for preoperative assessment.
      In this study, we found that the risk of disease progression or intervention over 4 years after uncomplicated cataract surgery increased according to the number of pachymetry map and posterior elevation map patterns present (0% when none were present, 50% when any 1 or 2 of the patterns were present, and 75% when all 3 were present), similar to the group overall and supporting our previous recommendation. In terms of prognosis for the overall group, CCT was a weak predictor of prognosis relative to the pachymetry map and posterior elevation map parameters (Table 3), especially when clinically definite edema was not visible (1 in 4 eyes with CCT less than the sample-specific optimal cutoff progressed or needed intervention). The present study further supports that an isolated measurement of CCT is not a helpful parameter for assessing FECD.
      Corneal backscatter increases in FECD
      • Amin S.R.
      • Baratz K.H.
      • McLaren J.W.
      • Patel S.V.
      Corneal abnormalities early in the course of Fuchs’ endothelial dystrophy.
      ,
      • Wacker K.
      • McLaren J.W.
      • Amin S.R.
      • et al.
      Corneal high-order aberrations and backscatter in Fuchs’ endothelial corneal dystrophy.
      because of the presence of corneal edema and also because of structural changes in response to chronic edema
      • Patel S.V.
      • McLaren J.W.
      In vivo confocal microscopy of Fuchs endothelial dystrophy before and after endothelial keratoplasty.
      ; it also correlates with, but is poorly predictive of, corneal endothelial function.
      • Wacker K.
      • McLaren J.W.
      • Kane K.M.
      • et al.
      Corneal hydration control in Fuchs’ endothelial corneal dystrophy.
      In these previous studies, we standardized corneal backscatter measurements to eliminate any confounding effect introduced by variations in the light source and camera of the instrument. Standardizing backscatter measurements requires imaging a fixed scatter source (standard) before every examination followed by adjustment of corneal measurements relative to the image intensity of the standard (which should remain the same over time).
      • McLaren J.W.
      • Bourne W.M.
      • Patel S.V.
      Standardization of corneal haze measurement in confocal microscopy.
      • McLaren J.W.
      • Wacker K.
      • Kane K.M.
      • Patel S.V.
      Measuring corneal haze by using Scheimpflug photography and confocal microscopy.
      • Patel S.V.
      • Winter E.J.
      • McLaren J.W.
      • Bourne W.M.
      Objective measurement of backscattered light from the anterior and posterior cornea in vivo.
      Unfortunately, this standardization step is not built into the instrument and therefore requires additional effort and external manipulation of the data, which is neither quick nor simple for routine clinical practice. In this study, we deliberately did not use standardized backscatter measurements to determine whether raw backscatter data from the instrument, which is easily available for clinical practice (and frequently but incorrectly used in investigational studies
      • Ni Dhubhghaill S.
      • Rozema J.J.
      • Jongenelen S.
      • et al.
      Normative values for corneal densitometry analysis by Scheimpflug optical assessment.
      • Droutsas K.
      • Lazaridis A.
      • Giallouros E.
      • et al.
      Scheimpflug densitometry after DMEK versus DSAEK: two-year outcomes.
      • Schaub F.
      • Enders P.
      • Bluhm C.
      • et al.
      Two-year course of corneal densitometry after Descemet membrane endothelial keratoplasty.
      • Chu H.Y.
      • Hsiao C.H.
      • Chen P.Y.
      • et al.
      Corneal backscatters as an objective index for assessing Fuchs’ endothelial corneal dystrophy: a pilot study.
      • Alnawaiseh M.
      • Zumhagen L.
      • Wirths G.
      • et al.
      Corneal densitometry, central corneal thickness, and corneal central-to-peripheral thickness ratio in patients with Fuchs endothelial dystrophy.
      ) may assist in predicting disease progression. We found that nonstandardized anterior corneal backscatter was an independent, but weak, predictor of FECD progression or intervention in multivariate analyses. However, with the HR being small (and very close to 1, Table 3) relative to those for loss of regular isopachs and displacement of the thinnest point and the difficulty in standardizing measurements, anterior corneal backscatter adds very little additional clinical value for predicting FECD prognosis.
