Antimetabolites in cataract surgery to prevent failure of a previous trabeculectomy

  • Review
  • Intervention

Authors


Abstract

Background

Patients having cataract surgery have often earlier undergone a trabeculectomy for glaucoma. However, cataract surgery may be associated with failure of the previous glaucoma surgery and antimetabolites may be used with cataract surgery to prevent such failure. There is no systematic review on whether antimetabolites with cataract surgery prevent failure of a previous trabeculectomy.

Objectives

To assess the effects of antimetabolites with cataract surgery on functioning of a previous trabeculectomy.

Search methods

We searched CENTRAL (which contains the Cochrane Eyes and Vision Group Trials Register) (2014, Issue 5), Ovid MEDLINE, Ovid MEDLINE In-Process and Other Non-Indexed Citations, Ovid MEDLINE Daily, Ovid OLDMEDLINE (January 1946 to June 2014), EMBASE (January 1980 to June 2014), Latin American and Caribbean Health Sciences Literature Database (LILACS) (January 1982 to June 2014), the metaRegister of Controlled Trials (mRCT) (www.controlled-trials.com), ClinicalTrials.gov (www.clinicaltrials.gov) and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en). We did not use any date or language restrictions in the electronic searches for trials. We last searched the electronic databases on 10 June 2014. We also searched the Science Citation Index database (July 2013) and reference lists of potentially relevant studies.

Selection criteria

Randomised controlled trials (RCTs) of antimetabolites with cataract surgery in people with a functioning trabeculectomy.

Data collection and analysis

Two review authors independently reviewed the titles and abstracts from the electronic searches. Two review authors independently assessed relevant full-text articles and entered data.

Main results

We identified no RCTs to test the effectiveness of antimetabolites with cataract surgery in individuals with the intention of preventing failure of a previous trabeculectomy.

Authors' conclusions

There are no RCTs of antimetabolites with cataract surgery in people with a functioning trabeculectomy. Appropriately powered RCTs are needed of antimetabolites during cataract surgery in patients with a functioning trabeculectomy.

Résumé scientifique

Antimétabolites dans la chirurgie de la cataracte pour prévenir l'échec d'une trabéculectomie précédente

Contexte

Les patients faisant l'objet d'une chirurgie de la cataracte ont souvent subi précédemment une trabéculectomie pour le glaucome. Cependant, la chirurgie de la cataracte peut être associée à l'échec de la chirurgie du glaucome précédente et des antimetabolites peuvent être utilisés dans la chirurgie de la cataracte pour prévenir un tel échec. Il n'existe pas de revue systématique établissant si l'association d'antimétabolites à la chirurgie de la cataracte permet de prévenir l'échec d'une trabéculectomie précédente.

Objectifs

Évaluer les effets des antimétabolites associés à une chirurgie de la cataracte sur l'efficacité d'une trabéculectomie précédente.

Stratégie de recherche documentaire

Nous avons effectué des recherches dans CENTRAL (qui contient le registre d'essais du groupe Cochrane sur l'ophtalmologie) (2014, numéro 5), Ovid MEDLINE, Ovid MEDLINE In-Process and Other Non-Indexed Citations, Ovid MEDLINE Daily, Ovid OLDMEDLINE (de janvier 1946 à juin 2014) , EMBASE (de janvier 1980 à juin 2014), LILACS (Latin American and Caribbean Health Sciences Literature Database) (de janvier 1982 à juin 2014), le méta-registre des essais contrôlés (mREC) (www.controlled-trials.com), ClinicalTrials.gov (www.clinicaltrials.gov) et dans le système d'enregistrement international des essais cliniques (ICTRP) de l'Organisation mondiale de la Santé (OMS) (www.who.int/ictrp/search/en). Nous n'avons appliqué aucune restriction concernant la langue ou la date lors des recherches électroniques d'essais. Notre dernière recherche dans les bases de données électroniques date du 10 juin 2014. Nous avons également consulté la base de données Science Citation Index (juillet 2013) et les bibliographies des études potentiellement pertinentes.

Critères de sélection

Essais contrôlés randomisés (ECR) d'antimétabolites associés à la chirurgie de la cataracte chez des personnes avec une trabéculectomie effective.

Recueil et analyse des données

Deux auteurs de la revue ont d'abord indépendamment passé en revue les titres et les résumés obtenus par les recherches électroniques. Ensuite, deux auteurs de la revue ont indépendamment évalué les articles pertinents en texte intégral et saisi les données.

Résultats principaux

Nous n'avons identifié aucun ECR évaluant l'efficacité des antimétabolites associés à la chirurgie de la cataracte dans l'intention de prévenir chez ces personnes l'échec d'une trabéculectomie précédente.

Conclusions des auteurs

Il n'existe a pas d'ECR sur des antimétabolites associés à la chirurgie de la cataracte chez des personnes avec une trabéculectomie effective. Des ECR d'une puissance statistique suffisante sur des antimétabolites lors de la chirurgie de la cataracte chez des patients avec une trabéculectomie effective sont nécessaires.

Notes de traduction

Traduction réalisée par le Centre Cochrane Français

Plain language summary

Using medications (antimetabolites) to avoid scarring after cataract surgery, which might cause failure of a previous glaucoma operation

Review question: We reviewed the evidence on the effect of using medications (antimetabolites) to avoid scarring after cataract surgery, which might cause failure of a previous glaucoma operation. We did not find any trials to include in this review.

Background: Patients with glaucoma who have had an operation to improve the pressure in their eyes often need a further operation to remove cataracts. Antimetabolites to prevent scarring can be used during the cataract operation with the intention of preventing failure of the previous glaucoma operation.

Search date: The evidence is current to 10 June 2014.

Key results: We did not find any trials to include in this review.

Résumé simplifié

L'utilisation de médicaments (antimétabolites) pour éviter les cicatrices après une chirurgie de la cataracte, ce qui pourrait provoquer l'échec d'une opération du glaucome précédente

Question de la revue : Nous avons examiné les éléments de preuve sur l'effet de l'utilisation de médicaments (antimétabolites) pour éviter les cicatrices après une chirurgie de la cataracte, ce qui pourrait provoquer l'échec d'une opération du glaucome précédente. Nous n'avons trouvé aucun essai à inclure dans cette revue.

Contexte : Les patients atteints de glaucome qui ont subi une opération pour améliorer la pression dans leurs yeux ont souvent besoin d'une nouvelle opération pour enlever les cataractes. Des antimetabolites visant à prévenir les cicatrices peuvent être utilisés lors de l'opération de la cataracte dans l'intention de prévenir l'échec de l'opération du glaucome précédente.

Date de recherche : Les preuves sont à jour jusqu'au 10 juin 2014.

Principaux résultats : Nous n'avons pas trouvé d'essai à inclure dans cette revue.

