Chapter 239 – Nonpenetrating Glaucoma Surgery
• Nonpenetrating glaucoma surgery (NPGS) refers to drainage procedures that restore aqueous humor filtration through a natural membrane, namely the trabecular meshwork. NPGS includes all the different surgical techniques previously named: trabeculectomy ab-externo, nonpenetrating deep sclerotomy, and viscocanalostomy.
• Schlemm’s canal.
• Trabecular meshwork.
• Descemet’s membrane.
The quest for a low-risk and effective glaucoma operation has prompted a growing interest in the nonpenetrating techniques because of their lower complication rate. During nonpenetrating glaucoma surgery (NPGS), the intraocular pressure (IOP) is gradually lowered ( Fig. 239-1 ), flat anterior chambers practically do not occur postoperatively, and complicated blebs are less common. A less obvious but important reason for developing NPGS is the fact that during the procedure, the surgeon unveils the site of pathology, namely the trabecular meshwork (TM). Whereas gonioscopy permits examination of the angle and the anterior part of the TM, NPGS allows examination of its posterior part, where the main resistance to outflow exists. Therefore, NPGS is not just another drainage operation because it also offers the opportunity of evaluating the TM functionality.
When Schlemm’s canal is unroofed and the TM is exposed, aqueous humor can be seen percolating in various amounts according to the severity of the disease. The experienced NPGS surgeon can recognize different types of TMs and tailor, accordingly, the next surgical steps to be taken to improve filtration. At present, peeling and thinning out the TM manually can improve filtration. In the future, drugs or lasers could be developed to restore normal filtration by direct topical application on the TM. At present, glaucoma surgery is deferred until medications and laser interventions fail to lower the IOP to acceptable levels. This current teaching is questioned because of the accumulating evidence that topical antiglaucoma medications have adverse effects on the success of glaucoma surgery. A certain percentage of glaucoma patients need surgery, sooner or later, to reach their safe target IOP. These patients could benefit from early surgery rather than wait for medications and laser failure. Early surgery in the form of NPGS in eyes with an unaltered conjunctiva will be beneficial for these patients who cannot reach their safe target IOP with medications or laser alone. The evolving NPGS has gradually approached this goal during the last decade with the advent of refinements in surgical techniques, improvements in surgical instruments, and modern surgical microscopes.
Figure 239-1 Comparison of intraocular pressure recordings during nonpenetrating and penetrating glaucoma surgeries. Intraoperative intraocular pressure recording during nonpenetrating glaucoma surgery (top graph) and during trabeculectomy (bottom graph). (Courtesy of Andre Mermoud, MD.)
The evolution of NPGS started relatively slowly with the original works of Epstein and Krasnov in the late 1950s and early 1960s. Epstein noticed aqueous humor oozing from the paralimbal sclera when he dissected deeply seated pterygiae and deducted that paralimbal deep sclerectomy could be performed, intentionally, to lower IOP in glaucoma patients. Epstein then described an operation that consisted of a paralimbal deep sclerectomy overlying Schlemm’s canal, over 180°, without
Figure 239-2 Early non-penetrating glaucoma surgeries. A, “Paralimbal deep sclerectomy” as described by Edward Epstein in 1959. B, “Sinusotomy” as described by Krasnov in the 1960s.
entering the anterior chamber ( Fig. 239-2 ). The deep sclerectomy was then covered with conjunctiva. Epstein performed this operation in South African black patients with severe advanced glaucoma. Krasnov described a similar operation during the same period and called it “sinusotomy” (see Fig. 239-2 ). Krasnov theorized that the resistance to outflow in glaucoma patients could be either upstream or downstream of Schlemm’s canal. Krasnov assumed that the sinusotomy operation would be efficient when the resistance to outflow is situated downstream of Schlemm’s canal.
Both authors suggested unroofing Schlemm’s canal as a means of reducing IOP. The longevity of effective filtration with these early methods was relatively short. The conjunctiva scarred over the bare TM, blocking effective filtration within a few months. High-quality surgical microscopes were not available yet and few surgeons could perform these filtration operations. Moreover, the classical trabeculectomy was introduced almost concurrently, or soon after, by Sugar and Cairns. The relative ease of performing a trabeculectomy and its efficacy overshadowed and held back the development of NPGS. In the early 1980s, the Russian school led by Fyodorov  and Koslov and the North American school led by Zimmerman made a comeback for NPGS and performed it under a scleral flap following the trabeculectomy model. Fyodorov and Koslov and their colleagues proposed creating a portion of deep sclerectomy adjacent to Schlemm’s canal under the superficial scleral flap. The deep sclerectomy was supposed to enhance intrascleral and uveal aqueous humor absorption. The scleral flap added some protection to the bare TM and somewhat improved the results of NPGS. At present, NPGS is still an evolving surgical technique, which has evoked growing interest in the last decade because of its low rate of complications.                      Proponents and opponents argue about the efficacy and longevity of NPGS versus the classical trabeculectomy. The opponents of NPGS claim that classical trabeculectomy yields lower IOP and has greater longevity.    The proponents of NPGS still prefer it to the classical trabeculectomy because of its superior safety profile.                        Because NPGS requires a long learning curve and more dexterity than the classical trabeculectomy, its proponents claim that the NPGS opponents have not yet mastered the technique. The long learning curve of NPGS explains the differences in the reported comparisons in the two different techniques.
