- Email: [email protected]
Intraocular pressure spike after YAG iridotomy in patients with pigment dispersion
CLINICAL STUDIES
Intraocular pressure spike after YAG iridotomy in patients with pigment dispersion Catherine M. Birt, MD, FRCSC ABSTRACT • RESUME Background: The role of laser peripheral iridotomy to break a suspected reverse pupil block in the long-term control of pigment dispersion is promising, but the usefulness of this procedure has not been completely established. The author examined whether patients with pigment dispersion are at higher risk for an intraocular pressure (lOP) spike after laser peripheral iridotomy due to possible compromise of trabecular meshwork function, compared with patients undergoing prophylactic peripheral iridotomy for an occludable angle. Methods: Data were collected prospectively on the first eye of 87 patients with occludable angles and 13 patients with pigment dispersion treated with peripheral laser iridotomy between November 1995 and October 1996 at the glaucoma service of a university-affiliated hospital in Toronto~ All patients received one drop of 0.5% apraclonidine before the procedure. lOP was measured before and I and 24 hours after the procedure. Results: There was no difference between the two groups in the distribution of right vs. left eyes, sex, race, the mean total energy required to produce a patent iridotomy, the mean number of medications used or the mean lOP before the procedure. The patients with pigment dispersion were significantly younger than those with occludable angles (mean age [and standard deviation (SD)] 40.5 [9.45] years vs.66.4 [10.78] years) (p < 0.00 I).There was no difference between the two groups in mean lOP at I hour or at 24 hours. Twenty-nine patients (33%) in the occludable angle group and seven (54%) in the pigment dispersion group had an lOP spike greater than 2 mm Hg after the procedure (p = 0.00 I). Among these patients, the mean lOP (36.4 [SD 10.83] mm Hg vs. 30.3 [SD 7.04] mm Hg, p = 0.05) and the mean rise in lOP (14.0 [SD 10.63] mm Hg vs. 8.7 [SD 4.73] mm Hg, p = 0.04) were significantly higher in those with pigment dispersion than in those with occludable angles. Among the patients who used antiglaucoma medications before the procedure or had a prelaser lOP level greater than 22 mm Hg, those with pigment dispersion were more likely than those with occludable angles to have an lOP spike at I hour (p ::; 0.005).
From Sunnybrook and Women's College Health Sciences Centre, Toronto, Ont. Presented in part as a poster paper at the Association for Research in Vision and Ophthalmology meeting held in Fort Lauderdale, Fla., May ll-15, 1997 Originally received May 6, 2003 Accepted for publication Oct. 10, 2003
234
YAG iridotomy in pigment glaucoma-Birt
Reprint requests to: Dr. Catherine M. Birt, M1 302a, Department of Ophthalmology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Ave., Toronto ON M4N 3M5; fax (416) 4804481, [email protected] This article has been peer-reviewed. Can J Ophthalmol 2004;39:234-9
YAG iridotomy in pigment glaucoma-Birt
Interpretation: Patients with pigment dispersion undergoing iridotomy to break a reverse pupil block should be carefully assessed after the procedure, as significant pressure spikes requiring treatment may occur.
Contexte : La technique de I'iridotomie peripherique au laser pour renverser ce qu'on croit etre un blocage pupillaire inverse dans la maitrise a long terme de la dispersion pigmentaire est prometteuse, mais I'on n'a pas encore etabli entierement I'utilite de ce procede. L'auteur s'est demande si les patients atteints de dispersion pigmentaire etaient plus a risque d'avoir une hausse de pression intraoculaire (PIO) apres une iridotomie peripherique au laser, acause de la compromission possible de la fonction du trabeculum, que ceux qui subissent une iridotomie peripherique prophylactique pour un angle etroit. Methodes: Les donnees ont ete recueillies de fa~on prospective sur Ie premier reil de 87 patients ayant des angles etroits et de 13 patients atteints de dispersion pigmentaire, qui avaient subis une iridotomie peripherique au laser entre les mois de novembre 1995 et octobre 1996 au service du glaucome d'un h6pital universitaire de Toronto. Les patients ont tous re~u une goutte d'apraclonidine a 0,5 % avant la procedure. La PIO a ete mesuree avant ainsi qu'une heure et 24 heures apres la procedure. Resultats: L'on n'a pas trouve d'ecart entre les deux groupes quant a la repartition entre les yeux droits et gauches, au sexe, a la race, a I'energie totale moyenne requise pour produire une iridotomie permeable, au nombre moyen de medicaments utilises et a la PIO moyenne avant la procedure. Les patients qui avaient une dispersion pigmentaire etaient beaucoup plus jeunes que ceux qui avaient des angles etroits (age moyen 40,5 ans [ecart type (ET) 9,45] c. 66,4 ans [ET 10,78]) (p < 0,00 I). II n'y avait pas d'ecart entre les deux groupes quant a la moyenne de PIO une heure ou 24 heures apres I'intervention. Vingt-neuf patients (33 %) du groupe avec angles etroits et sept (54 %) de celui avec dispersion pigmentaire avaient une hausse de PIO superieure a 2 mm Hg apres la procedure (p 0,00 I). Chez ces patients, la moyenne de PIO (36,4 mm Hg [ET 10,83] c. 30,3 mm Hg [ET 7,04], P = 0,05) et la hausse moyenne de la PIO (14,0 mm Hg [ET 10,63] c. 8,7 mm Hg [ET 4,73], P = 0,04) etaient significativement plus elevees chez Ie groupe avec dispersion pigmentaire qu'avec angles etroits. Parmi les patients qui etaient traites auparavant contre Ie glaucome ou dont la PIO etait superieure a 22 mm Hg avant I'intervention au laser, ceux qui avaient une dispersion pigmentaire etaient plus predisposes que les autres a avoir une hausse de PIO une heure apres I'operation (p :5 0,005). Interpretation : Les patients avec dispersion pigmentaire subissant une iridotomie pour renverser un blocage pupillaire inverse devraient etre evalues attentivement apres la procedure, car d'importantes hausses de pression necessitant un traitement peuvent survenir.
