Trabeculectomy Success Rate Among Glaucoma ...

2010 International Conference on Science and Social Research (CSSR 2010), December 5 - 7, 2010, Kuala Lumpur, Malaysia

Trabeculectomy: the success rate and its factors among glaucoma patients in Kelantan, Malaysia Khairil Anuar Md. Isa

Liza Sharmini Ahmad Tajudin

Department of Basic Sciences Universiti Teknologi Mara (UiTM) Kampus Puncak Alam, Puncak Alam, Malaysia [email protected]

Department of Ophthalmology Universiti Sains Malaysia (USM) Kampus Kesihatan, Kubang Kerian, Malaysia [email protected] intervention. Trabeculectomy, which is performed by creating a fistula to improve the aqueous outflow and reduced the IOP was introduced by Sugar [5] and improved by Cairns [6], is the preferred procedure for the surgical management for various types of glaucoma. After more than 30 years since the introduction of trabeculectomy, it still remains the most commonly used effective incisional surgery for glaucoma [7]. Based on various studies that compared the initial trabeculectomy with medical treatment found that trabeculectomy provides consistently lower IOP than medical therapy [8].

Abstract— Trabeculectomy success rate among Asians is not too promising. Thus, a survival analysis among our local population is very important to help future strategy of glaucoma patient management. Objective: To determine the five year trabeculectomy success rate among glaucoma patient treated at Hospital Universiti Sains Malaysia. Methodology: A retrospective record review study was conducted involving 96 glaucoma patients (96 eyes) who underwent trabeculectomy from 1st January 1990 to 31st July 2006. Additional follow up of one year after the recruitment of the study subjects was done from 1st August 2006 to 31st July 2007. All patients who fulfilled the selections criteria were included in the study. The medical records were reviewed by a single researcher with the help of senior consultant of ophthalmology and important information on the variables of interest and trabeculectomy outcome status were collected and recorded into a clinic form. The Kaplan Meire analysis was used in statistical analysis. Results: The median trabeculectomy success time was 18.8 months and the overall trabeculectomy success rate was 36.1% (95% CI:25.6; 46.8) at five year. The patients without ocular surgery [42.2% (95% CI:30.3; 53.5)] demonstrated higher trabeculectomy success rate compared to those with previous ocular surgery [0% (95% CI: 0; 0)] prior to trabeculectomy at five year post trabeculectomy and the difference was statistically significant (p=0.008).Conclusion: Trabeculectomy success rate for glaucoma patients at HUSM was acceptably low. The used of antimetabolites during trabeculectomy is highly recommended in our population especially those with previous ocular surgery.

The long-term success in term of IOP reduction post trabeculectomy ranged from 40% to 98%, depending on follow-up time [9], type of glaucoma [10], variables of interest [11], type of population [12] and the criteria used to define successful outcome [13]. Races seem to play an important factor. Blacks were believed to have poorer success rate [14] compared to Caucasians [15] and Asians [16]. The aim of this study was to evaluate the trabeculectomy success rate in terms of IOP among Kelantanese. II.

A. Patients and methods A retrospective records review study was carried out. The glaucoma patients who had undergone trabeculectomy in Hospital Universiti Sains Malaysia (HUSM) and those whose medical records available within the time frame between the 1 January 1990 and 31 July 2006 were included in this study. The selected patients were also required to complete follow up period of one year after the recruitment. Based on sample size calculation using Power and Sample Size software [17], 140 subjects were required [95% confidence interval, power of 80%, detectable hazard of 1.7, ratio of patients of 0.75, and median success time of 6 months for the variable type of glaucoma [18] (Angle closure glaucoma and Open angle glaucoma). Considering 20 % drop out (incomplete one year follow up), a total of 170 subjects were required. Systemic comorbidity such as hypertension, diabetes, hyperlipidemia, heart disease, and cerebrovascular accident were documented.

Keywords- Trabeculectomy; glaucoma; success rate

I.

