Outcome of Trabeculectomy with Intraoperative Subtenon Injection of Mitomycin C

Purpose: To evaluate the ecacy of trabeculectomy with intraoperative subtenon injection of Mitomycin C (MMC) in terms of control of Intraocular pressure post-operatively. Methods: A total of 49 patient’s medical records who underwent trabeculectomy with intraoperative subtenon injection of MMC with diagnosis of Primary Open Angle Glaucoma (POAG) from January 2017 to December 2018 were evaluated to see the post-operative outcomes in terms of control of intraocular pressure (IOP) with 12 months follow-up. The medical records were retrieved using the hospital information system. Age, gender, preoperative IOP, Best-Corrected Visual Acuity (BCVA), co-morbids, previous surgery, Central Corneal Thickness (CCT), fundus ndings, number of glaucoma medications and postoperative complications were obtained by using a proforma. Results: Total 72 eyes of 49 patient records were evaluated to see post operative outcomes at three, six and twelve months follow up. The mean IOP preoperatively was 24.68±13.66 mm Hg with maximum anti-glaucoma treatment. After the surgery the mean IOP was 13.69±6.68 mmHg at 3 months follow-up, and 12.68±4.04 and 13.33±4.8 mmHg at 6 and 12 months respectively (P-value 0.001). Preoperatively mean CCT was 529.81±28.75 and at 12 months follow-up after surgery was 530.45±29.43 with P-value 0.245. Best Corrected Visual Acuity outcomes were seen at each follow-up and results were found to be statistically signicant (P value ≤ 0.05). Conclusions: Twelve months follow-up of trabeculectomy show that intraoperative subtenon injection of MMC is effective in terms of control of IOP postoperatively with minimal complications in patients with POAG. MMC 0.2mg/ml varied between 1.9 to 17.3 microg. 15 Lee and colleagues rather than using sponge on scleral surface, injected MMC into subtenon tissue at the beginning of the trabeculectomy. 16 Further investigations have shown that this technique of delivering uniform dose of MMC results in good outcome in controlling the IOP with minimal adverse effects. 17,18 We conducted this study to evaluate the outcomes of trabeculectomy procedure after the use of intraoperative subtenon injection of MMC rather than MMC soaked sponges in cohort of patients with diagnosis of Primary Open Angle Glaucoma (POAG). The primary end point was control of IOP at one year postoperatively with success dened as IOP of ≤ 15 mmHg without any additional anti-glaucoma drops and secondary outcomes were change in Central Corneal Thickness (CCT) and in the Best Corrected Visual Acuity (BCVA). All surgeries were performed by the same surgeon (PSM) using peribulbar 2% xylocaine local anesthesia. Surface anesthesia was achieved with topical proparacaine (Alcaine – Alcon, Belgium). A 6 - O vicryl traction suture was inserted into clear cornea superiorly. A small snip incision was given in limbal conjunctiva at 11’O clock position with introduction of blunt 30 guage canula mounted on tuberculin syringe. MMC (Mitomycin-C, Kyowa – Japan) was injected in dose of 0.1 ml (0.2 mg/ml) given in subtenon space 10 mm away from the limbus. A cotton tip applicator was applied over the conjunctival opening to stop any reux of the drug. With weck-cell sponge, MMC was distributed over the large surface of conjunctiva superiorly. Peritomy was carried out immediately from 11’O clock to 1’O clock position with undermining of conjunctiva. The exposed scleral surface was irrigated with 10 ml of BSS. After light cauterization of any bleeding point, a triangular scleral ap measuring 4 X 4 mm was fashioned. Before creating 1 X 1 mm deeper corneo-sclerotomy with Kelly’s punch, anterior chamber was lled and maintained with 1% sodium


