Iridocorneal Endothelial (ICE) Syndrome
A spectrum of unilateral anterior segment disorders characterised by abnormal proliferating corneal endothelial cells (ICE cells) that migrate across the angle and iris, causing progressive angle closure, glaucoma, iris destruction, and corneal decompensation. The three clinical entities — Essential Iris Atrophy, Chandler Syndrome, and Cogan-Reese Syndrome — represent a pathological continuum.
Last updated: March 2026
Panel A — Anterior View: ICE Syndrome (Essential Iris Atrophy Variant)
Panel B — Cross-Section: ICE Cell Migration, Membrane Formation & Angle Closure
Essential Iris Atrophy
Iris holes, corectopia, ectropion uveae. Membrane contracts and stretches/perforates iris stroma.
Chandler Syndrome
Corneal oedema predominates. ICE cells cannot maintain endothelial pump function. Minimal iris changes.
Cogan-Reese (Iris Naevus)
Pedunculated iris nodules — membrane surrounds islands of normal iris creating naevus-like masses.
Panel A. Anterior view: Essential Iris Atrophy variant showing corneal oedema (A), ICE cell membrane on iris (B), iris atrophy holes (C), corectopia pulled superotemporal (D), progressive PAS (E), and ectropion uveae (F).
Panel B. Cross-section: ICE cells (red) on endothelium migrate across TM (forming PAS, blocking outflow → glaucoma) and extend onto iris surface as contractile membrane causing iris destruction and corectopia. Normal endothelium (blue) shown for comparison on right.
Iridocorneal Endothelial (ICE) syndrome is a unilateral, acquired, progressive disorder of the anterior segment characterised by a pathological population of corneal endothelial cells — termed ICE cells — that proliferate and migrate beyond their normal anatomical boundaries. ICE syndrome encompasses three clinical entities that share the same underlying pathophysiology but differ in which tissue manifests predominant disease: Essential Iris Atrophy (iris destruction and corectopia predominate), Chandler Syndrome (corneal oedema from endothelial pump failure predominates), and Cogan-Reese or Iris Naevus Syndrome (pedunculated melanocytic iris nodules from membrane surrounding islands of normal iris tissue). These are best understood as a clinical spectrum rather than distinct diseases.
ICE syndrome predominantly affects young-to-middle-aged women (typically 20–50 years) and is almost always unilateral. The most compelling proposed aetiology is transformation of corneal endothelial cells following subclinical Herpes Simplex Virus (HSV) infection — HSV DNA has been detected in the corneal endothelium of ICE syndrome patients using PCR methods. The vision-threatening complications are glaucoma (present in approximately 50% of cases, often refractory to medical management alone) and corneal endothelial failure (Chandler syndrome variant). Early diagnosis through specular microscopy, regular IOP monitoring, and timely surgical intervention for glaucoma are the cornerstones of management.
Herpes Simplex Virus (HSV) Endotheliitis — Primary Hypothesis
The strongest current evidence implicates HSV infection as the initiating trigger. HSV DNA has been detected by PCR in corneal endothelial specimens from ICE syndrome patients but not in controls. The proposed mechanism is that subclinical HSV endotheliitis induces transformation of corneal endothelial cells from their normal mesenchymal phenotype into an epithelial-like phenotype — an epithelial-to-mesenchymal transition in reverse (mesenchymal-to-epithelial transition). These transformed cells (ICE cells) lose contact inhibition, proliferate uncontrollably, and migrate outside their normal domain across the trabecular meshwork and iris surface. This hypothesis is supported by the observed elevated HSV seropositivity in ICE syndrome patients and by unilaterality (consistent with unilateral viral infection).
Congenital Neural Crest Cell Abnormality — Alternative Hypothesis
An alternative theory proposes that ICE syndrome arises from a developmental abnormality of neural crest-derived anterior segment cells, resulting in a subpopulation of endothelial cells with abnormal growth potential. This would explain the unilateral presentation and the fact that some patients are born with detectable specular abnormalities before any clinical signs. However, this hypothesis does not explain the lack of other congenital anterior segment anomalies, the predominantly female sex distribution, or the middle-age onset.
Why Unilateral?
