Retrospective Analysis of Long-term

8/10/2019 Retrospective Analysis of Long-term

1/8

A retrospective analysis of long-termoutcomes following a single episodeof transscleral cyclodiode laser

treatment in patients with glaucomaIvailo Zhekov, 1 Razia Janjua, 1 Humma Shahid, 1 Nicholas Sarkies, 1

Keith R Martin, 1,2 Andrew J R White 1,2,3

To cite: Zhekov I, Janjua R,Shahid H, et al . Aretrospective analysis oflong-term outcomesfollowing a single episode

of transscleral cyclodiodelaser treatment in patientswith glaucoma. BMJ Open 2013;3:e002793.doi:10.1136/bmjopen-2013-002793

Prepublication history forthis paper is available online.To view these files pleasevisit the journal online(http://dx.doi.org/10.1136/ bmjopen-2013-002793 ).

Received 26 February 2013Revised 27 May 2013Accepted 29 May 2013

This final article is availablefor use under the terms ofthe Creative CommonsAttribution Non-Commercial3.0 Licence; seehttp://bmjopen.bmj.com

1 School of Clinical Medicine,Cambridge UniversityTeaching Hospitals NHSFoundation Trust, Cambridge,UK2 NIHR Biomedical ResearchCentre, University ofCambridge, Cambridge, UK3 Centre for Vision Research,Westmead MillenniumInstitute, Sydney, Australia

Correspondence toDr Ivailo Zhekov;[email protected]

ABSTRACTObjectives: To investigate the efficacy of a singlecyclodiode laser photocoagulation treatment forrefractory glaucoma and its effect on visual outcome in

patients with good visual potential as well as toevaluate possible predictive factors in establishingoptimal treatment parameters.Design: Retrospective observational study.Setting: Tertiary referral centre.Participants: The notes of 87 patients with refractoryglaucoma who underwent cyclodiode photocoagulationas a first surgical intervention over a 7-year period.Main outcome measures: Maintenance ofintraocular pressure (IOP) reduction, number ofmedications and visual acuity outcomes post-treatment.Results: The mean IOP after a single treatment decreasedfrom 39.51.3 to 17.81.5 mm Hg after a 6-week

follow-up period (p0.05) with relatively good visualpotential (average vision preoperatively was 0.57 logMAR).Hypotony occurred in 5.3% of patients. No patientsrequired enucleation or evisceration.Conclusions: A single session of cyclodiode lasertherapy was associated with significant IOP reduction in amajority of patients with refractory glaucoma. The majoritywere able to maintain the IOP reduction over a 3-yearperiod without the need for a further surgical intervention.Additionally, over 80% of the patients in our study wereable to maintain their baseline visual acuity. These resultssupport the view that a single cyclodiode treatment can besufficient in achieving long-term IOP control and may beconsidered in eyes with relatively good visual potential.

INTRODUCTIONTransscleral diode laser cyclophotocoagula-tion (cyclodiode) has been established as a

relatively safe and effective intervention forglaucoma. 1

6 Cyclodiode is often used inrefractory glaucoma, where alternative surgi-cal approaches, such as antiproliferative aug-mented trabeculectomy and tube shunt surgery, may sometimes be judged less appro-priate, for example, in eyes with poor visualpotential. Additionally, cyclodiode has beenshown to be safer than other cyclodestructiveprocedures, such as Nd:YAG laser cyclopho-tocoagulation and cyclocryotherapy, which

ARTICLE SUMMARY

Article focus To investigate the efficacy of a single cyclodiode

laser photocoagulation treatment for refractoryglaucoma.

To determine the effect of a single cyclodiodelaser photocoagulation treatment on visualoutcome in patients with good visual potential.

To evaluate the possible predictive factors inestablishing optimal treatment parameters.

Key messages A single session of cyclodiode laser therapy was

associated with significant intraocular pressure(IOP) reduction in the majority of patients withrefractory glaucoma.

The majority of patients were able to maintain

the IOP reduction over a 3-year period withoutthe need for a further surgical intervention.

Over 80% of patients in our study were able tomaintain their baseline visual acuity.

Strengths and limitations of this study The retrospective nature of the study design

reflected the need to analyse multiple clinicalvariables and evaluate different types of out-comes as well as their progression over severalyears. We aimed to avoid selection bias byincluding all patients with clearly annotated pre-operative and postoperative parameters in theelectronic medical record system.