Notes de traduction

Traduction réalisée par le Centre Cochrane Français

Background

The global burden of vision impairment in 2010 was estimated as 39 million blind and 246 million with low vision (based on 39 surveys and interpolation of data for countries with missing data). The principal causes of visual impairment are estimated as uncorrected refractive errors (43%), cataracts (33%), undetermined causes (18%), glaucoma (2%), and age-related macular degeneration, diabetic retinopathy, trachoma and corneal opacities (all approximately 1%). The causes of blindness are estimated as cataract (51%), undetermined causes (21%), glaucoma (8%), age-related macular degeneration (5%), childhood blindness and corneal opacities (4%), uncorrected refractive errors and trachoma (3%) and diabetic retinopathy (1%) (Pascolini 2012).

Glaucoma is a disease that affects people of all nations and ethnicities (Quigley 2006). Glaucoma is an optic neuropathy with visual function loss. Elevated intraocular pressure (IOP) is one of the most important risk factors, but elevated IOP is not a pre-requisite for glaucoma (Kwon 2009). Glaucoma can occur as open angle (OAG) or angle closure (ACG) glaucoma. With OAG there is loss of ganglion cells and atrophy of the optic nerve but the trabecular meshwork and drainage angle appear normal on gonioscopy, and the pressure may be elevated or in the normal range. With ACG the drainage angle is closed and the pressure is invariably elevated.

Leske 2007 estimated that for individuals aged 40 and over, the prevalence rate in Europeans is 2.1% (Bayesian 95% credible interval 1.6% to 2.7%), in Asians 1.4% (95% credible interval 1.0% to 2.0%) and in Africans 4.2% (95% credible interval 3.1% to 5.8%), but with heterogeneity in African populations ranging from 1% in Nigeria to 8% in Ghana (Buhrmann 2000; Cook 2009; Leske 1994). Other studies estimate the prevalence of ACG in East Asians to be at least 10 times that in Caucasians (Congdon 1992; Hu 1989; Wang 2010), and high rates of normal pressure glaucoma in Japan (Iwase 2004).

Glaucoma can lead to impaired vision and ultimately blindness. If usual medications such as beta-blockers (e.g. timolol) or carbonic anhydrase inhibitors (e.g. dorzolamide, brinzolamide) or prostaglandin-F2α analogues (e.g. bimatoprost, latanoprost or travoprost) are not controlling IOP, treatment options for OAG include more medications, laser trabeculoplasty and guarded filtration surgery (e.g. trabeculectomy), and for ACG peripheral laser iridotomy or trabeculectomy. Medical therapy is typically first-line, but surgery such as trabeculectomy may often be indicated if the disease is advanced, or if the disease is progressing despite drug therapy. 

Description of the condition

Individuals with glaucoma and their surgeons thus have several options.

A Cochrane review compared medical and surgical interventions for OAG and identified four trials (published 1984, 1994, 1996 and 2001) with 888 participants with previously treated OAG (Burr 2012). The risk of progressive visual field loss (after adjustment for cataract surgery) was not significantly different for medications compared to surgery (odds ratio (OR) 0.74, 95% confidence interval (CI) 0.54 to 1.01). However, trabeculectomy was associated with a higher risk of reduced visual acuity (OR 0.47, 95% CI 0.31 to 0.74) and more eye discomfort (P value = 0.03) compared to medication. The review was based on four trials with methodological weaknesses.

The advantages of trabeculectomy are that the patient often no longer requires medication and that the reduction in IOP is usually long-standing; 85% of people after trabeculectomy maintain an IOP at 21 mm Hg or less (Chen 2009; Molteno 1999). Visual acuity does not improve after elective trabeculectomy.

To reduce pressure, the incisional filtering surgical procedure most commonly used is trabeculectomy. The trabeculectomy procedure involves removing a small portion of the tissue at the corneoscleral junction beneath a partial thickness scleral flap. The fluid of the eye (aqueous humour) drains through this new pathway (fistula) into the subconjunctival and sub-Tenon spaces. The scleral flap provides resistance to flow and prevents the eye from completely decompressing, which would lead to hypotony and other complications.

However, after cataract surgery the functioning of a trabeculectomy may diminish or cease due to postsurgical inflammation and scarring (Nguyen 1999; Sood 1983). Risk factors for postoperative failure include previous surgical interventions, prolonged exposure to eye drops (Lavin 1990) and a shorter interval between the trabeculectomy and the cataract surgery (Husain 2012). A functional filtering bleb is essential to a long-lasting and functional trabeculectomy. Scarring of the scleral flap or overlying conjunctiva reduces the flow of fluid. The advent of the antimetabolites 5-fluorouracil (5-FU) or mitomycin C (MMC) has resulted in greater surgical success and lower IOPs by reducing scarring (Green 2014; Wilkins 2005). These antimetabolites are applied through small, soaked sponges placed in direct contact with the sclera at the planned trabeculectomy site. The antimetabolites are applied to a large area surrounding the scleral flap location, and the duration of application is 0.5 to 5 minutes. The area is flushed because intraocular antimetabolite exposure may have deleterious effects on all eye structures. 5-FU can also be given by postoperative subconjunctival injections.

After glaucoma surgery individuals are at increased risk of developing visually significant cataracts (Hylton 2003; Jampel 2009). Surgical removal of these cataracts from eyes with previously functioning trabeculectomies may cause trabeculectomy failure and elevated IOPs (Casson 2002; Mathew 2011; Rebolleda 2002; Seah 1996), which in turn may result in glaucoma progression and vision loss. Buys 2008, Crichton 2005, Crichton 2010, El Sayyad 1999, Shin 1998 and Shingleton 2006 have discussed combined glaucoma and cataract operations, and Crichton 2005 and Crichton 2010 sequences of these operations.

There is no systematic review on whether using antimetabolites during cataract surgery prevents the cataract surgery causing failure of the previous trabeculectomy, and this review examines this question.

Description of the intervention

Any antimetabolite (5-FU or MMC) used during or after cataract surgery to reduce the risk of the cataract surgery causing the failure of an earlier trabeculectomy.

How the intervention might work

The most important cause of surgical failure is fibrosis of subconjunctival and episcleral tissue. Antimetabolites such as 5-FU or MMC may reduce complications such as scarring of the filtering bleb after a cataract operation (or after a trabeculectomy for glaucoma). Both antimetabolites modify the body's healing response to a wound by inhibiting fibroblasts: 5-FU inhibits pyrimidine synthesis (De Fendi 2013) and MMC inhibits DNA replication by acting as a DNA cross-linking alkylating agent (De Fendi 2013; Wilkins 2005).

The amount of scarring occurring after cataract extraction may be affected by the:

  1. interval between trabeculectomy and cataract extraction (Husain 2012);

  2. method of cataract surgery (extracapsular cataract extraction (ECCE) versus phacoemulsification (PHACO), or scleral tunnel versus clear corneal incision);

  3. type of glaucoma (OAG versus ACG);

  4. type and dosage of antimetabolite at cataract extraction;

  5. exposure to antimetabolites at the time of trabeculectomy;

  6. cataract surgical complications; and

  7. surgeon.

