Supra–Tenon Capsule Placement of Original Molteno vs Molteno 3 Tube Implants in Black Patients With Refractory Glaucoma: Title and subTitle BreakA Single-Surgeon Experience

Much evidence documents the poor response of black patients with glaucoma to standard filtering procedures, including full-thickness filters, trephination, and subsequent trabeculectomies. Obliteration of the filtering bleb by scar tissue is the primary cause of filter failure.^{1 - 2} Welsh³ performed gonioscopy on eyes of black patients following filtration failure and postulated that the overgrowth of fibrous tissue derived from the Tenon capsule blocked the external scleral outlet in most cases. Excision of the Tenon capsule has been reported to improve the success rate of filtration in black eyes.⁴ Gorin⁵ reported a high success rate of filtration in black patients by avoiding handling of the Tenon capsule and by using a thin conjunctival flap. Based on these studies, avoidance of the Tenon capsule in implant surgery seems a logical step in improving outcomes among patients with refractory glaucoma.

A new surgical procedure consisting of supra–Tenon capsule glaucoma implant placement was devised by one of us (J.F.) to assess the effect of eliminating the fibrotic potential of the Tenon capsule on bleb formation. Results using this technique with the original single-plate Molteno tube implant (Molteno Ophthalmic, Dunedin, New Zealand) have been previously reported.⁶ With subsequent introduction of the new single-plate Molteno 3 tube implant (with a larger, thinner, more curved, and flexible plate), it was thought that this plate would be more user-friendly for supra–Tenon capsule insertion. It was also believed that an additional advantage might be provided by its large plate size of 175 mm² compared with 130 mm² for the original Molteno tube implant.⁷ Black patients with refractory glaucoma do not respond well to treatment with glaucoma implants.⁸ Most individuals who have been treated with supra–Tenon capsule placement of Molteno tube implants have been black patients with refractory glaucoma. This study compares the efficacy of supra–Tenon capsule placement of original Molteno vs Molteno 3 tube implants when used to treat refractory glaucoma in black patients. The study assessed the effects of plate size and elimination of the fibrotic potential of the Tenon capsule on lowering intraocular pressure (IOP).

This retrospective study compares the results of supra–Tenon capsule placement of the original Molteno tube implant (130-mm² plate size) with those of the newer Molteno 3 tube implant (175-mm² plate size). Original Molteno tube implants were placed into 17 eyes (original Molteno–implanted group; hereafter referred to as the original Molteno group) and Molteno 3 tube implants into 27 eyes (Molteno 3–implanted group; hereafter referred to as the Molteno 3 group). All patients had undergone previous glaucoma filtration procedures that had failed. Seven patients in the original Molteno group and 5 patients in the Molteno 3 group had previously received Molteno tube implants, one of which was a Molteno 3 placed in a sub–Tenon capsule position that had failed because of bleb fibrosis. All surgical procedures were performed by one us (J.F.) using an identical technique for insertion. The study was approved by the Institutional Review Board of the Long Island College Hospital, Brooklyn, New York.

A supra–Tenon capsule location was formed by creating a fornix-based conjunctival flap and then separating the conjunctiva from the underlying Tenon capsule. This dissection is facilitated by positioning a 7-0 polyglactin suture into the edge of the Tenon capsule and placing it on tension, allowing easy separation from the overlying conjunctiva. The thickness of the conjunctiva can be titrated by creating a plane of dissection closer to the conjunctiva (thin) or closer to the Tenon capsule (thick). This method of titrating the conjunctival thickness is important to ensure that the conjunctiva does not become too thin, resulting in possible erosion over the plate before formation of the bleb, which occurs approximately 3 weeks after implantation. Although larger, the thin, curved Molteno 3 tube implant was found to fit more easily into the created pocket than the flatter original Molteno tube implant. The Tenon capsule anterior to the front edge of the implant is dissected off the underlying sclera and is removed by cautery to prevent bleeding. This exposes the sclera at the front edge of the plate, allowing for fixation of the plate to the exposed sclera. The plate is fixated with 8-0 nylon sutures placed through the 2 anteriorly placed holes in the plates. The silicone tube has a 3-0 polyamide suture placed into the lumen, and the lumen of the silicon tube is tied off posteriorly adjacent to the plate with a 7-0 polyglactin suture, totally occluding the tube. The silicone tube is covered with a pericardial patch before it is inserted into the anterior chamber via a 22-gauge needle entry at the limbus. A vertical slit 2 to 3 mm in length is made with a microsharp blade into the tube before covering it with the patch. This slit facilitates the passage of aqueous to control IOP in the immediate postoperative period before removing the 3-0 polyamide stent, which was usually done 3 weeks after surgery.

All eyes received implants in the upper quadrants superonasally or superotemporally. Postoperative medication consisted of topical antibiotic and corticosteroid drops for a minimum of 4 to 6 weeks, supplemented by antiglaucoma medication if needed before removing the 3-0 polyamide suture.

