Iridocorneal endothelial (ICE) syndrome which includes Chandler's syndrome, Progressive Iris Atrophy (PIA), and Cogan-Reese/Iris Nevus Syndrome (INS) manifests clinically with varying combinations of corneal edema, iris atrophy, peripheral anterior synechiae (PAS), and secondary glaucoma. Differential diagnosis includes multiple other ocular conditions presenting with one of more of these findings e.g. Fuchs’ endothelial dystrophy (FED), Posterior Polymorphous Dystrophy (PPD), Reiger's syndrome and iridoschisis.
Endothelial abnormalities have been reported to be a consistent feature across the varied manifestations of ICE syndrome. PAS, iris atrophy/nodules and glaucoma are believed to occur secondary to acquisition of epithelial-like structural and proliferative properties by endothelial cells which then migrate across the anterior surface of the iris into the angle. Clinically the “hammered silver” or “beaten metal” appearance of the posterior corneal surface on specular reflection, as classically described initially for Chandlers Syndrome, and later on also for PIA and INS, is however not diagnostic and may also be seen with the corneal guttata present in FED.
Ultra-structural assessment of the endothelium by in vivo techniques like specular microscopy and in vivo confocal microscopy (IVCM), can thus play a pivotal in confirming the clinical diagnosis. Vis a vis specular microscopy, IVCM has the advantage of superior resolution and less deterioration of image quality in the presence of corneal edema or mild scarring. Abnormal endothelial cells initially labeled “ICE cells” by Sherrard et al., have subsequently been identified on both specular and confocal microscopy by various groups.
Shield et al., have reported severity of corneal involvement and presence of secondary glaucoma to be prognostic factors in cases of ICE syndrome. Similarly, Laganowski et al., documented the utility of specular microscopic appearance of the posterior cornea in predicting likelihood of glaucoma development. In contrast, Liu and colleagues, found that specular microscopy did not reliably predict the prognosis with neither ICE grading nor endothelial cell density (ECD) correlating with corneal edema or intraocular pressure (IOP). In the background of conflicting reports in literature, the spectrum of micro structural endothelial alterations as visualized by in vivo confocal microscopy (IVCM) and their clinical implications in an Indian cohort are reported in this study.
This retrospective, observational study was carried out at a tertiary care referral center in North India. It adhered to tenets of the Declaration of Helsinki and institutional ethics committee approval was obtained. Records of patients diagnosed with ICE syndrome between January 2012 and April 2017 on the basis of clinical examination and endothelial imaging by IVCM, were reviewed. Clinical diagnosis of ICE syndrome was based on the presence of at least any 2 of the following 3 main features on slit lamp biomicroscopy and gonioscopy i.e. (a) typical iris changes of holes, corectopia, atrophy or ectropion uvea, (b) PAS, and (c) abnormal corneal endothelium on specular reflection [Fig. 1a-h]. IVCM had been performed bilaterally, using the Rostock Cornea Module of the Heidelberg Retinal Tomograph (HRT) III (Heidelberg Engineering, GmBH, Dossenheim, Germany) which uses a 670 nm helium neon diode laser beam to scan the cornea in a raster pattern and achieves magnification levels up to 800 times, with axial and lateral resolutions as low as 4 μm and 1-2 μm, respectively. As a routine practice, in order to facilitate acquisition of a larger number of scans in a shorter duration, the ‘volume scan’ option is chosen with the corneal apex being applanated first followed by the superior, nasal, inferior and temporal quadrants. On IVCM, presence of ‘epitheloid like endothelial cells similar to those described previously in literature, for ICE syndrome i.e. presence of prominent hyper reflective nuclei and loss of regularity of cellular shape and size in the clinically involved eye had been considered diagnostic of ICE syndrome in the patient records.