We planned to search for data on the effects of these co-interventions.

Why it is important to do this review

Glaucoma is one of the pivotal diseases in ophthalmology. Trabeculectomy and cataract surgery are common procedures that may interact in a negative manner so as to increase morbidity. A recent review of antimetabolites commented: "Although initial surgical success is usually observed with TRAB, long-term follow-up studies show a substantial rate of surgical failure in reducing IOP. Fibrosis of subconjunctival and episcleral tissue is considered the most important cause of surgical failure. In an attempt to prevent this complication, the antifibrotic agents mitomycin C (MMC) and 5-fluorouracil (5-FU) became adjuvant treatments to TRAB." (De Fendi 2013).

There is no systematic review of the outcomes of cataract surgery after an earlier surgical procedure for glaucoma, or of the interventions that may be undertaken to preserve the benefits of the prior glaucoma procedure. Reviewing the evidence on antimetabolites during or after cataract surgery, and thus helping to guide practice rationally so as to avoid trabeculectomy failure and glaucoma progression, is a worthwhile endeavour, as glaucoma and cataracts affect many people and these procedures are widely employed.

Objectives

To assess the effects of antimetabolites with cataract surgery on functioning of a previous trabeculectomy.

Methods

Criteria for considering studies for this review

Types of studies

All RCTs that met our inclusion criteria, with no date or language restrictions.

Types of participants

All individuals with any diagnosis of glaucoma, who have a functioning trabeculectomy bleb and who subsequently underwent cataract surgery. There were no restrictions by age or co-morbidities. Non-penetrating glaucoma surgery was excluded as antimetabolites are rarely used with this procedure.  

Types of interventions

Use of antimetabolites (5-FU or MMC in any dosing or duration) to reduce the risk of the failure of an earlier trabeculectomy as a result of the cataract surgery compared to no antimetabolite during or after cataract surgery. (The antimetabolite is usually given at the end of the cataract operation in the upper fornix). We planned to compare seven factors that may modify the effect of antimetabolites (listed in 'How the intervention might work').

Types of outcome measures

To make this review consistent with the parallel published Cochrane systematic reviews (Green 2014; Wilkins 2005), we selected the same primary and secondary outcomes.

Primary outcomes

Failed trabeculectomy 12 months after cataract surgery, defined as:

  • failure of the functioning trabeculectomy bleb post cataract surgery, leading to another surgery or intervention;

  • uncontrolled IOP (more than 22 mm Hg) requiring resuming eye medications or additional topical or systemic medications;

  • change in vision (loss of 3 or more lines of visual acuity on a logMAR chart).

We planned to record data for longer follow-ups if publications provided them.

Secondary outcomes
  1. Chronic hypotony

  2. Wound leaks, measured by the presence of a positive Seidel test (visible aqueous flow with the tear film stained with fluorescein)

  3. Infection: blebitis and endophthalmitis (defined as infection of the globe contents), since even with prompt aggressive treatment this can result in loss of visual function

  4. Corneal toxicity

  5. Change in mean IOP between baseline and 12 months in the treated compared to the control group

  6. Change in patient quality of life, pain or discomfort from the procedures, as assessed by any questionnaires used by the authors

  7. Any economic assessments of costs and benefits by the authors

Search methods for identification of studies

Electronic searches

We searched CENTRAL (which contains the Cochrane Eyes and Vision Group Trials Register) (2014, Issue 5), Ovid MEDLINE, Ovid MEDLINE In-Process and Other Non-Indexed Citations, Ovid MEDLINE Daily, Ovid OLDMEDLINE (January 1946 to June 2014), EMBASE (January 1980 to June 2014), Latin American and Caribbean Health Sciences Literature Database (LILACS) (January 1982 to June 2014), the metaRegister of Controlled Trials (mRCT) (www.controlled-trials.com), ClinicalTrials.gov (www.clinicaltrials.gov) and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en). We did not use any date or language restrictions in the electronic searches for trials. We last searched the electronic databases on 10 June 2014.

See: Appendices for details of search strategies for CENTRAL (Appendix 1), MEDLINE (Appendix 2), EMBASE (Appendix 3), LILACS (Appendix 4), mRCT (Appendix 5), ClinicalTrials.gov (Appendix 6) and the ICTRP (Appendix 7).

Searching other resources

We searched the reference lists of potentially relevant studies to further identify any relevant studies that were not found by the electronic search and we searched the Science Citation Index database (July 2013). We also contacted experts in the field for suggestions about other unpublished studies or studies in press.

Data collection and analysis

Selection of studies

The Trials Search Co-ordinator assessed the titles and abstracts for relevance and two authors (RET, BCT) independently reviewed the relevant full-text articles and checked them against the aforementioned pre-specified inclusion criteria. We excluded those that both authors agreed did not meet our inclusion criteria. If one or both review authors was unsure about exclusion/inclusion we planned for disagreements to be by arbitrated by the third author (AC) but there were no disagreements. We documented reasons for exclusion.

We did not find any RCTs that met our inclusion criteria. We plan to use the following methodology in review updates.

Data extraction and management

We planned that two review authors (RET, BCT) would extract data independently using the standard Cochrane Eyes and Vision Group data extraction form. We also planned that the statistics reported by the authors in each manuscript would be checked for any errors evident within the manuscript. We planned to correspond with authors when necessary to clarify study design and execution, and ask for additional data as needed. We planned to ask for individual patient data (IPD) if we needed to separate results for participants of interest if they were reported within a larger data set that included participants not all corresponding to our inclusion criteria.

When this review is updated, if RCTs are available, we will enter the outcome data planned in the 'Criteria for considering studies for this review' in The Cochrane Collaboration's statistical software, RevMan 2012.

  • Population: country of study, dates of procedures, numbers in intervention and control groups, demographics of participants (age, gender, ethnicity, recorded co-morbidities).

  • Interventions (glaucoma surgery): technique of trabeculectomy (if available); intra-or postoperative antimetabolites 5-FU or MMC by dose, frequency and duration of dosing.

  • Interventions (cataract surgery): method of cataract surgery (ECCE versus PHACO, and scleral tunnel versus clear corneal incision). Use of antimetabolites by dose, frequency and duration of dosing (if available).

  • Interval between trabeculectomy and subsequent cataract surgery: < 12 compared to ≥ 12 months. We will assess other durations recorded by authors as data permit.

  • The primary and secondary outcomes (which include adverse events) listed in the 'Types of outcome measures'.

Assessment of risk of bias in included studies

In a future update, if RCTs are identified, we will use The Cochrane Collaboration's tool for assessing risk of bias as described in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011a). We will assess studies to ascertain if there is evidence of high, unclear or low risk of bias: see Appendix 8 for further details.

Measures of treatment effect

We planned to analyse dichotomous outcomes with risk ratios (RRs with 95% confidence intervals (CIs)) and for continuous data we planned to compute mean differences (MDs with 95% CIs). Dichotomous outcomes are failed trabeculectomy, uncontrolled IOP, hypotony, wound leaks, infection and corneal toxicity, and continuous outcomes are visual acuity, IOPs, quality of life and economic assessments.