INTRODUCTION

Glaucoma is defined as a progressive optic neuropathy with characteristic optic disc cupping and visual field loss [1]. Glaucoma is the second leading cause of blindness worldwide after cataract [2]. Glaucoma is also known as the thief of sight, robbing the vision silently [1]. OAG was the most common form of glaucoma and it accounted for 19% of all blindness among African-Americans compared to 6% in Caucasians [3]. Currently, IOP is the only modifiable risk factor and lowering IOP proven to be beneficial in reducing the risk of progression of visual field loss in glaucoma [4]. The reduction of IOP is achievable by medical treatment, laser or surgical

978-1-4244-8986-2/10/$26.00 ©2010 IEEE

METHODOLOGY

154

histogram showed skewed graph, their median with interquartile range were then calculated. Some of the numerical variables, such as age at diagnosis and age at operation were then recoded into categorical variables to simplify the interpretation of results. Meanwhile, for categorical variables, frequency of observations and percentages were obtained.

The diagnosis was based on diagnosis made by the treating physician at the time of diagnosis. However there was also missing data as some patients were treated outside of HUSM. In case of diabetes, we didn’t obtain any parameter or indicator to divide them into good or poor control. Similarly, systemic blood pressure was not recorded. The patients who were confirmed to have glaucoma and indicated for trabeculectomy that have completed at least one year followup post trabeculectomy were selected. In cases of bilateral trabeculectomy the eye with more severe glaucoma staging based on Hodapp’s visual field [19] was chosen. Patients’ medical records that demonstrated pattern of infrequent follow up or defaulter, ambiguous diagnosis or any abnormality preventing reliability Goldmann applanation tonometry measurement were excluded. Based on surgical record books, more than 300 cases were identified but only 96 medical records were available due to many records had been destroyed. There was no probability sampling done since available patients’ medical records obtained were less than predetermined sample size calculated. These records were reviewed by a single researcher with the help of senior consultant ophthalmologist. Sociodemographic status, type of glaucoma, preoperative characteristics and intraoperative data were explored and documented.

The types of glaucoma were regrouped into two main groups: OAG and ACG. For the purposes of statistical analysis, ethnicity was divided into Malays and Non-Malays. Similarly history of previous systemic disease was also divided into absent and present due to large inequality frequency in each cell. Median trabeculectomy success time and the five year probabilities of trabeculectomy success in term of IOP ≤ 21 mmHg for all categorical independent variables were estimated using the Kaplan Meier analysis. The trabeculectomy success probability was changed to percentages and expressed as trabeculectomy success rate in results section. The statistical difference of the survival probabilities between all categorical independent variable was evaluated using the log-rank test. However for categorical variables that had more than two levels, p-value were obtained from pair wise comparison were then compared to α level that had been corrected using Bonferroni correction.

The survival time was defined as the success of trabeculectomy from the first postoperative day until the last date of filtration failure based on several defined criteria. “IOP more than 21 mmHg with or without medication”, or “requiring additional glaucoma surgery”, “phthisical eye” and “non perception to light” considered as trabeculectomy failures or terminal events. In order to prevent underestimation of survival time, censored observations were introduced. The observation is considered censored if the “IOP is ≤ 21mmHg at the end of the study period without any medication (successful trabeculectomy at the end of study period)” or “the patient passed away during the study period due to other causes unrelated to glaucoma or its complication” or “the patients defaulted from scheduled follow up”. This study received approval from the Research and Ethics Committee of Universiti Sains Malaysia: (ref.no.USMKK/PPP/JEPeM[194.4(3.2)]).

III.

RESULTS

A. Descriptive statistics The preoperative demographic characteristics of the patients were listed in table 1. A total of 96 patients were recruited, with slightly more males than females and majority were Malays, followed by Chinese, Indians and Siamese. The mean age of patients on the day of operation was 56 years (SD 19.6) with almost a year after the diagnosis was made [55 years (SD 19.8)]. More than two-thirds were bilateral cases with 21 patients underwent bilateral trabeculectomy but only one eye was selected. Only 5.2% of total recruited patients were “naive eye” or “virgin eye” without exposure to topical antiglaucoma drugs. Majority received more than one topical drugs. In addition more than two-thirds of them were taking oral antiglaucoma medication especially Acetalzolamide. In OAG group, almost half were diagnosed to have POAG, while PACG predominate the ACG group. The PACG composed of 33 eyes or 94.3% from total ACG group.

