Ahmed glaucoma valve implantation in uveitic glaucoma versus open-angle glaucoma patients Rony Rachmiel,* MD; Graham E. Trope,* MB, PhD, FRCSC; Yvonne M. Buys,* MD, FRCSC; John G. Flanagan,*† MCOptom, PhD; Mary L. Chipman,*‡ MA ABSTRACT • RÉSUMÉ
Background: The poor long-term success rate of repeat trabeculectomies in refractory uveitic glaucoma (UG) patients has led to the use of glaucoma drainage devices (GDDs). However, the success and complication rates of GDDs in UG patients utilizing a control group with standard demographic data, design, and surgical technique have never been evaluated. Methods: Fifteen patients (15 eyes) with chronic uveitis and 53 patients (53 eyes) with uncontrolled openangle glaucoma (OAG) who underwent Ahmed glaucoma valve (AGV) implantation were included in a retrospective, comparative, case-controlled study. Postoperative intraocular pressure (IOP), number of antiglaucoma medications, visual acuity, and complications were compared. Results: There was a significant difference between the UG versus the OAG group with respect to age only (59.3 years vs 68.4 years, p = 0.006). Regression analysis of the postoperative IOP controlled for age and glaucoma type, and preoperative IOP revealed significantly lower IOP in the UG group at 1 month (p = 0.04; 95% confidence interval [CI] –5.9 to 0.15) and 2 months (p = 0.008; 95% CI –6.0 to 0.97). No significant differences were found at 3, 6, 12, 24, and 30 months. The cumulative success rates at 3 to 30 months for the UG and OAG groups were 80% to 66.6% versus 84.9% to 57% (p = 0.713), respectively. The only complication between the 2 groups that was significantly different was tube removal, which occurred more often in the UG group (p = 0.018). Interpretation: AGV implantation is an effective and safe procedure in the management of UG, similar to primary OAG. Contexte : Le faible taux de réussite à long terme des reprises de la trabéculectomie chez les patients réfractaires atteints de glaucome uvéitique (GU) a mené à l’utilisation d’appareils de drainage du glaucome (ADG). Toutefois, on n’a jamais évalué les taux de réussite et de complication des ADG chez les patients GU en regard d’un groupe témoin avec données démographiques, modèles et techniques opératoires normaux. Méthodes : Quinze patients (15 yeux) atteints d’uvéite chronique et 53 patients (53 yeux) atteints de glaucome à angle ouvert (GAO) non maîtrisé qui avaient reçu l’implantation de la valve Ahmed pour le glaucome (VAG) ont été inclus dans l’étude rétrospective et comparative, avec cas témoins. La pression intraoculaire (PIO) postopératoire, le nombre de médicaments anti-glaucome, l’acuité visuelle et les complications ont été comparés. Résultats : Il y avait un écart important entre les groupes GU et GAO en regard de l’âge seulement (59,3 ans vs 68,4 ans, p = 0,006). L’analyse de régression de la PIO postopératoire en regard de l’âge et du type glaucome, et la PIO préopératoire a révélé une PIO significativement inférieure chez le groupe GU après un mois (p = 0,04; intervalle de confiance [IC] 95 % –5,9 à 0,15) et après 2 mois (p = 0,008; IC 95 % –6,0 à 0,97). Aucun écart important n’a été trouvé après 3, 6, 12, 24 et 30 mois. Les taux cumulatifs de réussite entre 3 et 30 mois pour les groupes GU et GAO ont été de 80 % à 66,6 % versus 84,9 % à 57 % (p = 0,713), respectivement. La seule complication qui s’est avérée significativement différente entre les 2 groupes a été le retrait du tube, qui fut plus fréquent chez le groupe GU (p = 0,018). Interprétation : L’implantation de la VAG est une procédure efficace et sûre dans le traitement du GU, semblable au GAO primaire.
From *the Department of Ophthalmology & Visual Sciences, Toronto Western Hospital, University of Toronto, Toronto, Ont.; †the School of Optometry, University of Waterloo, Waterloo, Ont.; and ‡the Department of Public Health Sciences, University of Toronto, Toronto, Ont. Originally received May 24, 2007. Revised Jan. 20, 2008 Accepted for publication Feb. 22, 2008 Published online July 8, 2008
462
CAN J OPHTHALMOL—VOL. 43, NO. 4, 2008
Correspondence to Rony Rachmiel, MD, Department of Ophthalmology, Toronto Western Hospital, 399 Bathurst Street, New East Wing 6-405, Toronto ON M5T 2S8;
[email protected] This article has been peer-reviewed. Cet article a été évalué par les pairs. Can J Ophthalmol 2008;43:462–7 doi:10.3129/i08-082
Ahmed glaucoma valve implantation—Rachmiel et al.
