Laser peripheral iridotomy across the spectrum ofprimary angle closureSurinder Singh Pandav, MD; Sushmita Kaushik, MD; Rajeev Jain, MD; Reema Bansal, MD;

Amod Gupta, MD

ABSTRACT • RÉSUMÉ

Background: To study the effectiveness of Nd:YAG laser peripheral iridotomy (LPI) for primary angle closurein Asian Indian patients.

Methods: Retrospective analyses of patients who underwent LPI and completed a minimum follow-up of 2years. Eyes were classified as primary angle-closure suspects (PACS), primary angle closure (PAC), andprimary angle-closure glaucoma (PACG).The indications for LPI, requirement of medication, and subsequentclinical course were studied in each group.

Results: 103 eyes of 55 patients were analyzed.The mean (SD) follow-up was 45.6 (2) months.The mean agein women was less than in men (55.7 [8.3] vs. 62.1 [7.8] years).Twenty-seven eyes were classified as PACS,43 eyes as PAC, and 33 eyes as PACG. After LPI, no eye with PACS progressed to PAC or PACG. Four of 43eyes (9.3%) with PAC progressed to PACG.Twenty-five of the 33 eyes (75.8%) with PACG did not progressafter LPI during the study period. Patients with ≤2 quadrants of angle closure at baseline had 7.7% odds ofprogression compared with 100% odds in patients with >2 quadrants of angle closure (risk ratio 12.9).

Interpretation: After LPI, the rate of progression from PAC to PACG was less than expectedfrom the reported natural course of the disease, and the majority of eyes with PACGremained stable. LPI appears to alter the natural course of PACS, PAC, and PACG favourably.

Contexte : Étude de l’efficacité de l’iridotomie périphérique au laser Nd:YAG (IPL) pour les cas de fermeturede l’angle primaire (CFAP) chez des patients indiens d’Asie.

Méthodes : Analyses rétrospectives de patients qui ont subi une IPL et complété un minimum de suivi de 2ans. On a réparti les yeux comme suit : soupçons de cas de fermeture de l’angle primaire (SFAP), cas defermeture de l’angle primaire (CFAP) et glaucome primitif par fermeture de l’angle (GPFA). Les indicationsd’IPL, de besoins de médication et de suivis cliniques subséquents ont été étudiées pour chaque groupe.

Résultats : En tout, 103 yeux de 55 patients ont été analysés. Le suivi moyen (ÉT) a été de 45,6 (2) mois. L’âgemoyen des femmes a été inférieur à celui des hommes (55,7 [8,3] c. 62,1 [7,8] ans).Vingt-sept yeux ont étéclassé parmi les SFAP, 43 yeux parmi les CFAP et 33 parmi les GPFA.Après l’IPL, aucun œil SFAP n’a progressécomme CFAP ou GPFA. Quatre des 43 yeux CFAP (9,3%) ont progressé vers le GPFA. Vingt-cinq des 33yeux atteints de GPFA (75,8%) n’ont pas progressé après l’IPL pendant la durée de l’étude. Les patients quiprésentaient une fermeture d’angle ≤2 quadrants au départ avaient 7,7% de chances de progressercomparativement à 100% chez les patients avec une fermeture d’angle >2 quadrants (ratio de risque 12,9).

Interprétation : Après l’IPL, le taux de progression des CFAP à GPFA était plus faible que prévuselon le rapport d’évolution naturelle de la maladie et la majorité des yeux avec GPFAdemeurèrent stables. L’IPL semble modifier favorablement l’évolution naturelle des SFAP,CFAP et GPFA.