Bilateral ICE syndrome is exceptionally rare (<5% of cases). The unilaterality strongly supports an acquired (infectious) rather than purely congenital aetiology, as bilateral involvement would be expected with a germline developmental defect.
- HSV endotheliitis (proposed trigger): Subclinical or resolved HSV infection induces endothelial cell transformation; transformed cells acquire epithelial-like characteristics — form desmosome-like intercellular junctions, express cytokeratin markers, and lose contact inhibition.
- Loss of contact inhibition: Normal corneal endothelial cells cease migration on contact with neighbouring cells (contact inhibition of proliferation and locomotion). ICE cells bypass this regulatory mechanism and proliferate continuously.
- Migration across trabecular meshwork: ICE cells extend as a thin fibrocellular membrane across the posterior trabecular meshwork surface. This membrane progressively covers trabecular beams, physically occluding aqueous outflow and forming peripheral anterior synechiae (PAS). PAS may be broad-based, tent-shaped, or sheet-like; they extend anteriorly toward or beyond Schwalbe line in advanced cases.
- Progressive angle closure from PAS: As PAS accumulate (progressive, unilateral, superior distribution often), aqueous outflow resistance increases → IOP elevation → secondary glaucoma. The PAS in ICE are distinct from those in PACG — they are caused by membrane contraction rather than appositional iris-angle contact.
- Membrane extension onto iris anterior surface: ICE cells continue migrating posteriorly across the angle onto the iris stroma, laying down a fibrovascular membrane over the iris.
- MMP secretion and iris destruction: The ICE cell membrane secretes matrix metalloproteinases (MMP-2, MMP-9) that degrade type IV collagen and laminin in the iris stroma, directly dissolving iris tissue.
- Membrane contraction and corectopia: The fibrovascular membrane contracts, exerting traction on the iris. The pupil is pulled (corectopia) in the direction of greatest membrane traction — typically toward the quadrant of heaviest PAS. Continued contraction stretches and eventually perforates the iris (Essential Iris Atrophy) or distorts its shape without perforation.
- Cogan-Reese variant: In the Cogan-Reese subtype, the ICE cell membrane surrounds but does not destroy islands of normal iris stroma. These entrapped islands appear clinically as pedunculated melanocytic nodules on the iris surface.
- Chandler variant: In Chandler syndrome, the ICE cells covering the corneal endothelium are metabolically insufficient — they cannot maintain the fluid-pumping function of normal endothelium. The resulting endothelial pump failure leads to progressive corneal stromal and epithelial oedema, even with IOP in the normal or mildly elevated range.
ICE syndrome is a clinical spectrum. All three variants share the same ICE cell pathology and may overlap in any individual patient.
| Variant | Predominant Feature | Iris Changes | Corneal Changes | Glaucoma Risk |
|---|---|---|---|---|
| Essential Iris Atrophy | Iris destruction | Corectopia, full-thickness holes, progressive iris loss | Subtle beaten-silver appearance; minimal oedema | High (~50%) |
| Chandler Syndrome | Corneal oedema | Mild; minimal corectopia or atrophy | Significant oedema even at low IOP; beaten-silver endothelium | Moderate; IOP may be only mildly elevated |
| Cogan-Reese (Iris Naevus) Syndrome | Iris nodules | Pedunculated, pigmented, mushroom-shaped iris nodules; diffuse iris stromal atrophy and darkening | Beaten-silver endothelium; moderate oedema | High (~50%) |
- Female sex: Strong female predominance; approximately 3:1 female-to-male ratio across all ICE syndrome variants
- Age 20–50 years: Typical presentation in young-to-middle-aged adults; very rare in childhood or after age 60
- Prior HSV infection: Elevated HSV seropositivity in ICE patients compared to age-matched controls; subclinical or resolved HSV endotheliitis proposed as the transforming trigger
- Unilateral anterior segment disease: Any unilateral progressive iris or corneal change in the appropriate demographic should raise suspicion for ICE syndrome
- White ethnicity: ICE syndrome more commonly reported in white populations, though it occurs across all ethnic groups
- Family history: Not a significant risk factor — ICE syndrome is not hereditary; bilateral cases are exceedingly rare and may represent independent disease rather than systemic predisposition
Specular Microscopy (Most Specific Test)
