Secondary Logo

Journal Logo


Ab interno trabeculectomy: Ultrastructural evidence and early tissue response in a human eye

Ferrari, Ettore MD; Ortolani, Fulvia PhD; Petrelli, Lucia; Contin, Magali PhD; Pognuz, Derri Roman MD; Marchini, Maurizio PhD; Bandello, Francesco MD

Author Information
Journal of Cataract & Refractive Surgery: October 2007 - Volume 33 - Issue 10 - p 1750-1753
doi: 10.1016/j.jcrs.2007.05.035
  • Free


Ab interno trabeculectomy, a recently introduced procedure, consists of removal of trabecular meshwork1 that anatomically includes the inner wall of Schlemm's canal. The procedure has resulted in a clinically relevant decrease in intraocular pressure (IOP) in a series of cases.2 We report the ab interno trabeculectomy–dependent structural implications in a living human eye with cataract and glaucoma. The eye was enucleated because of concurrent unappreciated choroidal melanoma. To our knowledge, this is the first ultrastructural analysis of the postoperative effects of ab interno trabeculectomy in a human eye as well as the first focused on the impact on tissues after surgery.


A 60-year-old woman with cataract and glaucoma was referred to the clinic for surgery. Two years earlier, she had unsuccessful filtering surgery and her glaucoma was poorly controlled despite maximum tolerated medical therapy (confirmed progression of visual field damage). At the time of surgery, she had a cataract that interfered with the preoperative fundus examination (presence of red reflex with poorly detectable fundus structures) and an IOP of 32 mm Hg under medical therapy. The fellow eye, which had successful filtering surgery 2 years earlier, had satisfactorily controlled glaucoma. The patient received a detailed description of both proposed procedures (phacoemulsification and ab interno trabeculectomy) and of alternative treatments and signed an informed consent form. She had phacoemulsification and ab interno trabeculectomy in the lower quadrant.

The procedure was performed as follows: A 2-step, tunnel-shaped corneal incision was prepared at the temporal quadrant. Two side ports were made, and the anterior chamber was filled with an ophthalmic viscosurgical device (OVD). A continuous curvilinear capsulorhexis was created, and cortical-cleaving hydrodissection was performed. Phacoemulsification and irrigation/aspiration (I/A) of cortical remnants were then performed. An OVD was used to reform the capsular bag, a foldable acrylic intraocular lens was implanted, and the tunnel was closed with a single 10-0 nylon suture. The anterior chamber was then refilled with an OVD to keep the anterior chamber angle open. Trabecular surgery was performed under direct observation of the anterior chamber angle with the operating microscope and a surgical gonioprism. A subretinal vitrectomy forceps (Thomas horizontal subretinal forceps, Synergetics, Inc.) was introduced into the anterior chamber through a limbal side port and directed toward the iridocorneal angle. The trabecular meshwork was then pinched and removed with a peeling-like approach. Great care was taken to selectively remove the trabeculum, avoiding damage of adjacent ocular structures such as the base of the iris or corneal endothelium. When modest bleeding occurred during the operation, the angular recess was filled with OVD until the bleeding stopped. At the end of the surgery, the OVD was removed with an automated I/A unit.

The day after surgery, the IOP was 11 mm Hg and the eye was calm; however, during the routine fundus examination, a midperipheral choroidal pigmented lesion was detected. Further ocular evaluation (echography, fluorescein angiography) confirmed the presence of a choroidal melanoma. The therapeutic options (conservative treatment versus enucleation) were discussed in detail with the patient, and she opted for enucleation. The patient was asked to authorize a histological examination of the anterior globe and a second ab interno trabeculectomy procedure for scientific purposes, both to be performed after the enucleation. The patient gave written informed consent for these procedures. Before enucleation, the IOP was 10 mm Hg. Enucleation was performed 10 days after phacoemulsification with ab interno trabeculectomy.