=
T
he best treatment for pigment dispersion glaucoma remains controversial. Aqueous suppression, miotic therapy and argon laser trabeculoplasty remain the preferred treatment(s) for intraocular pressure (lOP) control. Karickhoff,,2 suggested that pigment dispersion occurs because of a reverse pupil block that results in elevated pressure in the anterior chamber
with respect to the posterior chamber, pushing the iris backward onto the zonules. The role of peripheral iridotomy to break this reverse pupil block in the longterm control of pigment dispersion is promising, but the usefulness of this procedure has not been completely established. Some ophthalmologists, however, are performing laser peripheral iridotomy to obtain
CAN JOPHTHALMOL-VOL. 39, NO.3, 2004
235
YAG iridotomy in pigment glaucoma-Birt
any potential benefits to the patient, on the assumption that the procedure is simple and safe. 3 The current study was designed to examine whether patients with pigment dispersion are at higher risk for an IOP spike after laser peripheral iridotomy due to possible compromise of trabecular meshwork function, compared with patients undergoing prophylactic peripheral iridotomy for an occludable angle. METHODS
The study was conducted between November 1995 and October 1996. The Research Ethics Board of Sunnybrook and Women's College Health Sciences Centre, Toronto, approved the protocol for this prospective, nonrandomized study before the start of data acquisition. Patients were included if they required peripheral iridotomy for the presence of an occludable angle (Shaffer grade 0 or I) or had pigment dispersion (the presence of Krukenberg's spindle, iris transillumination defects and a heavily pigmented trabecular meshwor0· 5) with or without an elevated lOP or use of glaucoma medications. Evidence of glaucomatous optic nerve damage or visual field defects was not required. All patients gave informed consent after a discussion of the potential risks and benefits of the procedure. All patients were pretreated with one drop of 0.5% apraclonidine (Iopidine, Alcon Canada Inc., Mississauga, Ont.) and one drop of 1% pilocarpine about 5 minutes before the procedure, in addition to any
previously prescribed antiglaucoma medications. A LASAG (LASAG AG, Thun, Switzerland) neodymium:YAG laser was used in all cases. The power was adjusted to provide maximum iris cutting with minimum patient discomfort. The procedure was considered complete when a full-thickness defect was created. Movement of pigment forward into the anterior chamber was seen in most of the patients with a narrow angle, and backward into the posterior chamber in some of the patients with pigment dispersion. All procedures were performed by one surgeon (C.M.B.). The lOP was measured 1 hour and 24 hours after the procedure. Any rise in lOP of over 2 mm Hg greater than the baseline value was treated as clinically indicated, with topically administered ~-block ers, a-agonists or parasympathomimetics as well as orally given carbonic anhydrase inhibitors and hyperosmotic agents, as necessary. Differences in categoric variables were assessed with the X2 test, and in continuous variables with Student's t test. RESULTS
Data were collected prospectively on the first eye treated of 87 patients with occludable angles and 13 patients with pigment dispersion. The mean age (and standard deviation [SD]) was 66.4 (10.78) years and 40.5 (9.45) years respectively, a significant difference
Table I-Demographic characteristics of patients with occludable angles or pigment dispersion who underwent laser peripheral iridotomy Group
Characteristic Mean age (and standard deviation [SO]), yr % right eye % female % white Mean no. of glaucoma medications used (and SO) Mean intraocular pressure (lOP) before procedure (and SO), mm Hg Mean energy used (and SO), mJ
236
CAN J OPHTHALMOL-VOL. 39, NO.3, 2004
Occludable angle (n = 87)
Pigment dispersion (n = 13)
P value
66.4 (10.78) 47.1 55.0 67.8
40.5 (9.45) 30.1 38.5 84.6
< 0.001 0.27 0.09 0.45
0.77 (0.96)
0.69 (0.93)
0.78
20.81 (4.73) 81.6 (65.9)
20.7 (5.38) 51.4 (43.0)
0.08 0.13
YAG iridotomy in pigment glaucoma-Birt
Table 2-Changes in intraocular pressure in the two groups* Group Variable Mean lOP before procedure (and SD),mm Hg Mean lOP after procedure (and SD), mm Hg Ih 24 h Mean rise in lOP at I h (and SD), mm Hg No. (and %) of patients with rise in lOP> 2 mm Hg after procedure Mean lOP after procedure (and SD),mm Hg Mean rise in lOP after procedure (and SD), mm Hg
(t =8.20, p < 0.001). There was no difference between