- ICE cells: Pathognomonic finding — reversed dark-light pattern compared to normal endothelium; normal cells appear light with dark boundaries, ICE cells appear dark with light halos around a dark nucleus
- Irregular cell shapes (pleomorphism), loss of hexagonality, reduced cell density
- Abnormal cells extend beyond the clinically visible area of disease; may involve the entire corneal endothelium of the affected eye
Slit-Lamp Biomicroscopy
- Beaten-silver or hammered-silver corneal endothelium: Subtle diffuse endothelial irregularity visible on wide-beam specular illumination; may be present before overt corneal oedema
- Corneal oedema (Chandler): Microcystic epithelial oedema; stromal thickening; may cause reduced visual acuity and surface irregularity; classically worse in the morning and improves as the day progresses (endothelial pump activity increases with blinking and warming)
- Corectopia: Displaced, distorted pupil drawn toward the quadrant of maximal PAS/membrane traction; most prominent in Essential Iris Atrophy variant; key clinical distinguishing feature from other iris diseases
- Iris atrophy/holes: Full-thickness iris defects producing a well-demarcated hole through which the fundus is visible on retroillumination; characteristic of Essential Iris Atrophy variant
- Transillumination defects: Moth-eaten, moth-wing, or irregular pattern on iris retroillumination from stromal thinning
- Ectropion uveae: Posterior iris pigment epithelium visible on the anterior iris surface at the pupillary margin from membrane traction
- Pedunculated iris nodules (Cogan-Reese): Multiple small, raised, darkly pigmented mushroom-shaped nodules on the iris surface; represent islands of iris stroma surrounded by ICE cell membrane; do NOT transilluminate
- Diffuse iris darkening (Cogan-Reese): The affected iris appears darker overall due to membrane covering the iris stroma
Gonioscopy
- Extensive peripheral anterior synechiae (PAS) — often broad-based, may be 360° in advanced disease; extend to or above Schwalbe line
- Trabecular meshwork covered by a fibrovascular ICE cell membrane
- Progressive angle closure pattern differs from PACG — no appositional component when IOP is normalised; adhesions are permanent
- Relative sparing of some angle quadrants early in the disease
- Blurred vision (unilateral): From corneal oedema (Chandler syndrome — typically worse in the morning, improving as the day progresses with evaporation and improved endothelial pumping), or from glaucomatous optic nerve damage; may be the presenting complaint
- Halos around lights: Corneal oedema causing light scatter; worse in the morning (Chandler syndrome); also from elevated IOP
- Photophobia: Corneal oedema causing surface irregularity; also from abnormal pupil shape and size reducing the patient's ability to regulate light entry
- Ocular discomfort or pain: Epithelial bullae from advanced corneal oedema cause significant pain; also from elevated IOP (dull pressure or headache)
- Noticed pupil or iris change: The patient or a family member may notice a change in pupil shape (corectopia), iris colour, or iris appearance; may be the initial presenting concern, particularly in younger patients
- Glaucoma symptoms: Acute IOP elevation — headache, nausea, rainbow halos, reduced vision; chronic glaucoma — usually asymptomatic until advanced visual field loss
- Cosmetic concern: Iris nodules (Cogan-Reese), corectopia, or iris colour change causing psychosocial distress
Glaucoma (Primary Vision-Threatening Complication)
Occurs in approximately 50% of ICE syndrome patients overall. Caused by progressive PAS from ICE cell membrane occluding the trabecular meshwork. This form of glaucoma is notoriously difficult to control medically because PAS continue to form even with good IOP control — the angle closes not from elevated pressure but from ongoing membrane proliferation. Medical therapy typically fails long-term, and surgical intervention (tube shunt surgery) is usually required. Glaucoma may develop years to decades after the initial diagnosis of ICE syndrome.
Corneal Endothelial Decompensation
Particularly in Chandler syndrome. ICE cells replace the normal endothelium but function poorly as pump cells. Progressive corneal oedema leads to: reduced best-corrected visual acuity, painful epithelial bullae (bullous keratopathy), risk of microbial keratitis from ruptured bullae, and corneal scarring in end-stage disease. DSAEK or DMEK is required for visual rehabilitation, with the risk that ICE cells may recur on the donor graft over time.
