JOURNAL OF OCULAR PHARMACOLOGY AND THERAPEUTICS Volume 30, Number 7, 2014 ª Mary Ann Liebert, Inc. DOI: 10.1089/jop.2014.0008
Does the Use of Preoperative Antiglaucoma Medications Influence Trabeculectomy Success? ¨ ztu¨rker,1 Can O ¨ ztu¨rker,2 Sukru Bayraktar,3 Cigdem Altan,2 and Omer Faruk Yilmaz 2 Zeynep Kayaarasi O
Abstract
Purpose: To investigate the influence of preoperative antiglaucoma medications on trabeculectomy outcome. Methods: Two hundred fifteen eyes, which underwent primary trabeculectomy, were retrospectively analyzed. The average follow-up was 39.8 – 30.3 months. The only cases of primary open-angle glaucoma, with or without pseudoexfoliation (PXF), were included. ‘‘Complete success’’ was defined as intraocular pressure (IOP) < 18 mmHg without glaucoma medications, whereas relative success was defined as the same IOP target with medications. The influence of the preoperatively used glaucoma medications on surgical success was analyzed by univariate Pearson correlation and multivariate (ordinal) regression analysis. Results: There were 118 male (54.9%) and 97 female (45.1%) patients with a mean age of 66.9 – 9.3 years. PXF glaucoma (PXFG) was present in 93 eyes (43.3%). In 33 patients (15.3%), diabetes mellitus (DM) was present. Complete success was achieved in 116 eyes (54%), relative success in 81 eyes (37.6%), and failure in 18 eyes (8.4%). Neither the total number nor the duration of glaucoma medications used before trabeculectomy was found to have any statistically significant influence on surgical success. In statistical analysis, a combination of topical beta-blocker and carbonic anhydrase inhibitor (BB + CAI) used before surgery was found to be associated with statistically better outcome, whereas the preoperative use of topical beta-blockers alone could have a negative influence on success. PXF was shown to be independently associated with trabeculectomy outcome on multivariate regression analysis. Conclusion: The glaucoma medications used preoperatively were not found to have any statistically significant negative influence on the trabeculectomy outcome and use of the combined BB + CAI preparation could have a positive influence, whereas the use of topical beta-blockers alone could have a negative influence on success, although not statistically significant. The presence of PXF was independently associated with a better surgical outcome.
Introduction
D
espite the recent trend toward minimally invasive glaucoma surgery, trabeculectomy is still the most frequently used procedure for the treatment of glaucoma. According to statistics, the number of surgical treatments has been increasing as the last glaucoma drug entered the market more than 10 years ago, and no new candidate molecule is present on the horizon.1,2 The glaucoma specialist usually prefers trying all the available medical options before switching to surgery as the pressure-lowering effect of trabeculectomy usually decreases with time. This brings to the question of whether the prior use of glaucoma medications would have any adverse influence on the outcome of filtering surgery.
1 2 3
The primary aim of this study was to evaluate the influence of preoperative antiglaucoma medications on trabeculectomy success. In addition, the effect of some of the patient-related ocular and systemic factors such as age, PXF, and diabetes mellitus (DM) was also investigated by using univariate and multivariate analyses.
Methods The medical records of 391 patients, who underwent primary trabeculectomy without the use of intraoperative antimetabolites in the Glaucoma Service of Beyoglu Eye Training and Research Hospital, were retrospectively analyzed. The study group composed of those patients with a regular follow-up of more than 12 months. Only eyes with
Istanbul Education and Research Hospital, Istanbul, Turkey. Beyoglu Eye Training and Research Hospital, Istanbul, Turkey. Istanbul Surgery Hospital, Istanbul, Turkey.
554
GLAUCOMA MEDICATIONS’ EFFECT ON TRABECULECTOMY SUCCESS
Table 1.
Table 3. Postoperative Change in Intraocular Pressure
Patient Demographics
Gender (male/female) Mean age (years) Pseudoexfoliation (PXF) DM
555
118/97 66.9 – 9.3 93/215 (43.2%) 33/215 (15.3%)
DM, diabetes mellitus.
primary glaucoma with or without PXF were included. Those eyes with secondary glaucoma and any prior intraocular surgery such as trabeculectomy and/or cataract extraction were also excluded. According to those inclusion criteria, our study group composed of 215 eyes of 215 patients. Ethical approval and informed consent were obtained according to the Helsinki Declaration.
Trabeculectomy technique In all the study eyes, an identical surgical technique was used. After sub-Tenon’s anesthesia, a limbus-based conjunctival incision approximately 10 mm behind the limbus was made, and then a 3 · 4 mm rectangular scleral flap in semiscleral thickness was prepared. After corneal paracentesis, a deep scleral block excision measuring *1 mm2 was made, and peripheral iridectomy was performed. The scleral flap was closed with 2 or 3 10/0 nylon sutures. The conjunctiva was closed with absorbable 8/0 silk sutures in a continuous fashion. Topical antibiotic and corticosteroids were used in all eyes for at least 6 weeks postoperatively, whereas cycloplegics were given in eyes with significant inflammation and/or anterior chamber shallowing in the postoperative period. Digital massage and/or laser suture lysis were performed, if necessary, during the first postoperative week. None of the known antimetabolites such as mitomycine-C or 5-fluorouracil was used either during or after the operation.