INDICATIONS AND CONTRAINDICATIONS FOR NPGS
In general, the indications for NPGS are wider and more inclusive than those for classical trabeculectomies for two reasons: NPGS is safer but not less efficient than trabeculectomies,                      and NPGS is indicated in certain types of glaucoma in which trabeculectomies normally fail or are not possible.
Until the advent of NPGS, penetrating glaucoma surgery was generally regarded as the last resort in the treatment of glaucoma. When medical therapy and laser failed to lower IOP to an acceptable level, glaucomatologists explained to their patients that an operation was necessary to halt the progression of the disease. NPGS, with its lower complication rate, can be offered earlier in the course of the disease. In fact, NPGS can be offered as a first-line treatment in cases in which it is obvious that medical treatment will not lower IOP to acceptable levels. This factor is particularly important in glaucoma patients younger than 50 years of age who have a longer life span. Several decades of continuous medical treatment are not sustainable because of the numerous side effects of antiglaucoma drugs. Furthermore, glaucoma surgery in general and NPGS in particular are more successful in glaucoma patients who were not exposed to medical treatment.         The noxious effects of topical medications on the conjunctiva are well documented.        The conjunctival tissues undergo scarring processes when exposed to certain topical medications. Such histologically altered conjunctiva is less amenable to the formation of a healthy diffuse bleb than a “virgin” conjunctiva. It is possible that even the TM undergoes biochemical-structural changes after years of medical treatment, rendering it less responsive to NPGS. It is therefore logical to propose NPGS earlier rather than later when the chances of favourable outcomes are greater. 
Open-angle glaucoma is the most common type of glaucoma and NPGS targets the presumed site of pathology, namely the TM. During NPGS, the TM is exposed and examined by the surgeon, who can assess on site the amount of filtration in vivo. With experience, different types of TMs can be recognized and classified according to their appearances and filtration capacities. NPGS has the advantage of being less cataractogenic than trabeculectomy and eventually may replace it in phakic open-angle glaucoma patients.                     
GLAUCOMA PATIENTS WITH HIGH MYOPIA.
Trabeculectomies in patients with high myopia carry an especially high risk of complications because of their abnormal globe dimensions. Choroidal detachments and consequent shallow anterior chambers occur in 10–15% of trabeculectomies done in high-myopic glaucoma patients. The factors predisposing myopic eyes to choroidal effusion may be related to the larger intraocular volume of myopic eyes, to the thinner sclera, and to vulnerable choroidal blood vessels. NPGS appears to offer glaucoma patients with high myopia a safer outcome because of the gradual intraoperative IOP reduction (see Fig. 239-1 ).
NPGS may be the treatment of choice for pigmentary glaucoma because the condition is very resistant to medical treatment. Pigmentary glaucoma occurs more frequently in young myopic male adults, and it is better to offer a safe surgical solution without depending on complex combination medical treatment. NPGS targets the site of pathology, namely the pigment-loaded TM, which can be reconditioned to reestablish filtration.
Pseudoexfoliation glaucoma is a form of open-angle glaucoma in which there is accumulation of exfoliation material along the aqueous outflow pathways. Because the exfoliation material is found in abundance in the TM and Schlemm’s canal, NPGS may be the treatment of choice for this condition. Opening Schlemm’s canal in the pseudoexfoliation patient is spectacular. This material can be peeled away from the exposed TM to reestablish filtration. IOP drops to acceptable levels for several years, and when exfoliation
material accumulates again, the site of filtration can be revised to restore filtration. NPGS can be done alone or in conjunction with cataract extraction according to the patient’s age, cataract status, and refractive error.
APHAKIC AND PSEUDOPHAKIC GLAUCOMA.
In aphakic glaucoma, iridectomy is not desirable because the vitreous moves forward through the iridectomy and blocks the filtration site. Extensive basal vitrectomy is needed to prevent blockage, but it is difficult to accomplish. Traction retinal detachment is not an uncommon complication in these combined vitrectomy-trabeculectomies. NPGS does not require iridectomy; therefore it may be particularly indicated in aphakic glaucoma. The only drawback of NPGS in aphakic glaucoma is the status of the TM. When aphakia has been long-standing, the TM is often collapsed and scarred; restoration of its function depends on its status and on the surgeon’s experience and skill. In many instances, the conjunctiva and the limbus are severely scarred by previous surgeries. The appropriate site for NPGS in aphakic glaucoma should be free of previous surgical scars.
CONGENITAL AND JUVENILE GLAUCOMA.
Congenital and juvenile glaucoma is severe and results in rapid optic nerve damage and loss of vision. Surgery is practically the only treatment available for these patients and, whenever possible, NPGS may be attempted first because of its low complication rate. The degree of success of NPGS is a function of the angle structure malformation and the surgeon’s experience. NPGS will succeed more in cases in which the pathological anatomy of the angle is not extreme. When NPGS fails, it is always possible to revert to penetrating glaucoma surgery, particularly in cases where the anatomy is severely distorted.
Sturge-Weber syndrome, a cutaneous hemangiomatous disorder, is often associated with congenital or developmental glaucoma. The greater numbers and tortuosity of the conjunctival blood vessels can be an indicator of glaucoma. Choroidal effusions following fistulizing surgery are notoriously known in these patients, and NPGS offers a safer alternative.
ANIRIDIA AND ANTERIOR SEGMENT DYSGENESIS SYNDROMES.