B. Statistical analysis Statistical Program for Social Science Software (SPSS), version 12.0.1 [38], was used for data entry and Statistical Data Analysis software (STATA), version 9.0 was used for data analysis [20]. Data from SPSS was transferred to STATA using Statistical Transfer software version 6.0 [21]. All the data were double checked and cleaned to verify the entire variables were properly documented and to detect any missing or erroneous values during data entry. Exploration of data was then conducted to obtain descriptive statistics and necessary graphs for all the variables. The numerical variables were explored for normality distribution. For all numerical variables with normal distribution, their mean and standard deviation were obtained. However for numerical variables that

Majority of the eyes, were without ocular surgery whereas 13 (13.5%) of the eyes had previous ocular surgery prior to trabeculectomy. As this a retrospective study, the recruitment of subjects were done much later during the era when augmented trabeculectomy was not ‘in fashion’ as it recently. Currently, in clinical practice almost all patients were subjected to augmented trabeculectomy. Only 19 (19.8%) procedures used augmentation. Most of the filtration surgeries were performed by experienced ophthalmologists.

155

eyes without ocular surgery prior to trabeculectomy and the difference was statistically significant. Wong et al., (1998) suggested that the excessive fibrosis at the conjuctival-scleral interface results from disturbed conjunctiva from previous surgical intervention found to be related with subsequent ocular surgery failure.

B. Median trabeculectyomy success time The overall median trabeculectomy success time were 18.8 months (95% CI; 12.2, 38.7) with 60 eyes were considered as failure at the end of study period (Fig. 1).

This study also observed that those 60 years and older at trabeculectomy surgery demonstrated better five year trabeculectomy success rate compared to those younger than 60 years old. The better trabeculectomy success rate in those older patients was noted as the younger group contained fewer patients who were considered as censored at the end of the study. There was evidence that the children younger than seven were associated with poor trabeculectomy outcome [25]. However, there was no standardize age in previous studies which investigated the effect of age with trabeculectomy success. Trabeculectomy success was also noted to be significant in patients older than 60 years old, but was not statistically significant after exclusion of congenital or secondary glaucoma [26]. Gressel et al., (1984) found that, trabeculectomy was more likely to be successful in increasing age [27].

Figure 1. Kaplan Meire survival curve showing overall trabeculectomy success probability

C. Five year trabeculectomy success rate Based on the Kaplan-Meier curve, the trabeculectomy success rates was 61.2% (95% CI: 50.5, 70.3) at one year, 48.4% (95% CI: 37.7, 58.2) at two year, 41.5% (95% CI: 30.9, 51.7) at three year, 38.0% (95% CI: 27.5, 48.5) at four year, 36.1% ( 95% CI: 25.6, 46.8) at five year and 21.6% (95% CI: 10.9, 34.6) at 10 year.

Malays were found to have better trabeculectomy success rate, perhaps due to inappropriate racial distribution in Kelantan, which 95% were Malays [28]. Though the results was not as good as those report among Caucasians [15]. It is generally accepted that Blacks (the results of trabeculectomy) were not as successful compared to Whites [14]. In fact Asian was predicted to have slightly better outcome [11]. However the present study and previous published data in Malaysia [18, 24] and Singapore [16, 29] found that the plain trabeculectomy for primary glaucoma were even lower than Blacks [14], which suggest that Asians populations were more at risk in trabeculectomy failure. Since then, the augmented surgery was proposed to improve trabeculectomy success rate in Asian population [16].

The history of previous ocular surgery (p=0.008) was associated with significant difference of five-year success probability. The patients without ocular surgery (42.2% (95% CI: 30.3, 53.5)) demonstrated higher trabeculectomy success rate compared to those with previous ocular surgery (0.0% (95% CI: 0, 0)) prior to trabeculectomy at five year post trabeculectomy. The other variables not show any significant different of five year success rate (Table 1).

IV.

The glaucoma patients with systemic diseases had lower trabeculectomy success rate compared to those without any systemic disease during surgery. This result was comparable with a study by Shin et al. (1996) which reported diabetic patients was associated with a lower trabeculectomy outcome. Diabetes mellitus also has been reported to be associated with increased risk of elevated IOP and glaucomatous optic nerve damage in POAG [30], which was believed to be one of the risk factors. Glaucomatous nerve damage is future facilitated by poor optic nerve head circulation secondary to diabetic or hypertension. Diabetes mellitus is usually associated with poor wound healing, which is the most important factor to maintain the function of the bleb. Unlike other surgical procedures, which need good healing process, trabeculectomy success is determined by slower healing process. However diabetic increased the risk for wound infection which could be claustrophobic for any ocular surgery.

