不同切口超声乳化白内障吸除术联合小梁切除术的疗效和耐受性的荟萃分析

　　作者：刘鹤南，李 迅，聂庆珠，陈晓隆 作者单位：110004 中国辽宁省沈阳市，中国医科大学附属盛京医院眼科

　　【摘要】 目的：评价并比较一切口和二切口超声乳化白内障吸除术联合小梁切除术治疗白内障合并青光眼的疗效和耐受性。方法：按照Cochrane协作网方法全面检索符合纳入标准的比较一切口和二切口超声乳化白内障吸除术联合小梁切除术的临床对照研究，将其进行荟萃分析。临床疗效的评估包括：眼压下降百分比采用标准化均差(SMD)，术后最佳矫正视力≥0.5的患者百分比采用比值比(OR)，手术成功率采用相对危险度(RR)。临床耐受性的评估采用RR。所有结果均以95%可信区间表示。数据分析采用Stata 10.1。结果：降低眼压的临床疗效二切口术式明显优于一切口术式，差异具有统计学意义(SMD，0.19;95% CI，0.33到 0.04;P=0.01);术后最佳矫正视力≥ 0.5的患者百分比二切口术式大于一切口术式，但差异不具有统计学意义(OR，0.65;95% CI，0.30到1.39;P=0.26);术后不加用抗青光眼药物达到靶眼压的患者百分比二切口术式大于一切口术式，但差异不具有统计学意义(RR, 0.94; 95% CI, 0.84到1.04; P=0.22);两种术式在术后并发症方面差别无统计学意义。结论：二切口超声乳化白内障吸除术联合小梁切除术临床疗效优于一切口术式。两种术式的术后并发症没有明显差异。

　　【关键词】 超声乳化白内障吸除术联合小梁切除术;一切口;二切口;荟萃分析

　　INTRODUCTION

　　With the increasing elderly population and concurrent longevity in life expectancy, there is an increase in the incidence of coexisting visually significant cataract and glaucoma. One of the challenges in the management of surgical procedure is difficulty in solving these two problems simultaneously. Phacoemulsification alone may be beneficial in some cases, which results in better intraocular pressure (IOP) than planned extracapsular cataract extraction procedures, and performing the glaucoma filtering surgery first and the cataract surgery later may best serve others[1]. However, there is a widespread shift towards the use of combined phacotrabeculectomy as the surgical treatment of choice for coexisting cataract and glaucoma in recent years[24].

　　Phacotrabeculectomy can be performed either using onesite or twosite incisions[5]. The earliest clinical studies of phacotrabeculectomy which is known as a onesite procedure reported surgical results using the same the scleral tunnel incision for both the phacoemulsification and trabeculectomy parts of the surgery. The introduction of the temporal incision for phacoemulsification has allowed surgeons to perform twosite procedure, with a prelimbal filtering incision for the trabeculectomy and a separate clear cornea incision for phacoemulsification[5]. Comparing of the two surgical procedures, previous studies generally had small sample sizes and showed conflicting results, which greatly hindered researchers drawing correct conclusions.

　　A metaanalysis of controlled clinical trials (prospective or retrospective) was conducted to assess the efficacy and tolerability of two surgical procedures for the management of coexisting cataract and glaucoma: onesite and twosite phacotrabeculectomy. This metaanalysis was designed to help resolve ambiguity regarding optimal management of coexisting cataract and glaucoma by pooling the outcome of available studies. Our analysis controlled for differences in study sizes and patient characteristics. However, we recognize the limitations introduced by differences in study protocols, publication bias, and the quality of studies.