the two groups in the distribution of right vs. left eyes, sex, race, mean number of medications used or mean total energy required to create the iridotomy (Table 1). There was no difference between the groups in mean lOP before the procedure, or 1 hour or 24 hours after the procedure (Table 2). Since the use of apraclonidine just before the procedure produced a decrease in lOP in many patients (58 [67%] of those with occludable angles and 6 [46%] of those with pigment dispersion), in both groups the average lOP following the procedure was similar to the value before the procedure. However, significantly more patients with pigment dispersion than with occludab1e angles experienced a rise in lOP greater than 2 mm Hg (54% [7/13] vs. 33% [29/87]) (p = 0.001, X2 = 81). When the patients who experienced such a rise in lOP were examined separately, the mean lOP was 30.3 (SD 7.04) mm Hg for the patients with nonoccludable angles and 36.4 (SD 10.83) mm Hg for those with pigment dispersion (t = -1.6, p = 0.05), with a mean rise in lOP of 8.7 (SD 4.73) mm Hg and 14.0 (SD 10.63) mm Hg respectively (t = 1.84, P = 0.04) (Table 2). Among the patients who required glaucoma medications before the procedure, those with pigment dispersion were more likely than those with occludable angles to have a pressure rise at 1 hour (F = 5.69, p = 0.005); this interaction was not observed at 24 hours (F = 2.62, P = 0.08). Among the patients with
Occludable angle
Pigment dispersion
p value
20.8 (4.7)
20.7 (4.9)
0.93
22.5 (7.5) 17.1 (4.5)
25.8 (12.2) 18.4 (4.4)
0.18 0.35
1.6 (6.3)
6.0 (10.8)
0.23
7
0.001
29 (33.3)
(53.8)
30.3 (7.0)
36.4 (10.8)
0.05
8.7 (4.7)
14.0 (10.6)
0.04
a prelaser lOP reading greater than 22 mm Hg, those with pigment dispersion were more likely than those with occludable angles to have a pressure rise at 1 hour (F = 12.21, P < 0.001); again, this interaction was not observed at 24 hours (F = 2.76, p = 0.07). Among the patients who had a rise in lOP greater than 2 mm Hg at 1 hour, the mean prelaser lOP was 15.5 mm Hg for those with pigment dispersion and 24.9 mm Hg for those with occludable angles (p < 0.05, Wilcoxon's rank-sum test 1.97). INTERPRETATION
Pigment dispersion is a cause of secondary openangle glaucoma. Work by Campbell and Jeffrey4,5 with the electron microscope has indicated that contact between the pigment epithelium of the iris and the zonular packets behind this pigment epithelium liberates pigment granules that circulate within the anterior chamber and are deposited on the corneal endothelium and within the trabecular meshwork. Contact between iris and zonules in the living eye has been shown by ultrasound biomicroscopy (UBM).6 Although such contact is seen more frequently in younger myopes than in emmetropic SUbjects, the cause of the contact was for some time unclear. The original thought was that patients with pigment dispersion had a floppier iris base that fell backward onto the zonules behind. The report by Karickhoffl,2 of iris
CAN J OPHTHALMOL-VOL 39, NO.3, 2004
237
YAG iridotomy in pigment glaucoma-Birt flattening following an iridotomy procedure, confirmed by CampbelI,7 suggested a more ~omplex mechanism. This theory posits a "reverse pupil block" that traps aliquots of aqueous in the anterior ~hamber and actively pushes the iris backward. 5,6 The theory was tested with the use of UBM, which showed posterior iris bowing on accommodation8 and blinking 9 that disappeared when the eye was at rest. Pupil block is well established as a causative factor in another major form of glaucoma, acute pupil-block angle closure. UBM has also been used to establish the fact of pupil block in this condition, showing contact between the iris tip and the lens, with resultant iris bombe and angle closure.lO Well before UBM became available, treatment to break the pupil block, first with surgical iridectomy and later by producing a peripheral iridotomy with a YAG laser, was well established. 1l - 15 This technique depends on 9pening a path between the posterior and anterior chambers to allow aqueous flow to bypass the blocked pUpil path. If, therefore, pigment dispersion is also a result of a pupil block mechanism, peripheral iridotomy should be helpful. UBM studies in patients with pigment dispersion have shown that when a patent iridotomy is present, the posterior iris bowing disappears, even under conditions that previously elicited the bowing. 