Efficacy of carbonic anhydrase inhibitors in management of cystoid macular edema in retinitis pigmentosa: A meta-analysis
Efficacy of carbonic anhydrase inhibitors in management of cystoid macular edema in retinitis pigmentosa: A meta-analysis
Qinzhu Huang 0 1
Ru Chen 0 1
Xianping Lin 0 1
Zhenyang Xiang 0 1
0 Editor: Gianni Virgili, Universita degli Studi di Firenze , ITALY
1 Taizhou Hospital, Wenzhou Medical University , Taizhou, Zhejiang , PR China
Based on non randomized controlled clinical studies, RP patients with CME who were
treated with CAIs had better anatomical outcomes, but the effect on visual acuity was
contradictory across studies. Multicenter prospective randomized controlled trials would be
ideal to definitively test its clinical efficacy in RP patients.
Data Availability Statement: All relevant data are
within the paper and its Supporting Information
Funding: The author(s) received no specific
funding for this work.
Competing interests: The authors have declared
that no competing interests exist.
Retinitis pigmentosa (RP) is a heterogeneous group of inherited retinal disorders. According
to the inheritance pattern, it is usually classified into three subtypes: autosomal dominant,
autosomal recessive, and X-linked forms. There are specific forms of RP such as Usher
syndrome, which is characterized by congenital sensorineural hearing loss in conjunction with RP
]. Clinical symptoms of RP patients include night blindness and progressive visual field loss
resulting from degeneration of photoreceptors, which eventually leads to blindness.
Complications such as an epi-retinal membrane, cataracts, or cystoid macular edema (CME) can can
also cause early visual loss. According to clinic-based surveys, the prevalence of CME in
patients with RP ranges from 11% to 49% [
]. The wide variation could be partly
explained by the resolution quality of various examination methods such as ophthalmoscopy,
fluorescein angiography, and optical coherence tomography (OCT).
A number of interventions have been applied to treat CME in RP. Reports show that RP
patients with CME may benefit from the administration of reagents such as CAI's [
intravitreal anti-vascular endothelial growth factor (VEGF) agents [
], and intravitreal
]. Among these therapies, both topical and oral CAIs have been reported
to be useful in managing CME. However, the majority of reports are relatively small cases
series, efficacy rates vary greatly between different groups [
], and visual acuity (VA)
improvements after treatment are still uncertain. To our knowledge, there has been no
systematic review significant enough to evaluate the potential of CAI treatment. Therefore, we
undertook a meta-analysis to assess the efficacy of CAI for the management of CME in RP.
We conducted searches of the following electronic databases: PubMed, Cochrane Library, and
Embase without language restriction. We used the combinations of the following terms:
carbonic anhydrase inhibitors, ethoxzolamide, acetazolamide, dorzolamide, pigmentary
petinopathy/pigmentary retinopathies, retinopathies pigmentary/retinopathy pigmentary, retinitis
pigmentosa, and macular edema. The search strategy for PubMed can be found in Supporting
Information (S2 File). In addition, we manually screened the pending references of original
reports to identify studies not yet included in the previous literature search. If sequential reports
from one group which investigated the same cohort of patients were identified, only the latest
updated or informative one was included. The final search was carried out on October 2016.
Articles selected from this initial search were considered eligible for inclusion in the meta analysis
using the following criteria: (1) study design: Randomized Controlled Trials (RCTs),
Non-randomized comparative studies such as single-arm studies, cross-over studies and retrospective cohort
studies; (2) population: RP patients with CME; (3) intervention: topical and oral CAI; (4) outcome
variables: baseline and mean stopping VA or the central macular thickness (CMT) data obtained
by OCT was included. Reports were excluded using the following criteria: (1) full texts and
abstracts from conferences without raw data; (2) duplicate publications; (3) letters, comments, and
reviews; (4) subjects were of rebound macular edema; (5) patients receiving multiple treatments.
Two reviewers extracted data independently. Disagreement was resolved by discussion on all
items. The following information was extracted from the original studies: first author of each
2 / 10
study, publication year, information on study design, number of patients/eyes, sex,
intervention, mean age, VA, and CMT measured by OCT. If the trials reported raw data including all
phases of follow-up, only data from the last follow-up time were analyzed.