Introduction:
Glaucoma is the second leading cause of blindness worldwide effecting as many as 6 million individuals 1 . Almost 60% of the world total glaucoma cases are from Asia 2 . According to one estimate there are more than two million people suffering from this disease in Pakistan 3 . Long-term control of intraocular pressure (IOP) is the mainstay of glaucoma treatment and trabeculectomy still remains the surgical procedure of choice to reduce IOP 4 . A steady rate of failure has been reported after this surgical modality because of scarring external to the scleral ap. 5 The anti brotic agent such as Mitomycin C (MMC) has therefore been used to enhance its success rate. 6 MMC is an antineoplastic/antibiotic agent isolated from soil bacterium Streptomyces caespitosus. 7 It inhibits the broblast proliferation by acting as a deoxyribonucleic acid cross-linker. It is used universally in medicine as a chemotherapeutic agent to treat a variety of cancers. Its use and implementation in ophthalmology is common practice because of its modulatory effects on wound healing. 8 Chen and coworkers 9 were the rst one to use MMC topically in a group of patients with refractory glaucoma. Since then it has been regularly used during trabeculectomy. MMC inhibits the postoperative episcleral brosis, thus enhancing the successful outcome regarding the control of IOP. 10,11,12 Traditionally sponges soaked in MMC are applied over the sclerotomy site before or after the scleral ap formation. However, there is no uniformity among the surgeons on this approach in several ways. The concentration of MMC which is being used ranges from 0.1mg/ml to 0.5mg/ml among different patients. The exposure time of MMC is also variable from 1 minute to 5 minutes duration. There are different types of material used for sponges ranging from weck-cell to lter paper and cotton sticks with each material having different absorption property. The size of sponge used is also not uniform. The e cacy of MMC can also be effected if accidentally exposed to light in the operating room. 13 Moreover the amount of irrigating the site of MMC application with Balanced Salt Solution (BSS) is also variable. 14 All these factors contribute to surgeon's inability to determine the exact amount of drug delivered to the tissue. Mehel and colleagues estimated that actual dose delivered in sponge soaked with MMC 0.2mg/ml varied between 1.9 to 17.3 microg. 15 Lee and colleagues rather than using sponge on scleral surface, injected MMC into subtenon tissue at the beginning of the trabeculectomy. 16 Further investigations have shown that this technique of delivering uniform dose of MMC results in good outcome in controlling the IOP with minimal adverse effects. 17,18 We conducted this study to evaluate the outcomes of trabeculectomy procedure after the use of intraoperative subtenon injection of MMC rather than MMC soaked sponges in cohort of patients with diagnosis of Primary Open Angle Glaucoma (POAG). The primary end point was control of IOP at one year postoperatively with success de ned as IOP of ≤ 15 mmHg without any additional anti-glaucoma drops and secondary outcomes were change in Central Corneal Thickness (CCT) and in the Best Corrected Visual Acuity (BCVA).

Materials And Methods:
This observational study with retrospective data collection was carried out in section of ophthalmology, Aga Khan University Hospital, Karachi, Pakistan.
The permission of study was granted by hospital Ethical Research Committee and was carried out in accordance with declaration of Helsinki.
The charts of all consecutive patients with diagnosis of POAG, who underwent trabeculectomy with intraoperative subtenon injection of MMC from January 2017 to December 2018 were included. The patients with corneal, retinal pathologies and diagnosis of other than POAG were excluded. The calculated sample size was 49 patients (72 eyes) using World Health Organization (WHO) software, taking 95% con dence level and 5% margin of error.
The medical records were retrieved using hospital's information system. Patient's age, gender, Best Corrected Visual Acuity (BCVA) associated comorbid, previous surgery, preoperative applanation IOP, pachymetry, gonioscopy, fundus ndings, number of glaucoma medications and postoperative complications were obtained by using a proforma. Patient follow up was assessed at 3, 6 and 12 months postoperatively.
All surgeries were performed by the same surgeon (PSM) using peribulbar 2% xylocaine local anesthesia. Surface anesthesia was achieved with topical proparacaine (Alcaine -Alcon, Belgium). A 6 -O vicryl traction suture was inserted into clear cornea superiorly. A small snip incision was given in limbal conjunctiva at 11'O clock position with introduction of blunt 30 guage canula mounted on tuberculin syringe. MMC (Mitomycin-C, Kyowa -Japan) was injected in dose of 0.1 ml (0.2 mg/ml) given in subtenon space 10 mm away from the limbus. A cotton tip applicator was applied over the conjunctival opening to stop any re ux of the drug. With weck-cell sponge, MMC was distributed over the large surface of conjunctiva superiorly. Peritomy was carried out immediately from 11'O clock to 1'O clock position with undermining of conjunctiva. The exposed scleral surface was irrigated with 10 ml of BSS. After light cauterization of any bleeding point, a triangular scleral ap measuring 4 X 4 mm was fashioned. Before creating 1 X 1 mm deeper corneo-sclerotomy with Kelly's punch, anterior chamber was lled and maintained with 1% sodium hyaluronate (Provisc -Alcon, Belgium). After peripheral iridectomy, scleral ap was closed with 10 − 0 nylon suture one at apex and one on either side. The patency of ap was checked with BSS pushed through the paracentesis site. Conjunctiva was closed on either side with 10 -O nylon suture and one mattress suture was applied in the central area holding conjunctiva to cornea avoiding any leakage and to prevent conjunctival retraction postoperatively.
The postoperative care consisted of Moxi oxacin 0.3 % (Vigamox -Alcon, Pakistan) every hourly for rst 24 hours, 2 hourly for next 3 days followed by 4 times a day for 3 weeks. We used Dexamethasone drops 0.1 % (Maxidex -Alcon, Pakistan) every hourly for 24 hours, 2 hourly for 4 weeks followed by 4 times a day for another 6 weeks. All conjunctival nylon sutures were removed at 3 weeks postoperatively.