As planned, a second ab interno trabeculectomy was performed at the opposite site (approximately 180 degrees from the area of trabeculum where the first ab interno trabeculectomy was done) immediately after globe removal as follows: A sclerotomy was created 3.5 mm from the limbus at the opposite site where the choroidal lesion was located. A 4.0 mm standard 20-gauge infusion vitrectomy cannula was inserted and sutured to the sclera to maintain IOP. By adjusting the height of the bottle, the IOP was kept at approximately the lower normal value. With the globe wrapped in gauze and put on a surgical table, an assistant held the globe and oriented it properly. After debridement, 2 corneal side ports were made and the anterior chamber was filled with an OVD. Trabecular surgery was then performed as previously described using the operating microscope and a surgical gonioprism. At the end of the procedure, the OVD was replaced with balanced salt solution (BSS) using an I/A unit. One of the 2 corneal side ports was used to introduce a blunt cannula to replace the BSS with fixative agent.

The part of the globe subjected to the procedures was marked for prompt identification of the treated areas afterward. Two-step fixation was performed with 5% (w/v) glutaraldehyde in 0.1 M phosphate buffer, pH 7.4. The first step consisted of the above-mentioned fixative irrigation into the anterior chamber. After dissection 4.0 mm behind the limbus, the anterior eye portion was immersed in the same fixative agent for 3 days (second step). Three lots of 4 samples were excised from the fixed eye portion: (1) samples from an area not subjected to any procedure (control samples); (2) samples from the area that had ab interno trabeculectomy before enucleation; (3) samples from the area that had ab interno trabeculectomy immediately after enucleation. Morphological analysis of the area of the globe that had trabeculectomy 2 years previously was not done. The excised samples had postfixation with 2% osmium tetroxide dissolved in the same buffer used for the fixation agent, dehydration in graded ethanols, and embedding in Araldite/Epon. Semithin sections were stained with 0.5% toluidine blue in distilled water. Thin sections were put on polyvinyl formal (Formvar)-coated 2.0 mm × 1.0 mm slot copper grids and stained with uranyl acetate and lead citrate. Observations and photographic records were performed with a Philips CM12 STEM transmission electron microscope.


Semithin sections of all samples showed minor changes in the iridocorneal angle as a result of sample handling during excision, reduction, and subsequent processing stages. On control sample sections (Figure 1, A), the integrity of the trabecular meshwork and good preservation of the outer and inner Schlemm's canal walls were always evident, with weak enlargements of interlamellar spaces being the sole artifact-dependent changes. In most sections of ab interno trabeculectomy samples before and after enucleation, full-thickness removal of trabeculum strip of varying widths (Figure 1, B and C) had occurred. Apart from shape or size, these clefts always developed, causing an opening of Schlemm's canal lumen into the anterior chamber. Over the limit of surgical tissue removal, trabecular meshwork was sometimes reduced because some trabecular beams facing the anterior chamber had been marginally involved in the ablative process (Figure 1, C).

Figure 1
Figure 1:
Toluidine blue–stained semithin sections. A: Control samples. B: Post-E-AIT samples. C: Pre-E-AIT samples. Opening (arrowheads) of SC lumen () into the anterior chamber (AC) and concurrent SC outer wall preservation are evident in B and C (×110).

A comparison of semithin sections of samples having ab interno trabeculectomy before enucleation and those having it after enucleation showed the most remarkable feature to be a greater amount of trabecular debris in post-enucleation samples. These fragments were detectable along the inner part of the trabecular meshwork as well as within the Schlemm's canal lumen. Concurrent preservation of the endothelium lining the outer Schlemm's canal and of the other angle structures was observed in samples having ab interno trabeculectomy before enucleation and those having it after enucleation.

Ultrastructural pictures confirmed the integrity of the outer wall of Schlemm's canal in all ab interno trabeculectomy samples. In particular, in samples before trabeculectomy (Figure 2, A) and samples after trabeculectomy (not shown), endothelial cells showed unaltered organule features, intact cell junctions, and well-textured basal laminae.