Quality assessments were conducted independently by two authors, and disagreements were
resolved by discussion. RCTs were assessed using a Jadad scale, [
] while single arm studies
and cross-over studies and retrospective cohort studies were evaluated using the Newcastle
Ottawa Scale (NOS).[
] For Jadad scale, studies scoring 3 points were considered to be of low
quality. For the NOS scale, overall study quality was defined as poor (score, 0±3), fair (score,
4±6) or good (score, 7±9).
All data was analyzed using Review Manager (RevMan [Computer program], Version 5.3,
Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014).The main
outcome variables were VA and CMT measured by OCT. If the VA was reported in Snellen
VA form in original publication, the data was converted to logarithm of minimal angle of
resolution (logMAR) for statistical analysis. Based on a previous report, ªcounting fingersº (CF),
ªhand motionº (HM), ªlight perceptionº (LP), and ªno light perception (NLP)º were assigned
a logMAR value of 1.85, 2.3, 2.7 and 3.0 respectively [
]. The mean changes of VA or CMT
from baseline to final follow-up points were pooled and calculated using inverse variance
methods. Both outcomes were reported with a 95% confidence interval (CI). P<0.05 was
considered statistically significant. Heterogeneity was analyzed with the I2 statistic, and defined
as low (25% to 50%), moderate (50% to 75%), or high (>75%) . Subgroup analyses were
conducted based on follow-up period, intervention, sample size and baseline level. The I2
value > 50% was defined as heterogeneity and a random-effects model was applied to the data.
Otherwise, a fixed effects model was considered for pooling the data.
Overall characteristics of selected trials and quality assessment
A total of 112 reports were initially identified. Of these, 101 were excluded based on the
exclusion criteria listed above. The 11 remaining clinical reports (1 abstract and 10 full-text) that
met the inclusion criteria were analyzed [
]. Fig 1 provides a flow
chart of the search results. These 11 reports include 1 randomized controlled trial (RCT); 2
retrospective cohort studies, 2 cross-over studies and 6 before-after trials. The RCT study was not
pooled and kept separate in the meta-analyses. If classified with intervention, 1 has both oral
and topical CAI groups; 5 are treated with oral CAIs; the other 5 are treated with topical CAIs.
In total, there were 358 eyes of 194 patients included in this meta-analysis. 59 patients (115
eyes) were included in the oral CAIs group, and 135 patients (243 eyes) were included in the
topical CAIs group. In addition, the overall quality scores of the included studies were
presented in Table 1, with the exception of one study only with English abstract.
Data for the therapeutic effect of treatment with CAIs in RP patients with CME were available
for pooled analysis. The short-term treatment with CAIs was 3 weeks and the long-term
treatment was 58months [
]. The combined results showed reduction of macular edema
through OCT examination. The pooled mean difference in central macular thickness (CMT)
3 / 10
Fig 1. Flow chart of trials included in this meta-analysis.
reduced 46.02μm (95% CI: -60.96, -31.08, I2 = 65%) from baseline to the final follow-up points
(Fig 2). However, when we tried to combine all visual acuity results, it was found that the
Isquare was as high as 91%. Thus, we gave up on conducting such a meta-analysis. Fig 3 shows
the 95% CI of each study. With most studies favoring CAIs treatment, there were three studies
that gave contradictory estimates. Altogether, these analysis shows that CMT decreases with
CAIs treatment, but the effect on visual acuity was inconsistent across studies.
We conducted subgroup analysis to explore the source of heterogeneity in CMT and VA with
respect to follow-up period, intervention, sample size and baseline level. More CMT change
was observed in long term follow-up studies (mean difference = -58.25; 95% CI: -105.78,
-10.72), oral Acetazolamide (mean difference = -62.84, 95% CI: -120.21, -5.47) (Fig 4), studies
of 20 less sample size (mean difference = -54.84, 95% CI: -83.08, -26.61) and studies with
higher CMT baseline level (mean difference = -48.60, 95% CI: -60.43, -36.77). In the subgroup
analysis of mean VA difference, we found that the substantial heterogeneity still remained
(Table 2). One clue from the VA subgroup analysis is that studies with larger sample sizes
4 / 10
RCT: Randomized controlled trial, NR: Not reported
Central Macular Thickness
Fig 2. Forest plot shows the mean change of central macular thickness (CMT) from baseline. Data Liew
2015a (topical CAI group) and Liew 2015b (topical CAI group) were extracted from one report. SE = standard error;
IV = inverse variance; CI = confidence interval.