Primary angle-closure glaucoma (PACG) has beenrecently recognized as an important cause of blind-

ness worldwide, especially in Asian countries.1,2 PACGamong Asian people often develops insidiously without

the classical symptomatic episodes occurring.3–8 Eyeswith appositional closure or occludable angles are at riskfor peripheral anterior synechiae, elevated intraocularpressure (IOP), and angle closure glaucoma. Laser

From the Department of Ophthalmology, Postgraduate Institute ofMedical Education and Research, Chandigarh, India

Originally received June 14, 2006. Revised Oct. 23, 2006Accepted for publication Oct. 24, 2006

Correspondence to: Surinder Singh Pandav, MD, Department ofOphthalmology, Postgraduate Institute of Medical Education and Research,Chandigarh - 160012, India; fax 91-172-2744401; sspandav@yahoo.com

This article has been peer-reviewed. Cet article a été évalué par les pairs.

Can J Ophthalmol 2007;42:233–7doi: 10.3129/can j ophthalmol.i07-012

Laser iridotomy in primary angle closure—Pandav et al 233

peripheral iridotomy (LPI) is used as a therapeuticmodality for narrow, occludable angles at risk for closure.

Recently published reports have suggested that LPI isnot needed for all cases of primary angle-closure suspects(PACS) in India, given their vast numbers and low riskfor progression to PACG.9,10 However, there is paucityof data on the natural course of PACS eyes and the effectof LPI on these eyes. This study was carried out to ascer-tain the effect of LPI by Nd:YAG laser on the course ofPACS, primary angle closure (PAC), and PACG in AsianIndian patients.

METHODS

Records of patients from the Glaucoma Clinic of theDepartment of Ophthalmology, Postgraduate Institute ofMedical Education and Research, Chandigarh, India wereretrospectively analyzed. The study was approved by theethics committee of the Postgraduate Institute of MedicalEducation and Research, Chandigarh, and conforms tothe principles enshrined in the Declaration of Helsinki.

All patients who underwent Nd:YAG LPI between June2000 and June 2002 and completed a minimum follow-up of 2 years were included in the study. Written informedconsent was obtained from each patient before the proce-dure. Patients with significant cataract accounting forunaided vision less than 20/60, those with unreliablevisual fields, or patients who had undergone any ocularsurgical procedure were excluded from the analysis.

GonioscopyGonioscopy was performed by experienced gonio-

scopists (Drs. Pandav and Kaushik) in a semi-darkenedroom with the minimum possible slit-lamp illumination,using a Sussman four-mirror goniolens. Care was takento ensure that the slit-lamp beam did not fall across thepatient’s pupil, so that there was no artificial opening ofthe angle. The angles were graded numerically (0 = closedto 4 = wide open) according to Shaffer’s classification.

On the basis of the results of the gonioscopy examina-tions indicating stage of disease, patients were assignedto one of three groups, which were defined as follows:11

• PACS: an eye where appositional contact between theperipheral iris and posterior trabecular meshwork wasconsidered possible (i.e., an occludable angle); no evi-dence of trabecular obstruction by the peripheral iris,such as peripheral anterior synechiae, elevated IOP, orglaucomflecken; normal optic disc and visual fields.

• PAC: an eye with an occludable angle; features sug-gestive of trabecular obstruction by the peripheral irisas described above; normal optic disc and visual fields.

• PACG: an eye with occludable angles; evidence oftrabecular obstruction as in PAC; glaucomatousoptic neuropathy and (or) visual field defects.

Angles were considered occludable if the posterior partof the trabecular meshwork was visible in less than 180°of the angle on gonioscopy. Glaucomatous optic neu-ropathy was defined as a disc with a vertical cup-to-discratio >0.7, asymmetry of the interocular cup-to-discratio >0.2, or specific features such as a focal notch ordisc hemorrhage.11 A glaucomatous visual field defectwas considered in the presence of a glaucoma hemifieldtest graded “outside normal limits” and a cluster of 3 ormore contiguous non-edge points at the p < 5% level onthe pattern deviation plot using the threshold test strat-egy with the 24-2 test pattern of the Zeiss-HumphreyField Analyzer II.11

The pre-intervention data collected were patientdemography, family history, IOP, and extent of angleclosure defined by the number of quadrants per anglewhere no angle structures were visible (grade 0). Post-intervention variables were post-laser IOP, requirementfor medication, and subsequent clinical course.