Figure 2
Figure 2:
Thin sections of samples of ab interno trabeculectomy before enucleation. A: An endothelial cell (En) lining Schlemm's canal (Sc) outer wall, which shows an unaltered intercellular junction (arrow) and a regular basal lamina (F = fibroblast). B to D: Trabeculum beam stumps; endothelial gaps are indicated (asterisks) (F = unactivated fibroblast; Ly = lymphocyte; lsc = long-spaced collagen aggregates; M = macrophage; c = collagen fibrils; e = elastin-dense cores; f = fibrillin microfibrils (original magnification ×7500 [A]; ×6500 [B]; ×12 000 [C]; ×8500 [D]).

Usual ultrastructural features persisted within the unremoved trabecular meshwork beams, whereas the extracellular matrix (ECM) maintained its peculiar loose texture with the presence of (1) sparse 67 nm banded collagen fibrils with 40 nm diameters in average (types I, III, and V collagens); (2) 100 nm banded collagen aggregates (collagen type VI); (3) bundles of microfibrils with approximately 10 nm diameters (fibrillin); (4) the peculiar complexes formed by electron-dense cores (elastin) surrounded by clusters of fibrillin microfibrils (Figure 2, B and D). In addition, neither activated fibroblasts nor leucocyte infiltrates were present, although solitary lymphocytes near the surfaces of some trabecular stumps (Figure 2, B) or macrophages within their stroma (Figure 2, D) were occasionally found. Of interest, the surface of these trabecular stumps showed an alternation of areas covered by unaltered endothelium with uncoated areas, where the ECM was exposed (Figure 2, B).

Finally, neither hematic residues nor fibrin deposits were found at the trabecular meshwork or within Schlemm's canal lumen.


Although ab interno trabecular procedures have been widely used for some time, little information exists about postoperative effects on involved tissue as well as tissue responses. These procedures share the common purpose of improving outflow via conventional pathways. The rationale is to remove the largest portion of resistance to outflow exerted by the juxtacanalicular portion of the trabecular meshwork and the inner wall of Schlemm's canal.3–5 In fact, better aqueous percolation is achieved during ab externo nonpenetrating filtering surgery by peeling the inner wall of Schlemm's canal.6

Ab interno trabeculectomy is a novel trabecular procedure that, when associated with phacoemulsification, has been shown to produce a clinically significant decrease in IOP in eyes with a poor prognosis for filtration surgery.2 The present study allowed us to evaluate the impact on ocular structure resulting from this type of ablation as well as the short-term postoperative tissue responses to the ab interno trabeculectomy procedure in a living eye. Our observations show that this surgical approach causes direct communication between Schlemm's canal lumen and anterior chamber in both in vivo and post-enucleation fresh conditions, confirming previous results reported for eye-bank eyes.1,7 The surgery affected only the trabecular meshwork, including the inner wall of Schlemm's canal, with no apparent structural damage to the outer wall of Schlemm's canal or adjacent angle structures.

The morphological results are consistent with the IOP reduction observed in the immediate postoperative period, suggesting that trabeculum removal improved aqueous outflow. It can also be assumed that the observed absence of hematic residues may have been promoted by the enhanced outflow caused by ab interno trabeculectomy.

Further significant information was provided by ultrastructural analysis. Concerning the endothelium, electron microscopy revealed the presence of cells showing normal-featured organules as well as the absence of basal lamina alterations such as loss of usual texture or redundance/multilayering. It is worthwhile to note that endothelium is very responsive to mechanical deformations.

Interestingly, distinct morphological effects, such as microvillous generation, budding, cytoplasm vacuolization, tight junction loss or damage, and mitochondrium degeneration, have been reported to be induced by shear stress8 or pressure change.9–11 Although these endothelium damages have been reported for anatomical districts other than the anterior chamber angle, their absence in the samples having ab interno trabeculectomy before enucleation can indicate that a remarkable cell sufferance was not elicited by the surgical procedure.