5 / 10
Fig 3. Diagram shows the 95% CI of mean change of best corrected visual acuity from baseline in
provide more stable results (mean difference = -0.08, 95% CI: -0.10, -0.06) (Fig 5). Other forest
plots for subgroup analysis can be found in supplementary materials (Fig A-D in S3 File).
CME is a common manifestation of various eye diseases such as diabetic retinopathy, retinal
vein occlusion, chronic uveitis, and retinitis pigmentosa. Given the importance of macular
function in vision, the treatment of CME can never be over emphasized. Fortunately, CME is
treatable and can be alleviated by intravitreal corticosteroids, anti-VEGF agents, and other
medications. In this meta-analysis we reviewed studies that treated CME in RP patients with
topical or oral CAI's. The combined results showed a significant reduction in central macular
thickness after CAIs treatment (46.02 μm, 95% CI: -60.96, -31.08, I2 = 65%). The pooled VA
data indicate that most studies (with the exception of 3 studies) show improvements in VA
after CAIs treatment (Fig 3). In one RCT report, CAIs treatment in PR patients achieved
significant improvement in the first 6 months, however a regression was observed at 1 year
(Table 1) [
]. Ikeda et al recommended Humphrey Field Analyzer as a more sensitive
Fig 4. Forest plot shows subgroup analysis of CMT according to interventions (Topical VS Oral).
6 / 10
technique to evaluate macular sensitivity in RP patients with CME [
]. It is proposed that
reduced VA may be due to macular cell loss or irreversible functional damage during the
development of edema [
], thus it's difficult to recover. Rebound is always an issue when
dealing with CME, however, RP patients who show signs of recurring macular edema can
restart the CAIs treatment after a period of discontinued use and still have a favorable response
The exact mechanism of how CAIs reduce CME in RP patients remains unclear. Two
working models have been proposed. Some authors have proposed that CAIs restore a relatively
normal distribution of carbonic anhydrase activity and polarity of RPE cells by selectively
blocking the activity of different anhydrase isozymes located in basolateral membrane of RPE
cells, thus facilitating subretinal fluid resorption [
]. However, some studies suggest a direct
role of CAIs on the retinal vasculature as they show that CAIs are capable of enhancing retinal
blood flow and oxygen tension [
Fig 5. Forest plot shows subgroup analysis of VA based on sample size.
7 / 10
It has been reported that systemic CAIs can cause side-effects such as appetite loss, fatigue,
and even the development of kidney stones [
]. However, the side effects rate was low in the
published reports, and it could be reduced by a lower yet effective dosage (125 mg/d) [
Compared to oral CAIs, the side effects during topical administration were minimal. The most
common side-effect was a short-term stinging sensation in the eye immediately after instilling
]. Allergic blepharitis can be avoided by using preservative-free eye-drops [
Several limitations of this work are worth considering. First, we cannot fully exclude
publication bias. We only searched literature with English abstracts. Second, results are affected by
the quality of individual studies. These studies were carried out with small sample sizes and
had different duration of intervention, populations, and different follow-up period. Third, due
to the paucity of RCTs, this review primarily evaluated cohort studies and single-arm studies,
thus associations between intervention and outcomes are at some risk for bias, which is a
source of heterogeneity. The possibility of CME alleviation by itself during long follow-up
period also cannot be fully excluded.
In conclusion, the current meta-analysis has shown that treatment of CME in RP patients
with CAIs significantly reduces the central macular thickness, but the effects on visual acuity
are contradictory across studies. Thus, multicenter prospective randomized controlled trials
would be ideal to definitively test its clinical efficacy in RP patients.
S1 File. PRISMA checklist.
S2 File. Search strategy for PubMed.
S3 File. Supplementary figures.
We would like to thank Anish Bhandari for his help on English writing.
Conceptualization: Qinzhu Huang.
Data curation: Qinzhu Huang.
Investigation: Qinzhu Huang.
Project administration: Ru Chen.
Supervision: Xianping Lin, Zhenyang Xiang.
Validation: Ru Chen.
Writing ± original draft: Qinzhu Huang.
Writing ± review & editing: Xianping Lin, Zhenyang Xiang.