Statistical analysis:
Data was entered and analyzed by using SPSS V. 19.0 (IBM Corp, Armonk, NY). Categorical data were reported as frequencies and percentages and quantitative data as means and standard deviation. Paired sample t-test was applied to compute proportions for continuous variable like IOP and Pachymetry whereas Chi-square test was applied to compute proportions for categorical variables like BCVA. P-value ≤ 0.05 was considered as statistically signi cant.
The mean intraocular pressure preoperatively was 24.68 ± 13.66 mm Hg with maximum anti-glaucoma treatment which comprised of latanoprost 0.005%, combined timolol 0.5% with dorzolamide 2% and brimonidine 0.2%. After the surgery the mean IOP was 13.69 ± 6.68 mmHg at 3 months followup, and 12.68 ± 4.04 and 13.33 ± 4.8 mmHg at 6 and 12 months respectively (P-value 0.001). (Table -II). This IOP was achieved without addition of any topical anti-glaucoma drops.
One eye postoperatively had a conjunctival leak with positive seidle sign and 3 eyes had shallow anterior chamber during 1st week of surgery which were all settled with appropriate medical treatment. No other unwanted side effects were seen throughout the entire 12 months follow-up involving cornea or sclera.
Preoperatively mean CCT was 529.81 ± 28.75 microns and at 12 months follow-up after surgery was 530.45 ± 29.43 microns having no signi cant difference (P-value 0.245). (Table -III) BCVA outcomes were seen at each follow-up. There was a positive change noticed at third month follow-up as most of the patients moved towards good vision with a signi cant difference. (p-value 0.001). Similarly, improvement in visual acuity was also seen at sixth and 12 months follow-up as the frequency of patients with good vision increased and results were found to be statistically signi cant with a P-value 0.045 and 0.032 respectively. (Table  -IV) Although postoperatively patient's visual elds and optical coherence tomography (OCT) did not show any signi cant change from the preoperative results, we attribute the change in visual acuity mainly due to improved tear lm stability with clearing of cornea because of discontinuation of antiglaucoma drops.

Discussion:
MMC augmented trabeculectomy is the common ltration procedure routinely performed in patients with uncontrolled IOP after maximum medical treatment or when routine use of anti-glaucoma drops cannot be guaranteed due to nancial or compliance issues. Weck-cell sponges soaked in MMC applied over the scleral surface has been the standard practice with majority of the glaucoma surgeons. Due to the variation in sponge size, material and MMC concentration and exposure time, subtenon injection of MMC offers a technique where surgeons ensure exact amount of antimetabolite distributed over wide sub-conjunctival area.
Lee et al 16 injected MMC into subtenon tissue at the beginning of the procedure. They injected 0.15 ml of MMC in variable dose of 0.2 mg / ml to 0.5 mg / ml depending on the risk of bleb failure determined preoperatively, over superior part of the conjunctiva using 23-gauge needle. After 5 minutes, the perilimbal conjunctiva was opened and saline was injected to milk out the residual amount of MMC. This was followed with conventional fornix based trabeculectomy. Their 12 months outcome data showed reduction of IOP from preoperative value of 23.6 ± 5.8 mm Hg to 12.2 ± 3.9 mm Hg in 70 % of their patients. The range of complications in their cohort of 108 patients included hypotony (n = 23, 21%), hyphema (n = 16, 15%) and choroidal detachment (n = 17, 16%) which were all managed conservatively. Pakravan and colleagues 19 in comparative analysis used subtenon injection of MMC in dose of 0.1 ml in 40 patients while another group of 40 patients received MMC soaked sponges (0.2 mg / ml) for 1 to 3 minutes. The injection group received subtenon injection of MMC into superior conjunctiva diffused with blunt canula followed by peritomy 1 minute later. Both groups had successful reduction of IOP in 82.5% of patients. They found the drainage bleb more diffuse, less vascularized and shallow in injection group. Both groups did not show any signi cant difference with regard to adverse effects. Khouri and colleagues 20 in retrospective analysis of 2 groups of patients with POAG receiving either injection or MMC soaked sponges concluded that the injection group had overall low postoperative IOP and comparative treatment success de ned as achieving > 30% IOP reduction without glaucoma medication.
Though MMC soaked sponges are traditionally used during glaucoma drainage surger,y the advantage of intraoperative subtenon injection of MMC includes a standard dosage of drug delivered to the surgical site. The wider area of application results in diffuse bleb formation, as the localized drainage bleb created by sponge application tends to be functionally limited by encapsulation described by Khaw 21 as a ring of steel. Another advantage of this technique is easy dissection of conjunctiva and tenon capsule reducing any damage or button hole of conjunctiva. By injecting the MMC, one also eliminates the risk of intraoperative sponge loss 22 . One of the concerns raised by Grower et al 23 for injection technique was that by using the ne needle to inject MMC superiorly, one can pierce the sclera or this may cause conjunctival leak postoperatively. In our surgical method as we enter with blunt canula at limbus, these complications are bypassed.
The limitation of our study is that it is single centered and retrospective in nature with small sample size and without any control group.
Currently a pilot study is undergoing by Caprioli and colleagues 24 at university of California Los Angeles. This compares preoperative injection of MMC at bleb site 3-4 weeks prior to trabeculectomy vs intraoperative injection of MMC vs topical application of MMC as conventionally used. We await results of this study with great interest.

Conclusions:
The technique of application of subtenon injection of Mitomycin C is quick and easy to perform during trabeculectomy and is alternate to use of sponges. The 12-months follow-up showed this technique to be effective in terms of control of IOP in patients with POAG. However, further prospective randomized studies on large population and long term follow-up are needed to clarify the e cacy of this technique.