DISCUSSION

Trabeculectomy aims to promote a filtering conjunctival bleb where the created scleral flap must be loose enough to permit aqueous outflow but taut enough to prevent post surgical ocular hypotony, which cause shallowing or flat anterior chambers [22]. Bleb is a blister filled with aqueous humour and majority of trabeculectomy failure overtime was linked to bleb failure [22]. The overall trabeculectomy success rate in the present study was lower than the studies conducted on Caucasians [23], Indians [10] and Japanese [13] populations. However, the outcome of this study was almost similar to previous studies conducted in Malaysia [18, 24] and Singapore [16]. In Europeans, Indians and Singaporean studies, antimetabolites were used, which has increased the trabeculectomy success rate tremendously. In most of the Malaysian studies were without antimetabolites, however this study outcome was much better.

The difference of trabeculectomy outcome between two major type of glaucoma, ACG and OAG were also looked in the present study. We found that the ACG showed mostly similar five year success rate compared to OAG. Sharif and Selvarajah (1997) found that patients with ACG showed poorer

This study demonstrated that the eyes with previous ocular surgery had lower trabeculectomy outcome compared to those

156

trabeculectomy outcomes compared to OAG patients, which may be due to the advanced of the disease itself. Moreover most of them presented with acute attack, which indicate sudden intensive inflammation especially at the conjuctival level. Chronic conjunctival inflammation was found to be associated with bleb failure [3]. However this study failed to demonstrate any significant difference of trabeculectomy success between OAG and ACG.

mmHg may not prevent the glaucoma from progression. Nevertheless the criteria we used allow us to compare our outcome from study done by others. Despite these limitations, ours results confirmed the impression that our population has relatively poorer trabeculectomy success rates when compared with Caucasian population. We hope that our finding would suggest ophthalmologists or clinicians in determining what the prognostics factor based on initial clinical at presentation and predict their patient’s outcome that might be helping them to choose the best treatment for their patient management of glaucoma by trabeculectomy technique.

Long term treatment with topical antiglaucoma has been implicated to induce subclinical inflammation with increase in the number of fibroblasts, lymphocytes, macrophage and mast cells [31], which associated with bleb failure and lower trabeculectomy success rate [32]. Pilocarpine has shown to effects on lacrimal function, the conjunctiva and subsequent filtration surgery [33]. The present study also found that those who were treated with topical antiglaucoma drugs demonstrated with lower success rate of trabeculectomy compared to trabeculectomy without any initial topical antiglaucoma drugs. However a study by Liza-Shamini et al., (2007), proposed that the presence of pigment in the conjunctiva of Asian patients may have protected against any adverse effect of topical Timolol [34].

V.

CONCLUSION

The overall success rate of trabeculectomy in this group of patients is relatively low than Caucasians and is comparable with those reported in other Asian population using the same criteria for success. Augmentation trabeculectomy with antimetabolites is recommended in our population especially those with history of previous ocular surgery.

Success rate of trabeculectomy was also to be associated with severity of glaucoma. Based on Hodapp’s visual field staging, this study found that the trabeculectomy success rate was better in early glaucoma. However based on current practice, most trabeculectomy is reserved on more advanced glaucoma. Perhaps more emphasized should be given to promote trabeculectomy as the first line management, rather than as the last option. Thought, the recent report found that development of cataract post trabeculectomy as the major disadvantage. Topical antiglaucoma treatment as the favourite first choice treatment is not without this advantages [35].

ACKNOWLEDGEMENT The author would like to thank the staff at the Medical Record Unit of HUSM, the member of the Research and Ethical Committee of USM, the director of HUSM and not to forget to all the patients who participated in this study.