LPI was considered a success if the disease stabilizedand a failure if the disease progressed. Progression wasevidenced by worsening of disc or visual field parametersin the follow-up period or by either an increase in thenumber of drugs or the requirement of surgery for IOPcontrol. The target IOP was set at 22.0 mm Hg for PACand PACS patients, while the target IOP for PACGpatients was set according to the baseline IOP and stageof disease.

Worsening of the disc was assessed by disc photo-graphs where available or by reviewing the disc diagramsfrom the clinic records. Any evidence of thinning of theneuroretinal rim compared with baseline, or the appear-ance of rim notching, disc hemorrhage, or nerve fibrebundle defect was considered evidence of worsening ofdisc parameters. Worsening on the visual fields was con-sidered if the glaucoma change probability analysis(Statpac 2, Humphrey Field Analyzer) marked any testlocation showing significant deterioration (p < 0.05)from baseline on 2 consecutive occasions.

AnalysisThe results were analyzed using SPSS software for

Windows (version 10.0, SPSS Inc., Chicago, Ill.). Theresults were considered significant at p < 0.05.

RESULTS

A total of 103 eyes of 55 patients (28 men and 27

Laser iridotomy in primary angle closure—Pandav et al

234 CAN J OPHTHALMOL—VOL. 42, NO. 2, 2007

women) were analyzed. Forty-eight patients had a LPI inboth eyes. Of the 7 patients with only 1 eye included inthe study, the other eye was aphakic in 2 patients, 3patients had undergone trabeculectomy, 1 patient hadabsolute glaucoma, and 1 patient was one-eyed. The mean(SD) age in all patients was 59.2 (8.6) years (range 40–79years). The mean (SD) age was 55.7 (8.3) years in femalesand 62.1 (7.8) years in males. Twenty-seven eyes (26.2%)were classified as PACS, 43 eyes (41.7%) as PAC, and 33eyes (32.1 %) as PACG before LPI. The mean age ofpatients with PACS, PAC, and PACG was 54.3 (10.2)years, 60 (6.9) years, and 61.8 (7.6) years, respectively,with a significant difference (p = 0.006, Dunnett 2-sidedt test) between PACS and PAC (Table 1).

The mean follow-up after LPI was 45.6 (2) months. Inthis period, none of the 27 patients with PACS pro-gressed to PAC or PACG (Table 1).

The course of PAC is detailed in Fig. 1. Of the 43 eyes,there was no change in the status after LPI in 32 (74.4%)eyes. In 7 eyes (16.3%), the IOP was controlled even afterwithdrawing previously used medications. Four eyes

(9.3%) of 3 patients in this group progressed to PACG inspite of LPI, including both eyes in 1 patient. The discpicture was not available for 1 of these patients with uni-lateral disease. Progression in all 4 eyes was detected byrepeatable new visual field defects (early nasal steps). Themean age of those who progressed to PACG was 52.0(1.9) years compared with 62.1 (6.0) years in those whodid not (p = 0.005, ANOVA). Two of the 3 patientsmissed their post-treatment follow-ups and came 9months and 13 months after LPI, exhibiting high IOP atthat time. The third patient with bilateral disease had afamily history of PACG. She progressed despite regularfollow-up and IOP well within the set target.

Twenty-five of the 33 eyes (75.8%) with PACGremained stable after LPI. Of the 8 eyes that progresseddespite LPI, 4 eyes could be managed by increasing thenumber of drugs used for IOP control, while 4 eyes sub-sequently required filtering surgery (Table 1).