Unit of analysis issues

We planned to assess whether studies included one or two eyes per participant:

  • Studies with two eyes per participant: If the study included both eyes, we planned to assess whether they received the same treatment. If different treatments were given (within-person study) we planned to assess if there was any assessment of whether the antimetabolite administered to one eye could influence the outcome of the fellow eye.

  • Studies with one eye per participant: If the trial included only one eye per participant, we planned to assess whether the eye selected for treatment was chosen randomly, or the worse eye was chosen (i.e. non-randomly).

  • We planned to assess whether the statistical analysis of the study took account of the inclusion of one or two eyes per participant.

If multiple treatments were attempted, we planned to compute the 95% CIs using the number of participants randomised and not the number of treatment attempts.

For trials in which people are randomly allocated to treatment but two eyes are reported, if the trial was not corrected for within-person correlation (clustering), and if the author could not provide intraclass correlation coefficients (ICCs), we planned to search for ICCs in the relevant literature and apply the median value.

Dealing with missing data

If RCTs are identified in a future update, we plan to follow the advice from Chapter 16 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011b):

  1. contact the original investigators to request missing data;

  2. make explicit the assumptions of any methods the authors used to cope with missing data (e.g. that the data were assumed missing at random, or that missing values were assumed to have a particular value such as a poor outcome);

  3. perform sensitivity analyses to assess how sensitive results are to the changes in the assumptions that any authors made, and address the potential impact of missing data and any attempt the authors made to cope with missing data on the findings of the review in the Discussion section. We will not ourselves interpolate or replace any missing data.

Assessment of heterogeneity

We planned to assess statistical heterogeneity with the Chi2 test and I2 statistic.

If there was evidence of substantial between-study heterogeneity (I2 > 50%, Chi2 > 0.1), we planned to assess the reasons for heterogeneity and not pool data.

Assessment of reporting biases

We planned to construct funnel plots to explore publication bias if there were at least 10 studies (Sterne 2011).

Data synthesis

We planned to use a random-effects model and pool studies if (I2 < 50%, Chi2 < 0.1), and a fixed-effect model if there were three trials or fewer.

Subgroup analysis and investigation of heterogeneity

We planned to compare: (1) patients with different intervals between trabeculectomy and cataract extraction, (2) method of cataract extraction, (3) type of glaucoma (OAG versus ACG), (4) type and dosage of antimetabolite at cataract extraction, (5) exposure to antimetabolites at time of trabeculectomy, (6) cataract surgical complications and (7) surgeon.

(Wilkins 2005 noted that variations in the dosage of MMC resulted from variations in both concentration and duration of application time but did not analyse the effect of variations. Green 2014 compared regular and lower-dose 5-FU to control).

Sensitivity analysis

We planned to undertake a sensitivity analysis by trial risk of bias (comparing outcomes for trials at low risk versus unclear and high risk of bias from randomisation, allocation concealment, masking and attrition). (Green 2014 found 12 trials (1319 participants) evaluating 5-FU for glaucoma surgery (last update February 2014), and Wilkins 2005 found 11 trials which investigated intraoperative MMC for glaucoma surgery amongst 698 participants (last update 18 January 2010). The only sensitivity analysis Green 2014 undertook was to exclude trials with score of ≤ 2 on the quality measure used, and Wilkins 2005 included only the three studies rated at low risk of bias).

Results

Description of studies

We did not identify any RCTs of antimetabolites with cataract operations in individuals with a previous functioning trabeculectomy.

Results of the search

The electronic searches yielded a total of 589 references (Figure 1). The Trials Search Co-ordinator scanned the search results, removed 190 duplicates and then removed 389 references that were not relevant to the scope of the review. We screened the remaining 10 reports and deemed two as not relevant on the information given in the abstract. We obtained the full-text of eight reports but none were RCTs and we excluded them. We identified two more reports for Husain 2012, one report for Swamynathan 2004 and a retrospective chart review by Sharma 2007 not identified by the electronic search. These have been added to the excluded studies.

Figure 1.

Results from searching for studies for inclusion in the review

Included studies

We did not identify any RCTs that met our inclusion criteria.

Excluded studies

We excluded nine studies after reviewing the full-text of the reports (Chen 2012; Husain 2012; Hylton 2003; Jampel 2009; Law 2011; Patel 2013; Shahid 2010; Sharma 2007; Swamynathan 2004). Reasons for exclusion can be found in the 'Characteristics of excluded studies' table.

Risk of bias in included studies

We did not identify any RCTs that met our inclusion criteria.

Effects of interventions

We did not identify any RCTs that met our inclusion criteria.

Discussion

Summary of main results

We did not identify any RCTs that met our inclusion criteria.

Overall completeness and applicability of evidence

There are four studies that report outcomes for patients who underwent glaucoma and cataract surgery and received antimetabolites, but none tests the research question in an appropriate randomised controlled trial (RCT) (the reasons why they did not provide an answer to the research question are listed below).

Shahid 2010 (Table 1) performed a retrospective case control study of the use of antimetabolites during or after cataract surgery and whether there were failures of earlier trabeculectomies: there were no statistically significant differences in mean intraocular pressures (IOPs) between the 5-fluorouracil (5-FU) and non 5-FU groups pre-operatively or at 1, 6, 12 and 24 months post-cataract surgery. However, injections of 5 mg 5-FU at 2, 4 and 12 weeks after cataract surgery are very late to prevent any scarring due to the operation.

Table 1. Shahid 2010
  1. 5-FU: 5-fluorouracil
    IOP: intraocular pressure

Methods

Study design: retrospective case control

Total study duration: follow-up 2 years

Masking: not performed

Other concerns about bias: historical control group "24 consecutive patients from the same unit who underwent cataract surgery in the presence of a functioning trabeculectomy bleb, but who had not received 5-FU injections following cataract surgery."

Interval between intervention and when outcome was measured: 1, 6, 12 and 24-month chart follow-up 

Power computation: not performed

Statistics: Chi2 and Fisher's exact tests to compare intervention (cataract surgery) and control group outcomes; Mann-Whitney U test to compare mean values between groups; Kaplan-Meier survival analysis for failure rate of trabeculectomy after cataract surgery; log-rank test to for survival analysis of trabeculectomy. No correction of P values when mean IOP compared at multiple time points

Participants

Total number: 5-FU group n = 25 "consecutive patients with primary open-angle glaucoma and a functioning trabeculectomy bleb"; control n = 24

Setting: Oxford Radcliffe Hospital

Diagnostic criteria: 25 participants with functioning trabeculectomy who underwent phacoemulsification surgery

Gender: 5-FU group (13 male, 12 female); control group (13 male, 11 female)

Age: 5-FU average 70 (range 49 to 83); control 72.6 (range 55 to 84)

Country: England

Co-morbidity, Socio-demographics, Ethnicity - not stated
Date of study January 2004 to July 2006

Interventions

Purpose: "To determine whether the use of postoperative subconjunctival 5-FU reduces the risk of trabeculectomy bleb failure after uncomplicated small incisional cataract surgery."