Glaucoma sometimes coexists with other ocular pathology, especially cataract, which necessitates a combine procedure [36]. However, this combined procedure is usually associated with poorer trabeculectomy success rate [37]. This study found the similar outcome. Additional inflammation induced extra surgical intervention may associated with postoperative hypertony and excessive fibrosis which eventually lead to trabeculectomy failure. There are several limitations that need to be considered in any interpretation of finding in this study. The study data is obtained from the patients’ medical records but it cannot achieve the required sample size that was calculated due to incompleteness and non active records. The study was retrospective in nature, with variable of follow up. There were multi surgeon involvements in the surgery for the patients leading to different technique in implementation of surgery. Another limitation was small sample size of types of glaucoma and races, since this study done in Kelantan where Malays are the major race group. Therefore, generalizability of this present finding to other races population in uncertained. Detailed and standardized gonioscopy, preoperative and postoperative data were not present. The outcomes chosen to assess the success of surgical intervention which was IOP is not ideal. It is well known that lowering IOP to less than 21

157

TABLE I. N(%)

FIVE YEAR TRABECULECTOMY SUCCESS RATES Median success time (months)

Success rates %

95% CI Lower limit

Upper limit

Log rank statistic

p-value

Age at diagnosis, Mean (SD) < 60 years 60 years and above

55(19.8) 49(51.0) 47(49.0)

14.1 26.7

32.1 41.7

19.0 25.0

45.9 57.5

0.35

0.554

Age at operation, Mean (SD) < 60 years 60 years and above

56(19.6) 47(49.0) 49(51.0)

14.1 37.0

32.9 38.5

19.4 22.4

47.0 45.3

1.10

0.295

Sex Male Female

50(52.1) 46(47.9)

14.1 25.4

33.7 39.3

19.9 23.7

48.1 54.4

0.51

0.476

Race Malay Others

80(83.3) 16(16.7)

23.3 14.0

38.0 30.

26.2 10.2

49.7 53.0

0.69

0.406

Type of Glaucoma OAG ACG

61(63.5) 35(36.5)

14.1 25.4

35.4 38.3

22.7 20.5

48.3 55.8

0.67

0.414

Previous ocular surgery No previous surgery Previous ocular surgery

83(86.5) 13(13.5)

25.4 5.3

53.5 42.2 30.3 1.91 0.008 0 0 0 [13] Shigeeda T, et al., Long term follow up of initial trabeculectomy with mitomycin C for primary open angle glaucoma in Japanese patients. Journal of Glaucoma, 2006. 15(3): p. 195-196. [14] Miller RD and Barber JC, Trabeculectomy in black patients. Ophthalmic Surg, 1981. 12(1): p. 46-50. [15] Diestelhorst M, Khalili MA, and Krieglstein GK. Trabeculectomy: a rectrospective follow-up of 700 eyes. Int Ophthalmol, 1999(22): p. 211220. [16] Wong JS, et al., Trabeculectomy survival with and without intraoperative 5-fluorouracil application in an Asian population. Australian and New Zealand Journal of Ophtalmology, 1998. 26: p. 283-288. [17] Dupont W, and Plummer W, PS power and sample size program available for free on the Internet. . Controlled Clin Trials, 1997. 18: p. 274. [18] Sharif FM, and Selvarajah S. The outcome of trabeculectomy for primary glaucoma in adult patients in UKM. Med J Malaysia, 1997. 52(1): p. 17-25. [19] Hodapp E, Parrish RK, and Anderson DR, Clinical Decisions in Glaucoma. 1993, St Louis, Missouri: CV Mosby Co. 52-61. [20] StataCorp., Stata Statistical Software. 2005: College Station, TX:StataCorp LP. [21] STAT/TRANS, Statistical Transfer software. 2000, Statistical Transfer software. [22] Singh J, et al., Enhancement of post trabeculectomy bleb formation by laser suture lysis. Br. J. Ophthalmol, 1996. 80: p. 624-627. [23] Erhnooth P, et al. Long term outcome of trabeculectomy in term of intra ocular pressure. Acta Ophtalmol. Scand., 2002. 80: p. 267-271. [24] Hooi S. and Hooi S, Trabeculectomy outcomes in a Malaysian general hospital. Med J Malaysia, 2003. 58(4): p. 565-578. [25] Miller RD and Rice N, Trabeculectomy combined with beta irradiation for congenital glaucoma. Br J Ophtalmol, 1991. 1991(75): p. 584-590. [26] Inaba Z, Long term results of trabeculectomy in Japanese:an analysis by life table method. Jpn J Ophthalmol, 1982. 26: p. 361-373. [27] Gressel MG, Heuer DK, and Parrish RK. Trabeculectomy in young patients. Ophtalmology, 1984. 91: p. 1242-1246. [28] Perangkaan, Laporan taburan penduduk dan ciri-ciri asas demografi banci penduduk dan perumahan 2000, in Laporan taburan penduduk dan