8,10,16 In patients with pigment dispersion syndrome without elevated pressure, breaking of the pupil block and subsequent cessation of pigment release should be protective against the development of glaucoma, assuming that further meshwork damage does not occur. It is known that pigment dispersion can "burnout": in some patients, filtration improves with time.17 Therefore, even in patients with pigment dispersion glaucoma with elevated pressure, relief of the continual pigment deposition may allow the meshwork to heal enough to permit adequate filtration to resume. Long-term comparative studies are needed to determine whether, in addition to altering the iris-tozonule configuration, peripheral iridotomy is clinically helpful. The current study was not intended to answer the question of whether YAG laser peripheral iridotomy is a useful tool in the control or cure of pigment dispersion glaucoma. However, given the theoretical benefits of decreasing pigment load on the meshwork, YAG peripheral iridotomy can be offered to patients with pigment dispersion, just as it can to patients with an occludable angle and consequent risk for pupil-
238
CAN J OPHTHALMOL-VOL. 39, NO.3, 2004
block angle closure. The purpose of this study was to examine the immediate consequences of YAG peripheral iridotomy in patients with pigment dispersion in comparison to patients receiving prophylactic peripheral iridotomy for their narrow angle configuration. Although, taking the groups as a whole, the former were no more likely than the latter to have an lOP spike after the procedure, in patients in whom a spike did occur, the magnitude of the spike was significantly higher in those with pigment dispersion. Similarly, patients with pigment dispersion glaucoma (i.e., either using glaucoma medication or with a prelaser lOP of 22 mm Hg or higher) were more likely to experience an lOP spike than were those with pigment dispersion syndrome (using no medications and with an lOP less than 21 mm Hg) or occludable angles. The fact that the total laser energy required was less in the pigment dispersion group suggests that underlying trabecular function was more important than the amount of laser energy used, reflecting the fact that many of the occludable angles, although narrow, presumably had normally functioning trabecular meshwork and outflow facility. The patients with pigment dispersion had meshwork that was already compromised by the unusual amount of pigment deposited within it and could not handle the additional load produced by the iridotomy. CONCLUSION
The results suggest that a rise in lOP after laser peripheral iridotomy, should it occur, is greater in eyes with pigment dispersion glaucoma than in those with occludable angles. Laser iridotomy should not be considered a risk-free intervention, even with the prophylactic use of apraclonidine. Patients with pigment dispersion glaucoma undergoing an iridotomy procedure to break a reverse pupil block should therefore be carefully assessed after the procedure, as significant pressure spikes requiring treatment may occur. Failure to appreciate the presence of a pressure spike and to manage it appropriately may put vulnerable patients at unnecessary risk for further optic nerve and visual field damage. REFERENCES
1. Karickhoff JR. Pigment dispersion syndrome and pigmentary glaucoma: a new mechanism concept, a new treatment, and a new technique. Ophthalmic Surg 1992; 23:269-77.
YAG iridotomy in pigment glaucoma-Birt 2. Karickhoff JR. Reverse pupillary block in pigmentary glaucoma: follow up and new developments [letter]. Ophthalmic Surg 1993;24:562-3. 3. Pavlin CJ, Harasiewicz K, Foster FS. Posterior iris bowing in pigmentary dispersion syndrome caused by accommodation. Am J OphthalmoI1994;118:114-6. 4. Campbell DG, Jeffrey CPo Pigmentary dispersion in the human eye, SEM. Scan Electron Microsc 1979;3:329-33. 5. Campbell DG. Pigmentary dispersion and glaucoma. A new theory. Arch OphthalmoI1979;97:1667-72. 6. Potash SD, Tello C, Liebmann J, Ritch R. Ultrasound biomicroscopy in pigment dispersion syndrome. Ophthalmology 1994;101:332-9. 7. Campbell DG. Iridotomy, blinking, and pigmentary glaucoma [abstract]. Invest Ophthalmol Vis Sci 1993;34 (Suppl):993. 8. Pavlin CJ, Macken P, Trope GE, Harasiewicz K, Foster FS. Accommodation and iridotomy in the pigment dispersion syndrome. Ophthalmic Surg Lasers 1996;27: 113-20. 9. Liebmann 1M, Tello C, Chew SJ, Cohen H, Ritch R. Prevention of blinking alters iris configuration in pigment dispersion syndrome and in normal eyes. Ophthalmology 1995;102:446-55. lO. Pavlin CJ, Harasiewicz K, Foster FS. An ultrasound biomicroscopic dark-room provocative test. Ophthalmic Surg 1995;26:253-5.