Zhekov I, Janjua R, Shahid H, et al . BMJ Open 2013;3:e002793. doi:10.1136/bmjopen-2013-002793 1

Open Access Research

group.bmj.comon November 12, 2014 - Published by http://bmjopen.bmj.com/ Downloaded from
http://dx.doi.org/10.1136/bmjopen-2013-002793http://dx.doi.org/10.1136/bmjopen-2013-002793http://bmjopen.bmj.com/http://group.bmj.com/http://group.bmj.com/http://bmjopen.bmj.com/http://bmjopen.bmj.com/http://group.bmj.com/http://bmjopen.bmj.com/http://bmjopen.bmj.com/http://dx.doi.org/10.1136/bmjopen-2013-002793http://dx.doi.org/10.1136/bmjopen-2013-002793


2/8


3/8

or required further treatment or enucleation for painsymptoms ( gure 1B). The data for the patients whodeveloped hypotony such as initial diagnosis, treatment settings, IOP measurements and VA outcome are sum-marised in table 2.

Long-term maintenance of IOP reduction after a singlecyclodiode treatmentThe IOP reduction was maintained for a long term over aperiod of 3 years in the majority of our patients.Measurements were taken preoperatively (39.51.3 mm Hg) as well as postoperatively at 6 weeks (17.81.5 mm Hg) and 6 months (19.61.5 mm Hg) for allpatients ( gure 1C). Patients who required additional pro-cedures in that eye post-treatment were excluded fromfurther analysis from the time point of additional interven-tion as it would arti cially lower their IOP measurementsThe follow-up measurements at 1 (18.9 mm Hg),2 (22.1 mm Hg) and 3 years (21.7 mm Hg) were all after asingle cyclodiode treatment.

Kaplan-Meier analysis of IOP reduction survival

following a single cyclodiode treatmentSuccess was de ned as an IOP of 6 21 mm Hg at the last follow-up visit without the need for oral acetazolamideand an IOP reduction of at least 30% compared withpreoperatively ( gure 2 ). The proportion of patients tomaintain the desired IOP reduction after a single cyclo-diode treatment without the need for furtherIOP-lowering intervention in our study was 67.7% at 6 weeks postoperatively, 66.2% at 6 months, 63% at 1 year, 61.2% at 2 years and 61.2% at 3 years. Of thepatients who responded to treatment at 6 weeks, 90.4% were able to maintain the IOP reduction over a periodof 3 years.

Figure 1 (A) Change in intraocular pressure (IOP) followinga single cyclodiode treatment. (B) Safety and efficacy ofcyclodiode treatment. (C) IOP reduction was maintained longterm over a period of 3 years after single treatment.

Table 1 Baseline demographics of the patients andclinical parameters including diagnostic features,preoperative IOP and visual acuity assessments

Category Cases (n=87)

Mean age (years) 66.3Sex (n (%))

Male 49 (56.3)Female 38 (43.7)

Preoperative VA (mean) 0.57 LogMARIOP (mm Hg), mean (SD) 39.51.3 mm HgMean number of glaucoma medications 2.6Glaucoma type (n (%))

Primary open angle glaucoma (POAG) 33 (37.9)Primary angle closure glaucoma(PACG)

6 (6.9)

Neovascular glaucoma 25 (28.7)Uveitic glaucoma 5 (5.7)Secondary open angle glaucoma(SOAG) other

18 (20.8)

IOP, intraocular pressure; VA, visual acuity.


Long-term outcomes after a single cyclodiode treatment in patients with glaucoma

http://group.bmj.com/http://group.bmj.com/http://bmjopen.bmj.com/http://bmjopen.bmj.com/http://group.bmj.com/http://bmjopen.bmj.com/


4/8

Cyclodiode effect on glaucoma medications A signi cant proportion of the patients (61.5%) wereable to decrease the number of medications they aretaking for IOP control following cyclodiode treatment ( gure 3 A). A decrease of 2 medications or more wasachieved by 34.6% of patients, while 26.9% of patientsdecreased their medication by one. Overall, the averagenumber of medications decreased from 2.6 before cyclo-diode treatment to 1.5 after cyclodiode treatment (p0.05; gure 3B). The data ana-lysis was performed on 47 patients (54.1%) who hadclearly annotated numerical values for their VA measure-ments both preoperatively and postoperatively. Theboxed numerals in the graph represent the number of values at the same point.