The success of NPGS in aniridia and anterior segment dysgenesis cases is dependent on the degree of anatomical distortion in Schlemm’s canal and TM. Schlemm’s canal rudiments are often seen in these cases. In aniridia and anterior segment dysgenesis syndromes the trabeculum is abnormal and it is possible to predict intraoperatively which cases will respond to NPGS according to the amount of filtration observed during the operation.
GLAUCOMA SECONDARY TO UVEITIS.
When elevated IOP persists after uveitis has been under control, glaucoma surgery is indicated. NPGS is indicated in these cases because it explores the site of resistance to aqueous outflow. During the inflammatory phases, the TM ultrastructures undergo changes that interfere with their normal function. These changes are mostly temporary, but when they are permanent glaucoma results. The TM can be reconditioned to improve filtration. Nevertheless, in cases in which multiple peripheral anterior synechiae have occurred, NPGS may not offer an efficient solution.
Laser iridotomy or surgical iridectomy is mostly a temporary measure in narrow-angle glaucoma. Cataract extraction or removal of the crystalline lens deepens the anterior chamber and opens the angle of the eye. When narrow-angle glaucoma has persisted for a certain length of time, glaucoma surgery is indicated in combination with lens extraction. For these combined operations, NPGS may be attempted, even though the iris root is very close to the TM and effective filtration may not occur immediately.
STATUS AFTER LASER TRABECULOPLASTY.
In eyes previously treated by laser trabeculoplasty, the TM may not be intact and might rupture during surgery. NPGS can be then converted to classical trabeculectomy.
AFTER TRAUMA ANGLE-RECESSION GLAUCOMA.
In traumatic angle-recession glaucoma, the TM loses its functionality because of scarification processes. NPGS can be attempted, however, because damage to the TM is not always complete and its functionality might be restored by scraping and peeling its posterior surface.
Absolute Contraindications: Neovascular Glaucoma
NPGS will fail in neovascular glaucoma because new blood vessels invade the iridocorneal angle. The TM loses its filtering function because of the neovascularization. This type of glaucoma is the most difficult to treat, and until now only implantation of silicone tube valves has yielded favorable results. 
MECHANISMS OF FILTRATION IN NONPENETRATING GLAUCOMA SURGERY
The main goal in NPGS is to restore aqueous drainage through an existing natural membrane, namely the trabeculodescemetic membrane (TDM). The TDM consists of the TM and a portion of the adjacent peripheral Descemet’s membrane. The TDM acts as an outflow dampener. With the TDM in place, the intraoperative drop in IOP is gradual and the loss of anterior chamber is prevented. Once the aqueous has filtered through the TDM, it is momentarily collected in a reservoir created by the deep sclerectomy before it is absorbed by several routes. The gradual filtration of the aqueous humor through the TDM and its collection in the deep sclerectomy space reduce the importance of the subconjunctival route for aqueous reabsorption. Large subconjunctival filtering blebs, with their potential risks and complications, are unnecessary in NPGS.
The main site of aqueous outflow resistance in open-angle glaucoma, pseudoexfoliative glaucoma, and pigmentary glaucoma is located at the juxtacanalicular TM and the inner wall of Schlemm’s canal. The aim of NPGS is to expose the TDM and remove the internal wall of Schlemm’s canal and the juxtacanalicular meshwork by scraping and peeling in order to reduce the main outflow resistance.
Aqueous Outflow through the Trabeculodescemetic Membrane
Vaudaux et al. studied the aqueous outflow through the TDM in an experimental model. Experiments were performed on enucleated human eyes unsuitable for keratoplasty. The intraoperative decrease in IOP was recorded (see Fig. 239-1 ) and the TDM resistance was calculated. The mean rate of IOP decrease was 2.7 ± 0.6?mmHg/min. The ocular aqueous outflow resistance dropped from a mean of 5.34 ± 0.19?mmHg µl per minute preoperatively to a mean of 0.41 ± 0.16?mmHg µl per minute postoperatively.
In these experiments, the postoperative TDM resistance was found to be low enough to ensure a low postoperative IOP and yet sufficient to prevent the collapse of the anterior chamber with its ensuing complications.
In the same study, the surgical site was examined histologically using ocular perfusion with ferritin. It was found that the anterior TM was the most porous site and most of the outflow occurred through it. To a lesser degree, there was some outflow through the posterior TM and Descemet’s membrane.
Aqueous Humor Reabsorption
After aqueous humor filtration through the TDM, four sites for aqueous reabsorption may be postulated: the subconjunctival
space, the intrascleral space, the suprachoroidal space, and the episcleral aqueous veins via the open ends of Schlemm’s canal.
The Subconjunctival Space
Most patients have a shallow diffuse subconjunctival bleb on the first day following NPGS. Ultrasound biomicroscopic (UBM) studies have demonstrated that successful cases of NPGS show a low profile and diffuse subconjunctival filtering blebs even years after surgery ( Fig. 239-3 ). However, these blebs tend to be shallower and more diffuse than those seen after trabeculectomy.
