Comparison of the safety and efficacy of Mitomycin C 0.02% used intra-operatively by subconjunctival injection versus direct scleral application using sponges in phacotrabeculectomy: A prospective randomized controlled trial

Abstract

Purpose:

To compare the safety and efficacy of subconjunctival injection (MMC: 0.02%) to those with subconjunctival application of sponges soaked in Mitomycin C (MMC: 0.02%) intra-operatively in patients who underwent phacotrabeculectomy.

Methods:

A total of 68 patients undergoing phacotrabeculectomy were randomized into two groups; the sponge group received 0.02% MMC-soaked sponges (n = 38), and the injection group received subconjunctival injection of 0.02% MMC (n = 30). The primary outcome was post-operative mean intra-ocular pressure reduction (IOP). The secondary outcomes were bleb morphology using Indiana Bleb Appearance Grading Score (IBAGS) and anterior segment optical coherence tomography (AS-OCT), post-operative use of 5-fluorouracil, and complications rates. These were compared at 1 week and 1, 3, and 6 months post-operatively. Complete success was defined as ≥30% reduction in IOP without anti-glaucoma medications.

Results:

In sponge and injection groups, the mean pre-operative IOP was 29.1 ± 8.1 and 29.8 ± 8.8 mmHg, respectively. At 6 months, IOP in sponge and injection groups reduced by 52% (14 ± 3.6 mmHg, P < 0.001) and 50% (15.2 ± 4.1 mmHg, P < 0.001), respectively. Complete success was observed in 90.9% in the sponge group and 95.83% in the injection group. Both groups had diffuse, shallow, relatively avascular blebs at 6 months using IBAGS and AS-OCT. A few complications were seen in the sponge group during this period, which were not vision-threatening.

Conclusion:

Subconjunctival MMC injection is an effective, safe, convenient, and time-saving alternative to sponge-soaked delivery of MMC in phacotrabeculectomy.

Trabeculectomy is the most commonly performed glaucoma filtration surgery in developing countries.[1] Given the cost of medications and the poor compliance, medical management of glaucoma, although desired, remains limited. Moreover, most of these patients have visually significant cataract.[2,3] Hence, it is not uncommon to perform both cataract surgery and trabeculectomy as combined or sequential procedures, the outcomes being comparable in both.[2,3] The success of trabeculectomy is, however, limited by the occurrence of subconjunctival reactive fibrosis and scarring.[4] Anti-metabolites like Mitomycin C (MMC) in various concentrations (0.01–0.04%) have been applied intra-operatively under the conjunctiva to reduce scarring and thus maintaining a good filtering bleb.[5] MMC promotes apoptosis of fibroblasts present in the tenon’s capsule, along with inhibition of cellular migration and production of an extra-cellular matrix.[5,6,7] A Cochrane review of 11 clinical trials (698 patients) concluded that MMC reduced the risk of surgical failure in eyes undergoing primary trabeculectomy and in high-risk eyes.[8] The method and duration of application of MMC significantly affect the bleb morphology and long-term survival of filtering bleb, thus altering the success of the procedure.[9] The use of MMC over a larger surface area reduces the risk of formation of a localized, cystic bleb and premature failure caused by scarring.[10] However, the use of MMC as an adjunct in trabeculectomy is associated with sight-threatening complications like hypotony and blebitis.[11,12] The conventional method of direct scleral application using MMC-soaked sponges is also associated with specific complications such as losing sponges in the orbit, breaking off of a sponge fragment, and its accidental retention, all of which may result in scleral necrosis and perforation with its related complications.[13] There is also evidence of variability in drug delivery characteristics with the use of different brands of microsurgical sponges when observed in vitro, suggesting that the volume of MMC delivered to the tissues varies with the type of sponge.[14] Unlike the sponge method, the concentration and volume of MMC delivered via subconjunctival injection remain consistent along with a larger surface area exposure, thereby promising a more uniform and reliable post-operative bleb outcome.[13] Recent studies have compared the efficacy of subconjunctival/intra-tenon injection of MMC 0.01%,[15] MMC 0.02%[16,17,18] against direct scleral application of sponges soaked in MMC 0.02%,[15] MMC 0.04%[16,17,18] in patients who underwent primary trabeculectomy. Both the techniques have been shown to have comparable success rates in terms of IOP reduction[15,16,17,18] with a relatively better bleb morphology[15] and lower complication rates with intra-operative injection.[17] To the best of our knowledge, there are no studies comparing these two techniques using the same concentration of MMC in patients undergoing phacotrabeculectomy. This study aims to compare the efficacy and safety of intra-operative delivery of 0.02% MMC by direct scleral application using sponges versus subconjunctival injection in patients undergoing phacotrabeculectomy in Indian eyes.