Intervention 1: subconjunctival injections of 5 mg 5-FU at 2, 4 and 12 weeks after cataract surgery adjacent to the bleb

Historical controls: "24 consecutive patients from the same unit who underwent cataract surgery in the presence of a functioning trabeculectomy bleb, but who had not received 5-FU injections following cataract surgery." 

All participants received (1) subconjunctival steroid at the end of the cataract surgery (2) postoperative: topical chloramphenicol 4 times daily x 2 weeks and topical dexamethasone 0.1% drops 4 times daily x 4 weeks

OutcomesOutcome measured: trabeculectomy failure was defined as IOP > 21 mm Hg on or off treatment at any time after the first postoperative month
Time points from the study that are considered in the review or measured or reported in the study: chart reviews at 1, 6, 12 and 24 months postsurgery
Notes

Husain 2012 and Swamynathan 2004, after initial RCTs of antimetabolites during trabeculectomy, later performed cataract extraction on a non-randomly chosen cohort.

Husain 2012 (Table 2) found that trabeculectomy failure occurred in 22.7% of the cataract surgery group and 41.3% of the non-cataract surgery group. He used Cox regression models on risk factors for trabeculectomy failure and the only significant factor was cataract surgery taking place a shorter time after trabeculectomy (P value = 0.03) but 5-FU versus non 5-FU was not significant (P value = 0.22). The authors also concluded "There was no significant effect of cataract surgery on IOP failure defined at > 21 mmHg (P = 0.444), > 17 mmHg (P = 0.547), and > 14 mmHg (P = 0.853) between operated and nonoperated patients at 3 years follow-up. In addition, there was no significant difference measured after Bonferroni adjustment on the level of visual acuity (logarithm of the minimum angle of resolution) measured between 5-FU and placebo treatment groups before and 3 years after surgery (P = 1.0)."

Table 2. Husain 2012
  1. 5-FU: 5-fluorouracil
    IOP: intraocular pressure
    RCT: randomised controlled trial

Methods

Study design: cohort study (effect of cataract extraction) nested within a RCT of previous trabeculectomy operation for glaucoma in which participants were randomised to receive either 5-FU or no 5-FU

Statistics: Cox regression of time between trabeculectomy and cataract surgery to trabeculectomy failure, adjusted for age, glaucoma diagnosis, presence of diabetes mellitus, IOP before trabeculectomy, total number of antiglaucoma drops instilled before trabeculectomy, average number of steroid drops instilled taken during the first 6 months after trabeculectomy, and usage of 5-FU during trabeculectomy

=====

Study design for the original RCT by Wong TT 2009 and Wong MHY 2013 (Husain 2012 took the data for his cohort study from this RCT)

Sequence generation: "One eye of each patient was randomised to receive either intraoperative 5-FU or normal saline (placebo) during trabeculectomy."

Allocation sequence concealment: not stated

Masking: "double blind" (no further statement)

Other concerns about bias: was a post hoc cohort study within a RCT

Interval between intervention and when outcome was measured: day 1; weeks 1, 2, 3; months 6, 9, 12, 16, 20, 24, 28, 32, 36; yearly thereafter (at 3 years the study protocol stopped and IOPs were assessed by retrospective chart review until 8 years of follow-up)  

Power computation: 203 randomised participants were required, with alpha = 0.05 and power 0.90 to detect 18% difference in visual field progression between 2 treatment arms over a 3-year period, assuming visual field failure rate in placebo arm = 29% versus 11% in 5-FU arm (based on higher rate of loss in Singapore patients due to higher angle-closure incidence), and with expected losses to follow-up, plan was to recruit 239 participants

Statistics: Kaplan-Meier estimations of time to trabeculectomy failure; Cox relative risk regression with proportional hazard assumptions for time to events

Participants

Diagnostic criteria for cataract extraction in cohort study: "The indication for cataract surgery was (1) the presence of a lens opacity deemed to be a cause of decreased vision by the examining opthalmologist and (2) a request from the patient for the cataract to be operated on."

Gender: cataract surgery group male n = 56 (57.7%); non-cataract surgery n = 100 (72.5%)

Median age: cataract surgery group 67 (range 47 to 78); non-cataract surgery 61 (range 36 to 82)

Country: Singapore

Co-morbidity: Diabetes: cataract surgery group n = 30 (30.9%); non-cataract surgery n = 27 (19.6%); Hypertension: cataract surgery group n = 37 (38.1%); non-cataract surgery n = 46 (33.3%); Ethnicity: cataract surgery group Chinese n = 75 (77.3%); non-cataract surgery Chinese n = 110 (79.7%)

===

Total number in original RCT: 235 from 3 hospitals in Singapore (288 recruited, 45 could not be randomised due to other illness or inability to perform visual field tests); 228 completed 36 months of follow-up (4 failures requiring repeat glaucoma procedures; 1 lost to follow-up, 4 died)

Interventions

Purpose: "To determine whether the timing of cataract surgery after trabeculectomy has an effect on trabeculectomy function in terms of intraocular pressure control"

Intervention 1: cataract extraction if ophthalmologist deemed lens opacity caused decreased vision, and participant requested cataract extraction. "Data on the postoperative treatment regimen used were not recorded but typically consisted of topical steroid and antibiotic for a period of 1 month."

Method: phacoemulsification, or extracapsular cataract extraction (ECCE) if lens very hard

Control: no cataract extraction

Definition of trabeculectomy failure: IOP > 21 mmHg on second of 2 consecutive visits after the first 6 weeks

Median time from trabeculectomy to cataract extraction: 21.7 months (range 4.6 to 81.9 months)

Total study duration: median follow-up for cataract surgery group 60 months (range 28 to 84 months) and for group that did not receive cataract surgery 48 months (range 12 to 84 months)

Integrity of randomisation to intervention: 124 were randomised to cataract extraction but: "27 (21.7%) experienced trabeculectomy failure before their cataract extraction and, therefore, were included in the non-cataract surgery group for analyses of time to trabeculectomy failure."

Outcomes

Outcome measured in cohort study of cataract extraction: IOP > 14, 17 or 21 mmHg; secondary outcomes: progression of visual field loss, rates of adverse events, interventions after surgery

Definition of failure of trabeculectomy: IOP > 21 mmHg on second of 2 consecutive visits after the first 6 weeks
Time points from the study that are considered in the review or measured or reported in the study: 36 months follow-up

NotesFunding: National Medical Research Council of Singapore; Medical Research Council (UK); NIHR Biomedical Research Centre

Swamynathan 2004 (Table 3) found that beyond three months all mean IOP measurements in the phacoemulsification group (11.8 ± 4.2) were significantly higher than pre-operatively (8.7 ± 4.5; P = 0.00003) compared to the non-cataract group (9.6 ± 3.5 versus 9.1 ± 3.1 mm Hg (P value = 0.42), although these are all excellent pressures. However, after cataract surgery, "No statistically significant differences were found when comparing the 5-FU and MMC subgroups within either the phacoemulsification group or the control group."