The Intrascleral Space and Its Aqueous Veins
During NPGS, after the superficial flap is lifted, a certain volume of sclera ranging between 5 and 8?mm3 is removed, creating an intrascleral space. Unless the superficial scleral flap adheres to the deep sclerectomy, this intrascleral space may act as an intrascleral filtering bleb. Absorbable implants made of porcine collagen or reticulated hyaluronic acid and nonabsorbable hydrophilic implants have been used in order to maintain the volume of this intrascleral space. On UBM studies, the mean volume of the intrascleral bleb was 1.8?mm3 in patients with the collagen implant. Aqueous vein openings are almost invariably observed in the deep sclerectomy bed (see Fig. 239-10 ). In successful cases of NPGS, these aqueous veins were shown to become hypertrophic in fluorescein angiography and UBM studies.  Existing and newly formed aqueous veins in the deep sclerectomy space might provide one of the routes for aqueous humor reabsorption.
The Subchoroidal Space
The deep scleral flap reaches a depth of 90%, leaving only a thin layer of scleral tissue over the ciliary body and the choroid. Aqueous humor can most probably seep into the suprachoroidal space through the thinned sclera and be absorbed by the uveal route. UBM studies have demonstrated the existence of localized shallow detachment of the ciliary body and choroid under the deep sclerectomy space in 45% of the patients studied years after the deep sclerectomy. It is not yet known how much of the aqueous is reabsorbed by this route, and further studies of the aqueous dynamics following NPGS are needed to elucidate it.
Figure 239-3 Photographic (A) and ultrasonic biomicroscopy (B) images of subconjunctival filtering bleb 5 years after NPGS with collagen implant. Ultrasonic biomicroscopy image of filtering blebs with the T Flux implant 3 months (C) and 2 years (D) after NPGS.
The Episcleral Veins via the Open Ends of Schlemm’s Canal
During NPGS, the open ends of Schlemm’s canal could theoretically provide a route to the episcleral veins that exist on either side of the deroofed Schlemm’s canal. In Stegmann’s viscocanalostomy, viscoelastic material is injected into the open ends of Schlemm’s canal in order to dilate them and to increase aqueous humor reabsorption by this route. The amount of aqueous reabsorption by this route has not yet been quantified in vivo. 
NPGS is a very challenging technique because the surgical acts are confined to a few square millimeters and extremely delicate structures. NPGS is best performed under high magnification, and it involves a long learning curve. Even in experienced hands, NPGS takes at least twice as long to perform as a classical trabeculectomy. It is therefore advisable to view several of these operations performed by an experienced surgeon and practice on animal or cadaver eyes before attempting the first procedure in a patient.
Because the critical steps of the operation (unroofing of Schlemm’s canal and peeling of the TM) are performed under the highest magnification, the patient has to reach maximal stability. Effective local anesthesia that ensures adequate akinesia, such as sub-Tenon or retrobulbar injections, is mandatory. In a beginner’s hands, general anesthesia might not be superfluous because it ensures complete immobility of the patient.
Conjunctival and Superficial Scleral Flaps
The first steps of NPGS are similar to those of classical trabeculectomy. A 7?mm fornix or limbal-based conjunctival flap, preferably in the upper quadrant, is created ( Fig. 239-4 ). A 5 × 5 × 1.5?mm trapezoidal or a 5 × 5?mm square scleral flap of 40 to 50% depth is dissected into clear cornea ( Figs. 239-5 and 239-6 ). It is extremely important to extend the first scleral step dissection into clear cornea, past the vascular arcade, in order to allow the creation of a wide TDM when the second scleral flap is
lifted. This first scleral flap is everted over the cornea and pulled down with an 8-0 virgin silk that is fixed to the lower limbus at the 6 o’clock position (see Fig. 239-5 ). This temporary flap fixation improves the exposure during the next phase of the operation, which is performed under the highest magnification.
Deep Sclerectomy and Exposure of the Trabeculodescemetic Membrane
A second 3 × 3 × 1?mm trapezoidal or a 3 × 3?mm square scleral flap is dissected to a depth of 90%, creating a deep sclerectomy and leaving only a thin layer of scleral tissue over the underlying uvea (see Fig. 239-6 ). At the level of the scleral spur, Schlemm’s canal is unroofed, creating a 3?mm long fenestration in its lumen. The scleral spur is recognized by its glistening circumferential fibers that can be differentiated from the crisscross pattern fibers situated in the deep sclerectomy bed. The scleral spur is the anatomical landmark where Schlemm’s canal starts, and it is located at the transition between the white sclera and the blue sclera in the limbus area. When the posterior aspect of the TM and the adjacent Descemet’s membrane are exposed (see Fig. 239-6 ), it is advisable to lower the IOP in order to reduce the risk of rupture of the TDM. Stegmann and Mermoud suggested performing a paracentesis, releasing some aqueous from the anterior chamber, in order to reduce the IOP prior to the complete exposure of the TDM. Nonetheless, the concept of nonpenetration is not violated by the paracentesis because NPGS pertains to the site of filtration and not to the penetration or nonpenetration of the anterior chamber. Dahan used a 25-gauge anterior chamber maintainer (ACM) connected to a bottle of balanced salt solution 20?cm above the patient’s head. The ACM allows the surgeon to control the IOP during the different steps of the operation ( Fig. 239-7 ). The ACM is switched off during the opening of Schlemm’s canal in order to lower the IOP and prevent inadvertent rupture of the TDM. A dry cellulose sponge is used to assess the amount of aqueous oozing from the TDM (see Fig. 239-7 ).