　　MATERIALS AND METHODS

　　Search Strategy A computerized literature search was conducted in the PubMed, EMBASE, Scientific Citation Index and Cochrane Controlled Trials Register for relevant articles published up to May 2009. And extensive search for meeting archives, including the annual meeting abstracts of American Association of Ophthalmology (AAO) and Association for Research in Vision and Ophthalmology (ARVO) was also carried out up to May 2009. These databases were searched systematically using the following key words: phacotrabeculectomy, phacoemulsification and trabeculectomy, combined phacoemulsification and trabeculectomy, combined phaco/trabeculectomy, combined cataract and glaucoma surgery, combined cataractglaucoma surgery,onesite phacotrabeculectomy, twosite phacotrabeculectomy. The search strategy used both keywords and Medical Subject Headings (MeSH) terms. There were no limits placed on the language of publication. All potentially relevant nonEnglish publications were to be translated into English for further assessment. Literature reference proceedings were searched manually at the same time. The title and abstract of all potentially relevant articles were screened to determine their relevance. Then, full articles were scrutinized if the title and abstract were ambiguous. References identified from bibliographies of pertinent articles or books also were retrieved. References of included publications were reviewed until no further relevant studies were found.

　　Inclusion and exclusion criteria Only controlled clinical trials directly comparing between onesite and twosite phacotrabeculectomy in patients with coexisting cataract and glaucoma were included, antimetabolites could be used intraoperatively. Studies needed to have measured efficacy, tolerability or both in humans. Outcome variables included at least one of the following primary outcome variables: intraocular pressure reduction (IOPR), the percentage having a bestcorrected visual acuity (BCVA) of 0.5 or better after surgery, complete success rates and adverse events, or relevant data. Abstracts from conferences and full texts without raw data available for retrieval, duplicate publications, letter and review were excluded.

　　Studies selection The assessment of the titles and abstracts for eligibility was conducted by two independent reviewers (Liu HN and Nie QZ). Articles of potential interest were retrieved and their inclusion was reassessed. Disagreement at each step was resolved with discussion between the two reviewers. We obtained the full article of any study that seemed to fit the inclusion criteria.

　　Data extraction Two reviewers (Liu HN and Nie QZ) performed the data extraction that were included independently. Any differences were resolved by discussion to reach consensus among the investigators. A customized form was used to record authors of study, publication year, location, design, followup time, sample size, patient characteristics, interventions, baseline and endpoint values, and adverse events.

　　Outcome measures For efficacy, we used the percentage intraocular pressure reduction (IOPR%) in preoperative to postoperative IOP. Secondary efficacy measure was the percentage having a postoperative BCVA of 0.5 or better and complete success rate, which was defined as the proportion of patients achieved the target IOP without antiglaucoma medication at the end point. We assessed tolerability to phacotrabeculectomy by considering the proportions of patients with adverse events, including hyphema, choroidal detachment, bleb leak, hypotony, posterior capsule opacification and shallow anterior chamber.

　　Statistical analysis Extracted data were pooled for summary estimates using Stata 10.1 for Windows (StataCorp LP, College Station, TX, USA). Continuous outcomes were expressed as standardised mean difference (SMD), with values <0 favouring twosite phacotrabeculectomy,and dichotomous outcomes as odds ratio (OR) or relative risk (RR). Both outcomes were reported with 95% confidence interval (CI). P<0.05 was considered statistically significant on the test for overall effect. Intertrial statistical heterogeneity was explored using the DerSimonian and Laird Q test, with calculated I2 indicating the percentage of the total variability in effect estimates among trials that is due to heterogeneity rather than chance. If heterogeneity tests were nonsignificant, fixed effects models were used, as they provide narrower 95% CIs than the equivalent random effects models, which are more appropriate where significant heterogeneity is detected. The Begg and Egger tests were used to assess for publication bias.

　　For studies that only reported absolute values for IOP at baseline and end point, the IOPR, standard deviation (SD) of the IOPR (SDIOPR), IOPR% and SD of the IOPR% (SDIOPR%) were calculated as follows: IOPR = IOPbaselineIOPend point, SDIOPR= (SDbaseline2+ SDend point2 SDbaseline ×SDend point)1/2, IOPR% = IOPR/ IOPbaseline, SDIOPR%= SDIOPR/ IOPbaseline. The difference of IOPR and its SD between groups was then calculated for each individual study.