Progressive Vision Loss
The combination of untreated or inadequately controlled glaucoma and corneal endothelial failure can lead to profound, irreversible visual impairment. Advanced glaucomatous optic neuropathy with visual field loss to the central 10° (tubular vision) or beyond represents the worst visual outcome from untreated ICE glaucoma.
Psychological Impact
Progressive cosmetic changes (corectopia, iris nodules, heterochromia) and the awareness of a progressive, incurable condition in a young patient can cause significant psychological burden. Counselling and appropriate management of expectations are important components of care.
- Herpes Simplex Virus (HSV): The proposed systemic link; HSV seropositivity is elevated in ICE syndrome patients. No active systemic HSV disease is required — the virus likely causes ICE syndrome via a historical, subclinical ocular infection. Clinicians should enquire about prior herpetic oral or ocular disease
- No direct systemic organ involvement: ICE syndrome is an exclusively ocular disorder; it does not involve internal organs, the skin, or other systems
- No hereditary component: ICE syndrome does not run in families and is not associated with chromosomal abnormalities or recognised systemic genetic syndromes
- Bilateral ICE (rare): Exceedingly rare bilateral cases do not imply systemic disease; they likely represent bilateral sequential viral transformation events
Note: ICE syndrome must be distinguished from Axenfeld-Rieger syndrome (bilateral, congenital, systemic associations) and Fuchs endothelial dystrophy (bilateral, hereditary, no iris changes) which do have systemic or genetic correlates.
- Specular microscopy (gold standard): Identifies characteristic ICE cells with reversed dark-light pattern, pleomorphism, and reduced cell density; the single most specific test; abnormal cells extend throughout the corneal endothelium of the affected eye even before clinical signs are apparent
- Slit-lamp biomicroscopy: Comprehensive anterior segment evaluation including retroillumination (TIDs, iris holes), wide-beam specular illumination (beaten-silver cornea), corectopia assessment, iris nodules; document all findings with photography
- Gonioscopy: Documents extent of PAS, morphology (broad-based, tent-shaped), trabecular meshwork appearance; essential for staging glaucoma severity and planning treatment
- IOP measurement: At every visit; consider diurnal curves; Goldmann applanation with corneal thickness correction; may be elevated, normal (Chandler syndrome with corneal oedema despite normal IOP), or fluctuating
- Anterior segment OCT (AS-OCT): Documents angle closure, corneal thickness, anterior chamber depth; non-contact; excellent for monitoring structural progression; PAS appear as adhesions between iris and peripheral cornea
- Ultrasound biomicroscopy (UBM): 50 MHz imaging detects angle structures not visible clinically; particularly useful in opaque media; may visualise ICE cell membrane as a thin hyperreflective layer in the angle
- Corneal confocal microscopy: In vivo imaging detects ICE cells at the sub-basal nerve plexus level; non-invasive alternative or adjunct to specular microscopy; supports early diagnosis
- Optic nerve imaging and visual fields: OCT optic nerve head (RNFL, GCC), Humphrey visual field (24-2 or 10-2) to establish baseline and monitor for glaucomatous progression; repeat every 6–12 months depending on IOP control
- HSV serology: Supportive but not diagnostic; elevated IgG seropositivity in ICE syndrome; does not confirm active HSV as causative
- Aqueous humour PCR for HSV: Research tool; not part of routine clinical diagnosis; may be performed by vitreoretinal or corneal specialists in atypical cases to exclude active viral uveitis
- Fluorescein angiography: Not routine; used if iris neovascularisation or vascular anomaly is suspected; normal iris FFA in early ICE; may show fibrovascular membrane staining in advanced disease
1. Corneal Oedema Management (Chandler Syndrome)
- Hypertonic saline drops (5% NaCl): 4 times daily; osmotic dehydration of the corneal stroma; reduce to twice daily if tolerated once oedema stabilises