For this study 42 eyes of 21 patients were included, though records of endothelial evaluation on IVCM was available only for 41 eyes i.e. bilaterally in 20 patients and unilaterally in one patient, the other eye having significant corneal edema and hence precluding adequate visualization of the endothelial cells by the confocal microscope. Patient demographics (age and gender), anterior and posterior segment findings on slit lamp biomicroscopy and gonioscopy, presence/absence of corneal edema, intraocular pressure, severity of glaucoma, treatment received (medical or surgical), and follow-up duration were noted for each patient from the medical charts. Corneal edema was graded clinically on slit lamp biomicroscopy as mild (increased corneal thickness as compared to contralateral eye without presence of stromal striae/Descemets membrane folds and clearly visible iris pattern), moderate (increased corneal thickness, presence of stromal striae or Descemet's membrane folds with/without microcystic epithelial edema and visible iris pattern, albeit with some loss of finer details) and severe (increased corneal thickness, presence of bullae/subepithelial fibrosis, complete inability to visualize iris details or anterior segment). Severity of glaucoma was classified as: Grade I (mild)- IOP ≤ 21 mm Hg on topical anti glaucoma medication only; Grade II (moderate)-maintaining IOP ≤ 21 mm Hg required systemic anti glaucoma drugs and/or surgery despite being on maximal tolerable topical medication; and Grade III (severe)- the patient required > 3 topical anti glaucoma and/or systemic medication after initial surgery or needed resurgery for maintaining IOP ≤ 21 mm Hg.
Endothelial images were evaluated in accordance with the specular microscopic classification proposed by Sherrard et al., and labeled as “disseminated ICE,” i.e. ICE cells scattered individually or in small clusters amongst an otherwise normal endothelial mosaic, “subtotal ICE” when 25–75% of the endothelial surface was occupied by the abnormal ICE cells and “total ICE” where the entire endothelium was replaced by abnormal cells. Though Sherrard et al., categorized only cases of subtotal ICE into the “plus”(+) or “minus” (−) variants depending on whether the ICE cells were smaller or larger respectively than the apparently normal endothelium in the same eyes, due to the majority of patients having total ICE in our cohort, we compared ICE cell size/area of the affected eye with the contralateral normal eye. Due to unavailability of a more sophisticated software which could numerically quantify cellular dimensions, this contralateral comparison was primarily made on the basis of visual inspection, aided by a caliper tool (equivalent to 50 μm) available as an overlay on the IVCM images. Thus abnormal ICE cells were labeled as “minus (−) variant” if on visual inspection they appeared larger than the endothelial cells of the contra lateral unaffected eye with widely spaced nuclei present predominantly in the center of the cells [Fig. 2a] and as “plus (+) variant” if they appeared smaller with more eccentrically located yet closely packed nuclei, in the endothelial mosaic [Fig. 2b]. ECD was evaluated for both eyes using a semi-automated software provided with the confocal microscope, wherein a rectangular area identifying the region of interest was plotted, followed by manually marking all endothelial cells completely within rectangle as well as those not completely within the rectangle but touching the left and lower borders of the rectangle. The software then computed the endothelial cell count along with the standard deviation. Only frames where at least 50 cells could be marked were selected for calculating ECD to ensure a reliable cell count. Association of the cell variant i.e. ICE + or ICE – cells with the patient demographics (age and gender), presence of corneal edema and glaucoma severity was also noted.
Analysis was conducted using IBM SPSS statistics (version 22.0, Chicago, IL, USA). The normality of quantitative data was checked by measures of Kolmogorov-Smirnov test of normality. Continuous data was expressed in the form of its mean, standard deviation (SD) and range. Gender was compared using the chi square test. The Mann-Whitney test was applied for comparison of 2 groups. All the statistical tests were two-sided and were performed at a significance level of α = 0.05. A P value of < 0.05 was considered significant.