　　RESULTS

　　Description of studies Seventeen potentially relevant controlled clinical trials associated with onesite and twosite phacotrabeculectomy in the treatment of coexisting cataract and glaucoma were identified through the literature search. Among these, four articles without exact raw data available for retrieval according with the exclusion criteria were excluded; two abstract reports were found in the annual meeting abstracts of ARVO; eleven controlled clinical trials that fulfilled the eligibility criteria were included in the present metaanalysis[616]. These were published in 8 different journals in English, Chinese and Spanish and no unpublished data were identified(Table 1).

　　Efficacy Effect sizes (SMD in patients with onesite and twosite phacotrabeculectomy on IOPR%) from the fixed effects model for all are prospective and retrospective studies, respectively (Figure 1). Twosite phacotrabeculectomy was associated with numerically lower IOPR% relative to onesite in all studies, except for those by Mandic et al[9] and Buys

　　et al[15]. Both surgical procedures significantly decreased IOP. The pooled summary estimate for all 11 studies favoured

　　Table 1Characteristics of included studies

　　AuthorsYearCountryDesignFollowup

　　(χ,mo)Participants

　　(n)Age

　　(χ,yr)M/FEyes(n)1site2site

　　Wyse et al[6]1998USAPro16.53375.07/262013

　　el Sayyad et al[7]1999Saudi ArabiaPro127665.5NA3739

　　Borggrefe et al[8]1999GermanyPro195074.316/342525

　　Mandic et al[9]2000CroatiaPro125571.617/222731

　　Zou et al[10]2001ChinaRetro18.94561.229/162918

　　IsasiSaseta et al[11]2002SpainRetro63576.416/191916

　　Dong et al[12]2004ChinaRetro123560.916/191525

　　Shingletonet al[13]2006USARetro12130NANA7164

　　Cotran et al[14]2007USAPro367675.426/504343

　　Buys et al[15]2008CanadaPro247970.929/503940

　　Nassiri et al[16]2008IranRetro1811368.855/586152

　　NA: not available.

　　Figure 1SMD in patients with onesite and twosite phacotrabeculectomy on IOPR%from the fixed effects model.

　　twosite procedure, and showed twosite phacotrabeculectomy was more effective than onesite in lowering IOP (SMD, 0.19; 95% CI, 0.33 to 0.04; P=0.01). No significant heterogeneity was presented between studies in the onesite versus twosite groups (χ2= 8.86, P=0.55,I2=0.0%). Then, we divided the studies into two subgroups according to study design (prospective and retrospective). Both prospective and retrospective subgroups showed that twosite approach was associated with numerically lower IOPR relative to onesite procedure, but no significant difference was found. There was no significant heterogeneity in these analysis. Publication bias was also tested using the Begg test (P=0.28) and the Egger test (P=0.34), and both produced nonstatistically significant results, providing no evidence of publication bias.

　　Three studies involving 166 eyes compared onesite with twosite procedure in visual acuity after phacotrabeculectomy (69% onesite and 78% twosite)[7,8,12]. No statistical heterogeneity was observed between studies (χ2= 0.10, P= 0.95,I2=0.0%). The combined result showed there was nonsignificant statistically difference in the percentage having a BCVA of 0.5 or better (OR, 0.65; 95% CI, 0.30 to 1.39, P=0.26).Seven studies, involving 426 eyes, reported the proportions of twosite patients than onesite patients achieved the target IOP without antiglaucoma medication at the end point (73% onesite and 79% twosite)[610,14,15]. No statistical heterogeneity was showed between studies (χ2=8.71, P= 0.19,I2= 31.1%), and the difference between groups was not statistically significant (RR, 0.94; 95% CI, 0.84 to 1.04; P= 0.22).

　　Tolerability Adverse events in controlled clinical trials comparing between onesite and twosite phacotrabeculectomy are showed in Table 2. Hyphema was one of the most commonly reported postoperative adverse events. However, no

　　Table 2Adverse events between onesite and twosite phacotrabeculectomy

　　Adverse eventsStudies

　　(n)Crude event rate,n/nOnesiteTwosite

　　RR (95%CI)HeterogeneityQPI2significant differences comparing between onesite and twosite phacotrabeculectomy were found in the incidence of hyphema, choroidal detachment, hypotony, bleb leak, posterior capsule opacification and shallow anterior chamber, with the pooled RRs being 1.03 (95% CI 0.61 to 1.75), 0.80 (95% CI 0.36 to 1.80), 1.03 (95% CI 0.55 to 1.92), 1.74 (95% CI 0.87 to 3.48), 1.26 (95% CI 0.59 to 2.70) and 0.90 (95% CI 0.27 to 2.95), respectively.