Study population Patient age, gender, presence of PXF, and systemic disease such as DM were recorded (Table 1). The total number of glaucoma medications, duration of medication use, name of glaucoma medications, and the preoperative IOP with/ without medications were also recorded (Table 2). In 111 of the 215 patients, according to extensive cupping, visual field
IOP Preoperative 1st year 2nd year 3rd year 4th year 5th year 6th year 7th year 8th year 9th year
N
Minimum
Maximum
Mean
SD
215 215 136 90 65 50 41 33 23 6
9.00 1.00 4.00 1.00 2.00 3.00 2.00 8.00 7.00 8.00
55.00 26.00 24.00 23.00 36.00 28.00 23.00 20.00 19.00 21.00
20.79 12.37 13.34 13.52 14.11 14.16 12.32 13.48 13.09 15.00
7.66 4.23 3.69 4.63 5.43 5.10 3.60 2.68 3.07 4.82
IOP, intraocular pressure.
defects, and acute exacerbation of IOP elevation, trabeculectomy was performed as the first choice of treatment.
Success criteria ‘‘Complete success’’ was defined as postoperative IOP measuring less than 18 mmHg in every control visit without glaucoma medications, whereas ‘‘relative success’’ was defined as IOP below 18 mmHg, but glaucoma medications were needed to achieve that target. Surgery was considered to be a failure for eyes, in which IOP could not be kept below 18 mmHg, despite using glaucoma medications.
Statistical analysis The SPSS 11.5 for Windows was used for statistical analysis. The influence of patient-related factors and preoperative glaucoma medications was analyzed by using univariate and multivariate methods. Pearson correlation analysis was used for univariate analysis, whereas ordinal regression was used for multivariate analysis as success was defined on an ordinal scale. The Kaplan–Meier survival analysis and log-rank test were also performed for those factors found to be statistically significant in multivariate analysis. Statistical significance was confirmed as the P-value less than 0.05.
Results The average follow-up was 39.8 – 30.3 months. The intraocular pressure (IOP) was statistically reduced when
Table 2.
Preoperative Glaucoma Medications
Average number of medications Average duration of medication use (months) Number of medications
Sympathomimetics Beta-blockers (BB) Carbonic Anhydrase (CA) inhibitor Combined BB + CA inhibitor Prostaglandins Miotics
1.06 – 1.33
Table 4. Postoperative Change in Glaucoma Medications
10.1 – 20.7 None/111 One/31 Two/37 Three or more/36 18 76 18
(51.6%) (14.4%) (17.2%) (16.7%) (8.4%) (35.3%) (8.4%)
33 (15.3%) 38 (17.7%) 27 (12.6%)
Preoperative 1st year 2nd year 3rd year 4th year 5th year 6th year 7th year 8th year 9th year
N
Minimum
Maximum
Mean
SD
215 215 136 90 65 50 41 33 23 6
0 0 0 0 0 0 0 0 0 0
4 4 3 4 3 4 3 3 3 3
1.06 0.37 0.71 1.06 1.25 1.60 1.76 1.67 1.70 1.67
1.33 0.74 0.81 0.94 1.00 1.09 1.16 1.22 1.06 1.21
¨ ZTU ¨ RKER ET AL. O
556
Table 5.
Factors not Statistically Associated with Surgical Success Pearson correlation coefficient P-value
Number of medications Average duration of medication use At least 6 months of medication use Sympathomimetics Carbonic anhydrase (CA) inhibitor Prostaglandins Miotics Presence of DM
0.057 - 0.016 0.022 0.021 - 0.031 0.089 - 0.116 - 0.121
0.402 0.831 0.853 0.763 0.647 0.196 0.691 0.079
compared with preoperative levels during the postoperative visits (P < 0.05, Table 3). Postoperative changes in the number of glaucoma medications are shown in Table 4; there was a statistically significant reduction only during the first year postoperatively (P < 0.05). At the last visit, complete success was present in 116 (54%) eyes, relative success in 81 (37.6%) eyes, and failure in 18 (8.4%) eyes. The factors not associated with surgical success are shown in Table 5, whereas those significantly associated with success are shown in Table 6. In the univariate Pearson correlation analysis, age (P = 0.02), presence of PXF (P = 0.0001), and preoperative use of combined beta-blocker and carbonic anhydrase preparation, (P = 0.001) were found to be associated with better surgical outcome, whereas the preoperative use of topical beta-blockers alone had a negative influence on success (P = 0.01). In multivariate regression analysis, the presence of PXF (P = 0.001) and the preoperative use of combined beta-blocker and carbonic anhydrase preparation (P = 0.001) were found to be the statistically significant factors associated with better trabeculectomy outcome, whereas the age (P = 0.12) and use of preoperative beta-blockers (P = 0.205) did not influence surgical success. The Kaplan–Meier survival analysis was performed using PXF and the preoperative use of combined BB + CAI as factors. The survival curves for complete success are shown in Figs. 1 and 2. In the log-rank test, complete success (event was defined as ‘‘the initiation of glaucoma medications to keep IOP below 18 mmHg’’) was found to be statistically
FIG. 1. Kaplan–Meier survival analysis of complete success in eyes with and without pseudoexfoliation (PXF). better in eyes with PXF as compared with no PXF (P = 0.01). However, no statistically significant difference was found between those eyes, in which a combined BB + CAI preparation was used preoperatively as compared with those, in which no similar treatment was given (P = 0.161).