Mean age of the study cohort was 44.24 ± 15.8 years (range 8-79 years). A significant male preponderance was noted with 13 males (62%) and 8 females (38%). On endothelial evaluation with IVCM, the most prominent feature was the presence of rounded or kidney bean shaped prominent hyper reflective nuclei [Fig. 3a and b]. This was associated in varying combinations with other changes suggestive of an epitheloid/epithelial-like transformation of the endothelial cells e.g. doubling of nuclei [Fig. 3c], presence of light dark reversal [Fig. 3d] and diversity in cellular size and shape. IVCM features of “ICE” like cells were visualized in 22 of the 41 eyes examined with IVCM i.e. unilaterally in 20 patients and bilaterally in one patient. The demographic profile, relevant clinical features and IVCM findings, management and follow-up details of all patients included in the study have been outlined in Table 1.
Nineteen patients (90.5%) had unilateral involvement both clinically and on IVCM. Two patients had clinical features suggestive of bilateral and asymmetric ICE. One patient reported previously had bilateral but asymmetric involvement with presence of PAS, iris atrophy, corneal edema, and glaucoma with total ICE—pattern on IVCM in one eye, while the contralateral eye had a normal anterior segment and angle, a “hammered silver” appearance of the endothelium on slit lamp biomicroscopy and disseminated ICE cells on IVCM with an otherwise well preserved endothelial mosaic. In one patient clinical and IVCM features were consistent with the diagnosis of ICE (patches of iris atropy and peripheral iridocorneal adhesions on slit lamp biomicroscopy and large cells with widely spaced centrally placed nuclei, hence labelled ICE–cell variant on IVCM) while the contralateral eye had significant corneal edema with subepithelial fibrosis not allowing endothelial images to be captured by IVCM.
Total ICE pattern was seen in 20 eyes (91%) while 1 eye each (4.5% each) had subtotal ICE (–variant) and disseminated ICE tissue. The ICE – variant was seen in 14 of 20 eyes (70%) having total ICE, while the ICE + variant was seen in the remaining 6 (30% eyes). Twelve out of 15 patients (80%) with ICE minus cells (14 eyes with total ICE - and 1 eye with subtotal ICE -), were males while five of the 6 (83.3%) patients with ICE + endothelial cells were female. Mean age of patients with ICE – and ICE + endothelial cell morphology was 45.8 ± 17.8 years (range, 8–79 years) and 40.3 ± 9.2 years (range, 26–53 years) respectively (P = 0.02). Of the 15 eyes having the ICE- type of the endothelial cells, 4 eyes each (26.7% each) were associated with grade 2 and 3 glaucoma, 6 eyes (40%) had grade 1 glaucoma while 1 eye (6.7%) had no glaucoma. Amongst the 6 eyes classified as having ICE + plus cells, Grade 3 and Grade 2 glaucoma was seen in 50% (n = 3 eyes) and 33.3% (n = 2 eyes) eyes respectively while 1 eye had no evidence of raised IOP/glaucoma. Trabeculectomy with MMC was the primary procedure in eleven eyes while 3 eyes underwent glaucoma drainage device implantation in primary sitting. Due to presence of significant peripheral anterior synechiae and iris being plastered to cornea, the GDD tube was inserted into sulcus behind the iris.
The average ECD of the 22 affected eyes in whom the endothelium could be imaged with IVCM was 1446 ± 653 cells/mm2 (range, 519–2532 cells/mm2) while in unaffected eyes (n = 19), it was 2628 ± 239 cells/mm2 (range, 2201–3062 cells/mm2) (P < 0.001). Corneal edema was present in 8 eyes of which 7 could be imaged. Mean ECD of these 7 eyes with corneal edema was 1369 ± 683 cells/mm2 (range, 680–2252 cells/mm2) which was similar to the 15 affected eyes without corneal edema where the mean ECD was 1470 ± 595 cells/mm2 (range, 519 cells/mm2–2514 cells/mm2) (P = 0.42). Though the difference in mean ECD of the ICE – (1118 ± 489 cells/mm2; n = 15) and ICE+ (2088 ± 284 cells/mm2; n = 6) eyes was statistically significant (P = 0.003) the incidence of corneal edema was similar, being seen in 2 of 6 eyes (33.3%) eye with ICE + morphology and 5 of 15 eyes (33.3%) with ICE – morphology.