      The strengths of this study include the standardized manner in which these eyes were evaluated by 2 cornea surgeons (S.V.P., K.H.B.) involved in programmatic research of FECD. In addition, median follow-up for all eyes was 5 years, and three quarters of all eyes had 45 months or more of follow-up, which helped to provide meaningful prognostic information. Limitations of this study include the relatively small sample size; although this resulted in relatively wide CIs for some analyses, the lower limit of the CIs for loss of regular isopachs and displacement of the thinnest point of the cornea were still of high clinical importance. Only 27 eyes underwent cataract surgery (without concurrent EK) after enrollment, and although the presence of abnormal pachymetry map and posterior elevation map patterns significantly predicted disease progression or intervention over 4 years, a larger series is needed to quantify this risk after cataract surgery more precisely. In addition, Scheimpflug images and ultrasonic pachymetry were not acquired at a standardized time of day, given that corneal edema is usually worse in the morning; nevertheless, our data were representative of typical clinical practice when patients with FECD may be seen at any time of day. Finally, it is unknown whether the mildest cases of subclinical edema can be detected by tomography or if tomography is helpful for clinical decision making if vision is affected by guttae in the absence of edema. However, we are unaware of carefully designed studies that show a benefit of EK when vision is assumed to be affected by guttae in the absence of edema; this deserves further investigation and could require revision of our FECD classification.
      The findings of this study support using the revised classification of FECD, which is a more functional classification compared with morphologic grading of guttae distribution. This classification is simple to use in clinical practice, requiring slit-lamp confirmation of FECD and exclusion of other corneal pathologic features followed by Scheimpflug imaging to assess for subclinical edema.
      • Sun S.Y.
      • Wacker K.
      • Baratz K.H.
      • Patel S.V.
      Determining subclinical edema in Fuchs endothelial corneal dystrophy. Revised classification using Scheimpflug tomography for preoperative assessment.
      Pachymetry map and posterior elevation map interpretation is familiar to many ophthalmologists when assessing for corneal ectasia, and thus the method can be implemented easily in many practices (Fig 1) and does not require external data manipulation (unlike assessing corneal backscatter). The presence of even 1 or 2 of the pachymetry map and posterior elevation map features increased the probability of FECD progression or intervention and suggested that subclinical edema is already present. We strongly recommend assessing for these pachymetry map and posterior elevation map patterns in FECD when patients report visual symptoms that may be attributable to the disease or when patients may require cataract surgery. This method is of higher clinical relevance than simply measuring CCT, especially when a patient has not been evaluated previously for FECD and when corneal edema is not clinically obvious. In addition, as novel treatments for FECD are being investigated,
      • Wacker K.
      • Baratz K.H.
      • Fautsch M.P.
      • Patel S.V.
      Medical and semi-surgical treatments for Fuchs endothelial corneal dystrophy.
      ,
      • Garcerant D.
      • Hirnschall N.
      • Toalster N.
      • et al.
      Descemet’s stripping without endothelial keratoplasty.
      evaluating pachymetry map and posterior elevation map patterns should be considered for future interventional clinical trials because this method can help to define the disease state and identify eyes at risk of progression.
      • Patel S.V.
      Towards clinical trials in Fuchs endothelial corneal dystrophy: classification and outcome measures—the Bowman Club Lecture 2019.

      Supplementary Data

      References

        • Sun S.Y.
        • Wacker K.
        • Baratz K.H.
        • Patel S.V.
        Determining subclinical edema in Fuchs endothelial corneal dystrophy. Revised classification using Scheimpflug tomography for preoperative assessment.
        Ophthalmology. 2019; 126: 195-204
        • Repp D.J.
        • Hodge D.O.
        • Baratz K.H.
        • et al.
        Fuchs’ endothelial corneal dystrophy. Subjective grading versus objective grading based on the central-to-peripheral thickness ratio.
        Ophthalmology. 2013; 120: 687-694
        • van der Meulen I.J.
        • Patel S.V.
        • Lapid-Gortzak R.
        • et al.
        Quality of vision in patients with Fuchs endothelial dystrophy and after Descemet stripping endothelial keratoplasty.
        Arch Ophthalmol. 2011; 129: 1537-1542
        • Seitzman G.D.
        • Gottsch J.D.
        • Stark W.J.
        Cataract surgery in patients with Fuchs’ corneal dystrophy: expanding recommendations for cataract surgery without simultaneous keratoplasty.
        Ophthalmology. 2005; 112: 441-446
        • van Cleynenbreugel H.
        • Remeijer L.
        • Hillenaar T.
        Cataract surgery in patients with Fuchs’ endothelial corneal dystrophy: when to consider a triple procedure.
        Ophthalmology. 2014; 121: 445-453
        • American Academy of Ophthalmology
        Cataract in the adult eye Preferred Practice Pattern.