Change in visual fieldsThe average visual eld sensitivity remained almost unchanged with an average MD value of 8.74 dB pre-treatment compared to the MD value of 9.05 dB at 6 months post-treatment, which was not statistically sig-ni cant (p>0.05). The data analysis was performed on36 patients (41.3%) who had clearly annotated numer-ical values for their visual elds measurements both pre-operatively and postoperatively ( gure 3C).

Effect of laser energy on IOP reductionOf the patients who received high-energy treatment (90 J), 80.3% obtained pressure reduction of >30% of initial IOP compared with 56.8% in patients whoreceived low-energy treatment (45 J) ( gure 4 A).

Effect of laser energy on VAOf the patients who received high-energy treatment (90 J), 18.6% noted improvement equivalent to at least one Snellen line compared with 6.7% in patients whoreceived low-energy treatment ( gure 4B).

Effect of degrees treated on IOP reductionOf the patients who underwent 360 treatment, 71.4%obtained pressure reduction greater than 30% of initialIOP compared with 55% of patients who underwent 180 treatment ( gure 4C).

Effect of degrees treated on VAOf the patients who underwent 360 treatment, 20.9%

noted improvement equivalent to at least one Snellenline compared with 5.5% of patients who underwent 180 treatment ( gure 4D).

DISCUSSIONThe results of this retrospective analysis are supportiveof a single cyclodiode treatment being effective in themanagement of refractory glaucoma. Although this is aretrospective rather than a prospective study, we aimedto avoid selection bias by including all patients either assuccess or failure of treatment. The patients whorequired additional surgical interventions were taken asfailure from the time of additional intervention.

Table 2 Details of patients who developed hypotony including diagnostic and treatment features, peri-operative IOPassessments and duration of hypotony

DiagnosisEnergyused

Degreestreated

Peak IOP(mm Hg)

PostopIOP(mm Hg)

Hypotonyduration(months)

Last IOP(mm Hg)

Preop VA(LogMAR)

Last VA(LogMAR)

PT1 Neovascular 90J 180 58 2 6 10 0.78 PLPT2 Neovascular 90J 360 41 4 6 32 PL NPLPT3 Neovascular 120J 360 40 2 6 9 0.6 HMPT4 SOAG (other) 120J 360 26 3 6 8 1.0 CFPT5 SOAG (other) 120J 360 48 2 3 HM HMIOP, intraocular pressure; SOAG, secondary open angle glaucoma.

Figure 2 Kaplan-Meier analysis for a cumulative proportionof success in IOP control after a single cyclodiode treatment.Survival was defined as a sustained intraocular pressure(IOP) drop of at least 30% compared with pretreatment or sustained IOP of 6 21 mm Hg without the need for additionalIOP-lowering medications or further surgical interventions.

4 Zhekov I, Janjua R, Shahid H, et al . BMJ Open 2013;3:e002793. doi:10.1136/bmjopen-2013-002793




5/8

Our data show that 67.7% of patients achieved an IOPreduction greater than 30% of presenting IOP at 6 weeks and 61.5% could reduce the number of medica-tions by at least 1. Hypotony appears to be related to thetype of glaucoma (with neovascular glaucoma being associated with worse outcome) as well as to the numberof degrees treated and total energy used. Indeed, the

patients with neovascular glaucoma in our cohort werealmost four times more likely to develop hypotony com-pared to any other types of glaucoma.

In our patients, the mean IOP decreased signi cantly from 39.5 mm Hg (SE 1.26) before cyclodiode therapy to 17.8 mm Hg (SE 1.51) at 6 weeks post-treatment, anobservable reduction of 45.1% (p


6/8

higher risk of hypotony in eyes with neovascular glau-coma, particularly if the preoperative IOP was high andthe energy level used was greater. This correlationbetween hypotony and higher energy used in cyclodiode

is also observed in several other studies.12

However, uni- variate regression analysis performed by Murphy et al 14

shows that very high pretreatment IOP alone, possibly causing ciliary body ischaemia, could be responsible forthe high incidence of hypotony in patients receiving high-energy treatment in their study.

Additionally, in a retrospective analysis of 209 eyes by Bloom et al ,18 a very low incidence of hypotony (1%) was observed with a high-energy protocol when themean energy used was 90 J. Furthermore, Iliev e t al 17

report in their study that eyes developing hypotony hadnot received higher energies compared with eyes that maintained normal IOP. Therefore, the intention toreduce hypotony risk by using a lower laser power andfewer applications per treatment should be weighedagainst the possibility of a lower ef cacy of the cyclo-diode treatment.