REFERENCES [1]

Ritch R, Sheilds S, and Krupin T. The Glaucoma:Basic Science 2nd Edi. 1996. [2] Quigley HA. and Broman AT, The number of people with glaucoma worldwide in 2010 and 2020. Br J Ophthalmol. 2006, 90: p. 262–267. [3] Shields MB, Text Book of Glaucoma. 4 ed, D.B. Cooke. 1997: William & Wilkins. [4] Heijl A et al., Reduction of intraocular pressure and glaucoma progression: results from the Early Manifest Glaucoma Trial. Archives of Ophtalmology, 2002. 120(10): p. 1268-1279. [5] Sugar AS, Experimental trabeculectomy in glaucoma. Am J Ophthalmol, 1961. 51: p. 623–627. [6] Cairns JE, Trabeculectomy: Preliminary report of a new method. Am J Ophthalmol , 1969. 66: p. 673–679. [7] Lichter PR, Musch DC, and Gillespie BW, Interim clinical outcomes in the Collaborative Initial Glaucoma Treatment Study comparing initial treatment randomized to medications or surgery. Ophthalmology , 2001. 108: p. 1943–1953. [8] Burr JA, Azuara-Blanco, and Avenell A. Medical versus surgical interventions for open angle glaucoma. Cochrane Database of Systematic Review, 2004(2). [9] Molteno ACB, Bosma NJ, and Kittelson JM. Otago Glaucoma Surgery Outcome Study: Long Term Results of Trabeculectomy-1976 to 1995. Ophtalmology, 1999. 106: p. 1742-1750. [10] Sihota R, Gupta V, and Agarwal HC, Long-term evaluation of trabeculectomy in primary open angle and chronic primary angle closure in an Asian population. Clinical & Experimental Ophtalmology, 2004. 32: p. 23-28. [11] Shin DH, et al. Primary glaucoma triple procedure with or without adjunctive mitomycin: Prognostic factors for filtration failure. Ophtalmology, 1996. 103: p. 1925-1933. [12] Lachkar Y, et al., Trabeculectomy with intraoperative sponge 5fluorouracil in Afro Caribbeans. Br J Ophtalmol, 1997. 81: p. 555-558.

158

[29] [30] [31]

[32]

[33]

ciri-ciri asas demografi banci penduduk dan perumahan 2000. 2001, Jabatan Perangkaan Malaysia. Tan C, et al., Trabeculectomy-success rates in Singapore hospital. Singapore Med J, 1996. 37(5): p. 505-507. Katz J, and Sommer A. Risk factors for primary open angle glaucoma. Am J Prev Med, 1998. 4: p. 110-114. Sherwood MB, Grierson I, and Millar L. Long term morphologic effect of antiglaucoma drugs on the conjuctiva and Tenon capsule in glaucoma patients. Ophtalmology, 1989. 4: p. 51-57. Broadway DC, et al. Adverse effects of topical antiglaucoma medication II. The outcome of filtration surgery. Arch Ophthalmol, 1994. 112(11): p. 1446-1454. Broadway DC, Grierson I, and Hitchings R. Adverse effects on the conjuctiva and implications for filtration surgery. Journal of Glaucoma, 1995. 4: p. 136-148.

[34] Liza-Shamini AT, Mutalib OA, and Manoharan M. The effect of topical antiglaucoma drugs on the conjuctival cell profile of Asian patients. Asian J Ophtalmol, 2007. 9: p. 17-22. [35] AGIS, The relationship between control of intraocular pressure and visual field deterioration. Am J Ophthalmol, 2000. 130: p. 429-440. [36] Friedman DS, et al., Surgical strategies for coexisting glaucoma and cataract: An evident update. Ophtalmology, 2002. 109: p. 1902-1915. [37] Naveh N, Kottass R, and Glovinsky J. The long term effect on intraocular pressure of a procedure combining trabeculectomy and cataract surgery as compared with trabeculectomy alone. Ophtalmic Surg Laser, 1990. 21: p. 339-345. [38] SPSS, Statistical program for social science and software. 2003,Apache software foundation

159