Table 3. Swamynathan 2004
  1. IOP: intraocular pressure
    MMC: mitomycin C
    POAG: primary open angle glaucoma
    RCT; randomised controlled trial

Methods

Study design: cohort study of 29 participants in WuDunn 2002 study who subsequently underwent phacoemulsification, and control = 29 participants who did not undergo phacoemulsification

Purpose: "To evaluate the effect of temporal corneal phacoemulsification on intraocular pressure in eyes after primary trabeculectomy with intraoperative fluorouracil (5-FU) or Mitomycin C (MMC)."

======

WuDunn RCT 2002

Study design: RCT "comparing the efficacies of intraoperative fluorouracil ... and mitomycin ... during primary trabeculectomy and who underwent temporal corneal phacoemulsification after the trabeculectomy..."

Total study duration:

Sequence generation: "enrolled eyes were randomized in a double masked fashion"

Allocation sequence concealment: study co-ordinator not involved in surgical procedure or clinical care

Masking: "double masked"

Interval between intervention and when outcome was measured: 12 months 

Power computation: for power of 80%, two tailed alpha = 0.05, and 90% success rate at 1 year for MMC group, estimated needed 58 eyes/group to detect statistically significant 20% lower rate of success for the 5-FU group

Statistics: Student t test for parametric data, Fisher exact test for proportions

Participants

Total number: intervention: n = 29; control n = 29

Setting: Glaucoma Service, Indiana University Hospital

Diagnostic criteria: participants enrolled in RCT comparing intraoperative 5-FU and MMC during trabeculectomy

Gender: phacoemulsification group (15 male, 14 female); control (20 male, 9 female)

Age: phacoemulsification group 71.4 (range 52 to 83); control 61.9 (range 31 to 81)

POAG/non-POAG: phacoemulsification group 26/3; control 23/6

Country: USA

Co-morbidity, Socio-demographics - not stated; Ethnicity: black/non-black: phacoemulsification group 7/22; control 7/22
Date of study not stated

===

WuDunn 2002 is the original RCT from which data were taken for the cohort study of cataract extraction, and enrolled 115 eyes from 103 participants

Interventions

Intervention 1: phacoemulsification

Control: "29 patients enrolled in the same study but who did not undergo phacoemulsification after the trabeculectomy. We confined our patient selection to those enrolled in the study to ensure uniformity of the technique of the trabeculectomy."

======

WuDunn RCT 2002

Primary trabeculectomy either with topical 5-FU 50 mg/ml to 57 eyes x 5 minutes, or topical MMC 0.2 mg/ml to 58 eyes x 2 minutes

Outcomes

Outcome measured: IOP, visual acuity, number of glaucoma medications
Time points from the study that are considered in the review or measured or reported in the study: 6 (range 3 to 9), 12 months

======

WuDunn RCT 2002

Outcome number of eyes achieving target pressures of 21, 18, 15 and 12 mmHg at 6 and 13 months post-trabeculectomy; best-corrected visual acuity; number of glaucoma medications, postoperative interventions, complications

NotesFunding: Research to Prevent Blindness, Inc., New York

Sharma 2007 (Table 4) in a retrospective chart review identified 60 patients who had undergone trabeculectomy, and then identified 22 patients later who underwent phacoemulsification and received 5-FU, and 25 patients of a colleague who did not use 5-FU after phacoemulsification. Twelve months after trabeculectomy the mean IOPs before phacoemulsification were 14.7 ± 3.7 in the group that was to receive 5-FU after phacoemulsification and 15.6 ± 3.8 in the group that was not to receive 5-FU. Patients were evaluated 1 week and 3, 6 and 12 months after phacoemulsification. Postoperative IOPs were not significantly different at 1 week, 1, 3, 6 or 12 months. However, by 12 months after phacoemulsification 13.6% of the 5-FU group required increased glaucoma treatment and 36.4% in the non 5-FU group (P value = 0.03). Sharma concluded "that 5-FU has a protective effect on the functioning bleb and may be used routinely at the end of phacoemulsification." (but the study is not a RCT).

Table 4. Sharma 2007
  1. 5-FU: 5-fluorouracil
    IOP: intraocular pressure

Methods

Study design: retrospective chart review

Total study duration: 1-year follow-up

Masking: not stated

Other concerns about bias: retrospective chart review, possible non-comparability of cohorts; cases from 1 consultant who used 5-FU compared to those of a colleague who did not

Interval between intervention and when outcome was measured: 1 week; 1, 3, 6, 12 months 

Power computation: not performed

Statistics: independent sample t-tests

Participants

Total number: 5-FU during trabeculectomy n = 22; no 5-FU during trabeculectomy n = 25

Setting: Birmingham and Midland Eye Centre, UK

Diagnostic criteria: participants in retrospective chart review of 1 consultant who used 5-FU after trabeculectomy and participants of another consultant who did not

Gender: 5-FU group (11 male, 11 female); non 5-FU group (15 male, 10 female)

Age: 5-FU group (76.9 11 ± 7.6); non 5-FU group (77.5 ± 9.5)

Country: UK

Co-morbidity: "similar ...systemic hypertension and diabetes mellitus"; Sociodemographics - no data; Ethnicity: 5-FU group (13 Caucasian, 2 Asian, 4 Afro-Caribbean, 3 undetermined), non 5-FU (15 Caucasian, 2 Asian, 3 Afro-Caribbean, 5 undetermined)
Date of study not stated

Interventions

Purpose: "to evaluate the protective role of subconjunctival 5-FU on the pre-existing bleb from trabeculectomy in patients undergoing phacoemulsification."

Intervention: 5-FU after trabeculectomy

Control: no 5-FU after trabeculectomy

OutcomesOutcome measured: IOP before and after phacoemulsification
Time points from the study that are considered in the review or measured or reported in the study: 1 week; 1, 3, 6, 12 months 
NotesFunding: no statement

Potential biases in the review process

There were no limitations of date or language in the searches. The titles and abstracts were reviewed by the Trials Search Co-ordinator and two review authors (RET, BCT) independently assessed the nine full-text articles.

Agreements and disagreements with other studies or reviews

There are no other systematic reviews of cataract extraction with or without antimetabolites after trabeculectomy.

Authors' conclusions

Implications for practice

There are no RCTs of antimetabolites during cataract surgery in patients with a functioning trabeculectomy to guide practice.

Implications for research

Appropriately powered RCTs are needed of antimetabolites during cataract surgery in patients with a functioning trabeculectomy.