8 / 10
9 / 10
1. Rosenberg T , Haim M , Hauch AM , Parving A . The prevalence of Usher syndrome and other retinal dystrophy-hearing impairment associations . Clin Genet . 1997 ; 51 ( 5 ): 314 ± 21 . Epub 1997/05/01. PMID: 9212179 .
2. Salvatore S , Fishman GA , Genead MA . Treatment of cystic macular lesions in hereditary retinal dystrophies . Surv Ophthalmol . 2013 ; 58 ( 6 ): 560 ± 84 . Epub 2013/10/29. https://doi.org/10.1016/j.survophthal. 2012 . 11 .006 PMID: 24160730 .
3. Fishman GA , Maggiano JM , Fishman M. Foveal lesions seen in retinitis pigmentosa . Arch Ophthalmol . 1977 ; 95 ( 11 ): 1993 ± 6 . Epub 1977/11/01. PMID: 921578 .
4. Fishman GA , Fishman M , Maggiano J . Macular lesions associated with retinitis pigmentosa . Arch Ophthalmol . 1977 ; 95 ( 5 ): 798 ± 803 . Epub 1977/05/01. PMID: 860943 .
5. Pruett RC . Retinitis pigmentosa: clinical observations and correlations . Trans Am Ophthalmol Soc . 1983 ; 81 : 693 ± 735 . Epub 1983/01/01. PMID: 6676982; PubMed Central PMCID : PMC1312466 .
6. Adackapara CA , Sunness JS , Dibernardo CW , Melia BM , Dagnelie G . Prevalence of cystoid macular edema and stability in oct retinal thickness in eyes with retinitis pigmentosa during a 48-week lutein trial . Retina . 2008 ; 28 ( 1 ): 103 ± 10 . Epub 2008/01/11. https://doi.org/10.1097/IAE.0b013e31809862aa PMID: 18185146 .
7. Hajali M , Fishman GA . The prevalence of cystoid macular oedema on optical coherence tomography in retinitis pigmentosa patients without cystic changes on fundus examination . Eye (Lond) . 2009 ; 23 ( 4 ): 915 ± 9 . Epub 2008/04/22. https://doi.org/10.1038/eye. 2008 .110 PMID: 18425064 .
8. Cox SN , Hay E , Bird AC . Treatment of chronic macular edema with acetazolamide . Arch Ophthalmol . 1988 ; 106 ( 9 ): 1190 ± 5 . Epub 1988/09/01. PMID: 3415543 .
9. Fishman GA , Gilbert LD , Fiscella RG , Kimura AE , Jampol LM . Acetazolamide for treatment of chronic macular edema in retinitis pigmentosa . Arch Ophthalmol . 1989 ; 107 ( 10 ): 1445 ± 52 . Epub 1989/10/01. PMID: 2803090 .
10. Artunay O , Yuzbasioglu E , Rasier R , Sengul A , Bahcecioglu H . Intravitreal ranibizumab in the treatment of cystoid macular edema associated with retinitis pigmentosa . J Ocul Pharmacol Ther . 2009 ; 25 ( 6 ): 545 ± 50 . Epub 2009/12/24. https://doi.org/10.1089/jop. 2009 .0089 PMID: 20028262 .
11. Yuzbasioglu E , Artunay O , Rasier R , Sengul A , Bahcecioglu H . Intravitreal bevacizumab (Avastin) injection in retinitis pigmentosa . Curr Eye Res . 2009 ; 34 ( 3 ): 231 ± 7 . Epub 2009/03/11. https://doi.org/10. 1080/02713680802710692 PMID: 19274531 .
12. Ozdemir H , Karacorlu M , Karacorlu S . Intravitreal triamcinolone acetonide for treatment of cystoid macular oedema in patients with retinitis pigmentosa . Acta Ophthalmol Scand . 2005 ; 83 ( 2 ): 248 ± 51 . Epub 2005/04/01. https://doi.org/10.1111/j.1600- 0420 . 2005 . 00395 . x PMID : 15799743 .
13. Scorolli L , Morara M , Meduri A , Reggiani LB , Ferreri G , Scalinci SZ , et al. Treatment of cystoid macular edema in retinitis pigmentosa with intravitreal triamcinolone . Arch Ophthalmol . 2007 ; 125 ( 6 ): 759 ± 64 . Epub 2007/06/15. https://doi.org/10.1001/archopht.125.6.759 PMID: 17562986 .
