Glaucoma
The Silent Thief of Sight — Red Flags & Treatment Ladder

1. What is glaucoma?

Glaucoma is not one disease but a family of diseases, defined by structural damage to the optic nerve (deepening cup, thinning RNFL) and matching visual-field loss. Most forms are linked to intraocular pressure (IOP), but not all glaucomas have high IOP.

Why does IOP rise? The eye continuously produces clear aqueous humor that flows through the anterior chamber and drains via the trabecular meshwork at the iridocorneal angle. If outflow slows or is blocked, aqueous accumulates and IOP rises; sustained pressure damages the optic nerve and kills retinal ganglion-cell axons. Once these neurons die they do not regenerate — the core mechanism of glaucomatous damage.

A special quirk of glaucoma: visual-field defects typically start peripherally (especially nasally) and creep centrally. Because the brain fills in gaps and the other eye compensates, early unilateral defects are virtually undetectable to the patient. By the time someone notices 'I can't see well,' damage is usually already moderate-to-severe — which is why periodic comprehensive eye exams after 40 are so important.

2. Classification of glaucoma

By anterior-chamber-angle status and etiology, glaucoma is broadly classified as:

① Primary open-angle glaucoma (POAG) — most common

• Open angle, but impaired trabecular meshwork outflow; aqueous accumulates and IOP rises
• Chronic, asymptomatic until visual-field loss becomes substantial
• The largest share of adult glaucoma (worldwide about twice as common as PACG, though similar in some Asian populations)

② Normal-tension glaucoma (NTG) — especially common in Asians

• A subtype of POAG — same clinical picture but IOP consistently within the 'normal' range (≤ 21 mmHg)
• The AAO 2025 PPP cites that 30-92% of POAG patients (depending on study) have untreated IOP ≤ 21 mmHg. The Taiwanese Shihpai Eye Study likewise reported that most local glaucoma patients had IOP within normal range.
• Possible mechanisms: low ocular perfusion pressure, nocturnal hypotension, intrinsic optic-nerve vulnerability, abnormal translaminar pressure gradient
• Lowering IOP still works — the Collaborative Normal-Tension Glaucoma Study (1998) showed 30% IOP reduction significantly slowed visual-field deterioration

③ Primary angle-closure glaucoma (PACG) — relatively common in Asians

• Iris bows forward and contacts the trabecular meshwork, partially or fully closing the angle
• May be chronic (progressive peripheral anterior synechiae) or acute
• Acute angle-closure crisis is an ophthalmic emergency (see Section 4 red flags)
• High-risk groups: Asian, Inuit (indigenous Arctic peoples), female, elderly, hyperopic, short axial length, shallow anterior chamber, thickening lens (cataract)
• Global PACG: ~20.2 million in 2013, of which 15.5 million live in Asia

④ Secondary glaucomas

Not 'primary' but caused by other ocular or systemic conditions:

• Pseudoexfoliation syndrome (PXS) — flaky deposits in the anterior segment with trabecular dysfunction; common in older Asians as well
• Pigment dispersion syndrome — iris pigment shed onto trabecular meshwork; classic in young myopic males
• Steroid-induced — any route (oral, inhaled, topical, intravitreal) chronic steroid use can raise IOP; usually reversible after withdrawal
• Uveitic, traumatic, neovascular, iridocorneal endothelial (ICE) syndrome, post-surgical etc.

⑤ Ocular hypertension (OHT) — not yet glaucoma

IOP > 21 mmHg but normal optic nerve and visual fields. The Ocular Hypertension Treatment Study (OHTS) showed 5-year POAG conversion ~9.5% in untreated subjects; older age, higher IOP, larger vertical cup-to-disc ratio, greater visual-field pattern standard deviation (PSD) and thinner central cornea predicted conversion. In the untreated observation group, 36% (28 of 78) of those with baseline IOP > 25.75 mmHg and CCT ≤ 555 μm developed POAG over a median follow-up of 72 months. Treatment is individualized. NICE 2022 recommends 360° SLT (selective laser trabeculoplasty) or PGA (prostaglandin analogue) when IOP ≥ 24 mmHg and the patient has future visual-impairment risk.