The view expressed by Shields, for ICE syndrome that “the typical patient is a woman” was supported by results of series published by Hirst et al., (all 17 patients in their series being female), Liu and colleagues. (12 females, 3 males) and Le et al., (10 females, 2 males). Sherrard et al., in their series of 57 eyes with ICE syndrome however, noted a lack of sex discrimination by the disease with a male: female ratio 47%:53%. Similarly, Laganowski et al., in their series of 66 ICE patients had 31 males (47%) and 35 (53%) females. In the present series we noted an apparent male preponderance (61.9%, n = 13 patients), which may represent either a true variation from the western populations due to racial and genetic differences or could perhaps be attributable to differences in health seeking behavior of males versus females in India, with the former accessing health care services more frequently. A sampling bias may also be responsible for the apparent male preponderance seen in this small retrospective study.
The predominantly unilateral occurrence of clinically detectable signs, is usually considered as another well-established feature of ICE syndrome, and often used to differentiate it from PPD which is frequently bilateral. A few cases of clinically manifest bilateral ICE syndrome have also been reported with either the same variant occurring in both eyes, or two different forms presenting simultaneously. Published literature thus seems to suggest that ICE may occasionally be a bilateral disease with asymmetric presentation rather than a purely unilateral pathology. This is also reflected in the occurrence of bilateral involvement in 2 of our patients who had a varying spectrum of disease severity in the contralateral fellow eye.
Based on the morphological appearance of the endothelial cells seen on IVCM, several important differences were noted in our series as compared to that reported from western literature. While 91% of our patients (20 of 22 eyes) had total ICE, with only one eye each (4.5% each) having disseminated ICE and subtotal ICE, previous studies, have documented significantly higher occurrence of the subtotal variant (50–58% cases) as compared to total ICE (30–36%). The ICE – cells were predominant in our cohort, occurring in 70% (n = 14) of the 20 eyes with total ICE. Liu et al., in another Asian population, documented ICE- endothelial morphology in 5 eyes and the ICE + variant in 3 eyes with subtotal ICE. Conversely, both Sherrard et al. and Laganowski et al. had a reverse ICE- (30%, n = 10 eyes) to ICE + (70%, n = 23 eyes) ratio albeit in eyes with subtotal ICE. Though the number of eyes in both the present series from the Indian subcontinent and that of Liu et al., from Taiwan is relatively small, the differences in the extent (total versus subtotal) and morphology of abnormal cells (ICE– versus ICE + variant) amongst Asian and Caucasian populations, may have clinical implications and merits further evaluation. The ICE– variant was commoner in males (80%) and the ICE + variant commoner in females (83.3%) in the present series. To the best of the authors’ knowledge, this gender specific predilection for the morphological variants of ICE cells has not been reported previously in published literature.
Overall 86.4% (n = 19) of the 22 eyes with endothelial abnormalities included in the study, had evidence of raised IOP/glaucoma which was comparable to the 76.7% incidence reported from an Asian (Thai) population, but much higher than the 40%–50% incidence reported from Caucasian populations. While Teekhasaenee et al., did not evaluate endothelial morphology in their series of ICE eyes, a subgroup analysis of the series by Sherrard et al. and Laganowski et al. revealed that the incidence of glaucoma was much higher in patients with total ICE, ranging from 71% to 75%, as compared to eyes with subtotal ICE where glaucoma was noted in only 17–20% eyes. The overall higher incidence of glaucoma in the present cohort may thus be a reflection of the predominance of total ICE pattern which was seen in our patients.
