Second Glaucoma Drainage Devices in Refractory Pediatric Glaucoma: Failure by Fibrovascular Ingrowth




Purpose


To evaluate initial glaucoma drainage device failure and subsequent glaucoma drainage device placement for refractory pediatric glaucoma.


Design


Retrospective interventional case series.


Methods


The Duke University surgical records from a single surgeon from 1997 to 2012 were reviewed for patients having at least 2 glaucoma drainage devices for refractory pediatric glaucoma. Data collected included glaucoma diagnosis, age at surgery, surgical interventions, preoperative/postoperative IOP and medications, and complications. Failure was defined as having an IOP >21 mm Hg (or clinically inadequate), and/or IOP-reducing surgery/devastating complication.


Results


Forty-three eyes (37 patients) had 2 or more glaucoma drainage devices. Mean age at second glaucoma drainage device implantation was 9.2 ±7.1 years, with mean IOP 30 ±11 mm Hg, on 3 ± 1 IOP-lowering medications. Fibrovascular ingrowth was documented during second glaucoma drainage device surgery in 12 of 43 eyes (28%), occurring only in Ahmed devices, at a mean of 70 months (range 11–153) after initial implantation. Failure of the second glaucoma drainage device surgery occurred in 18 of 43 eyes (42%) at mean 26.1 ± 32.2 months (median 19.3 months). Surgical success of second glaucoma drainage devices (Kaplan-Meier analysis) at 1 year, 2 years, and 3 years was 81%, 62%, and 50%, respectively.


Conclusions


Single glaucoma drainage device surgery fails to control IOP in some eyes with refractory pediatric glaucoma. Second glaucoma drainage device implantation offers a treatment option with modest success over time. Fibrovascular ingrowth should be suspected as a relatively common cause of IOP elevation and failure after Ahmed device implantation.


Although some cases of pediatric glaucoma can be controlled with medical therapy and/or angle surgery, more refractory cases can often be successfully treated with glaucoma drainage devices. Glaucoma drainage devices for refractory pediatric glaucoma can fail to control intraocular pressure (IOP) by several mechanisms. In addition to encapsulation, fibrovascular ingrowth can limit outflow in valved Ahmed devices. In a subset of refractory glaucoma, one glaucoma drainage device may not adequately control IOP, and either a replacement glaucoma drainage device at the same site or a new glaucoma drainage device at another site may be warranted. This study evaluates initial glaucoma drainage device failure and subsequent glaucoma drainage device placement for refractory pediatric glaucoma.


Methods


The Duke University surgical records from a single surgeon from January 1, 1997 to August 1, 2012 were reviewed retrospectively for patients having at least 2 glaucoma drainage devices for refractory pediatric glaucoma. The minimum follow-up period after surgery was 6 months. The patient had to be younger than 18 years old at the time of the first glaucoma drainage device. The study was approved by the Duke University Health System Institutional Review Board. Data collected included glaucoma diagnosis, age at surgery, surgical interventions, preoperative/postoperative IOP and medications, and complications. Failure was defined as having an IOP >21 mm Hg (or clinically inadequate), requiring an IOP-reducing surgery, or devastating complication (eg, endophthalmitis, retinal detachment, phthisis).


Four patients who received a second glaucoma drainage device had hypotony following the first glaucoma drainage device. These patients were excluded from the study since the reason for the second glaucoma drainage device (downsizing a Baerveldt device to an Ahmed device) was hypotony rather than inadequate IOP control.


The implanted second glaucoma drainage devices were all placed by a single surgeon (S.F.F.). The decision to place an Ahmed (New World Medical Inc, Rancho Cucamonga, California, USA) or Baerveldt (Abbott Laboratories Inc, Abbott Park, Illinois, USA) glaucoma drainage device was a clinical decision made by the surgeon, as was the decision about whether to use the original site and remove the first device, rather than implanting a second device in a new location. The clinical algorithm that was implemented remained essentially stable over the time period of the study and was as follows: (1) If it was suspected that an Ahmed device was no longer functional (in other words, the tube tip was patent but there was a high IOP with low or no bleb on inspection/ultrasound in the clinic), then the second device was implanted in the same quadrant, after removal of the Ahmed device. If immediate IOP control was needed, an Ahmed device would be reimplanted; otherwise, and especially if there was not too much scarring, the preference was to replace an Ahmed by a Baerveldt device, either trimmed to the same size to fit in the “footprint” of the removed Ahmed or placed at the same site after the capsule of the prior Ahmed was at least partly removed to allow a “winged” reservoir to be placed. The Ahmed device was always removed, rather than simply “clearing” the fibrovascular tissue from the valve chamber. (2) If the original device was deemed to be encapsulated but still patent (in other words, it was a Baerveldt device with a patent tube tip, or it was an Ahmed device that still had fluid over the plate on ultrasound that was confirmed by bleb needling in the operating suite before surgery), then a second site was chosen for “additional” drainage, rather than replacing the encapsulated but patent initial drainage device.