　　DISCUSSION

　　Twosite phacotrabeculectomy now is used frequently as a primary intervention for the management of coexisting cataract and glaucoma[5]. However, it remains controversial as to whether it provides a better outcome than onesite phacotrabeculectomy in the treatment of coexisting cataract and glaucoma[616]. Previous studies have prospectively evaluated the efficacy and tolerability of onesite phacotrabeculectomy compared with twosite procedure[69,14,15]. The overwhelming majority of studies presented that twosite procedure was associated with a numerically lower but nonsignificant reduction in IOP efficaciously compared with onesite approach[68,14]. Variations of sample sizes and followup time within these studies prohibit attribution of treatment outcome to one type of intervention in these reports and make it difficult to draw a valid conclusion regarding the superiority of one procedure over another. We identified various studies that provided comparative treatment outcomes of onesite and twosite procedure and controlled for variations in study characteristics to identify a preferred intervention for the management of coexisting cataract and glaucoma.The results of this metaanalysis imply that, with available evidence from controlled clinical trials, the efficacy of twosite phacotrabeculectomy appears to be superior to onesite for the management of coexisting cataract and glaucoma, and there is nonsignificant difference in tolerability between two surgical procedures. Twosite phacotrabeculectomy was associated with numerically greater, and significant, efficacy than onesite in lowering IOP, numerically greater, but nonsignificant, proportions of twosite patients than onesite patients had a BCVA of 0.5 or better,and numerically greater,but nonsignificant, proportions of twosite patients than onesite patients achieved the target end point IOP. Twosite procedure was comparable with onesite in lowering adverse events. However, the greater IOPR effect and slightly greater BCVA increase effect of twosite procedure over onesite that we have shown does not necessarily indicate a greater surgical effect with twosite procedure. This is because IOP and BCVA merely are surrogate measures for phacotrabeculectomy, and the two surgical procedures may act through pathways independent of this mechanism. There are many preoperative and postoperative key factors to determine which surgical approach to carry out. Factors that may favor a onesite procedure are faster surgical time, less corneal endothelial cell loss, and surgeon experience with a superior approach. Factors that may favor a twosite approach are surgeon familiarity with temporal phacoemulsification, orbital physiognomy, reduced the surgicallyinduced astigmatism, conjunctival scar, limited superior access, ergonomic comfort for the surgeon, and absence of irrigation outflow underneath the conjunctival flap during phacoemulsification that might potentially affect intraoperative antimetabolite effect.

　　The results of our metaanalysis should be interpreted with caution because there may be some limitations in this metaanalysis. One limitation of our metaanalysis is that the analysis of clinically relevant outcome measures that were based on data pooled from trials and followup periods were not uniform. Another potential source of heterogeneity in the results is the assessment criteria of success. Success was defined as target end point IOP, and there were several different criteria of the normal IOP, such as IOP ≤18, ≤20, and ≤21mmHg. Although such assessments of success are widely used as outcome measures in clinical trials, further research is still needed to fully determine their validity, reliability, and sensitivity to choose the best one. A third limitation of this metaanalysis is that publication bias cannot be excluded fully, because with no sufficient studies, the Begg and Egger tests have a low power to detect publication bias. Finally, some of the controlled clinical trials included in the analysis are not prospective randomized controlled trials, but retrospective or prospective nonrandomized, which may fail to detect actual results. The likelihood of bias was minimized by developing a detailed protocol before initiating the study, by performing a meticulous search for published and unpublished studies, especially published in other languages, and by using explicit methods for study selection, data extraction, and statistical analysis.