The degree of angle closure before intervention wasascertained by determining the number of quadrantswhere the angle was completely closed (no angle structuresvisible). In the eyes that progressed, the mean (SD)number of closed quadrants per angle were 2.33 (0.9)compared with 0.82 (0.8) in those eyes that stabilized afterLPI (p ≤ 0.001, Mann–Whitney U test). Patients were alsoanalyzed by subset according to the degree of angle closureat presentation (Table 2). Those with ≤2 quadrants ofangle closure at baseline had 7.7% odds of progressioncompared with 100% in patients with >2 quadrants ofangle closure at presentation (risk ratio 13.0).

A cohort of patients who had never used any medica-tion before LPI was separately analyzed (Table 3). Of the

Laser iridotomy in primary angle closure—Pandav et al

CAN J OPHTHALMOL—VOL. 42, NO. 2, 2007 235

Table 1—Mean age of patients by diagnosis

Diagnosis nAge

mean (SD), y p value

(age difference) Progres-sion*

PACS 27 54.3 (10.2) 0PAC 43 60.0 (6.9) 0.006† (PACS vs. PAC) 4 PACG 33 61.8 (7.6) 0.42 (PAC vs. PACG) 8Total 103 59.2 (8.6) 12

Note: PACS = primary angle-closure suspect; PAC, primary angle clo-sure; PACG, primary angle-closure glaucoma. *Number of patients showing progression after laser peripheral iridotomy. †Difference in ages statistically significant by Dunnett 2-sided t test.

Fig. 1—Course of eyes with primary angle closure (PAC) after laser peripheral iridotomy.IOP = intraocular pressure; PACG, primary angle-closure glaucoma.

37 eyes in this group, 34 had an IOP <22.0 mm Hgbefore LPI. No eye in this group progressed, comparedwith 2 of the 3 eyes (whose IOP was ≥22.0 mm Hg atpresentation), which progressed (p < 0.001).

The patients’ family histories were analyzed withrespect to their role in progression. Of the 55 patients, 4had a history of angle closure in either parent or sibling.All 4 patients were treated bilaterally, of which both eyesof 1 patient progressed (25%). Ten of the 95 eyes ofpatients with no family history progressed despite LPI(10.5 %). The odds of progression in those patients witha family history were 33%, compared with 11.8% inthose with no family history. This translates into a riskratio of 2.8, or an increased risk of progression 2.8 timesgreater in those with a family history of glaucoma com-pared with those with no family history of glaucoma.

No patients developed any serious complications afterthe procedure. During the procedure, 5 eyes developedtransient bleeding, which was controlled in all eyes byapplying temporary pressure on the Abrahams lens.

INTERPRETATION

Angle closure is a significant problem in India.3,12,13

The Andhra Pradesh Eye Disease Study3 reported 2.21%of the population over 40 years of age had occludableangles at risk of angle closure and 1.08% had manifestPACG, a large proportion of whom were undiagnosedand untreated. In the Vellore Eye Study,12 manifest

PACG was as high as 4.3%. Angle closure glaucomaconstituted 46% of all primary adult glaucoma seen in atertiary care hospital in North India.13

Laser iridotomy eliminates pupillary block, enablingthe convex iris to flatten, and thus deepens the anteriorchamber angle. It is generally agreed that eyes withoccludable iridocorneal angles should undergo laser iri-dotomy.14,15 Since visual loss resulting from PACG ispotentially preventable if peripheral iridotomy is per-formed in the early stage, strategies for early detection ofPAC could reduce the risk of blindness resulting fromPACG in India.