To thin out and render the TM more permeable, trabecular forceps (HUCO 4.4475) are used to peel off Schlemm’s canal endothelium and the juxtacanalicular TM ( Fig. 239-8 ). In some cases, these structures can be first loosened using a TM scraper with a carbide-impregnated metal tip (Katena K3-1120 or HUCO 4.6030) ( Fig. 239-9 ). When these delicate steps are completed,
Figure 239-4 A 7?mm fornix-based conjunctival flap and trapezoidal marking for the superficial flap.
Figure 239-5 A 40–50% depth superficial flap is dissected into clear cornea and reflected back to allow a 90% depth deep sclerectomy.
the ACM can be switched on again gradually in order to assess the amount of aqueous filtration after the reconditioning of the TM.
The internal scleral flap is excised along its base 0.5?mm anterior to Schwalbe’s line to create the deep sclerectomy space ( Fig. 239-10 ). In high-risk cases, mitomycin C can be applied onto the deep sclerectomy only to prevent intrascleral scarring and enhance intrascleral aqueous reabsorption. It is advisable to prevent contact between the mitomycin C and the conjunctiva in order to prevent avascular blebs postoperatively. During the mitomycin C application, it is advisable to keep the ACM switched on to create positive IOP and to prevent mitomycin C seeping into the anterior chamber and suprachoroidal space.
Intrascleral Hydrophilic Implants
Different hydrophilic implants can be used to fill the deep sclerectomy space to create an intrascleral bleb. The implants can be absorbable (Aquaflow made of porcine collagen or SKGel made of reticulated hyaluronic acid) or nonabsorbable (T Flux made of 38% water content Poly-Megma) ( Fig. 239-11 ). The superficial scleral flap is reflected back in place and sutured with at least one suture. For a fornix-based approach the conjunctival flap is sutured back into place with one or two sutures at the limbus, whereas for a limbal-based conjunctival flap the conjunctiva is sutured with a continuous suture.
Postoperative treatment consists of a topical dexamethasone, neomycin, polymyxin B sulfates (Maxitrol), or dexamethasone and chloramphenicol (SpersadexCo) instilled four times a day for 2 weeks or until IOP =10?mmHg. The topical steroids are then replaced with a nonsteroidal anti-inflammatory agent such as diclofenac (Naclof) and continued for 4 to 8 weeks.
Stegmann has described a variant of NPGS and termed it viscocanalostomy to emphasize the importance of injecting viscoelastic material into Schlemm’s canal as a means of improving aqueous drainage by this route ( Fig. 239-12 ). However, in viscocanalostomy, peeling of the Schlemm’s canal endothelium and
Figure 239-6 90% depth deep sclerectomy with the unveiled trabeculodescemetic membrane. A, En face view and, B, sagittal view showing (1) scleral spur, (2) trabecular meshwork, and (3) Descemet’s membrane.
the juxtacanalicular TM has not been suggested as a means of improving filtration. Stegmann claimed that the denuded peripheral Descemet’s membrane is the main route of aqueous filtration, bypassing the resistance situated at the level of the TM. In viscocanalostomy, it is suggested to suture the superficial scleral flap tightly in order to prevent subconjunctival filtration blebs and to direct the aqueous from the deep sclerectomy space into the enlarged openings of Schlemm’s canal. These assertions have yet to be proved by further postoperative aqueous dynamics studies.
COMBINED CATARACT EXTRACTION AND NONPENETRATING GLAUCOMA SURGERY
NPGS can be combined with cataract extraction when the two conditions coexist and when the anterior chamber needs to be deepened in cases of narrow-angle glaucoma. For example, it is possible to perform a phacoemulsification via a temporal clear corneal approach and an NPGS at the 12 o’clock position. It is advisable to perform the cataract extraction first and then to proceed with the NPGS in order to prevent rupture of the TDM during the phacoemulsification. Combined NPGS and cataract extraction yields results comparable to those of trabeculectomy and cataract extraction but involves fewer complications.
Figure 239-7 Intraoperative video images of non-penetrating glaucoma surgery. A, A dry cellulose sponge is used to disinsert Descemet’s membrane and to assess filtration. B, A 25-gauge anterior chamber maintainer provides IOP control throughout the operation.
NPGS AS AN ADJUNCT TO GLAUCOMA IMPLANT SURGERY
In cases in which silicone tube implants are indicated, Freedman suggested associating NPGS with a Seton implant in order to prevent the early postoperative hypotony that occurs with these implants. NPGS can provide safe initial lowering of IOP, and when the IOP rises again the preplaced silicone tube can be activated to provide further IOP control. The activation of the Seton is performed by removing a preplaced stent from the silicone tube when sufficient scarring has occurred in the surrounding tissues.
COMPLICATIONS OF NONPENETRATING GLAUCOMA SURGERY
Although there is unanimous agreement that NPGS involves fewer complications than conventional penetrating glaucoma surgery, it is not totally devoid of complications. NPGS complications can occur intraoperatively, during the early postoperative period, or during the late postoperative period.
Figure 239-8 The Schlemm’s canal endothelium and the juxtacanalicular meshwork are peeled in order to improve filtration.
Figure 239-9 The trabecular meshwork is scraped with a carbide-incrusted metal tip (Trabecular Meshwork Scraper).
During the dissection of the deep scleral flap, the choroid can be inadvertently exposed in a small area, causing a small choroid herniation. These small choroid herniations can be ignored as long as the deep scleral flap dissection is redirected into the correct plane.