Grupcheva et al. using IVCM found less uniform cellular organization and greater multilayering of the endothelium in patients with ICE – type of endothelial cells as compared to patients with the ICE + variant. They suggested that the latter may represent early disease which usually did not require surgery, as also evidenced by the lack of histopathological studies demonstrating “small cells”. In the present series though the number of eyes with ICE + type of endothelial cells was relatively small to draw any definitive conclusion, this variant appeared to be associated with clinically more severe disease as compared to the ICE –cell type as reflected indirectly in the significantly younger age of the patients with ICE + type cells as compared to those with the ICE– cells, as also the greater proportion (83%) of ICE + patients suffering from a more severe grade of glaucoma (Grades 2 or 3) vis a vis those with ICE – pattern in whom nearly half (47%) had Grade 1 or no glaucoma.
The ECD overall was lower in the affected eyes as compared to the uninvolved eyes, as has also been reported by Le et al. However, the mean ECD between affected eyes with and without corneal edema, was comparable suggesting that the cell count did not accurately reflect endothelial function in ICE eyes. Further the ICE cell type i.e. the “+” or “−” variant was also not predictive of the occurrence of corneal edema as seen by the significantly different mean ECD's but comparable occurrence of corneal edema in ICE + and – eyes. Our results appear similar to those of Liu et al., who in their series of 15 patients with ICE syndrome used both Hirst and Sherrard, classifications but did not find any distinct correlation between either ICE grading and occurrence of corneal edema or endothelial function and cell density.
To summarize we evaluated 21 patients of Indian origin having ICE syndrome and noted certain differences compared to published western literature. These include a greater overall incidence of raised IOP and/or glaucoma and more frequent occurrence of both the total ICE pattern versus the subtotal disease and ICE – cells versus the ICE + variant. An interesting, previously unreported observation was a gender predilection of the ICE cell variants, with ICE – cells being commoner in males and ICE + cells in females. The latter also appeared to manifest with clinically evident disease earlier than patients with ICE – cells, which may in part account for the common perception of ICE being predominantly a disease of women in young to middle adulthood, as the early onset of glaucoma may lead to a quicker diagnosis in these patients. This observation however needs to be validated in future studies, due to the small number of patients with ICE + patients in our series. Drawbacks of the present study include its retrospective nature, small sample size and qualitative rather than quantitative categorization of ICE cells as the “+” or “−” variant. Another limitation of the study was that the corneal edema was defined clinically and was not quantified because the study was a retrospective study and corneal edema was not a part of the three criteria required for clinical diagnosis of ICE. A comparison of salient results of the present study with the previous studies is provided in Table 2.
In conclusion recognizing the association (or lack thereof) of cell morphology as seen on IVCM, with clinically relevant parameters of gender, age, corneal edema, and severity of glaucoma in the clinical setting may help in better long-term prognostication of these eyes even when seen at earlier stages of the disease process.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
1. Shields MB. Progressive essential iris atrophy, Chandler's syndrome, and the Iris Nevus (Cogan-Reese) syndrome: A spectrum of disease Surv Ophthalmol. 1979;24:3–20