VA remained unchanged for the majority of ourpatients who underwent cyclodiode treatment (69.9%).In our study, 16.4% reported deterioration in VA of oneSnellen line, while 13.7% of the patients reported animprovement in VA of one Snellen line after 6 weeks.The latter probably resulted either from resolution of preoperative corneal oedema or inter-observer variationin VA testing. Similar results are also reported in a study by Murphy e t al ,

14

where VA remained the same in

74.6% of patients at 6 months following cyclodiode treat-ment. In our study, most patients with dif cult tomanage glaucoma retained their good VA in the long term following cyclodiode treatment. The proportion

losing two Snellen lines is actually better than that reported after trabeculectomy or tube surgery. 21 Theseresults suggest a possible role for the use of transscleralcyclodiode treatment in eyes with relatively good visualpotential; however, further controlled prospective studies would be required to better de ne this role.

This study is the rst to measure MD values for thepatients both preoperatively and postoperatively in order toprovide a more objective assessment of glaucoma-related visual loss. The MD values in our patients remained virtu-ally unchanged after cyclodiode treatment from 8.74 dBpreoperatively to 9.05 dB postoperatively. Previously, it has been reported that the most frequent cause of visual loss was further progression of glaucoma, which was an attributable cause in over half of the cases. 22

The peri-operative visual eld assessment presented inthis study suggests that the maintenance of VA observedin the majority of the patients following cyclodiode treat-ment could be due to a good postoperative IOP control.There are no previous retrospective or prospectivestudies evaluating visual eld measurements peri-operatively in patients undergoing cyclodiode treatment and this most likely re ects the dif culties associated withperforming the tests in patients with poor VA. Our study therefore is the rst to provide a standardised measurableevidence of maintenance of patients visual elds as a

Figure 4 (A) Effect of laser energy on intraocular pressure (IOP) reduction. (B) Effect of laser energy on visual acuity (VA). (C)Effect of degrees treated on IOP reduction (D) Effect of degrees treated on VA.

6 Zhekov I, Janjua R, Shahid H, et al . BMJ Open 2013;3:e002793. doi:10.1136/bmjopen-2013-002793




7/8

marker of glaucoma progression following a single cyclo-diode treatment. Our study s main limitations are itsretrospective nature and the absence of a control groupto distinguish the adverse effects of the treatment fromthe natural history of the underlying disease.

The results of this study suggest that the IOP reductionafter cyclodiode treatment could potentially prevent

further deterioration in the vision of patients with glau-coma. Currently, according to the UK NationalCyclodiode Survey Study, only 60% of practitioners wouldperform cyclodiode procedure in the presence of good VA.12 Until recently, cyclodiode treatment has been asso-ciated with subsequent loss of VA in a signi cant percent-age of the patients. 1 8 23 24 However, cyclodiode use ineyes with useful vision has already been proposed inseveral studies. 19 22

25 Furthermore, the risk of VA loss inpatients after multiple cyclodiode treatments does not appear to be any greater than that after any other surgicalmodality used for treating patients with glaucoma asreported in a study by Rotchford et al .24

In our study, the better outcome in VA was most pro-nounced in the high-energy group, probably due to themore effective IOP control. Of the patients who receivedhigh-energy treatment (90 J), 18.6% noted improvement of at least one Snellen line compared with 6.7% inpatients who received low-energy treatment (45J). Similarresults are observed when patients were assessed inregard to the degrees treated. Here, of the patients whounderwent 360 treatment, 20.9% noted improvement equivalent to at least one Snellen line compared with5.5% of the patients who underwent 180 treatment.

In conclusion, conventional diode cyclophotocoagula-

tion is characterised by low incidence of complicationsand therefore appears to be a safe and effective treat-ment for refractory glaucoma. In our study, the IOPpressure was effectively reduced in patients with glau-coma after a single cyclodiode treatment without adverse effects on VA in the majority of the patients overa 3-year period. Hypotony seems to be the main risk of treatment and could be limited by reducing the laserenergy applied to less than 90 J, particularly in patients with neovascular glaucoma.Contributors IZ, RJ, KRM and AJRW conceived the idea of the study and wereresponsible for the design of the study. IZ, HS, KRM and AJRW were responsiblefor undertaking the data analysis and produced the tables and graphs. RJ and NSprovided input into the data analysis. The initial draft of the manuscript wasprepared by IZ, HS, KRM and AJRW and then circulated repeatedly among allauthors for critical revision. IZ and RJ were responsible for the acquisition of thedata, and IZ, RJ, HS, NS, KRM and AJRW contributed to the interpretation of theresults. All authors read and approved the final manuscript.