In our study, with a mean follow-up of nearly 4 yearsafter undergoing LPI, no patient with PACS progressedto PAC, and 9.3% with PAC progressed to PACG. Thisis significantly less than the natural history of progres-sion reported in the retrospective review of PACS9 andPAC.16

The mean age of presentation of the three stages of thedisease increased from 54.3 years for PACS to 61.8 yearsfor PACG, reflecting the progression over time.Increasing lens thickness with age may have been a con-tributing factor. The mean age of presentation was 6.4years earlier in females than in male patients. Youngerage was significantly associated with progression toPACG in spite of LPI, while those who normalized were,on average, 10 years older. A larger population-basedstudy is needed to answer the question of whether thedisease course is more severe if it manifests earlier.Another factor that appears to have contributed to pro-gression was the higher IOP at presentation, probablyrepresenting an increased extent of peripheral anteriorsynechiae. Of the 4 eyes with PAC who progressed toPACG, 3 had baseline IOP > 22.0 mm Hg, and 1 wason antiglaucoma medication before LPI. This was alsoindicated by the greater degree of angle closure seen inthe eyes of those who progressed, which was reflected inmore “grade 0” quadrants per angle compared with thosewho stabilized after LPI. A family history of angleclosure also contributed to increased rates of progression.Our study showed an odds ratio of 2.8 in favour of pro-gression in patients with a family history of glaucoma. Itmust be kept in mind, however, that we had to rely ondisc drawings to indicate progression in several patientswhere disc pictures were not available. Disc drawings arenotoriously poorly reproducible, and the fine detailsrequired to indicate progression in the few patients thatdid not have disc pictures may not have been accuratelyrecorded.

One limitation of this study is that there was nocontrol group for PACS patients. This group was not ini-tially randomized into a treated and non-treated control

Laser iridotomy in primary angle closure—Pandav et al

236 CAN J OPHTHALMOL—VOL. 42, NO. 2, 2007

Table 2—Disease progression by degree of angle closure at presentation

No. of patients (%) No. of angle quadrants closed at presentation Progressed Stabilized Total

Odds of progression

%

0 1 (3.8) 25 (96.2) 26 41 2 (4.5) 56 (96.5) 58 3.6 2 4 (28.6) 10 (71.4) 14 403 2 (100) 0 (0) 2 100 4 3 (100) 0 (0) 3 100Total 12 (11.6) 91 (88.3) 103 13.2 ≤ 2 7 (7.1) 91 (92.9) 98 7.7> 2 5 (100) 0 (0) 5 100

Table 3—IOP in patients on no medication before LPI

No. of patients on no medication (%)

IOP, mm Hg

Diagnosis n < 22 ≥ 22 Total

PACS 27 27 (100) 0 (0) 27 (100) PAC 43 7 (16.3) 2 (4.6) 9 (20.9) PACG 33 0 (0) 1 (3) 1 (3) Total 103 34 (33.0) 3 (2.9) 37 (35.9)

Note: LPI = laser peripheral iridotomy; IOP, intraocular pressure; PACS, primary angle-closure suspect; PAC, primary angle closure; PACG, primary angle-closure glaucoma.

group, and because this was a retrospective study utiliz-ing records of patients who underwent LPI, we have noinformation about PACS patients who may not havebeen offered or may have refused laser iridotomy. A pop-ulation-based study from South India revealed that 22%of PACS did progress to PAC in the absence of treat-ment,9 and 28.5% of PAC progressed to PACG16 after 5years. In the latter group, all 28 patients with PAC wereadvised to undergo LPI, of whom only 9 complied; 1 ofthe 9 with LPI progressed to PAC compared with 7 ofthe 19 who refused treatment. One needs to be cautiousin making direct comparisons, however, considering theshorter follow-up period of our study.

It has been argued that LPI is not indicated for allpatients with PACS because of the large numbers ofaffected patients and their relatively lower risk of pro-gression. Nonetheless, angle-closure glaucoma has apotential to cause severe irreversible visual loss, and LPIis a very effective one-time intervention to prevent suchan occurrence provided it is done in time. It may bebetter to offer LPI at the PACS stage in selected patientswho are at higher risk of developing PAC or PACG, suchas patients with a family history of PACG, those who areunlikely to return for follow-up, or where consequencesof developing acute attack are likely to be severe andunacceptable, such as in one-eyed persons with PACS.