The occasional finding of lymphocytes and macrophages suggests that no or only moderate leucocyte reaction may have occurred during the period between the ab interno trabeculectomy and enucleation, even if immune reactivity might have been influenced by the coexisting choroidal melanoma.12

The observation of discontinuous covering of trabecular stumps by undamaged endothelium might be due to migration and/or flattening of these cells. Although the histological effects of ab interno trabeculectomy were different from those caused by other laser or surgical trabecular procedures,13–15 that undesirable connective tissue overgrowth may occur later cannot be excluded. All these procedures remove little tissue, leaving close wound edges that might promote filling and scarring processes over time, with consequent closure of the trabecular opening.15 However, 3 years of experience in controlled IOP in eyes that had ab interno trabeculectomy2 suggests this procedure can be considered efficacious, at least in the medium term.

Our data confirm that the procedure can remove trabecular tissue selectively. Although the time between the first ab interno trabeculectomy procedure and enucleation was too short to allow us to state conclusive considerations, it is remarkable that the trabecular tissue appeared essentially calm. However, the absence of an evident inflammatory reaction in this case should be considered with caution because of possible tumor-induced immune suppression.


1. Ferrari E, Bandello F, Ortolani F, et al. Ab-interno trabeculo-canalectomy: surgical approach and histological examination. Eur J Ophthalmol. 2002;12:401-405.
2. Ferrari E, Bandello F, Roman-Pognuz D, Menchini F. Combined clear cornea phacoemulsification and ab interno trabeculectomy: three-year case series. J Cataract Refract Surg. 2005;31:1783-1788.
3. Nesterov AP, Batmanov YE. Study on morphology and function of the drainage area of the eye of man. Acta Ophthalmol (Copenh). 1972;50:337-350.
4. Nesterov AP, Batmanov YE. Trabecular wall of Schlemm's canal in the early stage of primary open-angle glaucoma. Am J Ophthalmol. 1974;78:639-647.
5. Lütjen-Drecoll E, Shimizu T, Rohrbach M, Rohen JW. Quantitative analysis of ‘plaque material’ in the inner and outer wall of Schlemm's canal in normal and glaucomatous eyes. Exp Eye Res. 1986;42:443-455.
6. Mermoud A, Ravinet E. Letter to the editor. J Glaucoma. 2001;10:365-367.
7. Francis BA, See RF, Rao NA, et al. Ab interno trabeculectomy: development of a novel device (Trabectome™) and surgery for open-angle glaucoma. J Glaucoma. 2006;15:68-73.
8. Masuda H, Kikuchi Y, Nemoto T, et al. Ultrastructural changes in the endothelial surface of the canine carotid artery induced by wall shear stress load. Biorheology. 1982;19:197-205.
9. Sageshima M, Kawamura K, Toda K, et al. An ultrastructural study of pulmonary capillary vessels in blood volume-overloaded rat. Adv Exp Med Biol. 1990;277:673-680.
10. von Mecklenburg C, Chang J-Y, Delgado T, et al. Ultrastructural cerebrovascular changes in a model of subarachnoid hemorrhage in baboon based on triple cisternal blood injection. Surg Neurol. 1990;33:195-201.
11. Milne AA, Bevan BH, Murphy WG, et al. Endothelial cell ultrastructure after aortic aneurysm rupture: an electron microscopy study. Br J Surg. 1995;82:635-637.
12. Gajewski TF, Meng Y, Harlin H. Immune suppression in the tumor microenvironment. J Immunother. 2006;29:233-240.
13. Melamed S, Pei J, Puliafito CA, Epstein DL. Q-switched neodymium:YAG laser trabeculopuncture in monkeys. Arch Ophthalmol. 1985;103:129-133.
14. van der Zypen E, Fankhauser F. The ultrastructural features of laser trabeculopuncture and cyclodialysis: problems related to successful treatment of chronic simple glaucoma. Ophthalmologica. 1979;179:189-200.
15. Ito S, Nishikawa M, Tokura T, et al., 1994. [Histopathological study of trabecular meshwork after trabeculotomy in monkeys] [Japanese], Nippon Ganka Gakkai Zasshi, 98, 811-819.
© 2007 by Lippincott Williams & Wilkins, Inc.