Discussion The current practice of glaucoma treatment almost always involves using all the medication and laser options before
Table 6. Statistically Significant Factors Related to Surgical Success Univariate analysis Pearson correlation
Age PXF Preoperative BB Preoperative combined BB + CAI
Multivariate analysis ordinal regression
Coefficient
P
P
0.215 0.256 - 0.176 0.219
0.02a 0.0001a 0.01a 0.001a
0.12 0.001a 0.205 0.001a
a p-Value < 0.05. BB + CAI, beta-blocker and carbonic anhydrase inhibitor; PXF, pseudoexfoliation.
FIG. 2. Kaplan–Meier survival analysis of complete success in eyes with and without preoperative use of a combined beta-blocker and carbonic anhydrase inhibitor.
GLAUCOMA MEDICATIONS’ EFFECT ON TRABECULECTOMY SUCCESS
proceeding with surgery. The rationale behind this approach is the slow but apparent rate of decrease in trabeculectomy success with time.3,4 Keeping trabeculectomy as the reserve treatment option, on the other hand, requires the use of glaucoma medications for long time periods and brings to the question of whether those medications have any adverse influence on the surgical outcome. In this retrospective study, we tried to analyze whether the preoperative use of glaucoma medications has any influence on trabeculectomy success. Our study group composed of patients who underwent primary trabeculectomy without the use of intraoperative antimetabolites. Today, we know that antimetabolites are used more often in trabeculectomy, so a group with this specificity may be included in future studies. A number of studies showed that long-term preoperative use of glaucoma drugs could change the conjunctival cell profile, induce inflammation, and then reduce the chance of further trabeculectomy success.5–9 Results of clinical studies, on the other hand, were controversial and have failed to demonstrate clear evidence supporting the histologic data.10,11 In approximately half of our study population, the total duration of glaucoma medication usage was less than 6 months—we called it as the no-medication group. According to extensive cupping, visual field defects, and acute exacerbation of IOP elevation, trabeculectomy was performed as the first choice of treatment in this group. Five different medication groups were used in the remaining eyes: miotics, sympathomimetics, beta-blockers, carbonic anhydrase inhibitors, and prostaglandins. The influence of the combined preparation of BB and CAI was analyzed separately as it was used in a relatively large proportion of eyes (more than 15% of eyes in the study). We could not find any statistically significant association with the total number of preoperative glaucoma medications and/or the average duration of medication use with the trabeculectomy success in the univariate correlation analysis. When we analyzed the medication groups individually, we found that the preoperative use of a combined BB + CAI drug was associated with a better outcome in both the univariate and multivariate analysis. There was a significant negative association between BB use and surgical success in the univariate correlation, but not in the multivariate regression. The influence of preoperative combined BB + CAI use on trabeculectomy outcome was also analyzed by Kaplan– Meier survival curves, but we failed to find an association in the log-rank test. We believe that a weak association could be present, but due to the relatively shorter follow-up of those eyes (17.5 – 7.9 months) receiving this preparation before surgery versus the rest of the eyes in the study (43.8 – 31.1 months), the Kaplan–Meier analysis might have failed to demonstrate it. The presence of DM was not found to be related to surgical success. Both patient age and the presence of PXF were found to be positively correlated with the success in the univariate correlation analysis, but only PXF was shown to be independently associated with trabeculectomy outcome on multivariate regression analysis. The significant influence of PXF on trabeculectomy outcome was also shown by using Kaplan–Meier survival analysis. There are controversial studies on PXFG and trabeculectomy in the literature.12–16 Our study clearly showed that the presence of PXF increased the trabeculectomy survival independently from other factors. In our study population,
557
the mean preoperative IOP was not too high (less than 21 mmHg with medications), but approximately half of them (43.2%) had PXF. Similar to the other Mediterranean countries, we usually see a lot of cases with PXFG and they usually require surgery.17,18 How could we explain the positive influence of PXF on trabeculectomy success? The disruption of blood–aqueous barrier has been shown to be somewhat greater and the postoperative inflammation was more severe in eyes with PXFG.19–21 Those events often led to decreased aqueous humor production rates. The chi-square test showed that the incidence of postoperative hypotony was statistically higher in those eyes with PXF (P = 0.0001). Although we did not measure the inflow or investigate bleb morphology in our study, we think that the decreased aqueous production might contribute to better surgical outcomes by forming more diffuse and favorable blebs without encapsulation. In summary, we found that preoperative use of glaucoma medications did not have any adverse influence on trabeculectomy success. In statistical analysis, a combination of topical beta-blocker and carbonic anhydrase inhibitor (BB + CAI) used before surgery was found to be associated with statistically better outcome, whereas the preoperative use of topical beta-blockers alone could have a negative influence on success. However, we advise taking this cautiously as the follow-up was relatively shorter in those groups of eyes. We also believe that the association between the presence of PXF and improved trabeculectomy success is apparent.