　　In summary, based on the findings of this metaanalysis, we conclude that the efficacy of twosite phacotrabeculectomy appears to be superior to onesite in IOP control, and the proportions of patients in both surgical procedures achieving BCVA of 0.5 or better were comparable, as well as complete success rate. Both twosite and onesite procedure were well tolerated. Pragmatic randomized controlled trials are needed to further evaluate the efficacy and tolerability of twosite phacotrabeculectomy in the treatment of patients with coexisting cataract and glaucoma. In particular, multicenter, longterm, large sample size, randomized, controlled trials are warranted.

　　【参考文献】

　　1 Shingleton BJ, Pasternack JJ, Hung JW, ODonoghue MW. Three and five year changes in intraocular pressures after clear corneal phacoemulsification in open angle glaucoma patients, glaucoma suspects, and normal patients. J Glaucoma 2006;15(6):494498

　　2 Tham CC, Li FC, Leung DY, Kwong YY, Yick DW, Lam DS. Microincision bimanual phacotrabeculectomy in eyes with coexisting glaucoma and cataract. J Cataract Refract Surg 2006;32(11):19171920

　　3 Bayer A, Erdem U, Mumcuoglu T, Akyol M. Twosite phacotrabeculectomy versus bimanual microincision cataract surgery combined with trabeculectomy. Eur J Ophthalmol 2009;19(1):4654

　　4 Tham CC, Kwong YY, Leung DY, Lam SW, Li FC, Chiu TY, Chan JC, Lam DS, Lai JS. Phacoemulsification versus combined phacotrabeculectomy in medically uncontrolled chronic angle closure glaucoma with cataracts. Ophthalmology 2009;116(4):725731

　　5 Kobayashi H, Kobayashi K. Randomized comparison of the intraocular pressurelowering effect of phacoviscocanalostomy and phacotrabeculectomy. Ophthalmology 2007;114(5):909914

　　6 Wyse T, Meyer M, Ruderman JM, Krupin T, Talluto D, Hernandez R, Rosenberg LF. Combined trabeculectomy and phacoemulsification: a onesite vs a twosite approach. Am J Ophthalmol 1998;125(3):334339

　　7 el Sayyad F, Helal M, elMaghraby A, Khalil M, elHamzawey H. Onesite vs 2site phacotrabeculectomy: a randomized study. J Cataract

　　Refract Surg 1999;25(1):7782

　　8 Borggrefe J, Lieb W, Grehn F. A prospective randomized comparison of two techniques of combined cataractglaucoma surgery. Graefes Arch

　　Clin Exp Ophthalmol 1999;237(11):887892

　　9 Mandic Z, Ivekovic R, Petric I, ZoricGeber M. Glaucoma triple procedure: a onesite vs a twosite approach. Coll Antropol 2000;24(2):367371

　　10 Zou Y, Lin Z, Zhou J. Comparison between onesite and twosite incision in phacotrabeculectomy. Chin J Ophthalmol 2001;37(5): 335337

　　11 IsasiSaseta MB, UrcelaySegura JL, ZamoraBarrios J, OrtegaUsobiaga J, Moreno GarcíaRubio B, CortésValdés C. Trabeculectomy and phacoemulsification. One site vs two site approach. A comparative study. Arch Soc Esp Oftalmol 2002;77(12):677680

　　12 Dong DQ, Chen G, Hou XW. Clinical observation of the combination of phacoemulsification and trabeculectomy. Chin J Ophthalmol 2004;40(5):295298

　　13 Shingleton BJ, Price RS, ODonoghue MW, Goyal S. Comparison of 1site vs 2site phacotrabeculectomy. J Cataract Refract Surg 2006;32(5):799802

　　14 Cotran PR, Roh S, McGwin G. Randomized comparison of 1site and 2site phacotrabeculectomy with 3year followup. Ophthalmology 2008;115(3):447454

　　15 Buys YM, Chipman ML, Zack B, Rootman DS, Slomovic AR, Trope GE. Prospective randomized comparison of one vs twosite Phacotrabeculectomy twoyear results. Ophthalmology 2008;115(7):11301133

　　16 Nassiri N, Nassiri N, Rahnavardi M, Rahmani L. A comparison of corneal endothelial cell changes after 1site and 2site phacotrabeculectomy. Cornea 2008;27(8):889894