G
laucoma is one of the most common complications of uveitis and occurs in approximately 20% of uveitis patients.1–3 The management of high intraocular pressure (IOP) refractory to medical therapy in uveitic glaucoma (UG) presents a challenge.4 Trabeculectomy with or without antimetabolites in UG has a variable success rate (IOP < 21 mm Hg) of 50% to 100% at 1 to 2 years5–8 and 50% to 67% at 5 years.8–11 The poor long-term success rate of repeat trabeculectomies in uveitis has led to the use of GDDs in this condition. Reported success rates with GDDs in UG patients are 83% to 94% at 2 years.10,12,13 The Ahmed glaucoma valve (AGV) (New World Medical, Inc, Rancho Cucamonga, Calif.) has advantages over nonvalved implants as it is designed to prevent hypotony in the immediate postoperative period and is implanted in a single-stage procedure.14 Previous reported studies with AGV in iritis, however, have not utilized a control group and were not standardized in terms of demographic data, design, and surgical technique. The purpose of this study was to evaluate the success and complication rates of AGV implantation in UG patients versus case-controlled primary open-angle glaucoma (POAG) patients. METHODS
After approval by the Research Ethics Board of the University Health Network, University of Toronto, and registration with the U.S. National Institutes of Health, a retrospective review was done of 141 medical records of patients who underwent AGV implant surgery in the glaucoma service at the Toronto Western Hospital, Toronto, Ont., from January 1, 1997, to July 31, 2004. The study population included 53 patients (53 eyes) with uncontrolled open-angle glaucoma (OAG) and 15 patients (15 eyes) with UG. In patients who underwent bilateral AGV implantation, only the first eye to undergo surgery was included in the study. After obtaining informed consent, patients underwent AGV implantation by 1 surgeon (Graham E. Trope) through a single approach (fornix-based) using the S2 and later (from 2003) the FP7 AGV. AGVs were implanted through a fornix-based incision, with the plate secured 8 to 12 mm from the corneoscleral limbus, as described elsewhere.15 The surgical procedure was only undertaken when there were no signs of active uveitis. Only patients who fulfilled the inclusion and exclusion criteria were enrolled in this study. Inclusion criteria included (i ) patients between the ages of 20 and 85 years on the date of the surgery; (ii ) uncontrolled UG or uncontrolled POAG on maximally tolerated medical therapy that did not, or were not expected to, respond to trabeculectomy; (iii ) visual acuity (VA) of 20/800 and better before the operation; and (iv ) regular postoperative follow-ups for at least 3 months. Exclusion criteria included (i ) a previous seton surgery in
the operated eye, (ii ) any ocular surface disease that could affect the healing of the conjunctiva after the surgery, (iii ) strabismus or complaint of diplopia prior to the seton surgery, (iv ) past history of retinal detachment surgery with scleral buckle in the operated eye, (v ) past history of endophthalmitis in the operated eye, and (vi ) a pars plana implant. No other concurrent surgical procedures (e.g., cataract, pars plana vitrectomy) were allowed at the time of the AGV implantation surgery. In addition, patients who received subsequent ocular procedures during their postoperative follow-up were excluded as soon as they were referred to the operation. The medical records were reviewed to determine demographic, preoperative, and postoperative data. Data collected included age, gender, glaucoma type, length of follow-up, diagnosis of uveitis, implantation quadrant, tube location, slit-lamp and fundus examination, VA, IOP, previous glaucoma surgeries, number of glaucoma medications used, and complications. The VAs throughout the period studied were converted to logMAR to facilitate statistical analysis. The preoperative IOP was determined as a mean of 3 measurements on 3 different visits before the operation and the last IOP measured before the AGV implantation procedure. Demographic, preoperative, and postoperative data were recorded. Postoperative data were recorded at 1 week, 3 weeks, 1 month, 2 months, 3 months, 6 months, 12 months, 18 months, 24 months, and 30 months. Data from visits close to the anniversary date were collected. Definitions (1) Hypotony: early hypotony was defined as IOP less