Author Disclosure Statement None of the authors has a financial or proprietary interest in any material or method mentioned.
References 1. Rotchford, A. What is practical in Glaucoma management? Eye 19:1125–1132, 2005. 2. Thomas, R., Paul, P., and Rao, G.N. Present status of Eye care in India. Survey Ophthalmol. 50:85–101, 2005. 3. Jerndal, T., and Lundstrom, M. 330 Trabeculectomies: a follow study through ½–3 years. Acta Ophthalmol. 55:52– 62, 1977. 4. Nouri-Mahadevi, K., Brigatti, L., Weitzman, M., and Caprioli, J. Outcomes of trabeculectomy for primary open angle glaucoma. Ophthalmology. 102:1760–1769, 1995. 5. Noecker, R.J., Herrygers, L.A., and Anwaruddin, R. Corneal and conjunctival changes caused by commonly used glaucoma medications. Cornea. 23:490–496, 2004. 6. Nenciu, A., Stefan, C., and Ardelean, C. Structural and immunohistochemical changes of conjunctiva induced by topical glaucoma medication. Oftalmologia. 48:35–42, 2004. 7. Turac¸li, E., Budak, K., Kaur, A., Mizrak, B., and Ekinci, C. The effects of long-term topical glaucoma medication on conjunctival impression cytology. Int. Ophthalmol. 21:27– 33, 1997. 8. Young, T.L., Higginbotham, E.J., Zou, X.L., and Farber, M.D. Effects of topical glaucoma drugs on fistulized rabbit conjunctiva. Ophthalmology. 97:1423–1427, 1990. 9. Lavin, M.J., Wormald, R.P.L., Migdal, C.S., and Hitchings, R.A. The influence of prior therapy on the success of trabeculectomy. Arch. Ophthalmol. 108:1543–1548, 1990.
558
10. D’Ermo, F., Bonomi, L., and Doro, D. A critical analysis of the long term results of trabeculectomy. Am. J. Ophthalmol. 88:829–835, 1979. 11. Watson, P.G., and Grierson, I. The place of trabeculectomy in the treatment of glaucoma. Ophthalmology. 88:175–196, 1981. 12. Serguhn, S., and Spiegel, D. Comparison of postoperative recovery after trabeculectomy for pseudoexfoliation glaucoma and chronic primary open angle glaucoma. Klin. Monbl. Augenheilkd. 215:281–286, 1999. 13. Landers, J., Martin, K., Sarkies, N., Bourne, R., and Watson, P. A twenty-year follow-up study of trabeculectomy: risk factors and outcomes. Ophthalmology. 119:694–702, 2012. 14. Konstas, A.G.P., Jay, J.L., Marshall, G.E., et al. Prevalence, diagnostic features, and response to trabeculectomy in exfoliation glaucoma. Ophthalmology. 100:619–627, 1993. 15. Popovic, V., and Sjostrand, J. Course of exfoliation and simplex glaucoma after primary trabeculectomy. Br. J. Ophthalmol. 83:305–310, 1999. 16. Tanihara, H., Negi, A., Akimoto, M., et al. Surgical effect of trabeculotomy ab externo on adult eyes with primary open angle glaucoma and pseudoexfoliation syndrome. Arch. Ophthalmol. 111:1653–1661, 1993. 17. Colin, J., Bonissent, J.F., and Resnikoff, S. Epidemiology of the exfoliation syndrome. 17th Congr Europ Soc Ophthalmol. Helsinki; 1985; p. 230–231.
¨ ZTU ¨ RKER ET AL. O
18. Colin, J., Le Gall, G., Le Jeune, B., et al. The prevalence of exfoliation syndrome in different areas of France. Acta. Ophthalmol. 66(Suppl 184):86–89, 1988. 19. Nguyen, N.X., Ku¨chle, M., Martus, P., and Naumann, G.O. Quantification of blood—aqueous barrier breakdown after trabeculectomy: pseudoexfoliation versus primary openangle glaucoma. J. Glaucoma. 8:18–23, 1999. 20. Ku¨chle, M., Nguyen, N.X., Hannappel, E., and Naumann, G.O. The blood-aqueous barrier in eyes with pseudoexfoliation syndrome. Ophthalmic. Res. 27(Suppl 1):136–142, 1995. 21. Ku¨chle, M., Vinores, S.A., Mahlow, J., and Green, W.R. Blood-aqueous barrier in pseudoexfoliation syndrome: evaluation by immunohistochemical staining of endogenous albumin. Graefes Arch. Clin. Exp. Ophthalmol. 234: 12–18, 1996.