Methods

Study design

This was a randomized, prospective, single-center, parallel-group, double-masked, active-controlled trial conducted from October 1, 2017 to October 2018 in a tertiary care medical college in South India. The study was conducted in compliance with the Declaration of Helsinki and the Institutional Review Board as per the Indian Council of Medical Research guidelines. All patients were included in the study after obtaining informed written consent.

Eligibility criteria

All patients registered to our glaucoma clinic, with primary open angle glaucoma (POAG), pseudoexfoliation glaucoma (PEXG), and chronic angle closure glaucoma (CACG), with a baseline IOP of more than 24 mm Hg with/without significant cataract, due to undergo phacotrabeculectomy/trabeculectomy with MMC were included in the study. Indications for surgery included high absolute IOP, progression of the disease despite maximally tolerated medical therapy, non-compliance to medical management, and inability to afford long-term anti-glaucoma medications (AGMs). Patients with normal tension glaucoma and secondary glaucoma other than PEX glaucoma were not included.

Randomization

Subjects were randomized in the operation theater on the day of surgery to receive either subconjunctival injection (injection group) or subconjunctival direct scleral application with impregnated cellulose sponges (sponge group) using the block randomization method with permuted block design of blocks 2 and 4. The allocation concealment was done using the opaque envelope method.

Surgical procedure

All the surgeries were performed by one of three glaucoma specialists in the department with more than 15 years of experience and using similar surgical techniques under peribulbar anesthesia. MMC injection was available as 4 mg in 10 ml solution (Zydus, Oncosciences, India). 1 ml of this was diluted with 1 ml sterile water for injection to give a final concentration of 0.2 mg/ml or 0.02% MMC. The same concentration was used for both the injection group and the sponge group. The trabeculectomy site or the area for the planned bleb was pre-determined to be between 12 o’clock and 2 o’clock positions. The eye to be operated was painted with povidone-iodine solution and draped with sterile drapes, and a superior rectus bridle suture was applied. The envelope was opened by one of the investigators. Those patients who were randomized to the injection group were given 0.1 ml of 0.02% MMC injection using a tuberculin syringe fitted with a 26-gauge needle, 6–8 mm behind the limbus [Fig. 1]. The injected fluid was distributed nasally, temporally, and toward the limbus using a cotton-tipped applicator over the conjunctiva. After 1 minute, a fornix-based conjunctival peritomy was done between 12 and 2 o’clock positions using blunt tipped Westcott scissors. Blunt dissection of the Tenons capsule was also performed, and a large sub tenon’s space was created. The area was thoroughly washed using 30 ml of balanced salt solution. In the sponge group, a fornix-based conjunctival peritomy was performed superiorly between 12 and 2 o’clock using blunt-tipped Westcott scissors. Sub tenon’s dissection was done over a wide area posteriorly. Using bipolar eraser cautery, the bleeding vessels were cauterized, taking precaution not to damage the conjunctiva. Three sponges 2 × 2 mm (cut from polyvinyl alcohol sponge spears; Omni lens Pvt. Ltd.), soaked in 0.2 ml of 0.02% MMC, were placed under the tenon’s capsule 4 to 6 mm behind the limbus over 4 clock hours for 90 seconds [Fig. 2]. The conjunctiva was pulled over this area to the limbus so that the edge of the conjunctiva does not touch the sponges. After 90 seconds, the sponges were meticulously removed, and the area was thoroughly washed with 30 ml of balanced salt solution. After cauterizing bleeding vessels, a partial thickness rectangular 3 × 2 mm scleral flap which is about 1/2-2/3^rd scleral depth was made with the help of a blade and a microcrescent. This lamellar dissection was continued beyond the limbus up to 1 mm into the cornea taking care not to inadvertently enter the anterior chamber.

Figure 1.

Intraoperative 0.02% MMC application using injection method

Figure 2.

Intraoperative 0.02% MMC application using sponge method

Phacoemulsification with foldable IOL implantation was performed using a clear corneal 2.8 mm incision superotemoprally and superonasally in the right and left eyes, respectively, before proceeding with trabeculectomy. The pupil was constricted using intracameral pilocarpine (0.5% w/v, Carpinol, Sunways India PVT LTD), and the anterior chamber was formed with air. The partial-thickness scleral flap was then lifted, and the anterior chamber was entered using a 1 mm blade at the anterior clear corneal limbus. A block of cornea and sclera was excised using Kelley’s punch ensuring an adequate overlap of the scleral flap. Surgical peripheral iridectomy was performed through the area where the block was excised. 10-0 nylon suture was used to secure the scleral flap by using one interrupted suture and one releasable suture at the posterior corners of the rectangular scleral flap. The conjunctiva was closed using 10-0 nylon sutures. Subconjunctival injection of dexamethasone and gentamicin was given at the end of surgery, and ointment atropine was applied to the inferior fornix.