Funding This research received no specific grant from any funding agency inthe public, commercial or not-for-profit sectors.

Competing interests None.

Ethics approval Cambridge University Hospitals NHS Foundation Trust EthicsCommittee.

Provenance and peer review Not commissioned; externally peer reviewed.

Data sharing statement No additional data are available.

REFERENCES1. Hennis HL, Stewart WC. Semiconductor diode laser transscleral

cyclophotocoagulation in patients with glaucoma. Am J Ophthalmol 1992;113:81 5.

2. Hawkins TA, Stewart WC. One-year results of semiconductor transscleral cyclophotocoagulation in patients with glaucoma. Arch Ophthalmol 1993;111:488 91.

3. Brancato R, Carassa RG, Bettin P. Contact transscleralcyclophotocoagulation with diode laser in refractory glaucoma. Eur J Ophthalmol 1995;5:32 9.

4. Threlkeld AB, Johnson MH. Contact transscleral diodecyclophotocoagulation for refractory glaucoma. J Glaucoma 1999;8:3 7.

5. Egbert PR, Fiadoyor S, Budenz DL. Diode laser transscleralcyclophotocoagulation as a primary surgical treatment for primaryopen angle glaucoma. Arch Ophthalmol 2001;119:345 50.

6. Mistlberger A, Liebmann JM, Tschiderer H. Diode laser transscleralcyclophotocoagulation for refractory glaucoma. J Glaucoma 2001;10:288 93.

7. Benson MT, Nelson ME. Cyclocryotherapy: a review of cases over a10-year period. Br J Ophthalmol 1990;74:103 5.

8. Schuman JS, Bellows AR, Shingleton BJ. Contact transscleral Nd:YAG laser cyclophotocoagulation. Ophthalmology 1992;99:1089 94.

9. Ulbig MW, McHugh DA, McNaught AI. Clinical comparison ofsemiconductor diode versus neodymium:YAG non-contactcyclophotocoagulation. Br J Ophthalmol 1995;79:569 74.

10. Pucci V, Tappainer F, Borin S, et al . Long term follow up after transscleral diode laser photocoagulation in refractory glaucoma.Ophthalmologica 2003;217:279 83.

11. Grueb M, Rohrbach JM, Bartz KU, et al . Transscleral diode laser cyclophotocoagulation as primary and secondary surgical treatmentin primary open angle and pseudoexfoliatve glaucoma: long termclinical outcomes. Graefes Arch Clin Exp Ophthalmol 2006;244:1293 9.

12. Agrawal P, Dulku S, Nolan W, et al . The UK National CyclodiodeLaser Survey. Eye 2011;25:168 73.

13. Walland MJ. Diode laser cyclophotocoagulation: longer term followup of a standardized treatment protocol. Clin Exp Ophthalmol 2000;28:263 7.

14. Murphy CC, Burnett CAM, Spry PDG, et al . A two centre study ofthe dose-response relation for transscleral diode laser cyclophotocoagulation in refractory glaucoma. Br J Ophthalmol 2003;87:1252 7.

15. Chang SH, Chen YC, Li CY. Contact diode laser transscleral

cyclophotocoagulation for refractory glaucoma: comparison of twotreatment protocols. Can J Ophthalmol 2004;39:511 16.16. Noureddin BN, Zein W, Haddad C. Diode laser transcleral

cyclophotocoagulation for refractory glaucoma: a 1 year follow-up ofpatients treated using an aggressive protocol. Eye 2006;20:329 35.

17. Iliev ME, Gerber S. Long-term outcome of trans-scleral diode laser cyclophotocoagulation in refractory glaucoma. Br J Ophthalmol 2007;91:1631 5.

18. Bloom PA, Tsai JC, Sharma K, et al . Cyclodiode trans-scleral diodelaser cyclophotocoagulation in the treatment of advanced refractoryglaucoma. Ophthalmology 1997;104:1508 20.