- Hypertonic saline ointment (5% NaCl nocte): Sustained overnight osmotic effect; more effective than drops alone for nocturnal oedema accumulation
- Hair dryer technique: Patient holds a hair dryer at arm's length, pointing away, to gently evaporate tear film and epithelial moisture in the morning — can reduce morning corneal oedema acutely
- Soft bandage contact lens: Relieves pain from epithelial bullae; does not treat underlying oedema; requires monitoring for microbial keratitis risk
- Endothelial keratoplasty (DSAEK/DMEK): Definitive treatment for corneal decompensation; DMEK provides superior visual outcomes; awareness that ICE cells may recur on the donor endothelium over time, potentially requiring repeat keratoplasty
2. Glaucoma Management
- Topical hypotensives: Prostaglandin analogues (latanoprost, bimatoprost, travoprost) as first-line; beta-blockers (timolol 0.5% bd), carbonic anhydrase inhibitors (dorzolamide, brinzolamide), alpha-2 agonists (brimonidine 0.2%); combination drops as needed
- Laser trabeculoplasty: SLT may be attempted but efficacy is reduced in ICE syndrome due to membrane covering the trabecular meshwork; not generally first-line surgical option
- Trabeculectomy with MMC: Higher failure rate than in POAG due to progressive ICE membrane formation over the filtering bleb and ongoing PAS formation; may be attempted as initial surgery in compliant patients without prior failed surgery
- Glaucoma drainage devices (tube shunts): Ahmed or Baerveldt tube shunt surgery is generally preferred for ICE-related glaucoma; provides more durable IOP control than trabeculectomy; the tube bypasses the blocked trabecular meshwork; considered the surgical treatment of choice by most glaucoma specialists for ICE syndrome
- Cyclophotocoagulation (CPC): Transcleral or endoscopic CPC for refractory ICE glaucoma where tube shunt has failed or is not feasible; reduces aqueous production; risk of hypotony and phthisis in advanced cases
3. Antiviral Treatment (Experimental — Not Standard Care)
Oral aciclovir (400–800 mg bd–tid) has been proposed to suppress the putative HSV trigger and potentially slow the progression of ICE syndrome. Small case series and case reports suggest possible benefit, but no randomised controlled trial has been conducted. This approach is not current standard of care but may be considered by corneal/glaucoma subspecialists in HSV-seropositive ICE patients with rapid disease progression.
4. Iris Reconstruction (Cosmetic — Specialist Only)
For severe corectopia causing monocular diplopia or significant cosmetic distress in visually stable patients, iris prosthetic implants (artificial iris devices) or cosmetic tinted contact lenses may be considered in specialist centres. These options carry inherent risks in an already compromised anterior segment and are not undertaken lightly.
5. Long-Term Monitoring
ICE syndrome is a lifelong progressive condition. Monitoring schedule: IOP measurement every 3–6 months; optic nerve OCT and visual fields every 6–12 months; specular microscopy annually; AS-OCT every 12 months or sooner if clinical change detected. Patients require indefinite follow-up under ophthalmology co-management.
Singapore Optometry Scope Note: Optometrists should recognise the beaten-silver corneal endothelial appearance and unilateral iris changes (corectopia, iris atrophy, iris nodules) as hallmarks requiring urgent specialist referral for specular microscopy. Do not assume unilateral corneal oedema in a young woman is Fuchs dystrophy — consider ICE syndrome and refer. Monitor IOP at every visit in diagnosed ICE syndrome patients; early glaucoma detection before optic nerve damage is critical for visual prognosis. Co-manage glaucoma medications under ophthalmology supervision and document IOP trends. Educate patients on the progressive nature of ICE syndrome and the importance of lifelong follow-up. Never initiate topical steroids alone assuming anterior uveitis without ruling out ICE syndrome — steroids will not halt ICE progression and may mask the diagnosis. Document and photograph iris changes at every visit as a baseline for monitoring corectopia progression.