14. Srour M , Querques G , Leveziel N , Zerbib J , Tilleul J , Boulanger-Scemama E , et al. Intravitreal dexamethasone implant (Ozurdex) for macular edema secondary to retinitis pigmentosa . Graefes Arch Clin Exp Ophthalmol . 2013 ; 251 ( 6 ): 1501 ± 6 . Epub 2013/01/01. https://doi.org/10.1007/s00417-012-2249-4 PMID: 23275039 .
15. Ahn SJ , Kim KE , Woo SJ , Park KH . The effect of an intravitreal dexamethasone implant for cystoid macular edema in retinitis pigmentosa: a case report and literature review . Ophthalmic Surg Lasers Imaging Retina . 2014 ; 45 ( 2 ): 160 ± 4 . Epub 2014/02/11. https://doi.org/10.3928/ 23258160 -20140131-03 PMID: 24506098 .
16. Genead MA , Fishman GA . Efficacy of sustained topical dorzolamide therapy for cystic macular lesions in patients with retinitis pigmentosa and usher syndrome . Arch Ophthalmol . 2010 ; 128 ( 9 ): 1146 ± 50 . Epub 2010/09/15. https://doi.org/10.1001/archophthalmol. 2010 .172 PMID: 20837798; PubMed Central PMCID : PMC3696579 .
17. Ikeda Y , Hisatomi T , Yoshida N , Notomi S , Murakami Y , Enaida H , et al. The clinical efficacy of a topical dorzolamide in the management of cystoid macular edema in patients with retinitis pigmentosa . Graefes Arch Clin Exp Ophthalmol . 2012 ; 250 ( 6 ): 809 ± 14 . Epub 2012/01/05. https://doi.org/10.1007/s00417- 011-1904-5 PMID: 22215259 .
18. Ikeda Y , Yoshida N , Notomi S , Murakami Y , Hisatomi T , Enaida H , et al. Therapeutic effect of prolonged treatment with topical dorzolamide for cystoid macular oedema in patients with retinitis pigmentosa . Br J Ophthalmol . 2013 ; 97 ( 9 ): 1187 ± 91 . Epub 2013/06/21. https://doi.org/10.1136/bjophthalmol-2012 - 303005 PMID: 23782868.
19. Jadad AR , Moore RA , Carroll D , Jenkinson C , Reynolds DJ , Gavaghan DJ , et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials . 1996 ; 17 ( 1 ):1± 12 . Epub 1996/02/01. PMID: 8721797 .
20. Wells GA , Shea BJ , O'Connell D , Peterson J , Welch V , Losos M , et al. The Newcastle±Ottawa Scale (NOS) for Assessing the Quality of Non-Randomized Studies in Meta-Analysis . Applied Engineering in Agriculture. 2014 ; 18 ( 6 ):paÂgs. 727 ± 34 .
21. Schulze-Bonsel K , Feltgen N , Burau H , Hansen L , Bach M. Visual acuities "hand motion" and "counting fingers" can be quantified with the freiburg visual acuity test . Invest Ophthalmol Vis Sci . 2006 ; 47 ( 3 ): 1236 ± 40 . Epub 2006/03/01. https://doi.org/10.1167/iovs.05-0981 PMID: 16505064 .
22. Chang JW , Kim JH , Kim SJ , Yu YS . Congenital aniridia: long-term clinical course, visual outcome, and prognostic factors . Korean J Ophthalmol . 2014 ; 28 ( 6 ): 479 ± 85 . Epub 2014/12/02. https://doi.org/10. 3341/kjo. 2014 . 28 .6.479 PMID: 25435751; PubMed Central PMCID : PMC4239467 .
23. Huedo-Medina TB , Sanchez-Meca J , Marin-Martinez F , Botella J . Assessing heterogeneity in metaanalysis: Q statistic or I2 index? Psychol Methods . 2006 ; 11 ( 2 ): 193 ± 206 . Epub 2006/06/21. https://doi. org/10.1037/ 1082 - 989X . 11 .2.193 PMID: 16784338 .