Figure 239-10 The deep scleral flap is cut off to create the deep sclerectomy space. Note the aqueous vein situated in the center of the deep sclerectomy. Aqueous veins in the deep sclerectomy space provide one of the several routes for aqueous reabsorption after nonpenetrating glaucoma surgery.
During the learning curve, the most common intraoperative complication is perforation of the TDM. It is acceptable to have a perforation rate of 30% during the first 10–20 cases. With proper training, the perforation rate can be lowered to less than 5%. When a perforation of the TDM occurs, the surgeon has to decide whether the NPGS has to be halted and converted into a conventional trabeculectomy or whether the perforation can be ignored. The key features in these instances are the size of the perforation and the presence of iris prolapse. Small holes in the TDM that do not cause iris prolapse can be ignored, whereas larger holes or tears with iris prolapse must lead to halting the NPGS and convert it to a trabeculectomy. A glaucoma implant such the T Flux can be used to tamponade medium-size holes with no iris prolapse.
Moderate hypotony with a deep anterior chamber is expected during the first week postoperatively after NPGS. In fact, a normal IOP on the first day postoperatively is not a good prognostic indicator because most of these patients develop high IOP later on. The transient hypotony subsides within a week or two with topical steroid medications. Patients must be warned that their visual acuity will be affected by the transient hypotony and the consequent transient astigmatism. Transient cystoid macular edema related to the hypotony can develop, but it normally subsides when IOP returns to normal levels.
High IOP on the first day postoperatively can be caused by insufficient dissection of the TDM. It occurs often in inexperienced hands, especially when the TDM is not wide enough and its
Figure 239-11 Absorbable and non-absorbable hydrophilic implants used in the deep sclerotomy space to increase the longevity of NPGS. A, Porcine collagen (Aquaflow) sutured in the deep sclerectomy space. B, Reticulated hyaluronic acid implant in the deep sclerectomy space. C, Nonabsorbable hydrophilic implant (T Flux) sutured in the deep sclerectomy space. D, Three-dimensional view of the T Flux nonabsorbable hydrophilic implant made of 38% water content Poly-Megma.
Figure 239-12 Viscoelastic material is injected into the Schlemm’s canal openings with a specially designed cannula (Grieshaber).
filtration capacity has not been sufficiently restored by scraping and peeling. This complication can be remedied by neodymium:yttrium-aluminum-garnet (Nd:YAG) laser goniopuncture. Presence of blood in the deep sclerectomy space can also hamper aqueous evacuation and cause early postoperative high IOP. When it is recognized, a short course of acetazolamide can solved this transient rise in IOP.
The TDM can rupture postoperatively at any stage if the patient rubs the eye vigorously or after a Valsalva maneuver. Following the TDM rupture, iris prolapse occurs and blocks the filtration site, causing a rise in IOP. In these cases, the filtration site has to be revised in order to convert the operation into a conventional trabeculectomy.
Peripheral anterior synechiae can form in the site of filtration without iris prolapse. These synechiae can hamper filtration and cause a rise in IOP. YAG laser iridoplasty can reopen the angle and be beneficial in some cases. In case of failure, NPGS can be repeated on another site.
Descemet’s membrane detachment is a rare complication of NPGS, occurring in about 1 out of 250–300 operated eyes. Descemet detachment is more inherent to viscocanalostomy because of the viscoelastic material injection into Schlemm’s canal. The inexperienced surgeon can be too vigorous when injecting the viscoelastic material and cause a Descemet detachment. Descemet detachment can also occur when aqueous humor from the deep sclerectomy space is forced into the sub-Descemet space following vigorous ocular massage or trauma. The patient can experience a drop in visual acuity due to corneal edema. Generally, Descemet detachments are transient and visual acuity returns to normal. In severe cases, Descemet reattachment can be attempted with the use of viscoelastic material in the anterior chamber.
A single case of scleral ectasia following NPGS has been reported in a 12-year-old girl with chronic arthritis complicated with glaucoma secondary to a chronic uveitis. Scleral ectasia can occur in any type of glaucoma surgery in patients with a tendency to scleromalacia. Antimetabolites should be used with extreme caution in these cases.
Because NPGS involves a long learning curve, the reports on its efficacy and longevity are not unanimous. In general, the NPGS pioneers with extensive experience and long series report on more favorable results than the novices.                           Reports that compare trabeculectomy with viscocanalostomy show particular discord between the proponents of NPGS and its opponents. There is nevertheless a common trend that indicates that the NPGS efficacy is initially good but its longevity is limited, as expected in any kind of glaucoma surgery.
Dahan and Drusedau analyzed a series of 86 eyes operated by NPGS over a mean follow-up of 46 months. Their analysis was particularly significant because no implants were used and no antiglaucoma treatment was added postoperatively in their study. The IOP dropped in average by 50%, from a mean preoperative value of 30.4?mmHg to a mean postoperative value of 15.35?mmHg without medication. Postoperatively, when IOP rose above 20?mmHg, instead of adding medical treatment, the filtration site was revised to reestablish filtration. During the study period, the filtration site was revised in 48 eyes (56%), in order to maintain the IOP below 21?mmHg without medication, after a mean period of successful filtration of 29.9 months. These data indicate that the efficacy of NPGS starts to fade in the third year postoperatively. Their series confirmed previous reports on the adverse effects of antiglaucoma medication on glaucoma surgery. The reoperation rate was 4.7 times higher in previously treated patients than in untreated patients.