19. Vernon SA, Koppens JM, Menon GJ, et al . Diode laser cycloablationin adult glaucoma:long-term results of a standard protocol andreview of current literature. Clin Exp Ophthalmol 2006;34:411 20.

20. Kosoko O, Gaasterland DE, Pollack IP. Long-term outcome ofinitial ciliary ablation with contact diode laser transscleralcyclophotocoagulation for severe glaucoma. The diode laser ciliaryablation study group. Ophthalmology 1996;103:1294 302.

21. Gedde SJ, Schiffman JC, Feuer WJ, et al . Treatment outcomes inthe tube versus trabeculectomy study after one year of follow-up. Am J Ophthalmol 2007;143:9 22.

22. Rotchford AP, Jayasawal R, Madhusudhan S, et al . Transscleraldiode laser cycloablation in patients with good vision. Br J Ophthalmol 2010;94:1180 3.

23. Youn J, Cox TA, Allingham RR. Factors associated with visual acuityloss after noncontact transscleral Nd:YAG cyclophotocoagulation.J Glaucoma 1996;5:390 94.

24. Wilensky JT, Kammer J. Long-term visual outcome of transsclerallaser cyclotherapy in eyes with ambulatory vision. Ophthalmology 2004;111:1389 92.

25. Ansari E, Gandhewar J. Long-term efficacy and visual acuityfollowing transscleral diode laser photocoagulation in casesof refractory and non-refractory glaucoma. Eye 2007;21:936 40.





8/8

patients with glaucomatransscleral cyclodiode laser treatment inoutcomes following a single episode ofA retrospective analysis of long-term

Martin and Andrew J R WhiteIvailo Zhekov, Razia Janjua, Humma Shahid, Nicholas Sarkies, Keith R

doi: 10.1136/bmjopen-2013-0027932013 3:BMJ Open

http://bmjopen.bmj.com/content/3/7/e002793Updated information and services can be found at:

These include:

References #BIBLhttp://bmjopen.bmj.com/content/3/7/e002793

This article cites 25 articles, 5 of which you can access for free at:

Open Access

http://creativecommons.org/licenses/by-nc/3.0/legalcode andhttp://creativecommons.org/license s/by-nc/3.0/ with the license. See:

properly cited, the use is non commercial and is otherwise in complia ncedistribution, and reproduction in any medium, provided the original work isCommons Attribution Non-commercial License, which permits use,This is an open-access article distributed under the terms of the Creative

serviceEmail alerting

box at the top right corner of the online article.Receive free email alerts when new articles cite this article. Sign up in the

CollectionsTopic Articles on similar topics can be found in the following collections

(48)Ophthalmology

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:

http://bmjopen.bmj.com/content/3/7/e002793http://bmjopen.bmj.com/content/3/7/e002793http://bmjopen.bmj.com/content/3/7/e002793#BIBLhttp://bmjopen.bmj.com/content/3/7/e002793#BIBLhttp://creativecommons.org/licenses/by-nc/3.0/legalcodehttp://creativecommons.org/licenses/by-nc/3.0/http://creativecommons.org/licenses/by-nc/3.0/legalcodehttp://creativecommons.org/licenses/by-nc/3.0/http://creativecommons.org/licenses/by-nc/3.0/legalcodehttp://creativecommons.org/licenses/by-nc/3.0/http://bmjopen.bmj.com//cgi/collection/bmj_open_ophthalmologyhttp://group.bmj.com/group/rights-licensing/permissionshttp://group.bmj.com/group/rights-licensing/permissionshttp://journals.bmj.com/cgi/reprintformhttp://journals.bmj.com/cgi/reprintformhttp://group.bmj.com/subscribe/http://group.bmj.com/subscribe/http://group.bmj.com/http://group.bmj.com/http://bmjopen.bmj.com/http://bmjopen.bmj.com/http://group.bmj.com/http://bmjopen.bmj.com/http://group.bmj.com/subscribe/http://journals.bmj.com/cgi/reprintformhttp://group.bmj.com/group/rights-licensing/permissionshttp://bmjopen.bmj.com//cgi/collection/bmj_open_ophthalmologyhttp://creativecommons.org/licenses/by-nc/3.0/legalcodehttp://creativecommons.org/licenses/by-nc/3.0/http://bmjopen.bmj.com/content/3/7/e002793#BIBLhttp://bmjopen.bmj.com/content/3/7/e002793

Documents

Retrospective Analysis of Long-term