- Progressive and lifelong: ICE syndrome has no known cure; disease progression continues over years to decades, though the rate varies considerably between individuals
- Glaucoma prognosis: The most common cause of vision loss; glaucoma becomes increasingly difficult to control as angle progressively closes; tube shunt surgery (Ahmed/Baerveldt) provides longer-term IOP control than trabeculectomy; visual prognosis depends critically on the degree of optic nerve damage at the time of glaucoma diagnosis
- Corneal prognosis: DSAEK/DMEK provides excellent visual rehabilitation for Chandler syndrome corneal decompensation; however, ICE cells may recur on the donor graft endothelium (estimated recurrence rate 20–40% over 10 years), requiring repeat keratoplasty
- Impact of early diagnosis: Patients diagnosed before significant optic nerve damage or corneal decompensation — with prompt institution of IOP-lowering therapy and regular monitoring — have a substantially better visual prognosis than those presenting with advanced disease
- Age at diagnosis: Younger patients at diagnosis face a longer lifetime disease burden; the decision to proceed to surgical glaucoma management should be made earlier in younger patients with progressive PAS
- Bilateral disease (rare): The minority of bilateral cases face double the risk of visual compromise; require even more intensive monitoring and earlier surgical intervention
| Condition | Key Differentiator from ICE Syndrome |
|---|---|
| Fuchs endothelial corneal dystrophy | Bilateral; guttata (central endothelial excrescences) on specular microscopy; no iris changes, no corectopia; no PAS; glaucoma not caused by membrane; normal-morphology (guttata pattern) endothelium on specular |
| Posterior polymorphous corneal dystrophy (PPCD) | Bilateral; autosomal dominant; hereditary; vesicular, band-like, or geographic lesions at the level of Descemet membrane; iris changes mild if present; specular microscopy shows a different pattern from ICE cells |
| Axenfeld-Rieger syndrome | Bilateral; congenital; posterior embryotoxon (anteriorly displaced Schwalbe line); iris strands to Schwalbe line; associated systemic anomalies (dental hypodontia, craniofacial); normal specular microscopy; no ICE cells |
| Fuchs heterochromic iridocyclitis | Unilateral; diffuse iris atrophy with heterochromia; stellate KPs scattered diffusely on entire endothelium; Amsler sign (bleeding from iris vessels on gonioscopy); no ICE cells; associated with toxoplasma; no corectopia or iris holes |
| Iridoschisis | Elderly patients; splitting (schisis) of anterior iris stromal layers; free-floating iris fibrils in the anterior chamber; normal corneal endothelium; shallow anterior chamber; no corectopia; no ICE cells on specular |
| Iris melanoma | Unilateral elevated pigmented iris lesion with intrinsic vascularity; does not cause beaten-silver cornea; may cause secondary glaucoma by direct angle invasion; UBM and FFA help differentiate; no ICE cells |
| Primary angle closure glaucoma (PACG) | Bilateral predisposition; appositional PAS (opens with IOP normalisation initially); no ICE cells; no iris holes or corectopia; peripheral iris crowding on UBM; typically older patients |
- Shields MB. Textbook of Glaucoma. 4th ed. Baltimore: Williams & Wilkins; 1998. ICE syndrome chapter.
- Eagle RC Jr, Font RL, Yanoff M, Fine BS. Proliferative endotheliopathy with iris abnormalities: the iridocorneal endothelial syndrome. Arch Ophthalmol. 1979;97(11):2104–2111.
- Laganowski HC, Sherrard ES, Muir MG, Buckley RJ. Distinguishing features of the ICE syndrome and posterior polymorphous dystrophy: value of endothelial specular microscopy. Br J Ophthalmol. 1991;75(4):212–216.
- Huna-Baron R, Barnett Mizrahi E, Shapiro A. Secondary glaucoma following iridocorneal endothelial syndrome. Ophthalmic Surg Lasers Imaging. 2010;41:S82–S86.
- Kidd MN, Grierson I, Hogg P. The cellular origin of pigment nodules in the iris in the iridocorneal endothelial syndrome. Curr Eye Res. 1988;7(9):933–943.
- Alvarado JA, Murphy CG, Juster RP, Hetherington J Jr. Pathogenesis of Chandler's syndrome, essential iris atrophy and the Cogan-Reese syndrome. Invest Ophthalmol Vis Sci. 1986;27(6):853–872.
- Tsai CS, Ritch R, Straus SE, et al. Antibodies to Epstein-Barr virus in iridocorneal endothelial syndrome. Arch Ophthalmol. 1990;108(10):1477–1480.
- Chandler PA. Atrophy of the stroma of the iris; endothelial dystrophy, corneal edema, and glaucoma. Am J Ophthalmol. 1956;41(4):607–615.
- Cogan DG, Reese AB. A syndrome of iris nodules, ectopic Descemet's membrane, and unilateral glaucoma. Doc Ophthalmol. 1969;26:424–433.