2. Hirst LW, Quigley HA, Stark WJ, Shield MB. Specular microscopy of iridocorneal endothelial syndrome Am J Ophthalmol. 1980;89:11–21
3. Alvarado JA, Murphy CG, Maglio M, Hetherington J. Pathogenesis of Chandler's syndrome, essential iris atrophy and the Cogan-Reese syndrome. I. Alterations of the corneal endothelium
Invest Ophthalmol Vis Sci. 1986;27:853–72
4. Alvarado JA, Murphy CG, Juster RP, Hetherington J. Pathogenesis of Chandler's syndrome, essential iris atrophy and the Cogan-Reese syndrome. II. Estimated age at disease onset Invest Ophthalmol Vis Sci. 1986;27:873–82
5. Grupcheva CN, McGhee CN, Dean S, Craig JP. In vivo
confocal microscopic characteristics of iridocorneal endothelial syndrome Clin Exp Ophthalmol. 2004;32:275–83
6. Campbell DG, Shields MB, Smith TR. The corneal endothelium
and the spectrum of essential iris atrophy Am J Ophthalmol. 1978;86:317–24
7. Howell DN, Damms T, Burchette JL Jr, Green WR. Endothelial metaplasia in the iridocorneal endothelial syndrome Invest Ophthalmol Vis Sci. 1997;38:1896–901
8. Chiou AG, Kaufman SC, Beuerman RW, Ohta T, Yaylali V, Kaufman HE. Confocal microscopy in the iridocorneal endothelial syndrome Br J Ophthalmol. 1999;83:697–70
9. Levy SG, Kirkness CM, Moss J, Ficker L, McCartney ACE. On the pathology of the Iridocorneal-endothelial syndrome: The ultrastructural appearances of ‘Subtotal ICE’ Eye. 1995;9:318–23
10. Chandler PA. Atrophy of the stroma of the iris, endothelial dystrophy, corneal edema and glaucoma Am J Ophthalmol. 1956;41:607–15
11. Cavanagh HD, Petroll WM, Alizadeh H, He YG, McCulley JP, Jester JV. Clinical and diagnostic use of in vivo
confocal microscopy in patients with corneal disease Ophthalmology. 1993;100:1444–54
12. Sherrard ES, Frangoulis MA, Kerr Muir MG, Buckley RJ. The posterior surface of the cornea in the irido-corneal endothelial syndrome: A specular microscopical study Trans Ophthalmol Soc UK. 1985;104:766–74
13. Lucas-Glass TC, Baratz KH, Nelson LR, Hodge DO, Bourne WM. The contralateral corneal endothelium
in the iridocorneal endothelial syndrome Arch Ophthalmol. 1997;115:40–4
14. Kupfer C, Kaiser-Kupfer MI, Datiles M, McCain L. The contralateral eye in the iridocorneal endothelial (ICE) syndrome Ophthalmology. 1983;90:1343–50
15. Malhotra C, Pandav SS, Gupta A, Jain AK. Phenotypic heterogeneity of the corneal endothelium
in iridocorneal endothelial syndrome by in vivo
confocal microscopy Cornea. 2014;33:634–7
16. Liu YK, Wang IJ, Hu FR, Hung PT, Chang HW. Clinical and specular microscopic manifestations of iridocorneal endothelial syndrome Jpn J Ophthalmol. 2001;45:281–7
17. Le QH, Sun XH, Xu JJ. In-vivo confocal microscopy of Iridocorneal endothelial syndrome Int Ophthalmol. 2009;29:11–8
18. Sherrard ES, Frangoulis MA, Muir MG. On the morphology of cells of posterior cornea in the iridocorneal endothelial syndrome Cornea. 1991;10:233–43
19. Laganowski HC, Kerr Muir MG, Hitchings RA. Glaucoma and the iridocorneal endothelial syndrome Arch Ophthalmol. 1992;110:346–50
20. Niederer R L, Mcghee CN. Clinical in vivo
confocal microscopy of the human cornea in health and disease Prog Retin Eye Res. 2010;29:30–58
21. Laganowski HC, Sherrard ES, Muir MG, Buckley RJ. Distinguishing features of the iridocorneal endothelial syndrome and posterior polymorphous dystrophy: Value of endothelial specular microscopy Br J Ophthalmol. 1991;75:212–6
22. Hemady RK, Patel A, Blum S, Nirankari VS. Bilateral iridocorneal endothelial syndrome: Case report and review of literature Cornea. 1994;13:368–72
23. Keiser-Kupfer M, Kuwabara T, Kupfer C. Progressive bilateral essential iris atrophy Am J Ophthalmol. 1977;83:340–6
24. Huna R, Barak A, Melamed S. Bilateral iridocorneal endothelial syndrome presented as Cogan-Reese and Chandler's syndrome J Glaucoma. 1996;5:60–2
25. Teekhasaenee C, Ritch R. Iridocorneal endothelial syndrome in Thai patients Arch Ophthalmol. 2000;118:187–92