11. Krupin T, Mitchell KB, Johnson MF, Becker B. The longterm effects of iridectomy for primary acute angle-closure glaucoma. Am J OphthalmoI1978;86:506--9. 12. Playfair TJ, Watson PG. Management of acute primary angle-closure glaucoma: a long-term follow-up of the results of peripheral iridectomy used as an initial procedure. Br J OphthalmoI1979;63: 17-22. 13. Ritch R. Should patients with anatomically narrow angles have prophylactic iridectomy? II. Definitive signs and gonioscopic visualization of appositional angle closure are indications for prophylactic laser iridectomy. Surv OphthalmoI1996;41:33-6. 14. Snow JT. Value of prophylactic peripheral iridectomy on the second eye in angle-closure glaucoma. Trans Ophthalmol Soc UK 1977;97:189-91. 15. Wilensky JT. Narrow angles accompanied by slit-lamp and gonioscopic evidence of risk are indications for prophylactic laser iridectomy. Surv Ophthalmol 1996;41: 31-2. 16. Pavlin CJ, Macken P, Trope G, Feldman F, Harasiewicz K, Foster FS. Ultrasound biomicroscopic features of pigmentary glaucoma. Can J OphthalmoI1994;29:187-92. 17. Speakman JS. Pigmentary dispersion. Br J Ophthalmol 1981;65:249-51. Key words: glaucoma, pigment dispersion, YAG laser iridotomy
CAN J OPHTHALMOL-VOL. 39, NO.3, 2004
239
Intraocular pressure spike after YAG iridotomy in patients with pigment dispersion Catherine M. Birt, MD, FRCSC ABSTRACT • RESUME Background: The role of laser peripheral iridotomy to break a suspected reverse pupil block in the long-term control of pigment dispersion is promising, but the usefulness of this procedure has not been completely established. The author examined whether patients with pigment dispersion are at higher risk for an intraocular pressure (lOP) spike after laser peripheral iridotomy due to possible compromise of trabecular meshwork function, compared with patients undergoing prophylactic peripheral iridotomy for an occludable angle. Methods: Data were collected prospectively on the first eye of 87 patients with occludable angles and 13 patients with pigment dispersion treated with peripheral laser iridotomy between November 1995 and October 1996 at the glaucoma service of a university-affiliated hospital in Toronto~ All patients received one drop of 0.5% apraclonidine before the procedure. lOP was measured before and I and 24 hours after the procedure. Results: There was no difference between the two groups in the distribution of right vs. left eyes, sex, race, the mean total energy required to produce a patent iridotomy, the mean number of medications used or the mean lOP before the procedure. The patients with pigment dispersion were significantly younger than those with occludable angles (mean age [and standard deviation (SD)] 40.5 [9.45] years vs.66.4 [10.78] years) (p < 0.00 I).There was no difference between the two groups in mean lOP at I hour or at 24 hours. Twenty-nine patients (33%) in the occludable angle group and seven (54%) in the pigment dispersion group had an lOP spike greater than 2 mm Hg after the procedure (p = 0.00 I). Among these patients, the mean lOP (36.4 [SD 10.83] mm Hg vs. 30.3 [SD 7.04] mm Hg, p = 0.05) and the mean rise in lOP (14.0 [SD 10.63] mm Hg vs. 8.7 [SD 4.73] mm Hg, p = 0.04) were significantly higher in those with pigment dispersion than in those with occludable angles. Among the patients who used antiglaucoma medications before the procedure or had a prelaser lOP level greater than 22 mm Hg, those with pigment dispersion were more likely than those with occludable angles to have an lOP spike at I hour (p ::; 0.005).
From Sunnybrook and Women's College Health Sciences Centre, Toronto, Ont. Presented in part as a poster paper at the Association for Research in Vision and Ophthalmology meeting held in Fort Lauderdale, Fla., May ll-15, 1997 Originally received May 6, 2003 Accepted for publication Oct. 10, 2003
234
YAG iridotomy in pigment glaucoma-Birt
Reprint requests to: Dr. Catherine M. Birt, M1 302a, Department of Ophthalmology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Ave., Toronto ON M4N 3M5; fax (416) 4804481, [email protected] This article has been peer-reviewed. Can J Ophthalmol 2004;39:234-9
YAG iridotomy in pigment glaucoma-Birt
Interpretation: Patients with pigment dispersion undergoing iridotomy to break a reverse pupil block should be carefully assessed after the procedure, as significant pressure spikes requiring treatment may occur.
Contexte : La technique de I'iridotomie peripherique au laser pour renverser ce qu'on croit etre un blocage pupillaire inverse dans la maitrise a long terme de la dispersion pigmentaire est prometteuse, mais I'on n'a pas encore etabli entierement I'utilite de ce procede. L'auteur s'est demande si les patients atteints de dispersion pigmentaire etaient plus a risque d'avoir une hausse de pression intraoculaire (PIO) apres une iridotomie peripherique au laser, acause de la compromission possible de la fonction du trabeculum, que ceux qui subissent une iridotomie peripherique prophylactique pour un angle etroit. Methodes: Les donnees ont ete recueillies de fa~on prospective sur Ie premier reil de 87 patients ayant des angles etroits et de 13 patients atteints de dispersion pigmentaire, qui avaient subis une iridotomie peripherique au laser entre les mois de novembre 1995 et octobre 1996 au service du glaucome d'un h6pital universitaire de Toronto. Les patients ont tous re~u une goutte d'apraclonidine a 0,5 % avant la procedure. La PIO a ete mesuree avant ainsi qu'une heure et 24 heures apres la procedure. Resultats: L'on n'a pas trouve d'ecart entre les deux groupes quant a la repartition entre les yeux droits et gauches, au sexe, a la race, a I'energie totale moyenne requise pour produire une iridotomie permeable, au nombre moyen de medicaments utilises et a la PIO moyenne avant la procedure. Les patients qui avaient une dispersion pigmentaire etaient beaucoup plus jeunes que ceux qui avaient des angles etroits (age moyen 40,5 ans [ecart type (ET) 9,45] c. 66,4 ans [ET 10,78]) (p < 0,00 I). II n'y avait pas d'ecart entre les deux groupes quant a la moyenne de PIO une heure ou 24 heures apres I'intervention. Vingt-neuf patients (33 %) du groupe avec angles etroits et sept (54 %) de celui avec dispersion pigmentaire avaient une hausse de PIO superieure a 2 mm Hg apres la procedure (p 0,00 I). Chez ces patients, la moyenne de PIO (36,4 mm Hg [ET 10,83] c. 30,3 mm Hg [ET 7,04], P = 0,05) et la hausse moyenne de la PIO (14,0 mm Hg [ET 10,63] c. 8,7 mm Hg [ET 4,73], P = 0,04) etaient significativement plus elevees chez Ie groupe avec dispersion pigmentaire qu'avec angles etroits. Parmi les patients qui etaient traites auparavant contre Ie glaucome ou dont la PIO etait superieure a 22 mm Hg avant I'intervention au laser, ceux qui avaient une dispersion pigmentaire etaient plus predisposes que les autres a avoir une hausse de PIO une heure apres I'operation (p :5 0,005). Interpretation : Les patients avec dispersion pigmentaire subissant une iridotomie pour renverser un blocage pupillaire inverse devraient etre evalues attentivement apres la procedure, car d'importantes hausses de pression necessitant un traitement peuvent survenir.