than 5 mm Hg in the first postoperative month. Late hypotony was defined as IOP less than 5 mm Hg any time after the first postoperative month. (2) Visual loss was considered for any VA less than 3 logMAR lines from the baseline measured more than 3 months after the operation. (3) Success was defined as postoperative IOP between 5 and 21 mm Hg and at least a 20% reduction in IOP from preoperative levels with or without medications use. (4) Failure was defined as IOP reduction of less than 20% on 2 consecutive visits, IOP equal to, or higher than, 22 mm Hg or lower than 5 mm Hg with or without medications after 3 months, VA worse than hand motion, explantation of glaucoma drainage device (GDD), or reoperation due to high IOP or other tube surgery complications. (5) Serious complications were those that necessitated reoperation or other surgical intervention or resulted in significant deterioration of vision (as defined by “vision loss”). Statistical analysis was performed with Stata software (v. 2005, Stata, College Station, Tex.).16 The demographic data and preoperative data of the 2 groups were compared using t tests with equal variances or χ2 tests. The VA was CAN J OPHTHALMOL—VOL. 43, NO. 4, 2008
463
Ahmed glaucoma valve implantation—Rachmiel et al. analyzed using the Mann-Whitney rank sum test. The IOP comparison between the 2 groups was analyzed using a 2-sample t test with unequal variance. Survival curves for success in each group were calculated using actuarial methods and compared with the log-rank test.17 The reason for using actuarial methods for survival analysis was that only the interval of time within which IOP deteriorated could be identified. RESULTS
The demographic data of the study are presented in Table 1. Fifty-three patients (53 eyes) with uncontrolled OAG and 15 patients (15 eyes) with a history of uveitis underwent AGV implantation for uncontrolled glaucoma between January 1997 and July 2004. The mean age of the UG patients was significantly younger (p = 0.006) than the mean age of the OAG patients (59.26 [SD 14.7] years [95% confidence interval {CI} 51.1–67.4] vs 68.39 [SD 9.7] years [95% CI 65.7–71.0], respectively) (Table 1). The 2 groups had a nonsignificant difference concerning the time of follow-up from the diagnosis of glaucoma to the AGV implantation date (19.7 [SD 10.3] years in the UG group vs 15.1 [SD 7.3] years in the OAG group; p = 0.08). The number of prior surgeries was 1.86 (SD 1.30) in the UG group versus 2.56 (SD 1.58) in the OAG group (p = 0.12, 95% CI –0.19 to 1.59). The difference in the number of prior surgeries was not statistically significant between the 2 groups but the sample size may not have been large enough to show any difference. There was a nonsignificant difference between the groups with regard to the mean preoperative IOPs (p = 0.08; Table 1). However, the IOP in the uveitic group was much more variable (mean 24.56 [SD 10.2] mm Hg; 95% CI 17.9–21.0) than the IOP in the OAG group (mean 19.46 [SD 5.6] mm Hg; 95% CI 17.9–21.0). There was a statistically significant (p = 0.04) mean range of pressure difference between the IOPs of the groups (–5.1 mm Hg; 95% CI –10.8 to 0.67). The uveitic etiologies were idiopathic anterior uveitis,
7 eyes; Fuchs heterochromic cyclitis, 3 eyes; herpes zoster uveitis, 1 eye; juvenile rheumatoid arthritis–related uveitis, 1 eye; rheumatoid arthritis–related sclerouveitis, 1 eye; lowgrade uveitis following penetrating keratoplasty, 1 eye. The success rates in the UG and OAG groups are presented in an actuarial life table analysis of the cumulative success rates (Fig. 1). There was no evidence of a difference between the 2 survival curves (p = 0.713) when compared using the log-rank test. At 30 months of follow-up, 66.6% of the UG and 57% of the OAG group remained successful. There was no change in the success rate in the UG group from the 1-year follow-up until the end of the follow-up period. The success rate in the OAG group dropped from 68.9% at 1 year to 57% at 30 months (statistically nonsignificant). At the end of the follow-up period, patients in the UG and OAG groups had an IOP reduction of 45.2% and 31.7%, respectively, from their preoperative levels. The mean IOP was similar in the 2 groups during the study period except for the IOP at 2 months (p = 0.01), when the UG group had an IOP of 10.9 (SD 3.6) mm Hg versus 13.9 (SD 4.1) mm Hg in the OAG group (Fig. 2).
Fig. 1—Actuarial life table analysis of the cumulative success rates for the uveitic glaucoma (UG) and open-angle glaucoma (OAG) patients with Ahmed valve implants.
Table 1—Demographic data of uveitic glaucoma versus openangle glaucoma patients Demographic Age, y
UG (n = 15), mean (SD)
OAG (n = 53), mean (SD)
p value
59.3 (14.7)
68.4 (9.7)
0.006
Sex Female
8 (53.3)
26 (49.1)
1.000
Male
7 (46.7)
27 (50.9)
1.000
Glaucoma duration, y
19.7 (10.3)
15.1 (7.3)
0.080
Postoperative follow-up, y
31.7 (17.7)
32.4 (19.7)
0.910
Prior glaucoma surgeries, n
1.86 (1.30)
2.56 (1.58)
0.120
Preoperative medications, n
3.0 (1.31)
3.47 (1.14)
0.180
24.56 (10.2)
19.46 (5.6)
0.080
0.50 (0.44)
0.49 (0.43)
0.930
Preoperative IOP, mm Hg Preoperative logMAR visual acuity
Note: UG, uveitic glaucoma; OAG, open-angle glaucoma; IOP, intraocular pressure.