The following preoperative variables were evaluated for the original Molteno group (Table 1) and for the Molteno 3 group (Table 2): age, sex, type of glaucoma, lens status, number of previous surgical procedures, number of glaucoma medications before and after surgery, IOP before and after surgery, and best-corrected visual acuity. The only statistically significant difference in preoperative variables between groups was the number of previous surgical procedures. The median numbers of previous surgical procedures were 1 for the original Molteno group and 2 for the Molteno 3 group (P < .05, Mann-Whitney test).

The median age of patients was 70 years in the original Molteno group, which was composed of 70% men (n = 12). The median age of patients was 67 years in the Molteno 3 group, which was composed of 48% men (n = 13). The difference in age between groups was not statistically significant by the Mann-Whitney test, and the difference in sex was not significant by the Fisher exact test. Within the original Molteno group, the lens status among 9 of 17 patients was pseudophakic, 6 were phakic, and 2 were aphakic. Within the Molteno 3 group, the lens status among 14 of 27 patients was pseudophakic, 11 were phakic, and 2 were aphakic. The median follow-up periods were 24 months (range, 12-48 months) for the original Molteno group and 18 months (range, 12-48 months) for the Molteno 3 group; the difference was not statistically significant by the Mann-Whitney test. The median preoperative IOPs were 30 mm Hg for the original Molteno group and 24 mm Hg for the Molteno 3 group; the difference was not statistically significant by the Mann-Whitney test (P = .08). The median postoperative IOPs were 13 mm Hg for the original Molteno group and 14 mm Hg for the Molteno 3 group; the difference was not statistically significant by the Mann-Whitney test (P = .03). Compared with preoperative values, postoperative IOPs of both groups were statistically significantly reduced (P < .001, Wilcoxon signed rank test).

The difference in the median number of glaucoma medications used before surgery (3 in both groups) vs after surgery (1 in the original Molteno group and 2 in the Molteno 3 group) was also statistically significant by the Wilcoxon signed rank test (P < .001). However, the difference was not statistically significant by the Mann-Whitney test (P = .08 and P = .13, respectively). A hypertensive phase occurred 3 weeks after surgery in 30% of the original Molteno group (n = 5) and in 22% of the Molteno 3 group (n = 6); the difference was not statistically significant by Fisher exact test.

Success was defined as an IOP of 18 mm Hg or less with or without adjuvant medical therapy. Using Kaplan-Meier analysis, survival percentages were 71% for the original Molteno group and 88% for the Molteno 3 group (Figure). Log-rank test for comparison of survival indicated no significant difference between the 2 groups (P >> .25). At 1 year, the cumulative probabilities of failure were 6% in the original Molteno group and 11% in the Molteno 3 group. However, there were more late failures in the original Molteno group, as the cumulative probability of failure increased to 29% at 3 years compared with an unchanged value of 11% in the Molteno 3 group.

Other glaucoma studies^{9 - 10} have defined total success as a 20% IOP reduction when patients are not receiving adjuvant glaucoma medication and partial success as a 20% IOP reduction when patients are receiving adjuvant glaucoma medication. Using these criteria in the original Molteno group, 11 of 17 patients (65%) achieved partial success and 6 (35%) achieved total success; there were no failures. In the Molteno 3 group, 23 of 27 patients (85%) achieved partial success and 2 patients (7%) achieved total success; there were 2 failures. By the Fisher exact test, there was no significant difference between groups in achievement of 20% IOP reduction (P = .51). However, the difference in achievement of total success, 35% for the original Molteno group (6 of 17 patients) vs 7% for the Molteno 3 group (2 of 27 patients), was statistically significant by the Fisher exact test.

Complications consisted of 3 choroidal effusions in the original Molteno group and 2 choroidal effusions in the Molteno 3 group, all of which resolved spontaneously. Two tube exposures and 1 tube blockage that occurred among patients in the Molteno 3 group were resolved surgically.

To our knowledge, clinical use of the newly designed single-plate Molteno 3 tube implant has not yet been described in the literature. It differs from the original single-plate Molteno tube implant in that the plate has a larger surface area (175 vs 130 mm²) and is more curved and flatter without a circumferential ridge, facilitating insertion. Ten years ago, one of us (J.F.) developed a technique for glaucoma implant surgery to remove the effect of the Tenon capsule on the thickness of the bleb wall using supra–Tenon capsule placement as opposed to sub–Tenon capsule placement. Implant placement in a supra–Tenon capsule position, eliminating part of the fibrotic process that contributes to bleb fibrosis, has been shown to be effective when used with the smaller-surfaced original Molteno tube implant.⁶ The present study was conducted to ascertain whether the use of a larger-surfaced Molteno 3 tube implant placed in a supra–Tenon capsule position would be more effective because of the larger plate size, confirming the concept that larger implants and larger blebs are more effective in lowering IOP. Alternatively, our investigation aimed to assess whether the mere act of eliminating the fibrotic potential of the Tenon capsule is as effective as increasing the surface area of the plate in lowering IOP.