=
T
he best treatment for pigment dispersion glaucoma remains controversial. Aqueous suppression, miotic therapy and argon laser trabeculoplasty remain the preferred treatment(s) for intraocular pressure (lOP) control. Karickhoff,,2 suggested that pigment dispersion occurs because of a reverse pupil block that results in elevated pressure in the anterior chamber
with respect to the posterior chamber, pushing the iris backward onto the zonules. The role of peripheral iridotomy to break this reverse pupil block in the longterm control of pigment dispersion is promising, but the usefulness of this procedure has not been completely established. Some ophthalmologists, however, are performing laser peripheral iridotomy to obtain
CAN JOPHTHALMOL-VOL. 39, NO.3, 2004
235
YAG iridotomy in pigment glaucoma-Birt
any potential benefits to the patient, on the assumption that the procedure is simple and safe. 3 The current study was designed to examine whether patients with pigment dispersion are at higher risk for an IOP spike after laser peripheral iridotomy due to possible compromise of trabecular meshwork function, compared with patients undergoing prophylactic peripheral iridotomy for an occludable angle. METHODS
The study was conducted between November 1995 and October 1996. The Research Ethics Board of Sunnybrook and Women's College Health Sciences Centre, Toronto, approved the protocol for this prospective, nonrandomized study before the start of data acquisition. Patients were included if they required peripheral iridotomy for the presence of an occludable angle (Shaffer grade 0 or I) or had pigment dispersion (the presence of Krukenberg's spindle, iris transillumination defects and a heavily pigmented trabecular meshwor0· 5) with or without an elevated lOP or use of glaucoma medications. Evidence of glaucomatous optic nerve damage or visual field defects was not required. All patients gave informed consent after a discussion of the potential risks and benefits of the procedure. All patients were pretreated with one drop of 0.5% apraclonidine (Iopidine, Alcon Canada Inc., Mississauga, Ont.) and one drop of 1% pilocarpine about 5 minutes before the procedure, in addition to any
previously prescribed antiglaucoma medications. A LASAG (LASAG AG, Thun, Switzerland) neodymium:YAG laser was used in all cases. The power was adjusted to provide maximum iris cutting with minimum patient discomfort. The procedure was considered complete when a full-thickness defect was created. Movement of pigment forward into the anterior chamber was seen in most of the patients with a narrow angle, and backward into the posterior chamber in some of the patients with pigment dispersion. All procedures were performed by one surgeon (C.M.B.). The lOP was measured 1 hour and 24 hours after the procedure. Any rise in lOP of over 2 mm Hg greater than the baseline value was treated as clinically indicated, with topically administered ~-block ers, a-agonists or parasympathomimetics as well as orally given carbonic anhydrase inhibitors and hyperosmotic agents, as necessary. Differences in categoric variables were assessed with the X2 test, and in continuous variables with Student's t test. RESULTS
Data were collected prospectively on the first eye treated of 87 patients with occludable angles and 13 patients with pigment dispersion. The mean age (and standard deviation [SD]) was 66.4 (10.78) years and 40.5 (9.45) years respectively, a significant difference
Table I-Demographic characteristics of patients with occludable angles or pigment dispersion who underwent laser peripheral iridotomy Group
Characteristic Mean age (and standard deviation [SO]), yr % right eye % female % white Mean no. of glaucoma medications used (and SO) Mean intraocular pressure (lOP) before procedure (and SO), mm Hg Mean energy used (and SO), mJ
236
CAN J OPHTHALMOL-VOL. 39, NO.3, 2004
Occludable angle (n = 87)
Pigment dispersion (n = 13)
P value
66.4 (10.78) 47.1 55.0 67.8
40.5 (9.45) 30.1 38.5 84.6
< 0.001 0.27 0.09 0.45
0.77 (0.96)
0.69 (0.93)
0.78
20.81 (4.73) 81.6 (65.9)
20.7 (5.38) 51.4 (43.0)
0.08 0.13
YAG iridotomy in pigment glaucoma-Birt
Table 2-Changes in intraocular pressure in the two groups* Group Variable Mean lOP before procedure (and SD),mm Hg Mean lOP after procedure (and SD), mm Hg Ih 24 h Mean rise in lOP at I h (and SD), mm Hg No. (and %) of patients with rise in lOP> 2 mm Hg after procedure Mean lOP after procedure (and SD),mm Hg Mean rise in lOP after procedure (and SD), mm Hg
(t =8.20, p < 0.001). There was no difference between