3. Epidemiology & risk factors

Epidemiology:

• Globally: ~52.7 million POAG cases in 2020, projected to 79.8 million by 2040 (aging population). Glaucoma is a major cause of blindness worldwide, responsible for 11-14% of all blindness (AAO 2025 PPP).
• Taiwan (Shihpai Eye Study, ≥72 years): overall glaucoma 8.7%, POAG 3.7%, PACG 4.8%. PACG share is markedly higher than in Western populations. Most local patients have IOP within the normal range — IOP-only screening misses them.
• Racial/ethnic differences: POAG ~3× more prevalent in Black and Latino/Hispanic individuals than non-Hispanic Whites; glaucoma blindness ≥6× more common in Black populations. Angle-closure highest in Inuit, East Asian and South-East Asian groups.
• Low awareness: about half of community-detected glaucoma cases were previously undiagnosed in multiple studies — underscoring the value of comprehensive periodic exams.

POAG risk factors (per AAO 2025 PPP):

① Strongly associated (non-modifiable)

• Age — prevalence rises sharply each decade after 40
• Family history — Rotterdam Eye Study showed a 9.2× higher odds of POAG in those with a first-degree relative affected
• Race/ethnicity — Black, Latino, East Asian (more PACG)
• Myopia (especially high myopia) — long axial length weakens optic-nerve support
• Thinner central cornea (in OHTS, risk was highest with CCT ≤ 555 μm) — also under-measures true IOP
• Low corneal hysteresis — a newer independent risk factor
• Larger cup-to-disc ratio
• Optic-disc hemorrhage — often heralds upcoming progression

② Moderately associated (partially modifiable)

• Diabetes mellitus — 40-100% increased POAG odds across population studies
• Low ocular perfusion pressure — diastolic perfusion pressure < 50 mmHg significantly raises POAG risk; nocturnal BP overshoot from antihypertensives may impair optic-nerve perfusion
• Long-term corticosteroids — any route
• Migraine, Raynaud's — vasospastic tendency, more linked to NTG
• Sleep apnea (OSA) — recent evidence links it to POAG and NTG
• Metabolic syndrome — associated with glaucoma in a Taiwan Biobank cross-sectional study of ~94,000 adults aged 40+ (adjusted OR 1.15; self-reported glaucoma)

PACG risk factors (per AAO 2025 ACG PPP): female, Inuit/East-Asian/South-Asian descent, older age (especially 50+), hyperopia, short axial length, shallow anterior chamber, family history, thickening lens (cataract). Asians' shorter axial lengths and shallower chambers explain the higher PACG share.

4. Symptoms & warning signs

① Chronic open-angle (POAG / NTG) — usually asymptomatic

• Early stages are silent — peripheral (nasal) field defects are filled in by the brain and the fellow eye
• In moderate stages: needing more deliberate side-mirror checks while driving, tripping on stairs, bumping into furniture, missed sports cues
• Late stage: 'tunnel vision' — only a small central island remains; poor night vision and reduced contrast sensitivity
• Visual acuity is preserved until very late — '20/20' does not rule out glaucoma

② Chronic / intermittent angle-closure

• May have intermittent ocular ache, mild headache, transient blur — typically in dim light when the pupil dilates (cinemas, evenings)
• Occasional halos around lights
• These are often dismissed as fatigue or eye strain — be sure to mention them

③ Acute angle-closure crisis (AACC) — ophthalmic emergency

A true ophthalmic emergency. The AAO 2025 ACG PPP lists the classic features:

Common triggers: dim environments dilating the pupil (cinemas, evening), certain medications (some cold meds, anticholinergics, some psychiatric drugs), emotional stress, prolonged head-down posture.

5. Diagnostic evaluation

Glaucoma diagnosis requires multiple tests interpreted together — never one measurement alone. A standard initial workup includes:

① Intraocular pressure (IOP)

• Gold standard: Goldmann applanation tonometry (slit-lamp mounted) — AAO and NICE both prefer it
• Non-contact (air-puff) tonometry is less accurate — NICE 2017 explicitly advises not basing referral decisions on air-puff alone
• One reading is not enough — IOP has diurnal variation; multiple measurements at different times of day are needed
• Corneal thickness affects measurement — thin corneas underestimate, thick corneas overestimate IOP