Outcome measures and follow-up

The primary outcome measure of the study was mean IOP reduction in both groups. Secondary outcomes studied were bleb morphology using the Indiana Bleb Appearance Grading Scale (IBAGS)[19] and anterior segment optical coherence tomography (AS-OCT),[20] the number of post-operative interventions, and rate of complications. All patients were admitted to the hospital for 5 days post-operatively as per the existing standard operating procedure for phacotrabeculectomy and were treated with prednisolone acetate eye drops every hour, chloramphenicol eye drops 4 times daily, and atropine drops at night. Patients were examined twice daily for signs of inflammation, IOP, and the presence of a filtering bleb. They were followed up weekly for 6–8 weeks to check IOP and examine the characteristics of the bleb. In those eyes where the IOP was higher than target IOP and in those where there was no filtering bleb, indirect massage was given while examining the patient on the slit lamp. The releasable suture was either released or trimmed within the first 3 weeks. Subconjunctival injections of 5-fluorouracil (5-FU, 0.10 ml, 50 mg/ml) were administered posterior to the bleb for those patients who had increased vascularization and early scarring. Patients were followed up at 3 months and 6 months. Patient data were collected pre-operatively and post-operatively at 1 week, 1 month (±1 week), 3 months (±2 weeks), and 6 months (±4 weeks) [Fig. 1]. Baseline demographics and clinical data included pre-operative AGMs and compliance as well as routine clinical examination and investigations. These included best-corrected visual acuity (BCVA), slit-lamp examination using Haag Streit model (BM 900), baseline IOP using Goldmann applanation tonometry (GAT), gonioscopy using Goldmann 2 mirror gonioscope, stereo biomicroscopic examination of the optic nerve head using a 78D condensing lens, automated perimetry using Humphrey’s Field Analyser model 750i, OCT using Topcon 3D Optical Coherence Tomography DRI OCT Triton (plus), and target IOP. Post-operative data included BCVA, slit lamp examination, GAT, bleb morphology (Indiana Bleb Appearance Grading Scale), use of AGMs, and complications at each follow-up.

Standard AS-OCT images were obtained for each bleb as follows: scan 1, a radial section perpendicular to the limbus, and scan 2, a section tangential to the limbus, both through the site of maximal bleb elevation. Bleb height on AS-OCT was compared to scleral thickness and classified as high, moderate, and low blebs. High blebs were defined as total bleb height (TH) exceeding twice scleral thickness. Moderate height blebs were defined as bleb height equal to or less than double of scleral thickness. Low blebs were defined as total bleb height of less than the scleral thickness.

Complete success was defined as a 30% or more reduction in IOP with no evidence of hypotony without AGM at a 6-month follow-up.[18] Those with a 30% reduction in IOP with AGM were considered as a qualified success. All others were categorized as a treatment failure.

Statistical analysis

Considering a non-inferiority margin of 4, assuming an SD of 6.5 mmHg, we required a sample of 35 patients in each arm with 80% power and 5% error.[21] While performing the study, patients undergoing trabeculectomy and phacotrabeculectomy were included in the randomization, giving us a total of 39 patients in the sponge group and 35 in the injection group. Given the considerably small number of patients who underwent only trabeculectomy (n = 6), we decided to exclude those from the final analysis. Five of those patients were in the injection group. Thus we had a total of 38 and 30 patients in the sponge and injection groups, respectively.

Continuous variables expressed as mean (SD)/median (IQR) depending on the normality and categorical data were expressed as frequency along with percentages. Normality was assessed using P-P plots and box plots. The change in continuous outcome over time and between the groups was assessed using generalized estimating equations (GEEs) with identity link and exchangeable correlation, and for categorical outcomes, the log link with a binomial family with exchangeable correlation was used. The association of complete success with groups was performed using Chi-square statistics. All the analyses were performed using STATA/IC 16.0.

Results

A total of 74 eyes of 74 patients were randomized to either of the two study arms [Fig. 3]. 68 eyes (38: sponge group and 30: injection group) of 68 patients who completed 6 months follow-up were included in the analysis. Table 1 shows the demographics and baseline characteristics of the groups. There was no statistically significant difference in age, gender, the eye operated, and the number of surgeries done by each surgeon. POAG was the most common diagnosis in both groups (P = 0.85) and the most common indication for surgery is high IOP despite maximally tolerated medical therapy (P = 0.67). The total number and duration of medical treatment were also similar (P = 0.41).