the two groups in the distribution of right vs. left eyes, sex, race, mean number of medications used or mean total energy required to create the iridotomy (Table 1). There was no difference between the groups in mean lOP before the procedure, or 1 hour or 24 hours after the procedure (Table 2). Since the use of apraclonidine just before the procedure produced a decrease in lOP in many patients (58 [67%] of those with occludable angles and 6 [46%] of those with pigment dispersion), in both groups the average lOP following the procedure was similar to the value before the procedure. However, significantly more patients with pigment dispersion than with occludab1e angles experienced a rise in lOP greater than 2 mm Hg (54% [7/13] vs. 33% [29/87]) (p = 0.001, X2 = 81). When the patients who experienced such a rise in lOP were examined separately, the mean lOP was 30.3 (SD 7.04) mm Hg for the patients with nonoccludable angles and 36.4 (SD 10.83) mm Hg for those with pigment dispersion (t = -1.6, p = 0.05), with a mean rise in lOP of 8.7 (SD 4.73) mm Hg and 14.0 (SD 10.63) mm Hg respectively (t = 1.84, P = 0.04) (Table 2). Among the patients who required glaucoma medications before the procedure, those with pigment dispersion were more likely than those with occludable angles to have a pressure rise at 1 hour (F = 5.69, p = 0.005); this interaction was not observed at 24 hours (F = 2.62, P = 0.08). Among the patients with
Occludable angle
Pigment dispersion
p value
20.8 (4.7)
20.7 (4.9)
0.93
22.5 (7.5) 17.1 (4.5)
25.8 (12.2) 18.4 (4.4)
0.18 0.35
1.6 (6.3)
6.0 (10.8)
0.23
7
0.001
29 (33.3)
(53.8)
30.3 (7.0)
36.4 (10.8)
0.05
8.7 (4.7)
14.0 (10.6)
0.04
a prelaser lOP reading greater than 22 mm Hg, those with pigment dispersion were more likely than those with occludable angles to have a pressure rise at 1 hour (F = 12.21, P < 0.001); again, this interaction was not observed at 24 hours (F = 2.76, p = 0.07). Among the patients who had a rise in lOP greater than 2 mm Hg at 1 hour, the mean prelaser lOP was 15.5 mm Hg for those with pigment dispersion and 24.9 mm Hg for those with occludable angles (p < 0.05, Wilcoxon's rank-sum test 1.97). INTERPRETATION
Pigment dispersion is a cause of secondary openangle glaucoma. Work by Campbell and Jeffrey4,5 with the electron microscope has indicated that contact between the pigment epithelium of the iris and the zonular packets behind this pigment epithelium liberates pigment granules that circulate within the anterior chamber and are deposited on the corneal endothelium and within the trabecular meshwork. Contact between iris and zonules in the living eye has been shown by ultrasound biomicroscopy (UBM).6 Although such contact is seen more frequently in younger myopes than in emmetropic SUbjects, the cause of the contact was for some time unclear. The original thought was that patients with pigment dispersion had a floppier iris base that fell backward onto the zonules behind. The report by Karickhoffl,2 of iris
CAN J OPHTHALMOL-VOL 39, NO.3, 2004
237
YAG iridotomy in pigment glaucoma-Birt flattening following an iridotomy procedure, confirmed by CampbelI,7 suggested a more ~omplex mechanism. This theory posits a "reverse pupil block" that traps aliquots of aqueous in the anterior ~hamber and actively pushes the iris backward. 5,6 The theory was tested with the use of UBM, which showed posterior iris bowing on accommodation8 and blinking 9 that disappeared when the eye was at rest. Pupil block is well established as a causative factor in another major form of glaucoma, acute pupil-block angle closure. UBM has also been used to establish the fact of pupil block in this condition, showing contact between the iris tip and the lens, with resultant iris bombe and angle closure.lO Well before UBM became available, treatment to break the pupil block, first with surgical iridectomy and later by producing a peripheral iridotomy with a YAG laser, was well established. 1l - 15 This technique depends on 9pening a path between the posterior and anterior chambers to allow aqueous flow to bypass the blocked pUpil path. If, therefore, pigment dispersion is also a result of a pupil block mechanism, peripheral iridotomy should be helpful. UBM studies in patients with pigment dispersion have shown that when a patent iridotomy is present, the posterior iris bowing disappears, even under conditions that previously elicited the bowing. 8,10,16 In patients with pigment dispersion syndrome without elevated pressure, breaking of the pupil block and subsequent cessation of pigment release should be protective against the development of glaucoma, assuming that further meshwork damage does not occur. It is known that pigment dispersion can "burnout": in some patients, filtration improves with time.17 Therefore, even in patients with pigment dispersion glaucoma with elevated pressure, relief of the continual pigment deposition may allow the meshwork to heal enough to permit adequate filtration to resume. Long-term comparative studies are needed to determine whether, in addition to altering the iris-tozonule configuration, peripheral iridotomy is clinically helpful. The current study was not intended to answer the question of whether YAG laser peripheral iridotomy is a useful tool in the control or cure of pigment dispersion glaucoma. However, given the theoretical benefits of decreasing pigment load on the meshwork, YAG peripheral iridotomy can be offered to patients with pigment dispersion, just as it can to patients with an occludable angle and consequent risk for pupil-