Received: February 19, 2014 Accepted: April 23, 2014 Address correspondence to: Dr. Zeynep Kayaarasi O¨ztu¨rker Istanbul Education and Research Hospital Istanbul 34488 Turkey E-mail: [email protected]
Does the Use of Preoperative Antiglaucoma Medications Influence Trabeculectomy Success? ¨ ztu¨rker,1 Can O ¨ ztu¨rker,2 Sukru Bayraktar,3 Cigdem Altan,2 and Omer Faruk Yilmaz 2 Zeynep Kayaarasi O
Abstract
Purpose: To investigate the influence of preoperative antiglaucoma medications on trabeculectomy outcome. Methods: Two hundred fifteen eyes, which underwent primary trabeculectomy, were retrospectively analyzed. The average follow-up was 39.8 – 30.3 months. The only cases of primary open-angle glaucoma, with or without pseudoexfoliation (PXF), were included. ‘‘Complete success’’ was defined as intraocular pressure (IOP) < 18 mmHg without glaucoma medications, whereas relative success was defined as the same IOP target with medications. The influence of the preoperatively used glaucoma medications on surgical success was analyzed by univariate Pearson correlation and multivariate (ordinal) regression analysis. Results: There were 118 male (54.9%) and 97 female (45.1%) patients with a mean age of 66.9 – 9.3 years. PXF glaucoma (PXFG) was present in 93 eyes (43.3%). In 33 patients (15.3%), diabetes mellitus (DM) was present. Complete success was achieved in 116 eyes (54%), relative success in 81 eyes (37.6%), and failure in 18 eyes (8.4%). Neither the total number nor the duration of glaucoma medications used before trabeculectomy was found to have any statistically significant influence on surgical success. In statistical analysis, a combination of topical beta-blocker and carbonic anhydrase inhibitor (BB + CAI) used before surgery was found to be associated with statistically better outcome, whereas the preoperative use of topical beta-blockers alone could have a negative influence on success. PXF was shown to be independently associated with trabeculectomy outcome on multivariate regression analysis. Conclusion: The glaucoma medications used preoperatively were not found to have any statistically significant negative influence on the trabeculectomy outcome and use of the combined BB + CAI preparation could have a positive influence, whereas the use of topical beta-blockers alone could have a negative influence on success, although not statistically significant. The presence of PXF was independently associated with a better surgical outcome.
Introduction
D
espite the recent trend toward minimally invasive glaucoma surgery, trabeculectomy is still the most frequently used procedure for the treatment of glaucoma. According to statistics, the number of surgical treatments has been increasing as the last glaucoma drug entered the market more than 10 years ago, and no new candidate molecule is present on the horizon.1,2 The glaucoma specialist usually prefers trying all the available medical options before switching to surgery as the pressure-lowering effect of trabeculectomy usually decreases with time. This brings to the question of whether the prior use of glaucoma medications would have any adverse influence on the outcome of filtering surgery.
1 2 3
The primary aim of this study was to evaluate the influence of preoperative antiglaucoma medications on trabeculectomy success. In addition, the effect of some of the patient-related ocular and systemic factors such as age, PXF, and diabetes mellitus (DM) was also investigated by using univariate and multivariate analyses.
Methods The medical records of 391 patients, who underwent primary trabeculectomy without the use of intraoperative antimetabolites in the Glaucoma Service of Beyoglu Eye Training and Research Hospital, were retrospectively analyzed. The study group composed of those patients with a regular follow-up of more than 12 months. Only eyes with
Istanbul Education and Research Hospital, Istanbul, Turkey. Beyoglu Eye Training and Research Hospital, Istanbul, Turkey. Istanbul Surgery Hospital, Istanbul, Turkey.
554
GLAUCOMA MEDICATIONS’ EFFECT ON TRABECULECTOMY SUCCESS
Table 1.
Table 3. Postoperative Change in Intraocular Pressure
Patient Demographics
Gender (male/female) Mean age (years) Pseudoexfoliation (PXF) DM
555
118/97 66.9 – 9.3 93/215 (43.2%) 33/215 (15.3%)
DM, diabetes mellitus.
primary glaucoma with or without PXF were included. Those eyes with secondary glaucoma and any prior intraocular surgery such as trabeculectomy and/or cataract extraction were also excluded. According to those inclusion criteria, our study group composed of 215 eyes of 215 patients. Ethical approval and informed consent were obtained according to the Helsinki Declaration.
Trabeculectomy technique In all the study eyes, an identical surgical technique was used. After sub-Tenon’s anesthesia, a limbus-based conjunctival incision approximately 10 mm behind the limbus was made, and then a 3 · 4 mm rectangular scleral flap in semiscleral thickness was prepared. After corneal paracentesis, a deep scleral block excision measuring *1 mm2 was made, and peripheral iridectomy was performed. The scleral flap was closed with 2 or 3 10/0 nylon sutures. The conjunctiva was closed with absorbable 8/0 silk sutures in a continuous fashion. Topical antibiotic and corticosteroids were used in all eyes for at least 6 weeks postoperatively, whereas cycloplegics were given in eyes with significant inflammation and/or anterior chamber shallowing in the postoperative period. Digital massage and/or laser suture lysis were performed, if necessary, during the first postoperative week. None of the known antimetabolites such as mitomycine-C or 5-fluorouracil was used either during or after the operation.