A group of researchers could undertake a multicentre trial and obtain an answer about antimetabolites within several years. A co-ordinating centre could maintain lists of individuals undergoing trabeculectomy and the patients could be invited to participate in ongoing monitoring with data collected at the time of trabeculectomy and updated with each visit. Subsequent appointments for cataract extraction could be notified to the co-ordinating centre. Patients should be grouped by prognostic factors (time since trabeculectomy, number of ophthalmic surgeries, age, co-morbidities) and the centre could provide the randomisation sequence for the use of antimetabolites. The primary and secondary outcomes measures could be those used in the reviews by Green 2014, Wilkins 2005 and this review. A power computation should be performed for the key outcomes specified in this review. The centre should standardise surgical techniques, medications and dosages and data collected at time of operations and all follow-ups. Subgroup analyses should be performed as recommended in that section of this review.

Acknowledgements

The Cochrane Eyes and Vision Group (CEVG) created and executed the electronic searches. We thank Catey Bunce for her comments on the protocol and Augusto Azuara-Blanco, Scott Fraser, Richard Wormald, Jennifer Evans and Anupa Shah for their comments on the review.

Data and analyses

Download statistical data

This review has no analyses.

Appendices

Appendix 1. CENTRAL search strategy

#1 MeSH descriptor Cataract Extraction
#2 cataract* near/4 (extract* or aspirat* or operat* or remov* or surg* or excis* or implant*)
#3 lens* near/4 (extract* or aspirat* or operat* or remov* or surg* or excis* or implant*)
#4 phacoemulsif* or capsulorhexis or lensectomy
#5 (#1 OR #2 OR #3 OR #4)
#6 MeSH descriptor Mitomycin
#7 mitomycin*
#8 mytomycin*
#9 mitomicin*
#10 mytomicin*
#11 MeSH descriptor Fluorouracil
#12 fluorouracil*
#13 flourouracil*
#14 fluoro uracil*
#15 5FU*
#16 5 FU*
#17 (#6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16)
#18 MeSH descriptor Trabeculectomy
#19 trabeculectom*
#20 (#18 OR #19)
#21 (#5 AND #17 AND #20)

Appendix 2. MEDLINE (OvidSP) search strategy

1. randomized controlled trial.pt.
2. (randomized or randomised).ab,ti.
3. placebo.ab,ti.
4. dt.fs.
5. randomly.ab,ti.
6. trial.ab,ti.
7. groups.ab,ti.
8. or/1-7
9. exp animals/
10. exp humans/
11. 9 not (9 and 10)
12. 8 not 11
13. exp cataract extraction/
14. ((extract$ or aspirat$ or operat$ or remov$ or surg$ or excis$ or implant$) adj3 cataract$).tw.
15. ((extract$ or aspirat$ or operat$ or remov$ or surg$ or excis$ or implant$) adj3 lens$).tw.
16. (pha?oemulsif$ or capsulorhexis or lensectomy).tw.
17. or/13-16
18. exp mitomycin/
19. mitomycin$.tw.
20. mytomycin$.tw.
21. mitomicin$.tw.
22. mytomicin$.tw.
23. exp fluorouracil/
24. fluorouracil$.tw.
25. flourouracil$.tw.
26. fluoro uracil$.tw.
27. 5FU$.tw.
28. 5 FU$.tw.
29. or/18-28
30. trabeculectomy/
31. trabeculectom$.tw.
32. or/30-31
33. 17 and 29 and 32
34. 12 and 33

The search filter for trials at the beginning of the MEDLINE strategy is from the published paper by Glanville et al (Glanville 2006).

Appendix 3. EMBASE (OvidSP) search strategy 

1. exp randomized controlled trial/
2. exp randomization/
3. exp double blind procedure/
4. exp single blind procedure/
5. random$.tw.
6. or/1-5
7. (animal or animal experiment).sh.
8. human.sh.
9. 7 and 8
10. 7 not 9
11. 6 not 10
12. exp clinical trial/
13. (clin$ adj3 trial$).tw.
14. ((singl$ or doubl$ or trebl$ or tripl$) adj3 (blind$ or mask$)).tw.
15. exp placebo/
16. placebo$.tw.
17. random$.tw.
18. exp experimental design/
19. exp crossover procedure/
20. exp control group/
21. exp latin square design/
22. or/12-21
23. 22 not 10
24. 23 not 11
25. exp comparative study/
26. exp evaluation/
27. exp prospective study/
28. (control$ or prospectiv$ or volunteer$).tw.
29. or/25-28
30. 29 not 10
31. 30 not (11 or 23)
32. 11 or 24 or 31
33. exp cataract extraction/
34. ((extract$ or aspirat$ or operat$ or remov$ or surg$ or excis$ or implant$) adj3 cataract$).tw.
35. ((extract$ or aspirat$ or operat$ or remov$ or surg$ or excis$ or implant$) adj3 lens$).tw.
36. (pha?oemulsif$ or capsulorhexis or lensectomy).tw.
37. or/33-36
38. mitomycin/
39. mitomycin c/
40. mitomycin$.tw.
41. mytomycin$.tw.
42. mitomicin$.tw.
43. mytomicin$.tw.
44. fluorouracil/
45. fluorouracil.tw.
46. flourouracil.tw.
47. fluoro uracil.tw.
48. 5FU$.tw.
49. 5 FU$.tw.
50. or/38-49
51. trabeculectomy/
52. trabeculectom$.tw.
53. or/51-52
54. 37 and 50 and 53
55. 32 and 54

Appendix 4. LILACS search strategy

cataract$ and mitomycin or fluorouracil and trabeculectom$

Appendix 5. metaRegister of Controlled Trials search strategy

cataract and antimetabolite and trabeculectomy

Appendix 6. ClinicalTrials.gov search strategy

Cataract AND Antimetabolite AND Trabeculectomy

Appendix 7. ICTRP search strategy

Cataract AND Antimetabolite AND Trabeculectomy

Appendix 8. Parameters for assessing risk of bias in included studies

Was the allocation sequence randomly generated?

  • Low risk (random number table, computer random number generator, coin tossing, shuffling cards or envelopes, throwing dice, drawing lots, or minimisation).

  • Unclear risk (insufficient information).

  • High risk (date of birth, date of admission, record number, clinician judgement, participant preference, based on the results of tests, or availability of the intervention).

Was the allocation adequately concealed?

  • Low risk (central allocation by telephone, web-based or independent expert, or sequentially numbered, opaque, sealed envelopes).

  • Unclear risk (insufficient information).

  • High risk (open random allocation schedule such as a list of random numbers, unsealed or non­opaque or not sequentially numbered envelopes, alternation or rotation, date of birth or record number).

Was knowledge of the allocated interventions adequately prevented during the study?

  • Low risk (no masking or incomplete masking, but the review authors judge that the outcome is not likely to be influenced by lack of masking; or masking of participants and key study personnel ensured, and unlikely that the masking could have been broken).

  • Unclear risk (insufficient information to permit judgement of 'Low risk' or 'High risk').