464
CAN J OPHTHALMOL—VOL. 43, NO. 4, 2008
Fig. 2—Mean intraocular pressure comparison between uveitic glaucoma (UG) and open-angle glaucoma (OAG) groups with Ahmed glaucoma valve implants.
Ahmed glaucoma valve implantation—Rachmiel et al. Regression analysis of the postoperative IOP controlled for age and glaucoma type, and preoperative IOP revealed significantly lower IOP in the UG group at 1 month (p = 0.04; 95% CI –5.9 to 0.15) and 2 months (p = 0.008; 95% CI –6.0 to 0.97). At 18 months, the difference was significant (p = 0.05) with a CI that had lack of power (95% CI –0.05 to 5.36). No statistically significant differences were found at 3, 6, 12, 24, and 30 months. The median and interquartile range (25th and 75th percentiles) of VA was analyzed with the Mann-Whitney rank sum test. Similar logMAR VA was found in the 2 groups, with no significant difference noticed over the study period (p > 0.05; Fig. 3). In the uveitic group, although it looks like the vision started to deteriorate after 18 months, the median was between 0.4 and 0.55 and there was a nonsignificant difference (p = 0.11 to p = 0.13) in comparison with the OAG group. The causes of failure were similar in the 2 groups. There were several reasons for loss of VA in both groups, among them chronic uveitis (33.3% in the UG group vs 24.5% in the OAG group) and choroidal effusion/detachment (6.6% in the UG group vs 20.7% in the OAG group). The number of glaucoma medications was similar between the groups except for months 2, 3, 6, and 12
(Fig. 4), when a significantly lower number of medications was found in the UG group (p = 0.02, p = 0.001, p = 0.001, and p = 0.007, respectively). Several complications were noted (Table 2) in the 2 groups. Only tube removal (3 out of 15 eyes, 20%) was found to be statistically significant (p = 0.018) in the UG group. INTERPRETATION
Comparable intermediate-term postoperative outcomes are reported in this study in patients with UG versus OAG undergoing AGV implantation. There are few studies reporting outcomes of AGV implantation in refractory UG patients.12,13,18 These studies report a beneficial effect utilizing AGV in UG, with success rates around 90% after 1 year and 60% to 68% after 2 years. 12,13,18 However, these studies differ in terms of their study group demographic data, design, definitions, and results. None of them had a control or a comparison group and no standardization for inclusion and exclusion criteria, making their conclusions difficult to interpret. Molteno et al.19 reported on the long-term results of Molteno implants in 40 eyes of 35 patients with UG at 5 and 10 years. At 5 years, 87% of the eyes were controlled on 0.44 medications and at 10 years, 93% of the eyes maintained IOP controlled on 0.33 medications. Ceballos et al.10 reported good short-term results with the Baerveldt glaucoma drainage device (2-year success rate of 91.7%) in eyes with UG. Krishna et al.20 reported success rates similar to ours, namely, 60% at 2 years follow-up in the UG group. The preoperative demographic data were similar in our Table 2—Postoperative complications in the open-angle glaucoma versus the uveitic glaucoma group
Fig. 3—LogMAR median visual acuity comparison between uveitic glaucoma (UG) and openangle glaucoma (OAG) patients with Ahmed glaucoma valve implants.
Fig. 4—Comparison between the number of postoperative medications taken by uveitic glaucoma (UG) and open-angle glaucoma (OAG) patients following Ahmed glaucoma valve implantation.
UG (n = 15), n (%)
OAG (n = 53), n (%)
Wound dehiscence
3 (20.0)
12 (22.6)
1.000
Plate erosion or exposure
1 (6.6)
1 (1.8)
0.395
Complication
p value
Tube erosion or exposure
2 (13.3)
1 (1.8)
0.120
Tube/cornea touch
1 (6.6)
7 (13.2)
0.674
Corneal edema/insufficiency
0 (0.0)
4 (7.5)
0.569
Penetrating keratoplasty
2 (13.3)
1 (1.8)
0.120
Hyphema
3 (20.0)
9 (17.0)
0.719
Uveitis
5 (33.3)
13 (24.5)
0.519
Uveitis within 6 weeks
2 (13.3)
6 (11.3)
1.000
Uveitis after 6 weeks
3 (20.0)
10 (18.8)
1.000
Overfiltration
1 (6.6)
3 (5.6)
1.000
Transient hypotony
4 (26.6)
5 (9.4)
0.200
Cataract
0 (0.0)
6 (11.3)
0.326
Choroidal effusion/detachment
1 (6.6)
11 (20.7)
0.276
Choroidal hemorrhage
0 (0.0)
1 (1.8)
1.000
Transient diplopia
1 (6.6)
8 (15.1)
0.672
Tube removal
3 (20.0)
0 (0.0)
0.018
New tube inserted
3 (20.0)
3 (5.6)
0.116
Note: More than 1 complication may have occurred in an eye. Test of significance not done if total number of occurrences was