Study population Patient age, gender, presence of PXF, and systemic disease such as DM were recorded (Table 1). The total number of glaucoma medications, duration of medication use, name of glaucoma medications, and the preoperative IOP with/ without medications were also recorded (Table 2). In 111 of the 215 patients, according to extensive cupping, visual field
IOP Preoperative 1st year 2nd year 3rd year 4th year 5th year 6th year 7th year 8th year 9th year
N
Minimum
Maximum
Mean
SD
215 215 136 90 65 50 41 33 23 6
9.00 1.00 4.00 1.00 2.00 3.00 2.00 8.00 7.00 8.00
55.00 26.00 24.00 23.00 36.00 28.00 23.00 20.00 19.00 21.00
20.79 12.37 13.34 13.52 14.11 14.16 12.32 13.48 13.09 15.00
7.66 4.23 3.69 4.63 5.43 5.10 3.60 2.68 3.07 4.82
IOP, intraocular pressure.
defects, and acute exacerbation of IOP elevation, trabeculectomy was performed as the first choice of treatment.
Success criteria ‘‘Complete success’’ was defined as postoperative IOP measuring less than 18 mmHg in every control visit without glaucoma medications, whereas ‘‘relative success’’ was defined as IOP below 18 mmHg, but glaucoma medications were needed to achieve that target. Surgery was considered to be a failure for eyes, in which IOP could not be kept below 18 mmHg, despite using glaucoma medications.
Statistical analysis The SPSS 11.5 for Windows was used for statistical analysis. The influence of patient-related factors and preoperative glaucoma medications was analyzed by using univariate and multivariate methods. Pearson correlation analysis was used for univariate analysis, whereas ordinal regression was used for multivariate analysis as success was defined on an ordinal scale. The Kaplan–Meier survival analysis and log-rank test were also performed for those factors found to be statistically significant in multivariate analysis. Statistical significance was confirmed as the P-value less than 0.05.
Results The average follow-up was 39.8 – 30.3 months. The intraocular pressure (IOP) was statistically reduced when
Table 2.
Preoperative Glaucoma Medications
Average number of medications Average duration of medication use (months) Number of medications
Sympathomimetics Beta-blockers (BB) Carbonic Anhydrase (CA) inhibitor Combined BB + CA inhibitor Prostaglandins Miotics
1.06 – 1.33
Table 4. Postoperative Change in Glaucoma Medications
10.1 – 20.7 None/111 One/31 Two/37 Three or more/36 18 76 18
(51.6%) (14.4%) (17.2%) (16.7%) (8.4%) (35.3%) (8.4%)
33 (15.3%) 38 (17.7%) 27 (12.6%)
Preoperative 1st year 2nd year 3rd year 4th year 5th year 6th year 7th year 8th year 9th year
N
Minimum
Maximum
Mean
SD
215 215 136 90 65 50 41 33 23 6
0 0 0 0 0 0 0 0 0 0
4 4 3 4 3 4 3 3 3 3
1.06 0.37 0.71 1.06 1.25 1.60 1.76 1.67 1.70 1.67
1.33 0.74 0.81 0.94 1.00 1.09 1.16 1.22 1.06 1.21
¨ ZTU ¨ RKER ET AL. O
556
Table 5.
Factors not Statistically Associated with Surgical Success Pearson correlation coefficient P-value
Number of medications Average duration of medication use At least 6 months of medication use Sympathomimetics Carbonic anhydrase (CA) inhibitor Prostaglandins Miotics Presence of DM
0.057 - 0.016 0.022 0.021 - 0.031 0.089 - 0.116 - 0.121
0.402 0.831 0.853 0.763 0.647 0.196 0.691 0.079
compared with preoperative levels during the postoperative visits (P < 0.05, Table 3). Postoperative changes in the number of glaucoma medications are shown in Table 4; there was a statistically significant reduction only during the first year postoperatively (P < 0.05). At the last visit, complete success was present in 116 (54%) eyes, relative success in 81 (37.6%) eyes, and failure in 18 (8.4%) eyes. The factors not associated with surgical success are shown in Table 5, whereas those significantly associated with success are shown in Table 6. In the univariate Pearson correlation analysis, age (P = 0.02), presence of PXF (P = 0.0001), and preoperative use of combined beta-blocker and carbonic anhydrase preparation, (P = 0.001) were found to be associated with better surgical outcome, whereas the preoperative use of topical beta-blockers alone had a negative influence on success (P = 0.01). In multivariate regression analysis, the presence of PXF (P = 0.001) and the preoperative use of combined beta-blocker and carbonic anhydrase preparation (P = 0.001) were found to be the statistically significant factors associated with better trabeculectomy outcome, whereas the age (P = 0.12) and use of preoperative beta-blockers (P = 0.205) did not influence surgical success. The Kaplan–Meier survival analysis was performed using PXF and the preoperative use of combined BB + CAI as factors. The survival curves for complete success are shown in Figs. 1 and 2. In the log-rank test, complete success (event was defined as ‘‘the initiation of glaucoma medications to keep IOP below 18 mmHg’’) was found to be statistically
FIG. 1. Kaplan–Meier survival analysis of complete success in eyes with and without pseudoexfoliation (PXF). better in eyes with PXF as compared with no PXF (P = 0.01). However, no statistically significant difference was found between those eyes, in which a combined BB + CAI preparation was used preoperatively as compared with those, in which no similar treatment was given (P = 0.161).