Surgical Techniques


The glaucoma drainage devices were all placed with a similar technique in the chosen quadrant, with modifications when glaucoma drainage device replacement in the same location was being performed. After a lid speculum was placed in the eye, a traction suture was placed to expose the quadrant in question. A fornix incision through conjunctiva and Tenon capsule was made approximately 8 mm from the limbus in the area between the rectus muscles. In cases of repeat surgery, care was taken to preserve both conjunctival and Tenon layers for later double-layer closure. For initial glaucoma drainage devices, or if the tube location was to be altered with the replacement glaucoma drainage device, blunt dissection was then carried out to the limbus anteriorly, and posteriorly between the respective rectus muscles.


For replacement glaucoma drainage devices, the tubing was generally exposed behind the sclera patch graft and carefully unroofed, and then was completely ligated just in front of the plate, with a 6-0 Vicryl suture (Ethicon, Inc., Somerville, New Jersey, USA), which was also attached to nearby Tenon or scar tissue as an anchor. The tube was then cut separate from the plate to allow its removal without pulling the tubing from the eye unintentionally. For new glaucoma drainage devices, blunt dissection was then carried out between the 2 respective rectus muscles. For replacement glaucoma drainage devices, the plate was then exposed by penetrating the capsule, with care taken to inspect the valve chamber and to identify any tissue that might be entering the chamber from the capsule tissue. Most of the capsule was then removed anteriorly, transecting tissue that spanned the holes in the plates of the Baerveldt and Ahmed flexible plate reservoirs; the plates were then removed. Ahmed plates were further inspected to identify fibrovascular tissue within the valve chamber, although sometimes the tissue was clearly removed during the process of capsule penetration and plate removal described above. When a larger plate was intended to be placed, the lateral and posterior “walls” of the capsule were also carefully opened, and sometimes partially removed, with care taken to identify the adjacent rectus muscles and to avoid injuring them.


For the Baerveldt devices, the 2 rectus muscles in that quadrant were hooked. Next, the Baerveldt device was inspected and the tube was flushed to ensure patency. A 6-0 Vicryl suture was then placed across the tube near the plate in order to completely ligate the tube. The Baerveldt device was then placed into the chosen quadrant with the wings tucked under the 2 rectus muscles. For replacement Baerveldt devices, the plate was often trimmed so that the wings were blunted, making it easier to place the plate where the prior glaucoma drainage device reservoir had been located. An 8-0 nylon suture was used to secure the plate to the sclera. The tube was then trimmed to an appropriate length with a bevel upward for tubes placed in the anterior chamber and sulcus, and a bevel downward for tubes placed pars plana.


For initial Baerveldt tube placement, a 30 gauge needle was then inserted at the intended site of tube placement (anterior chamber, sulcus, or pars plana). Once the needle was noted to be in the correct position, the 30 gauge needle was removed and replaced by a 23 gauge needle to create a large enough opening for the tube to be placed with the tube introducer. After the tube was noted to be in the correct position, it was secured with a 9-0 nylon suture to the globe. Three venting slits were then placed in the tubing just anterior to the ligation suture, except in cases where early hypotony was to be avoided (eg, Sturge-Weber-associated glaucoma). For replacement Baerveldt devices, the tubing was advanced into the “sleeve” of the prior glaucoma drainage device tube, after removing the tubing of the prior device from its location in the anterior chamber, sulcus, or pars plana.


For the Ahmed devices, the procedure was identical to that described above, except that no tube ligation was carried out, and the insertion of the reservoir was carried out with care not to compress the valve chamber at any time. The tube was then placed in a manner identical to that for the Baerveldt devices, whether a primary or a replacement Ahmed device was being placed.


After placement of the glaucoma drainage device, donor sclera or cornea was the used to cover the site of tube entry into the globe. For replacement glaucoma drainage devices, donor tissue was sometimes not required as the initial patch material was still in place. The Tenon capsule and conjunctiva were then closed separately with running 8-0 Vicryl sutures on a vascular needle to ensure a watertight closure.


Results are reported as mean (standard deviation) unless otherwise noted. Statistical comparisons were performed using a 2-tailed t test for continuous variables, the χ 2 test for ratios, and Kaplan-Meier survival curve analysis for surgical success. P values are considered significant if ≤ .05.




Results


Forty-three eyes (37 patients) having 2 or more glaucoma drainage devices were identified. Glaucoma diagnoses included primary congenital glaucoma (20/43, 47%), glaucoma following cataract surgery (aphakic) (12/43, 28%), glaucoma with acquired condition (uveitis) (5/43, 12%), and other (6/43, 14%). At the time of the second glaucoma drainage device implantation, the average age was 9.2 ± 7.1 years. Mean IOP was 30 ± 11 mm Hg, and the eyes were on 3 ± 1 IOP-lowering medications. Failed first glaucoma drainage devices included Ahmed (n = 33), Baerveldt (n = 8), and Molteno (n = 2). Implanted second glaucoma drainage devices were Ahmed at the same site replacing an Ahmed (n = 8), Ahmed at a different site (n = 20), Baerveldt at the same site replacing an Ahmed (n = 6), and Baerveldt at a different site (n = 9). Of the total second glaucoma drainage devices, 14 were placed at the same site and 29 were placed at a second site ( Table ).


Jan 8, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on Second Glaucoma Drainage Devices in Refractory Pediatric Glaucoma: Failure by Fibrovascular Ingrowth

Full access? Get Clinical Tree

Get Clinical Tree app for offline access