② Optic-nerve head assessment

• Dilated stereoscopic biomicroscopy at the slit lamp — direct optic-nerve view
• Observe: cup-to-disc ratio, neuroretinal rim thickness, the ISNT rule (normal: Inferior > Superior > Nasal > Temporal), disc hemorrhage, parapapillary atrophy, RNFL wedge defects
• Disc hemorrhages often herald upcoming progression — AAO PPP notes that periodic photography can reveal hemorrhages missed on examination, and a lower target IOP may be justified when they are present

③ Visual field testing

• Standard automated perimetry (Humphrey Field Analyzer most common)
• 24-2 program (24-degree field) — mainstay for early-to-moderate disease
• 10-2 program (central 10 degrees) — detects central defects; vital in advanced disease
• Patterns: nasal step, arcuate scotoma, Bjerrum scotoma; tunnel vision in advanced disease
• New or worsening defects should be confirmed by repeat testing with the same strategy

④ Optical coherence tomography (OCT)

• RNFL thickness — peripapillary retinal nerve fiber layer
• Macular GCC / GCL — sensitive to early ganglion-cell loss
• OCT & VF are complementary — OCT measures structure, VF measures function; they don't always change in parallel and together give the fullest picture

⑤ Gonioscopy — especially important in Asians

• Distinguishes open vs closed angles — drives treatment choice
• Must be done in dim light without strong illumination to avoid false widening
• May be supplemented with AS-OCT or UBM

⑥ Central corneal thickness (CCT)

CCT affects IOP measurement (thin cornea underestimates true IOP). CCT is also an independent POAG risk factor — in OHTS, thinner CCT predicted conversion from ocular hypertension to POAG (≤ 555 μm vs > 588 μm: hazard ratio 3.4).

6. Treatment goal: target IOP

Key trial findings:

• EMGT: 25% IOP reduction with topical drugs + ALT slowed disc and field damage in newly diagnosed POAG.
• CNTGS: a 30% IOP reduction slowed visual-field loss even in normal-tension glaucoma — NTG also benefits from IOP lowering.
• AGIS: patients whose IOP was < 18 mmHg at every visit (mean 12.3) had no further field loss on average — moderate-severe disease may need a lower target.
• UKGTS: latanoprost vs placebo reduced visual-field progression over 2 years.

Individualization: AAO PPP suggests an initial target 20-30% below baseline IOP, adjusted up or down by disease course and severity; more severe or faster-progressing disease warrants a lower target (a less aggressive target may be reasonable when life expectancy is limited or treatment risks outweigh benefits). EGS 6th-edition reference values: about 18-20 mmHg with at least a 20% reduction in early glaucoma, about 15-17 mmHg with at least 30% in moderate glaucoma, and lower (e.g. 8-12 mmHg) in advanced disease, though EGS stresses that target IOP is an upper limit, that little evidence supports any particular algorithm, and that it must be estimated for each eye and reviewed over time. Actual targets are individualized.

7. Treatment ladder

Modern glaucoma treatment centers on IOP lowering via three modalities: laser, medication, surgery.

① First-line: 360° selective laser trabeculoplasty (SLT)

• Principle: 532 nm Q-switched laser delivers 360° spots to the trabecular meshwork, stimulating cell turnover and outflow. Non-scarring, non-destructive (unlike older ALT).
• Effect: in the LiGHT trial, 69.8% of SLT-arm eyes remained at or below target IOP at 6 years without medication or surgery; repeatable with retained efficacy.
• Procedure: outpatient slit-lamp, ~5 minutes, painless, return home same day
• Side effects: mild anterior-chamber inflammation; transient IOP spikes 4.5-27%; higher in heavily pigmented or narrow angles
• Not suitable for: pigment dispersion, advanced glaucoma, patients unable to position

② Topical IOP-lowering drops

Prescribing principles:

• If drugs are chosen first, PGA is usually preferred (once daily, strongest IOP reduction, fewest systemic effects)
• If PGA is insufficient, intolerable or contraindicated (active uveitis, CME risk, asymmetric appearance from unilateral use) — switch to β-blocker (confirm no asthma/heart disease first)
• If still insufficient → add α2 agonist or CAI (fixed combinations reduce drop frequency)
• Switch to preservative-free formulations for BAK allergy or significant ocular surface disease (NICE 2022)
• Failure on 3 drug classes → consider laser (if not yet done) or surgery

③ Minimally invasive glaucoma surgery (MIGS)