        American Academy of Ophthalmology, San Francisco2016
        • Patel S.V.
        Towards clinical trials in Fuchs endothelial corneal dystrophy: classification and outcome measures—the Bowman Club Lecture 2019.
        BMJ Open Ophthalmol. 2019 Jul 11; 4e000321https://doi.org/10.1136/bmjophth-2019-000321. eCollection 2019
        • Louttit M.D.
        • Kopplin L.J.
        • Igo Jr., R.P.
        • et al.
        A multicenter study to map genes for Fuchs endothelial corneal dystrophy: baseline characteristics and heritability.
        Cornea. 2012; 31: 26-35
        • Amin S.R.
        • Baratz K.H.
        • McLaren J.W.
        • Patel S.V.
        Corneal abnormalities early in the course of Fuchs’ endothelial dystrophy.
        Ophthalmology. 2014; 121: 2325-2333
        • Wacker K.
        • McLaren J.W.
        • Amin S.R.
        • et al.
        Corneal high-order aberrations and backscatter in Fuchs’ endothelial corneal dystrophy.
        Ophthalmology. 2015; 122: 1645-1652
        • Liu Z.
        • Huang A.J.
        • Pflugfelder S.C.
        Evaluation of corneal thickness and topography in normal eyes using the Orbscan corneal topography system.
        Br J Ophthalmol. 1999; 83: 774-778
        • McLaren J.W.
        • Bourne W.M.
        • Patel S.V.
        Standardization of corneal haze measurement in confocal microscopy.
        Invest Ophthalmol Vis Sci. 2010; 51: 5610-5616
        • McLaren J.W.
        • Wacker K.
        • Kane K.M.
        • Patel S.V.
        Measuring corneal haze by using Scheimpflug photography and confocal microscopy.
        Invest Ophthalmol Vis Sci. 2016; 57: 227-235
        • Patel S.V.
        • Winter E.J.
        • McLaren J.W.
        • Bourne W.M.
        Objective measurement of backscattered light from the anterior and posterior cornea in vivo.
        Invest Ophthalmol Vis Sci. 2007; 48: 166-172
        • Patel S.V.
        • McLaren J.W.
        In vivo confocal microscopy of Fuchs endothelial dystrophy before and after endothelial keratoplasty.
        JAMA Ophthalmol. 2013; 131: 611-618
        • Wacker K.
        • McLaren J.W.
        • Kane K.M.
        • et al.
        Corneal hydration control in Fuchs’ endothelial corneal dystrophy.
        Invest Ophthalmol Vis Sci. 2016; 57: 5060-5065
        • Ni Dhubhghaill S.
        • Rozema J.J.
        • Jongenelen S.
        • et al.
        Normative values for corneal densitometry analysis by Scheimpflug optical assessment.
        Invest Ophthalmol Vis Sci. 2014; 55: 162-168
        • Droutsas K.
        • Lazaridis A.
        • Giallouros E.
        • et al.
        Scheimpflug densitometry after DMEK versus DSAEK: two-year outcomes.
        Cornea. 2018; 37: 455-461
        • Schaub F.
        • Enders P.
        • Bluhm C.
        • et al.
        Two-year course of corneal densitometry after Descemet membrane endothelial keratoplasty.
        Am J Ophthalmol. 2017; 175: 60-67
        • Chu H.Y.
        • Hsiao C.H.
        • Chen P.Y.
        • et al.
        Corneal backscatters as an objective index for assessing Fuchs’ endothelial corneal dystrophy: a pilot study.
        J Ophthalmol. 2017; 2017: 8747013https://doi.org/10.1155/2017/8747013
        • Alnawaiseh M.
        • Zumhagen L.
        • Wirths G.
        • et al.
        Corneal densitometry, central corneal thickness, and corneal central-to-peripheral thickness ratio in patients with Fuchs endothelial dystrophy.
        Cornea. 2016; 35: 358-362
        • Wacker K.
        • Baratz K.H.
        • Fautsch M.P.
        • Patel S.V.
        Medical and semi-surgical treatments for Fuchs endothelial corneal dystrophy.
        Klin Monbl Augenheilkd. 2018; 235: 709-713
        • Garcerant D.
        • Hirnschall N.
        • Toalster N.
        • et al.
        Descemet’s stripping without endothelial keratoplasty.
        Curr Opin Ophthalmol. 2019; 30: 275-285

      Linked Article