24. Liew G , Moore AT , Webster AR , Michaelides M. Efficacy and prognostic factors of response to carbonic anhydrase inhibitors in management of cystoid macular edema in retinitis pigmentosa . Invest Ophthalmol Vis Sci . 2015 ; 56 ( 3 ): 1531 ± 6 . Epub 2015/02/12. https://doi.org/10.1167/iovs.14-15995 PMID: 25670491 .
25. Alkin Z , OÈ zkaya A , Karataş G , Yazici AT , Demirok A . Brinzolamide Therapy in Cystoid Macular Edema Secondary to Retinitis Pigmentosa . Retina-Vitreus. 2013 .
26. Lemos Reis RF , Moreira-Goncalves N , Estrela Silva SE , Brandao EM , Falcao-Reis FM . Comparison of topical dorzolamide and ketorolac treatment for cystoid macular edema in retinitis pigmentosa and Usher's syndrome . Ophthalmologica . 2015 ; 233 ( 1 ): 43 ± 50 . Epub 2014/11/28. https://doi.org/10.1159/ 000368052 PMID: 25428176 .
27. Orzalesi N , Pierrottet C , Porta A , Aschero M. Long-term treatment of retinitis pigmentosa with acetazolamide. A pilot study . Graefes Arch Clin Exp Ophthalmol . 1993 ; 231 ( 5 ): 254 ± 6 . Epub 1993/05/01. PMID: 8319913 .
28. Chung H , Hwang JU , Kim JG , Yoon YH . Optical coherence tomography in the diagnosis and monitoring of cystoid macular edema in patients with retinitis pigmentosa . Retina . 2006 ; 26 ( 8 ): 922 ± 7 . Epub 2006/ 10/13. https://doi.org/10.1097/01.iae. 0000250008 .83779.23 PMID: 17031294 .
29. Moldow B , Sander B , Larsen M , Engler C , Li B , Rosenberg T , et al. The effect of acetazolamide on passive and active transport of fluorescein across the blood-retina barrier in retinitis pigmentosa complicated by macular oedema . Graefes Arch Clin Exp Ophthalmol . 1998 ; 236 ( 12 ): 881 ± 9 . Epub 1998/12/29. PMID: 9865617 .
30. Grover S , Apushkin MA , Fishman GA . Topical dorzolamide for the treatment of cystoid macular edema in patients with retinitis pigmentosa . Am J Ophthalmol . 2006 ; 141 ( 5 ): 850 ± 8 . Epub 2006/03/21. https:// doi.org/10.1016/j.ajo. 2005 . 12 .030 PMID: 16546110 .
31. Thobani A , Fishman GA . The use of carbonic anhydrase inhibitors in the retreatment of cystic macular lesions in retinitis pigmentosa and X-linked retinoschisis . Retina . 2011 ; 31 ( 2 ): 312 ± 5 . Epub 2010/10/23. https://doi.org/10.1097/IAE.0b013e3181e587f9 PMID: 20966823; PubMed Central PMCID : PMC3690283 .
Wolfensberger TJ , Dmitriev AV , Govardovskii VI . Inhibition of membrane-bound carbonic anhydrase decreases subretinal pH and volume . Doc Ophthalmol . 1999 ; 97 ( 3 ±4): 261 ± 71 . Epub 2000/07/15.
33. Pedersen DB , Koch Jensen P , la Cour M , Kiilgaard JF , Eysteinsson T , Bang K , et al. Carbonic anhydrase inhibition increases retinal oxygen tension and dilates retinal vessels . Graefes Arch Clin Exp Ophthalmol . 2005 ; 243 ( 2 ): 163 ± 8 . Epub 2005/03/03. https://doi.org/10.1007/s00417-003-0817-3 PMID: 15742212 .
34. Siesky B , Harris A , Brizendine E , Marques C , Loh J , Mackey J , et al. Literature review and meta-analysis of topical carbonic anhydrase inhibitors and ocular blood flow . Surv Ophthalmol . 2009 ; 54 ( 1 ): 33 ± 46 . Epub 2009/01/28. https://doi.org/10.1016/j.survophthal. 2008 . 06 .002 PMID: 19171209 .
35. Kass MA , Kolker AE , Gordon M , Goldberg I , Gieser DK , Krupin T , et al. Acetazolamide and urolithiasis . Ophthalmology . 1981 ; 88 ( 3 ): 261 ± 5 . Epub 1981/03/01. PMID: 7231915 .