  • High risk (not all of the study's pre-specified primary outcomes have been reported; or one or more primary outcomes is reported using measurements, analysis methods or subsets of the data (e.g. subscales) that were not pre-specified; or one or more reported primary outcomes were not pre-specified (unless clear justification for their reporting is provided, such as an unexpected adverse effect); or one or more outcomes of interest in the review are reported incompletely so that they cannot be entered in a meta-analysis; or the study report fails to include results for a key outcome that would be expected to have been reported for such a study).

Surgeons: It is unlikely that surgeons are masked, as it is likely that the same surgeon will assess the patients' need for surgery, decide on the preferred treatment and follow the patients up after the interventions. Absence of masking could affect surgeons' decisions, e.g. if they know antimetabolites were used they might be less inclined to respond rapidly to an increase in intraocular pressure.

Outcome assessors: We will examine whether in any study those who assess outcomes were masked to any aspect of the interventions or outcomes.

Statistical analysts: We will assess whether the statistical analysts were masked to the interventions and outcomes.

Patients: We will assess whether patients were masked to any aspect of the intervention or outcome. Masking is unlikely, because ethical surgery requires that the benefits and complications of surgery are fully explained to patients so that they may make decisions in line with their own preferences. However, we plan to assess, if there are descriptions of disclosures and consultations with patients about the benefits and complications of the procedures, how much detail was provided, and if the study report provides details of signed ethical consents.

Were incomplete outcome data adequately addressed?

  • Low risk (no missing outcome data; reasons for missing outcome data unlikely to be related to true outcome (for survival data, censoring unlikely to be introducing bias); missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk is not enough to have a clinically relevant impact on the intervention effect estimate).

  • Unclear risk (insufficient information).

  • High risk (reason for missing outcome data likely to be related to true outcome, with either imbalance in numbers or reasons for missing data across intervention groups; for dichotomous outcome data, the proportion of missing outcomes compared with observed event risk enough to induce clinically relevant bias in intervention effect estimate for continuous outcome data, plausible effect size (difference in means or standardised difference in means) among missing outcomes enough to induce clinically relevant bias in observed effect size, 'as-treated' analysis done with substantial departure of the intervention received from that assigned at randomisation; or potentially inappropriate application of simple imputation).

For dichotomous outcomes, the potential impact of missing outcomes depends on the frequency (or risk) of the outcome. The Cochrane Handbook for Systematic Reviews of Interventions discusses using sensitivity analysis to approach this issue and provides this example: "a particular study has 20% risk in the intervention arm and 15% risk in the control arm. An available case analysis implicitly assumes that the same fractions apply in the missing data, so three suitable sensitivity analyses to compare with this analysis might consider the risk in the missing data to be 15% in both arms, or 15% and 10% in the experimental and control arms respectively, or 20% and 10% respectively" (Higgins 2011a). We planned to examine the data and conduct appropriate sensitivity analyses to assess the potential effects of missing data.

For continuous outcomes, the potential impact of missing data increases with the proportion of participants with missing data and the intervention effects among participants with missing data. If there are enough data we plan to conduct sensitivity analyses to assess the potential effects of missing data.

There is no satisfactory solution for interpolating missing data (Altman 2007). Patients excluded after randomisation are unlikely to be representative of those remaining (Nűesch 2009). The Cochrane Handbook for Systematic Reviews of Interventions advises mapping any methods for handling missing data closely to the known characteristics of the data-sets, and to other data-sets in the literature that are likely to have comparable outcomes (Higgins 2011a).

Individuals who miss appointments after surgery may not be representative of those who attend all their appointments. Thus intention-to-treat solutions such as baseline carried forward (BOCF), last observation carried forward (LOCF), and complete case analysis (excludes participants with incomplete outcome data) are inappropriate because they require that the mechanisms governing dropout are independent of future unobserved measurements (Kenward 2009; Molenberghs 2004). We do not know whether those who miss monitoring appointments have personal, medical, financial, social or cultural factors that relate to missing appointments.

Thus we will not plan to impute any missing data. We will plan to conduct sensitivity analyses to assess the potential effects of missing data.

Are reports of the study free of suggestion of selective outcome reporting?

  • Low risk (study protocol is available and all of the study's pre-specified (primary and secondary) outcomes of interest in the review have been reported in the pre-specified way; or study protocol is not available but it is clear that the published reports include all expected outcomes, including those that were pre-specified.

  • Unclear risk (insufficient information).

  • High risk (not all the study's pre-specified primary outcomes have been reported; or one or more primary outcomes is reported using measurements, analysis methods or subsets of the data (e.g. subscales) that were not pre-specified; or one or more reported primary outcomes were not pre-specified (unless clear justification for their reporting is provided, such as an unexpected adverse effect); or one or more outcomes of interest in the review are reported incompletely so that they cannot be entered in a meta-analysis; or the study report fails to include results for a key outcome that would be expected to have been reported for such a study.

Was the study apparently free of other problems that could put it at a risk of bias?

  • Low risk (free of other sources of bias).

  • Unclear risk (insufficient information).

  • High risk (potential source of bias related to the specific study design; or stopped early due to a data-dependent process (including a formal-stopping rule); or extreme baseline imbalance).

Contributions of authors

RET and AC drafted the protocol. RET and BCT independently assessed the full-text articles and abstracted the data. RET wrote the review text. All authors reviewed and approved the final text.

Declarations of interest

RET, BCT and AC have no declarations of interest.

Sources of support

Internal sources

  • No sources of support supplied

External sources

  • National Institute of Health Research (NIHR), UK.

    Richard Wormald (Co-ordinating Editor for CEVG) acknowledges financial support for his CEVG research sessions from the Department of Health through the award made by the NIHR Research to Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology for a Specialist Biomedical Research Centre for Ophthalmology. The views expressed in this publication are those of the authors and not necessarily those of the NIHR, National Health Service (NHS) or the Department of Health.

Differences between protocol and review

The following changes to the text have been made:

Primary outcomes: change in vision (loss of 0, 1, 2, 3 ... lines of vision on a Snellen chart) to change in vision (loss of three or more lines of visual acuity on a logMAR chart).

Secondary outcomes:
"5. Higher mean IOP in the treated group compared to the control group
6. Reduction in patient quality of life, pain or discomfort from the procedures, as assessed by any questionnaires used by the authors"

We have reworded as:

"5. Change in mean IOP between baseline and 12 months in the treated compared to the control group
6. Change in patient quality of life, pain or discomfort from the procedures, as assessed by any questionnaires used by the authors"

Characteristics of studies

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    RCT: randomised controlled trial

Chen 2012Letter commenting on Husain 2012
Husain 2012Cohort study (no randomisation) within prior RCT
Hylton 2003Review of cataract extraction after filtration surgery
Jampel 2009Article assessing whether trabeculectomy places patients at higher risk for cataract surgery
Law 2011Management of cataracts in patients with glaucoma
Patel 2013Management of cataracts in patients with glaucoma
Shahid 2010Retrospective case control study
Sharma 2007Retrospective chart review
Swamynathan 2004Cohort study (no randomisation) of patients within a previous RCT

Ancillary