Koslov et al. proposed a porcine collagen implant to keep a filtration space under the superficial scleral flap. Sourdille et al. developed an absorbable implant made of reticulated hyaluronic acid. This absorbable implant is left under the superficial scleral flap in order to create a space for aqueous humor reabsorption. Until recently, all the proposed implants were absorbable and showed certain value in improving the outcomes of NPGS.                          The glaucoma unit of the Jules Gonin Eye Hospital in Lausanne reported a complete success rate of 34.6% without a collagen implant versus a success rate of 63.4% with a collagen implant. Similarly, other reports confirm the value of other absorbable and nonabsorbable implants in NPGS.
NPGS can be combined with cataract extraction, especially when there is evidence of a narrow angle. The patient will benefit from the lens removal because the anterior chamber invariably deepens after combined surgery. Several studies reported on the favorable outcomes of combined NPGS with cataract extraction.
CONCLUSIONS AND FUTURE PERSPECTIVES
NPGS is an evolving technique that is associated with fewer complications than classical trabeculectomy. The experienced NPGS surgeon finds it at least as effective as trabeculectomy, whereas the novice claims that trabeculectomy yields better results. Undoubtedly, NPGS still needs further improvements to prolong its efficacy and longevity. The available implants have proved to be beneficial in prolonging successful filtration. At present, the deep sclerectomy and the exposure of the TDM are done manually. These steps are challenging and require a long learning curve. Various lasers have been tried in performing the deep sclerectomy, but they have not yet matched dissection. Future attempts to use lasers in performing the deep sclerectomy will probably be more successful and will help in popularizing this promising technique.
1. Epstein E. Communications. Fibrosing response to aqueous. Its relation to glaucoma. Br J Ophthalmol. 1959;43:641–7.
2. Krasnov MM. Externalisation of Schlemm’s canal (sinusotomy) in glaucoma. Br J Ophthalmol. 1968;52:157–61.
3. Sugar HS. Experimental trabeculectomy in glaucoma. Am J Ophthalmol. 1961;51: 623–7.
4. Cairns JE. Trabeculectomy-preliminary report of a new method. Am J Ophthalmol. 1968;66:673–9.
5. Fyodorov SN, Ioffe DI, Ronkina TI. Deep sclerectomy: technique and mechanism of a new glaucomatous procedure. Glaucoma. 1984;6:281–3.
6. Koslov VI, Bagrov SN, Anisimova SY, et al. Nonpenetrating deep sclerectomy with collagen. Ophthalmic Surg. 1990;3:44–6.
7. Zimmerman TJ, Kooner KS, Ford VJ, et al. Trabeculectomy vs. non-penetrating trabeculectomy. A retrospective study of two procedures in phakic patients with glaucoma. Ophthalmic Surg. 1984;15:734–40.
8. Hara T, Hara T. Deep sclerectomy with Nd: YAG laser trabeculotomy ab interno: two stage procedure. Ophthalmic Surg. 1988;19:101–6.
9. Kershner RM. Non-penetrating trabeculectomy with placement of collagen device. J Cataract Refract Surg. 1995;21:608–11.
10. Demailly P, Jeanteur-Lunel MN, Berkani M, et al. La sclérectomie profonde non-perforante associée a la pose d’un implant de collagen dans le glaucome primitif a angle ouvert. Resultats retrospectifs a court terme. Ophthalmologie. 1995;9:666–70.
11. Chiou AG, Mermoud A, Underdahl JP, Schnyder CC. An ultrasound biomicroscopic study of eyes after deep sclerectomy with collagen implant. Ophthalmology. 1998;105:746–50.
12. Mermoud A, Karlen ME, Schnyder CC. Nd:YAG goniopunctures after deep sclerectomy with collagen implant. Ophthalmic Surg Lasers. 1999;30:120–25.
13. Mermoud A, Vaudaux J. Aqueous humor dynamics in non-penetrating filtering surgery (deep sclerectomy). Invest Ophthalmol Vis Sci. 1997;38(Suppl):S1064.
14. Sanchez E, Schnyder CC, Sickenberg M, et al. Deep sclerectomy: results with and without collagen implant. Int Ophthalmol. 1997;20:157–62.
15. Welsh NH, Delange J, Wassermann P, Ziemba L. The “deroofing” of Schlemm’s canal in patients with open-angle glaucoma through placement of a collagen device. Ophthalmic Surg Lasers. 1998;29:216–26.
16. Mermoud A, Chiou AGY, Jewelewicz DA. Post-operative inflammation following deep sclerectomy with collagen implant versus standard trabeculectomy. Graefes Arch Clin Exp Ophthalmol. 1998;236:593–6.
17. Ambresin A, Shaarawy T, Mermoud A. Deep sclerectomy with collagen implant in one eye compared with trabeculectomy in the other eye of the same patient. J Glaucoma. 2002;11:214–20.
18. Shaarawy T, Karlen ME, Sanchez E, et al. Long term results of deep sclerectomy with collagen implant. Acta Ophthalmol Scand. 2000;78:323–28.
19. Stegmann R, Piennar A, Miller D. Viscocanalostomy for open-angle glaucoma in black African patients. J Cataract Refract Surg. 1999;25:316–22.