238
CAN J OPHTHALMOL-VOL. 39, NO.3, 2004
block angle closure. The purpose of this study was to examine the immediate consequences of YAG peripheral iridotomy in patients with pigment dispersion in comparison to patients receiving prophylactic peripheral iridotomy for their narrow angle configuration. Although, taking the groups as a whole, the former were no more likely than the latter to have an lOP spike after the procedure, in patients in whom a spike did occur, the magnitude of the spike was significantly higher in those with pigment dispersion. Similarly, patients with pigment dispersion glaucoma (i.e., either using glaucoma medication or with a prelaser lOP of 22 mm Hg or higher) were more likely to experience an lOP spike than were those with pigment dispersion syndrome (using no medications and with an lOP less than 21 mm Hg) or occludable angles. The fact that the total laser energy required was less in the pigment dispersion group suggests that underlying trabecular function was more important than the amount of laser energy used, reflecting the fact that many of the occludable angles, although narrow, presumably had normally functioning trabecular meshwork and outflow facility. The patients with pigment dispersion had meshwork that was already compromised by the unusual amount of pigment deposited within it and could not handle the additional load produced by the iridotomy. CONCLUSION
The results suggest that a rise in lOP after laser peripheral iridotomy, should it occur, is greater in eyes with pigment dispersion glaucoma than in those with occludable angles. Laser iridotomy should not be considered a risk-free intervention, even with the prophylactic use of apraclonidine. Patients with pigment dispersion glaucoma undergoing an iridotomy procedure to break a reverse pupil block should therefore be carefully assessed after the procedure, as significant pressure spikes requiring treatment may occur. Failure to appreciate the presence of a pressure spike and to manage it appropriately may put vulnerable patients at unnecessary risk for further optic nerve and visual field damage. REFERENCES
1. Karickhoff JR. Pigment dispersion syndrome and pigmentary glaucoma: a new mechanism concept, a new treatment, and a new technique. Ophthalmic Surg 1992; 23:269-77.
YAG iridotomy in pigment glaucoma-Birt 2. Karickhoff JR. Reverse pupillary block in pigmentary glaucoma: follow up and new developments [letter]. Ophthalmic Surg 1993;24:562-3. 3. Pavlin CJ, Harasiewicz K, Foster FS. Posterior iris bowing in pigmentary dispersion syndrome caused by accommodation. Am J OphthalmoI1994;118:114-6. 4. Campbell DG, Jeffrey CPo Pigmentary dispersion in the human eye, SEM. Scan Electron Microsc 1979;3:329-33. 5. Campbell DG. Pigmentary dispersion and glaucoma. A new theory. Arch OphthalmoI1979;97:1667-72. 6. Potash SD, Tello C, Liebmann J, Ritch R. Ultrasound biomicroscopy in pigment dispersion syndrome. Ophthalmology 1994;101:332-9. 7. Campbell DG. Iridotomy, blinking, and pigmentary glaucoma [abstract]. Invest Ophthalmol Vis Sci 1993;34 (Suppl):993. 8. Pavlin CJ, Macken P, Trope GE, Harasiewicz K, Foster FS. Accommodation and iridotomy in the pigment dispersion syndrome. Ophthalmic Surg Lasers 1996;27: 113-20. 9. Liebmann 1M, Tello C, Chew SJ, Cohen H, Ritch R. Prevention of blinking alters iris configuration in pigment dispersion syndrome and in normal eyes. Ophthalmology 1995;102:446-55. lO. Pavlin CJ, Harasiewicz K, Foster FS. An ultrasound biomicroscopic dark-room provocative test. Ophthalmic Surg 1995;26:253-5.
11. Krupin T, Mitchell KB, Johnson MF, Becker B. The longterm effects of iridectomy for primary acute angle-closure glaucoma. Am J OphthalmoI1978;86:506--9. 12. Playfair TJ, Watson PG. Management of acute primary angle-closure glaucoma: a long-term follow-up of the results of peripheral iridectomy used as an initial procedure. Br J OphthalmoI1979;63: 17-22. 13. Ritch R. Should patients with anatomically narrow angles have prophylactic iridectomy? II. Definitive signs and gonioscopic visualization of appositional angle closure are indications for prophylactic laser iridectomy. Surv OphthalmoI1996;41:33-6. 14. Snow JT. Value of prophylactic peripheral iridectomy on the second eye in angle-closure glaucoma. Trans Ophthalmol Soc UK 1977;97:189-91. 15. Wilensky JT. Narrow angles accompanied by slit-lamp and gonioscopic evidence of risk are indications for prophylactic laser iridectomy. Surv Ophthalmol 1996;41: 31-2. 16. Pavlin CJ, Macken P, Trope G, Feldman F, Harasiewicz K, Foster FS. Ultrasound biomicroscopic features of pigmentary glaucoma. Can J OphthalmoI1994;29:187-92. 17. Speakman JS. Pigmentary dispersion. Br J Ophthalmol 1981;65:249-51. Key words: glaucoma, pigment dispersion, YAG laser iridotomy
CAN J OPHTHALMOL-VOL. 39, NO.3, 2004
239