• Many devices (iStent inject, Hydrus, XEN gel stent, Kahook Dual Blade, PreserFlo MicroShunt, etc.)
• Small wound, safer, fast recovery, fewer complications — suits mild-to-moderate OAG, often combined with cataract surgery
• Less aggressive IOP reduction than trabeculectomy but lower risk

④ Conventional glaucoma surgery

• Trabeculectomy — gold-standard, most powerful IOP reduction (often <10 mmHg), but bleb-related risks (leak, infection, hypotony). For advanced disease or refractory cases. NICE 2022 recommends glaucoma surgery with MMC augmentation as first choice for advanced COAG.
• Aqueous shunt (Ahmed/Baerveldt) — for trabeculectomy failure or complex cases (neovascular, uveitic, post-surgical)
• Cyclodiode laser — destroys ciliary body, used in refractory or end-stage eyes

⑤ Acute angle-closure crisis management

Per the AAO 2025 ACG PPP algorithm:

• Step 1 (immediate): oral/IV acetazolamide, topical timolol + α2 agonist, pilocarpine (after IOP starts to fall), and consider anterior-chamber paracentesis
• Step 2： Laser peripheral iridotomy (LPI) — creates a hole in the peripheral iris allowing aqueous to bypass the pupil. Standard treatment after AACC; for primary angle-closure suspects (PACS), the AAO advises considering LPI, with regular follow-up as an alternative, discussed according to risk.
• Step 3 (if needed): lens extraction — the EAGLE trial showed early lens extraction outperforms LPI + meds for PACG and PAC with elevated IOP
• The fellow eye must also receive prophylactic LPI — without it, the unaffected eye is at very high risk of an acute attack (per AAO PPP)

8. Follow-up

Glaucoma is chronic and requires lifelong follow-up. AAO 2025 PPP and NICE 2022 recommend:

• Stable patients: by severity and duration of control, the AAO PPP suggests IOP follow-up about every 3-12 months (about 3-6 if severe) and visual-field and optic-nerve evaluation about every 6-18 months (about 6-12 if severe). NICE suggests reassessing controlled, non-progressing chronic open-angle glaucoma (COAG) at about 6-18 months; your doctor sets the actual interval
• After treatment change: the AAO PPP mentions a visit at 2-8 weeks, depending on severity, to assess response and side effects
• Progressing: closer follow-up; intensify treatment
• VF and OCT are complementary: function vs structure — they don't always change in parallel and both should be tracked
• New disc hemorrhage is an early progression sign — periodic photography can reveal it, and a lower target IOP may be considered

9. Home self-care

Glaucoma care doesn't end at the clinic. Adherence + correct technique + sensible lifestyle adjustments maximize treatment benefit.

① Drop adherence — the most important home task

• Phone alarms; pair drops with toothbrushing or showers
• Punctal occlusion: after instilling, gently press the inner corner of the eye for 1-2 minutes — markedly reduces systemic absorption (especially β-blockers) and prolongs ocular contact time.
• Wait 5 minutes between different drops so the second doesn't wash out the first
• Don't stop on your own — a 'normal' IOP only reflects the drug's effect; stopping causes rebound and new damage

② Exercise: moderate aerobic is good; avoid certain postures

• Beneficial: regular aerobic exercise (brisk walking, jogging, swimming, cycling) — IOP may fall transiently after moderate aerobic exercise, with magnitude and duration varying between people; it also supports cardiovascular health
• Avoid: head-down or inverted yoga poses (such as headstand and downward dog) — in a small study, head-down poses raised IOP by about 4-12 mmHg within a minute, returning to baseline within 2 minutes of sitting up; headstand has been reported to roughly double IOP. Also avoid Valsalva (heavy weightlifting with held breath), prolonged head-down posture, inversion tables.
• Swimming: don't strap goggles too tight — pressure can elevate IOP

③ Systemic health: cardiovascular, BP, smoking

• Quit smoking — linked to progression and impaired optic-nerve perfusion
• Control BP but avoid nocturnal hypotension — if you take antihypertensives, tell your ophthalmologist. Nocturnal BP dipping > 10 mmHg (over-dipping) reduces optic-nerve perfusion and may accelerate NTG progression. Discuss with the internist about shifting antihypertensives to morning if appropriate.
• Manage diabetes and metabolic syndrome

④ Watch out for chronic steroid use

Any form of long-term corticosteroid can raise IOP:

• Asthma inhalers, nasal sprays, skin creams, oral prednisolone, intravitreal injections, joint injections, hemorrhoidal creams (with steroids)
• If you have glaucoma and must take chronic steroids (e.g., autoimmune disease), inform your ophthalmologist and monitor IOP closely
• Some prescriptions also elevate IOP or trigger angle closure (anticholinergics, certain antihistamines, antidepressants, topiramate) — tell every prescriber you have glaucoma

⑤ Not worth over-restricting

• Caffeine: a meta-analysis found IOP rose by about 2 mmHg on average 1-1.5 hours after caffeine in glaucoma/OHT patients, with a smaller effect in other people; the clinical significance remains unclear — no need to quit. In UK Biobank, higher caffeine intake was associated with higher IOP and glaucoma prevalence in people genetically predisposed to elevated IOP, so they may wish to discuss moderation with their doctor (cited by AAO PPP)
• Screen time: phones, computers, TV do not affect glaucoma
• Water: no need to drink huge amounts or restrict — normal intake is fine
• Lutein/zeaxanthin/blueberry/supplements: AREDS2 supports use in AMD but no evidence for glaucoma — save your money

⑥ Get family screened

First-degree relatives of POAG patients have 9.2× higher odds (Rotterdam Eye Study). Recommend:

• AAO recommends regular comprehensive eye exams for people with glaucoma risk factors such as family history: every 1-3 years at 40-54, every 1-2 years from 55, and every 2-5 years under 40 (with IOP, optic-nerve, VF, OCT and gonioscopy testing as needed)
• With additional risks (diabetes, high myopia, thin cornea) or higher-risk ancestry (e.g., African descent), more frequent exams may be needed — your ophthalmologist will set the interval
• Remind family: '20/20 vision' does not exclude glaucoma — VF testing is essential

10. Taiwan NHI Coverage (updated 22 October 2026)

Key rules from the Taiwan NHI drug reimbursement rules (updated 2026-10-22), section 14.1 'Anti-glaucoma drops' (discuss details with your doctor):

• 1st line: β-blockers (e.g., Timolol) — NHI default first choice
• 2nd line (restricted): PGAs (Latanoprost, Travoprost, Bimatoprost), topical CAIs (Dorzolamide, Brinzolamide), α-2 agonist (Brimonidine), Omidenepag — NHI restricts to patients for whom β-blockers are ineffective, unsuitable or contraindicated; PGAs and omidenepag should start as monotherapy, with other IOP-lowering drops added only if control is insufficient, and omidenepag may not be combined with a PGA (14.1.1)
• Fixed combinations (Cosopt, Combigan, DuoTrav, etc.): restricted to 'inadequate IOP control after monotherapy', 2nd-line or later
• No co-administration within same class (can't use two PGAs or two β-blockers simultaneously)
• Chronic refills limited to ophthalmologists (since June 2025) with the IOP from that visit documented; non-ophthalmologists can only prescribe 1 month at a time, must instruct the patient to have IOP measured, refer to ophthalmology for IOP measurement within 3 months at most, and arrange regular optic-nerve and visual-field follow-up. In NHI-designated under-served areas and mountain/offshore-island areas, a local doctor may also issue chronic refills under the NHI telemedicine program after a joint video consultation with an ophthalmologist, with IOP data attached
• Preservative-free single-dose units also covered; quantity caps by daily frequency per 4 weeks, one or both eyes (QD ≤ 30, BID ≤ 60, TID ≤ 90, QID ≤ 120) — good for long-term use or comorbid DED
• Multi-dose bottles have quantity caps by bottle size and dosing frequency — e.g., a once-daily bottle of ≤ 3 mL (such as Xalatan): 1 bottle per 4 weeks for one eye, 1 per 3 weeks for both eyes (4 per 3 months); discuss details with your doctor

For the full class-by-class table, prescription caps, and coverage of SLT laser / MIGS / traditional surgery, see the NHI section in the Glaucoma Treatment Selection article.

* Source: Taiwan NHI drug reimbursement rules, updated 22 October 2026 (ROC 115-10-22), section 14.1 (14.1.1 single-agent, 14.1.2 combinations). Rules update periodically; verify latest at NHIA.