Flow of aqueous across the bleb wall, and subsequent IOP lowering related to the bleb, depends on the thickness of the bleb wall. The Tenon capsule has a major role in determining bleb wall thickness.^{11 - 13} The results of our study indicate that IOP lowering obtained with different-sized plates did not statistically significantly vary, although the raw figures indicated slightly better IOP lowering with the larger-surfaced plate. If the difference in IOP lowering between the 2 plates of different sizes is minimal and the Tenon capsule is eliminated, the effect on IOP lowering attributed to the difference in plate size is also decreased and bleb function improves and may be more important than plate size.

Supra–Tenon capsule placement of the glaucoma implant was initially reserved for eyes in which previous glaucoma surgery had failed. It was thought that there was substantial tissue reaction to surgery in these eyes, which suggested less conjunctival erosion when the implant was placed mainly just under the conjunctiva. More recently, with the technique of titrating the conjunctival thickness, the procedure has been performed as a primary filter, without any conjunctival erosion. Following supra–Tenon capsule formation, it is mandatory to ensure that the conjunctiva is sufficiently loose and abundant to reach the limbus and cover the plate without any tension. A lack of adequate conjunctiva precludes implant placement into a supra–Tenon capsule location, as this will result in erosion over the plate, mandating implant removal. Two patients who were not included in this study because their surgery was performed more recently experienced conjunctival erosion due to what subsequently was realized to be a lack of flexible and adequate conjunctiva. Both patients received the Molteno 3 tube implant, suggesting that more careful titration of the flap thickness needs to be performed when using this implant because the surface plate is larger and requires a greater area of flexible conjunctiva.

“Glaucomatous” aqueous has been shown to contain proinflammatory substances, particularly transforming growth factor β (TGFβ), which will decrease the efficacy of the bleb as a result of inflammation induced in the bleb wall.^{11 ,14} The potential effect of glaucomatous aqueous on bleb outcome was eliminated in both study groups herein by occluding the tube for 2 to 3 weeks before allowing aqueous of normal IOP to enter the plate area. Aqueous subjected to normal IOP does not contain discernible proinflammatory cytokines.¹⁵ As described by Molteno and Dempster,¹⁶ the transient hypertensive phase usually starts approximately 4 weeks after implantation and lasts for as long as 16 weeks. This phase appears to be more common and intense following valved implant insertion.^{17 - 18} The hypertensive phase seems to be due to allowance of cytokine-rich aqueous, found in glaucomatous eyes, to reach the plate surface. Obstructing aqueous flow to the plate until IOP is normalized is thought to result in less inflammation and occurrence of a hypertensive phase after surgery. The incidence of hypertensive phases in our trial was 30% in the original Molteno group (n = 5) and 22% in the Molteno 3 group (n = 6). The hypertensive phase occurred on average 3 weeks after implantation.

In a study¹⁷ comparing valved with nonvalved implants, the incidence of a hypertensive phase in the nonvalved implant, where aqueous was kept from the plate surface in the immediate postoperative period, was 27.1% compared with 60.4% in the valved implant, where aqueous reached the plate surface immediately, suggesting that glaucomatous aqueous was responsible for the higher incidence of bleb fibrosis. However, the incidence of hypertensive phases in our study seemed to indicate that supra–Tenon capsule placement does not alter the occurrence or frequency of a hypertensive phase.

Transforming growth factor β level has been reported to be pressure related, increasing with greater IOP.⁸ Before allowing aqueous onto the plate surface in our study, tubes were occluded until IOP levels in the eye had been normal for 3 to 4 weeks. Although TGFβ levels were not measured before allowing aqueous onto the plate surface, they were assumed to have decreased, with a resulting low incidence of hypertensive phases. The objective of combining supra–Tenon capsule implant surgery with initial occlusion of the implant tube was to eliminate the proinflammatory effect of glaucomatous aqueous and its cytokine content, as well as to limit the fibrotic potential of the Tenon capsule, which has been shown to contain messenger RNA for TGFβ.¹² Allowing the glaucomatous aqueous to immediately react with the tissue over the implant plate results in the development of a thicker and less functional bleb. In an attempt to prove this concept, we have begun a study to assess cytokine levels in preoperative aqueous of patients receiving Molteno tube implants vs cytokine levels in aqueous obtained at the time of stent removal from the tube, when IOP has normalized.

The findings herein suggest the usefulness of eliminating the fibrotic potential of the Tenon capsule, which results in a thinner and more efficiently draining bleb over the glaucoma implant plate. Furthermore, eliminating this fibrotic potential lessens the advantage of using larger-surfaced plates. Our results compare the use of different-sized single-plate implants using a new surgical procedure in a cohort of black patients with glaucoma, although a weakness of the study is its retrospective design. A randomized controlled prospective investigation among the same patients that uses equal group sizes may be superior as a comparative study.