Discussion The current practice of glaucoma treatment almost always involves using all the medication and laser options before
Table 6. Statistically Significant Factors Related to Surgical Success Univariate analysis Pearson correlation
Age PXF Preoperative BB Preoperative combined BB + CAI
Multivariate analysis ordinal regression
Coefficient
P
P
0.215 0.256 - 0.176 0.219
0.02a 0.0001a 0.01a 0.001a
0.12 0.001a 0.205 0.001a
a p-Value < 0.05. BB + CAI, beta-blocker and carbonic anhydrase inhibitor; PXF, pseudoexfoliation.
FIG. 2. Kaplan–Meier survival analysis of complete success in eyes with and without preoperative use of a combined beta-blocker and carbonic anhydrase inhibitor.
GLAUCOMA MEDICATIONS’ EFFECT ON TRABECULECTOMY SUCCESS
proceeding with surgery. The rationale behind this approach is the slow but apparent rate of decrease in trabeculectomy success with time.3,4 Keeping trabeculectomy as the reserve treatment option, on the other hand, requires the use of glaucoma medications for long time periods and brings to the question of whether those medications have any adverse influence on the surgical outcome. In this retrospective study, we tried to analyze whether the preoperative use of glaucoma medications has any influence on trabeculectomy success. Our study group composed of patients who underwent primary trabeculectomy without the use of intraoperative antimetabolites. Today, we know that antimetabolites are used more often in trabeculectomy, so a group with this specificity may be included in future studies. A number of studies showed that long-term preoperative use of glaucoma drugs could change the conjunctival cell profile, induce inflammation, and then reduce the chance of further trabeculectomy success.5–9 Results of clinical studies, on the other hand, were controversial and have failed to demonstrate clear evidence supporting the histologic data.10,11 In approximately half of our study population, the total duration of glaucoma medication usage was less than 6 months—we called it as the no-medication group. According to extensive cupping, visual field defects, and acute exacerbation of IOP elevation, trabeculectomy was performed as the first choice of treatment in this group. Five different medication groups were used in the remaining eyes: miotics, sympathomimetics, beta-blockers, carbonic anhydrase inhibitors, and prostaglandins. The influence of the combined preparation of BB and CAI was analyzed separately as it was used in a relatively large proportion of eyes (more than 15% of eyes in the study). We could not find any statistically significant association with the total number of preoperative glaucoma medications and/or the average duration of medication use with the trabeculectomy success in the univariate correlation analysis. When we analyzed the medication groups individually, we found that the preoperative use of a combined BB + CAI drug was associated with a better outcome in both the univariate and multivariate analysis. There was a significant negative association between BB use and surgical success in the univariate correlation, but not in the multivariate regression. The influence of preoperative combined BB + CAI use on trabeculectomy outcome was also analyzed by Kaplan– Meier survival curves, but we failed to find an association in the log-rank test. We believe that a weak association could be present, but due to the relatively shorter follow-up of those eyes (17.5 – 7.9 months) receiving this preparation before surgery versus the rest of the eyes in the study (43.8 – 31.1 months), the Kaplan–Meier analysis might have failed to demonstrate it. The presence of DM was not found to be related to surgical success. Both patient age and the presence of PXF were found to be positively correlated with the success in the univariate correlation analysis, but only PXF was shown to be independently associated with trabeculectomy outcome on multivariate regression analysis. The significant influence of PXF on trabeculectomy outcome was also shown by using Kaplan–Meier survival analysis. There are controversial studies on PXFG and trabeculectomy in the literature.12–16 Our study clearly showed that the presence of PXF increased the trabeculectomy survival independently from other factors. In our study population,
557
the mean preoperative IOP was not too high (less than 21 mmHg with medications), but approximately half of them (43.2%) had PXF. Similar to the other Mediterranean countries, we usually see a lot of cases with PXFG and they usually require surgery.17,18 How could we explain the positive influence of PXF on trabeculectomy success? The disruption of blood–aqueous barrier has been shown to be somewhat greater and the postoperative inflammation was more severe in eyes with PXFG.19–21 Those events often led to decreased aqueous humor production rates. The chi-square test showed that the incidence of postoperative hypotony was statistically higher in those eyes with PXF (P = 0.0001). Although we did not measure the inflow or investigate bleb morphology in our study, we think that the decreased aqueous production might contribute to better surgical outcomes by forming more diffuse and favorable blebs without encapsulation. In summary, we found that preoperative use of glaucoma medications did not have any adverse influence on trabeculectomy success. In statistical analysis, a combination of topical beta-blocker and carbonic anhydrase inhibitor (BB + CAI) used before surgery was found to be associated with statistically better outcome, whereas the preoperative use of topical beta-blockers alone could have a negative influence on success. However, we advise taking this cautiously as the follow-up was relatively shorter in those groups of eyes. We also believe that the association between the presence of PXF and improved trabeculectomy success is apparent.