20. Sourdille P, Santiago PY, Villain F, et al. Reticulated hyaluronic acid implant in non-perforating trabecular surgery. J Cataract Refract Surg. 1999;25:332–9.
21. Dahan E, Drusedau MUH. Nonpenetrating filtration surgery for glaucoma: control by surgery only. J Cataract Refract Surg. 2000;26:695–701.
22. Drusedau MUH, Von Wolff K-D, Bull H, Von Barsewisch B. Viscocanalostomy for primary open-angle glaucoma: the Gross Pankow experience. J Cataract Refract Surg. 2000;26:1367–73.
23. Hamel M, Shaarawy T, Mermoud A. Deep sclerectomy with collagen implant in patients with glaucoma and high myopia. J Cataract Refract Surg. 2001;27:1410–17.
24. Li M. [Nonperforating trabecular surgery with reticulated hyaluronic acid implant]. Chin J Ophthalmol. 2001;37:404–8.
25. Shaarawy T, Karlen M, Schnyder C, et al. Five-year results of deep sclerectomy with collagen implant. J Cataract Refract Surg. 2001;27:1770–8.
26. El Sayyad F, Helal M, El-Kholify H, et al. Nonpenetrating deep sclerectomy versus trabeculectomy in bilateral primary open-angle glaucoma. Ophthalmology. 2000;107:1671–4.
27. Chiselita D. Non-penetrating deep sclerectomy versus trabeculectomy in primary open-angle glaucoma surgery. Eye. 2001;15:197–201.
28. Tan JC, Hitchings RA. Non-penetrating glaucoma surgery: the state of play. Br J Ophthalmol. 2001;85:234–7.
29. Jonescu-Cuypers C, Jacobi PC, Konen W, Krieglstein GK. Primary viscocanalostomy versus trabeculectomy in white patients with open-angle glaucoma. A randomised clinical trial. Ophthalmology. 2001;108:254–8.
30. O’Brart DP, Rowlands E, Islam N, Noury AM. A randomised, prospective study comparing trabeculectomy augmented with antimetabolites with a viscocanalostomy technique for the management for open-angle glaucoma uncontrolled by medical therapy. Br J Ophthalmol. 2002;86:748–54.
31. Watson PG, Jakeman C, Ozturk M, et al. Complications of trabeculectomy (20 years follow-up). Eye. 1990;4:425–38.
32. Dahan E, Rivett K, Michiels X. Comparison of early postoperative complications in trabeculectomies alone versus trabeculectomies with cataract extraction. Eur J Implant Refract Surg. 1994;6:18–21.
33. Jay JL, Allan D. The benefit of early trabeculectomy vs. conventional management in primary open-angle glaucoma relative to severity of disease. Eye. 1989;3:528–35.
34. Migdal C, Gregory W, Hitchings RA. Long-term functional outcome after early surgery compared with laser and medicine in open-angle glaucoma. Ophthalmology. 1994;101:1651–7.
35. Lavin MJ, Worvald RFL, Migdal CS, Hitchings RA. The influence of prior therapy on the success of trabeculectomy. Arch Ophthalmol. 1990;108:1543–8.
36. Broadway DC, Grierson I, O’Brien C, Hitching RA. Adverse effects of topical antiglaucoma medications I—the conjunctival cell profile. Arch Ophthalmol. 1994;112:1437–45.
37. Broadway DC, Grierson I, O’Brien C, Hitching RA. Adverse effects of topical antiglaucoma medications II—The outcome of filtration surgery. Arch Ophthalmol. 1994;112:1446–54.
38. Baudouin C, Pisella PJ, Fillacier K, et al. Ocular surface inflammatory changes induced by topical antiglaucoma drugs: human and animal studies. Ophthalmology. 1999;106:556–63.
39. Broadway DC, Chang LP. Trabeculectomy, risk factors for failure and the preoperative state of the conjunctiva. J Glaucoma. 2001;10:237–49.
40. Ruderman JM, Harbin TS Jr, Campbell DG. Postoperative suprachoroidal hemorrhage following filtration procedures. Arch Ophthalmol. 1986;104:201–5.
41. Freedman J. Clinical experience with the Molteno dual chamber single-plate implant. Ophthalmic Surg. 1992;23:238–41.
42. Mermoud A, Salmon JF, Alexader P, et al. Molteno tube implantation for neovascular glaucoma. Ophthalmology. 1993;100:897–902.
43. Vaudaux J, Uffer S, Mermoud A. Aqueous dynamics after deep sclerectomy: in vitro study. Ophthalmic Pract. 1999;16:204–9.
44. Johnson DH, Johnson M. How does non-penetrating glaucoma surgery work? Aqueous outflow resistance and glaucoma surgery. J Glaucoma. 2001;10:55–67.
45. Gianoli F, Mermoud A. Cataract-glaucoma surgery: comparison between phacoemulsification combined with deep sclerectomy or trabeculectomy. Klin Monatsbl Augenheilkd. 1999;210:256–60.
46. Sherwood MD, Smith MF. Prevention of early hypotony associated with Molteno implants by a new occluding stent technique. Ophthalmology. 1993;100:85–90.
47. Ravinet E, Shaarawy T, Schnyder CC, et al. Descemet membrane detachment after nonpenetrating filtering surgery. J Glaucoma. 2002;11:244–52.
48. Milazzo S. Scleral ectasia as a complication of deep sclerectomy. J Cataract Refract Surg. 2000;26:785–7.