Figure 3.

Flow diagram showing randomization of patients into the two study groups along with the ‘n’ number on subsequent follow-ups

Table 1.

Baseline Demographics

	Sponge group (n=38)	Injection group (n=30)	P
Age (years) (mean±SD)	64.4±(7.9)	64.8±(8.1)	0.85
Male/Female [n (%)]	19 (50)/19 (50)	17 (54.7)/13 (43.3)	0.58
Right eye/Left eye [n (%)]	21 (55.3)/17 (44.7)	15 (50)/15 (50)	0.66
Surgeon AB/LMA/AS [n(%)]	12 (31.6)/11 (29)/15 (39.4)	12 (40)/12 (40)/6 (20)	0.22
Diagnosis POAG/CACG/PEXG [n(%)]	23 (60.5)/8 (21)/7 (18.4)	16 (53.3%)/6 (20)/8 (26.7)	0.85
IOP in mmHg (mean±SD)	29.1+/- 8.1	29.8+/- 8.8	0.60

POAG, primary open angle glaucoma; CACG, chronic angle closure glaucoma; PEXG, pseudoexfoliation glaucoma.

IOP

Mean baseline IOP was similar in the sponge and injection groups (P = 0.60). The IOP at each post-operative follow-up was statistically significantly lower than baseline IOP in both groups.

At 6 months, the percentage reduction in IOP was 52% in the sponge group (mean: 14 ± 3.6 mm Hg, P < 0.001) and 50% in the injection group (mean: 15.2 ± 4.1 mm Hg, P < 0.001). However, between the two groups, there was no statistically significant difference in the mean IOP at any point during the post-operative follow-up of up to 6 months.

Surgical success

Complete success was achieved in 30 (90.9%) of patients in the sponge group compared to 23 (95.8%) in the injection group. Qualified success was seen in 2 (6%) patients in the sponge group compared to none in the injection group. One patient each in the sponge group and injection group had failure of surgery on follow-up.

Bleb morphology: IBAGS and AS-OCT grading

32 patients in the sponge group and 30 patients in the injection group showed no statistically significant difference in the mean bleb height, vascularity, and extent at 3 months. At 6 months follow-up, though we have data of only 22 patients in the injection group, there was no difference [Table 2]. AS-OCT grading of the bleb on 29 patients in the sponge group and 23 patients in the injection group did not show any statistically significant difference [Table 3]. The discrepancy in the ‘n’ number between IBAGS and AS-OCT grading at 3 and 6 months was due to technical issues in the machine due to which bleb photographs/AS-OCT images for a few patients could not be captured during post-operative follow-up.

Table 2.

Bleb morphology based on Indiana Bleb Appearance Grading Scale

IBAGS Grading	Sponge group		Injection group		P
	n	(Mean±SD)	n	(Mean±SD)
3 months	32		30
Height	1.3±0.50			1.2±0.6	0.28
Vascularity	1.9±0.6			2±0.50	0.96
Extent	2.1±0.4			2±0.4	0.73
6 months	32		22
Height	1.2±0.5			1.1±0.6	0.05
Vascularity	1.9±0.7			1.9±0.2	0.70
Extent	2.1±0.4			2.1±0.3	0.32

Table 3.

Morphology of the bleb using Anterior Segment OCT

AS-OCT Grading	Sponge group		Injection group		P
	n	Mean±SD	n	Mean±SD
3 months	34	1.6±0.6	30	1.6±0.6	0.49
6 months	29	1.5±0.5	23	1.3±0.6	0.19

Post-operative additional procedures and complications

Both groups underwent similar post-operative procedures including releasable suture modification, laser suturolysis, bleb needling, and subconjunctival 5-fluorouracil injection [Table 4]. 5.3% in the sponge group underwent choroidal drainage. None of the study patients had any major post-operative complications. Minor post-operative complications like early hyphema in 2 patients (5.26%), early hypotony in 2 patients (5.26%), late hypotony in 2 patients (5.26%), early choroidal detachment in 2 patients (5.26%), late choroidal detachment in 1 patient (2.63%), and tenon’s cyst in 2 patients (5.26%) were observed among patients in the sponge group within 6 weeks.

Table 4.