11. Common myths — Q&A

Key references

1. [AAO 2025 主要指引] American Academy of Ophthalmology. Primary Open-Angle Glaucoma Preferred Practice Pattern. 2025. Available at: https://www.aao.org/preferred-practice-pattern/primary-open-angle-glaucoma-ppp
2. [AAO 2025 急性閉角型] American Academy of Ophthalmology. Primary Angle-Closure Disease Preferred Practice Pattern. 2025. Available at: https://www.aao.org/preferred-practice-pattern/primary-angle-closure-ppp
3. [NICE 2022] National Institute for Health and Care Excellence. Glaucoma: diagnosis and management (NG81). Last updated 2022. Available at: https://www.nice.org.uk/guidance/ng81
4. [歐洲指引] European Glaucoma Society. Terminology and Guidelines for Glaucoma, 6th Edition. Br J Ophthalmol. 2025;109(Suppl 1):1-212.
5. [台灣本土資料] Kuo CY, Ko YC, Kuang TM, et al. Prevalence of glaucoma in the elderly population in Taiwan: The Shihpai Eye Study. J Chin Med Assoc. 2020;83(9):880-884.
6. [LiGHT trial] Gazzard G, Konstantakopoulou E, Garway-Heath D, et al. Selective laser trabeculoplasty versus eye drops for first-line treatment of ocular hypertension and glaucoma (LiGHT): a multicentre randomised controlled trial. Lancet. 2019;393(10180):1505-1516（NICE 與 AAO PPP 引述）.
7. [LiGHT 6 年延伸] Gazzard G, Konstantakopoulou E, Garway-Heath D, et al. Laser in Glaucoma and Ocular Hypertension (LiGHT) Trial: Six-Year Results of Primary Selective Laser Trabeculoplasty versus Eye Drops for the Treatment of Glaucoma and Ocular Hypertension. Ophthalmology. 2023;130(2):139-151（AAO PPP 引述）.
8. Heijl A, Leske MC, Bengtsson B, et al; Early Manifest Glaucoma Trial Group. Reduction of intraocular pressure and glaucoma progression: results from the Early Manifest Glaucoma Trial. Arch Ophthalmol. 2002;120(10):1268-1279（AAO PPP 引述）.
9. Collaborative Normal-Tension Glaucoma Study Group. The effectiveness of intraocular pressure reduction in the treatment of normal-tension glaucoma. Am J Ophthalmol. 1998;126(4):498-505（AAO PPP 引述）.
10. Wormald R, Virgili G, Azuara-Blanco A. Systematic reviews and randomised controlled trials on open angle glaucoma. Eye (Lond). 2020;34(1):161-167.
11. American Academy of Ophthalmology. Comprehensive Adult Medical Eye Evaluation Preferred Practice Pattern. 2025. Ophthalmology. 2026;133(4):P202-P236.
12. Chang SH, Lee JI, Chen SC, Huang SP, Geng JH. A Blurred Vision of Health: Metabolic Syndrome as a Risk Factor for Glaucoma in a Large Taiwanese Population Study. Int J Med Sci. 2026;23(1):216-226.
13. Li M, Wang M, Guo W, Wang J, Sun X. The effect of caffeine on intraocular pressure: a systematic review and meta-analysis. Graefes Arch Clin Exp Ophthalmol. 2011;249(3):435-442.
14. Jiwani AZ, Rhee DJ, Brauner SC, et al. Effects of caffeinated coffee consumption on intraocular pressure, ocular perfusion pressure, and ocular pulse amplitude: a randomized controlled trial. Eye (Lond). 2012;26(8):1122-1130.
15. Simanjuntak OP, Jocelyn O, Simanjuntak KT, et al. Caffeine Intake Effects on Intraocular Pressure in Healthy and Glaucomatous Populations: A Systematic Review and Meta-Analysis. J Curr Ophthalmol. 2026;38(1):1-16.
16. Kass MA, Heuer DK, Higginbotham EJ, et al. The Ocular Hypertension Treatment Study: a randomized trial determines that topical ocular hypotensive medication delays or prevents the onset of primary open-angle glaucoma. Arch Ophthalmol. 2002;120(6):701-713.
17. Gordon MO, Beiser JA, Brandt JD, et al. The Ocular Hypertension Treatment Study: baseline factors that predict the onset of primary open-angle glaucoma. Arch Ophthalmol. 2002;120(6):714-720.