Author Disclosure Statement None of the authors has a financial or proprietary interest in any material or method mentioned.
References 1. Rotchford, A. What is practical in Glaucoma management? Eye 19:1125–1132, 2005. 2. Thomas, R., Paul, P., and Rao, G.N. Present status of Eye care in India. Survey Ophthalmol. 50:85–101, 2005. 3. Jerndal, T., and Lundstrom, M. 330 Trabeculectomies: a follow study through ½–3 years. Acta Ophthalmol. 55:52– 62, 1977. 4. Nouri-Mahadevi, K., Brigatti, L., Weitzman, M., and Caprioli, J. Outcomes of trabeculectomy for primary open angle glaucoma. Ophthalmology. 102:1760–1769, 1995. 5. Noecker, R.J., Herrygers, L.A., and Anwaruddin, R. Corneal and conjunctival changes caused by commonly used glaucoma medications. Cornea. 23:490–496, 2004. 6. Nenciu, A., Stefan, C., and Ardelean, C. Structural and immunohistochemical changes of conjunctiva induced by topical glaucoma medication. Oftalmologia. 48:35–42, 2004. 7. Turac¸li, E., Budak, K., Kaur, A., Mizrak, B., and Ekinci, C. The effects of long-term topical glaucoma medication on conjunctival impression cytology. Int. Ophthalmol. 21:27– 33, 1997. 8. Young, T.L., Higginbotham, E.J., Zou, X.L., and Farber, M.D. Effects of topical glaucoma drugs on fistulized rabbit conjunctiva. Ophthalmology. 97:1423–1427, 1990. 9. Lavin, M.J., Wormald, R.P.L., Migdal, C.S., and Hitchings, R.A. The influence of prior therapy on the success of trabeculectomy. Arch. Ophthalmol. 108:1543–1548, 1990.
558
10. D’Ermo, F., Bonomi, L., and Doro, D. A critical analysis of the long term results of trabeculectomy. Am. J. Ophthalmol. 88:829–835, 1979. 11. Watson, P.G., and Grierson, I. The place of trabeculectomy in the treatment of glaucoma. Ophthalmology. 88:175–196, 1981. 12. Serguhn, S., and Spiegel, D. Comparison of postoperative recovery after trabeculectomy for pseudoexfoliation glaucoma and chronic primary open angle glaucoma. Klin. Monbl. Augenheilkd. 215:281–286, 1999. 13. Landers, J., Martin, K., Sarkies, N., Bourne, R., and Watson, P. A twenty-year follow-up study of trabeculectomy: risk factors and outcomes. Ophthalmology. 119:694–702, 2012. 14. Konstas, A.G.P., Jay, J.L., Marshall, G.E., et al. Prevalence, diagnostic features, and response to trabeculectomy in exfoliation glaucoma. Ophthalmology. 100:619–627, 1993. 15. Popovic, V., and Sjostrand, J. Course of exfoliation and simplex glaucoma after primary trabeculectomy. Br. J. Ophthalmol. 83:305–310, 1999. 16. Tanihara, H., Negi, A., Akimoto, M., et al. Surgical effect of trabeculotomy ab externo on adult eyes with primary open angle glaucoma and pseudoexfoliation syndrome. Arch. Ophthalmol. 111:1653–1661, 1993. 17. Colin, J., Bonissent, J.F., and Resnikoff, S. Epidemiology of the exfoliation syndrome. 17th Congr Europ Soc Ophthalmol. Helsinki; 1985; p. 230–231.
¨ ZTU ¨ RKER ET AL. O
18. Colin, J., Le Gall, G., Le Jeune, B., et al. The prevalence of exfoliation syndrome in different areas of France. Acta. Ophthalmol. 66(Suppl 184):86–89, 1988. 19. Nguyen, N.X., Ku¨chle, M., Martus, P., and Naumann, G.O. Quantification of blood—aqueous barrier breakdown after trabeculectomy: pseudoexfoliation versus primary openangle glaucoma. J. Glaucoma. 8:18–23, 1999. 20. Ku¨chle, M., Nguyen, N.X., Hannappel, E., and Naumann, G.O. The blood-aqueous barrier in eyes with pseudoexfoliation syndrome. Ophthalmic. Res. 27(Suppl 1):136–142, 1995. 21. Ku¨chle, M., Vinores, S.A., Mahlow, J., and Green, W.R. Blood-aqueous barrier in pseudoexfoliation syndrome: evaluation by immunohistochemical staining of endogenous albumin. Graefes Arch. Clin. Exp. Ophthalmol. 234: 12–18, 1996.
Received: February 19, 2014 Accepted: April 23, 2014 Address correspondence to: Dr. Zeynep Kayaarasi O¨ztu¨rker Istanbul Education and Research Hospital Istanbul 34488 Turkey E-mail: [email protected]