Our study demonstrates that LPI may favourably alterthe natural course of angle closure. PACS and patientswith PAC are likely to be cured, and those with PACGmay benefit in terms of disease stabilization. In theabsence of a control group in which LPI was not per-formed, however, these conclusions are at best specula-tive. Larger prospective studies on the course of treatedand untreated eyes with narrow angles are needed toprovide the final answer to the question of which factorsat baseline are predictive of those at greatest risk of pro-gression.

The authors have no proprietary or financial interest in anyproduct mentioned in the article.

REFERENCES

1. Thylefors B, Négrel AD. The global impact of glaucoma. BullWorld Health Organ 1994;72:323–6.

2. Quigley HA. Number of people with glaucoma worldwide. BrJ Ophthalmol 1996;80:389–93.

3. Dandona L, Dandona R, Mandal P, et al. Angle-closure glau-coma in an urban population in southern India. The AndhraPradesh eye disease study. Ophthalmology 2000;107:1710–6.

4. Salmon JF, Mermoud A, Ivey A, Swanevelder SA, HoffmanM. The prevalence of primary angle-closure glaucoma andopen angle glaucoma in Mamre, Western Cape, South Africa.Arch Ophthalmol 1993;111:1263–9.

5. Congdon N, Quigley HA, Hung PT, Wang TH, Ho TC.Screening techniques for angle-closure glaucoma in ruralTaiwan. Acta Ophthalmol 1996;74:113–9.

6. Foster PJ, Baasanhu J, Alsbirk PH, Munkhbayar D,Uranchimeg D, Johnson GJ. Glaucoma in Mongolia. A pop-ulation-based survey in Hövsgöl Province, northern Mongolia.Arch Ophthalmol 1996;114:1235–41.

7. Foster PJ, Oen FT, Machin DS, et al. The prevalence of glau-coma in Chinese residents of Singapore: a cross-sectional pop-ulation survey of the Tanjong Pagar district. Arch Ophthalmol2000;118:1105–11.

8. Oh YG, Minelli S, Spaeth GL, Steinman WC. The anteriorchamber angle is different in different racial groups: a gonio-scopic study. Eye 1994;8:104–8.

9. Thomas R, George R, Parikh R, Muliyil J, Jacob A. Five yearrisk of progression of primary angle closure suspects to primaryangle closure: a population based study. Br J Ophthalmol2003;87:450–4.

10. Thomas R, Sekhar GC, Kumar RS. Glaucoma managementin developing countries: medical, laser, and surgical options forglaucoma management in countries with limited resources.Curr Opin Ophthalmol 2004;15:127–31.

11. Foster PJ, Buhrmann R, Quigley HA, Johnson GJ. The defi-nition and classification of glaucoma in prevalence surveys. BrJ Ophthalmol 2002;86:238–42.

12. Jacob A, Thomas R, Koshi SP, Braganza A, Muliyil J.Prevalence of primary glaucoma in an urban south Indianpopulation. Indian J Ophthalmol 1998;46:81–6.

13. Sihota R, Agarwal HC. Profile of the subtypes of angle closure glau-coma in a tertiary hospital in north India. Indian J Ophthalmol1998;46:25–9.

14. Slamovits T, Dutton JJ. Should patients with anatomicallynarrow angles have prophylactic iridectomy? Surv Ophthalmol1996;41:31–6.

15. Ritch R. Assessing the treatment of angle closure. Ophthalmology2003;110:1867–8.

16. Thomas R, Parikh R, Muliyil J, Kumar RS. Five-year risk ofprogression of primary angle closure to primary angle closureglaucoma: a population-based study. Acta Ophthalmol Scand2003;81:480–5.

Key words: laser iridotomy, primary angle closure

Laser iridotomy in primary angle closure—Pandav et al

CAN J OPHTHALMOL—VOL. 42, NO. 2, 2007 237