Post-operative additional procedures

Post-operative Procedures	Sponge group (%)	Injection group (%)	P
Releasable suture released	34 (91.9%)	26 (86.7%)	0.49
Releasable suture trimmed	3 (8.1%)	4 (13.3%)
Argon laser suturolysis	4 (10.5%)	6 (20.0%)	0.27
Needling	6 (15.8%)	2 (6.7%)	0.25
No. of postop 5-FU injections			NA
1	4 (10.5%)	9 (30.0%)
2	6 (15.8%)	6 (20.0%)
>2	5 (13.1%)	0 (0.0%)
Choroidal drainage	2 (5.3%)	0 (0.0%)	NA
Conjunctival advancement	0 (0.0%)	1 (3.3%)	NA
Re-suturing of the scleral flap	1 (2.63)	1 (3.33)	NA

NA: Cannot present P values.

Discussion

Patients diagnosed to have POAG, PEXG, and CACG who were planned to undergo trabeculectomy or phacotrabeculectomy with MMC 0.02% were prospectively randomized to either the sponge group or injection group and followed up for 6 months. However, the number of patients who underwent trabeculectomy alone was considerably less (n = 6). Of these patients, the majority (5 out of 6) were in the sponge group and could potentially confound final results. Hence, those data were removed from the final analysis. This resulted in a less than the required number in the injection group as was calculated initially. Nevertheless, even with this limitation, the difference between sponge and injection groups for change in IOP from baseline to 6 months is -1.84 (-7.49, 3.80). This clearly includes the non-inferiority margin. The power analysis for 6 months was calculated as 88.78%, thus ensuring that the sample size of 38 and 30 in both groups has adequate power. The success of phacotrabeculectomy using the same concentration of MMC (0.02%) was similar in both the groups, 91% in the sponge group and 96% in the injection group. The mean IOP reduction at 6 months was 15.14 (52%) and 14.61 (50%) mmHg in the sponge and injection groups, respectively. This reduction is alike to similar studies on trabeculectomy alone done in Caucasian and mixed ethnicities by Pakravan et al.[15] (50% and 52.7% reduction in sponge and injection groups, respectively) and Do et al.[16] (48.6 and 47.8% reduction in sponge and injection groups, respectively).

In a similar study done by Jiun et al.,[22] both methods of MMC delivery (subconjunctival injection and scleral MMC soaked sponge application) had comparable surgical outcomes similar to our study. However, the reduction in IOP at 6 months from baseline was more in our study patients, even though it was comparable among the groups. Most of our patients underwent phacotrabeculectomy compared to only trabeculectomy in their study, and that might have resulted in better reduction in IOP.

Maheshwari et al.,[17] in a study on Indian eyes undergoing trabeculectomy using 0.04% MMC, have also demonstrated a similar reduction (57.5%) in IOP. Our study showed a much higher success rate than other reported rates of complete success in both MMC sponge (25–82.5%) and injection techniques (41.7–82.5%).[10,15,16,17,18] The high short-term success rate in our hands compared to other published data is possibly attributable to the intense post-operative follow-up that is routinely practiced in our glaucoma clinic. This is accompanied by appropriate interventions such as needling of the bleb and also active massage techniques taught to patients and relatives to encourage good bleb formation within the first 6 months. Our study did not have sufficient data to determine whether this approach results in increased long-term success. We also do not have any data on the cost of such an intensive follow-up. We found that both the groups had similar bleb morphologies at the end of 6 months follow-up as graded using not only the IBAGS clinical grading system but also AS-OCT characterization of the bleb. This is in contrast to the finding of a significantly shallower and less vascularized bleb with an intra-tenon injection of MMC in patients with POAG at 3 years follow-up post trabeculectomy.[15,21] Although both groups underwent similar post-operative additional procedures, complications were seen only in the sponge group. These results are concordant with the study published by Maheshwari et al.[17]

The main limitation of our study was the attrition rate of the patients during follow-up. Due to the technical malfunction of the machines, we could not document bleb photographs for all patients during post-operative follow-up resulting in lesser numbers. The bleb grading using IBAGS and AS-OCT was performed by a single masked investigator (NA). Grading by at least two masked authors would have increased the strength of the study.

Conclusion

In conclusion, subconjunctival MMC (0.02%) injection is an effective, safe, and convenient alternative method to the conventional direct scleral application using sponges soaked in MMC (0.02%) in phacotrabeculectomy. Both the techniques provided similar efficacy in lowering IOP. However, the complication rates, though visually non-threatening, were higher in the sponge group as compared to the injection group. A longer follow-up period is necessary to ascertain the long-term safety and efficacy of this alternative technique.

Financial support and sponsorship:

Nil.

Conflicts of interest:

There are no conflicts of interest.