- Email: [email protected]
Early Clinical Experience with the Baerveldt 350-mm2 Glaucoma Implant and Associated Extraocular Muscle Imbalance
Early Clinical Experience with the Baerveldt 350--mm2 Glaucoma Implant and Associated Extraocular Muscle Imbalance Shannon L. Smith, MD, 1 Richard]. Starita, MD, 2•3 Ronald L. Fellman, MD, 2•3 John R. Lynn, MD 2•3 Purpose: The Baerveldt glaucoma implant is a new commercially available aqueous drainage device. Clinical data on the performance of the implant are relatively limited. The purpose of this investigation is to explore potential advantages and disadvantages of this implant over the more widely used Molteno implant. Methods: Retrospective analysis was performed on 37 eyes of 36 patients with refractory glaucoma who underwent placement of a mid-sized Baerveldt 350-mm 2 implant. Patient selection was based on previous failure of conventional medical, laser, and surgical management, with poor prognosis for further glaucoma surgery and adjunctive antifibrosis therapy. Results: Significant postoperative extraocular motility restriction and heterotropia were found in the operated eyes. Twenty-three (77%) of 30 eyes measured have significant heterotropia in primary gaze and restriction of gaze into the quadrant of the implant. As a result, 11 (65%) of 17 functionally binocular patients have diplopia in primary gaze. Conclusions: Pending further study, the authors are avoiding placement of the Baerveldt 350-mm 2 implant in both binocular and monocular patients due to the high frequency of induced heterotropia and motility restriction. Ophthalmology 1993;100:914-918
Extraocular muscle imbalance has been well-documented after scleral buckling procedures, with the most significant factors being the size of the material used and its location under a rectus muscle. 1- 4 Muscle imbalance after glau-
Originally received: September 24, 1992. Revision accepted: December 28, 1992. 1 Lehmann Eye Center, Nacogdoches, Texas. 2 Glaucoma Associates of Texas, Dallas. 3 Department of Ophthalmology, University of Texas Health Science Center at Dallas, Southwestern Medical School, Dallas. Presented in part at the Association for Research in Vision and Ophthalmology Annual Meeting, Sarasota, May 1992. Reprint requests to Shannon L. Smith, MD, Lehmann Eye Center, 4848 NE Stallings, Suite 102, Nacogdoches, TX 75961.
914
coma drainage implants is thought to be rare when the device is placed superiorly. Among the available modernday ocular drainage devices, the Molteno implant has proven to be more popular. It comes in single ( 130-160 mm 2 ), and double (265-320 mm 2 ) polypropylene plate varieties. The calculated surface area of the plates varies, depending on the method used for calculation. 5 •6 The Molteno plates respect the extraocular muscles in their positioning on the globe. The plates are positioned over the equatorial sclera in one or two adjacent quadrants, between the rectus muscles, with the interplate tube straddling a muscle. The space beneath the muscle remains undisturbed (Fig 1). The actual incidence of muscle imbalance after placement of the Molteno implant is unreported. Our retrospective review of 100 of our patients who underwent
Smith et al · Clinical Experience with the Baerveldt Glaucoma Implant
Figure 1. Molteno implant.
Figure 2. Baerveldt implant.
placement of double-plate Molteno implants indicates that it may be as high as 6%, and is much lower if inferior implants are excluded. The Baerveldt single-plate implant (Iovision, Inc, Irvine, CA) is manufactured in three sizes, based on the presumed surface area of the collecting reservoir: 200, 350, and 500 mm 2 • The surface area of the mid-sized Baerveldt 350-mm2 implant compares well with that of the double-plate Molteno implant. The Baerveldt singleplate implant lies in one scleral quadrant. It extends beneath two adjacent rectus muscles (Fig 2). We were highly motivated to use the Baerveldt 350-mm2 implant due to its potential advantage over the double-plate Molteno implant of relative ease of insertion (requiring only one quadrant of dissection). Also, the plate is silicone, which is less inflammatory than polypropylene, 6 and more flexible, lending ease of manipulation into the quadrant between and under the muscles.
propylene, 8 mm from the surgical limbus, with the plate extensions lying 2 mm posterior to the rectus muscle insertions. The plate was checked to ensure its placement 2 mm posterior to the muscle insertions. The tube was cut to an appropriate length, and was intubated with a 50 or 6-0 nylon suture, which was buried subconjunctivally in the inferior fornix, to be pulled at a later date. One or two 6-0 and/or 7-0 polyglactin ligatures were tied around the tube as an additional safeguard against hypotony, as described by Latina. 7 These ligatures were adjusted to
Methods After obtaining informed written consent, the Baerveldt 350-mm 2 implant was placed in 37 eyes of 36 patients with complicated glaucoma, having previously failed conventional medical, laser, and surgical management. Five patients had evidence of preexisting heterotropia or heterophoria; four were functionally monocular and one was binocular. None had preoperative complaints of diplopia or poor cosmesis. Each patient had an average of three prior intraocular surgeries. The diagnostic categories of these patients were varied (Table 1). The patients' mean age was 57 years (range, 8-86 years). The standard surgical technique was as follows: a conjunctival incision was made 6 mm posterior to the surgical limbus, in the planned quadrant of implantation. Dissection was carried anteriorly to the limbus. Two adjacent rectus muscles were isolated, and the Baerveldt 350-mm2 implant was inserted to extend beneath the muscles. Tenon's capsule was not violated, and the muscle sleeves were not stripped in the course of the procedure. The anterior lip of the plate was sutured to the sclera with 9-0 poly-
Table 1. Diagnostic Categories Diagnosis
No. of Eyes
OAG Association with Pseudophakia Uveitis Aphakia Juvenile glaucoma POAG 2° ACG Association with Rubeosis Fibrous ingrowth Pseudophakia ICE Aphakia Aphakia, PK Juvenile glaucoma Traumatic glaucoma Congenital glaucoma Juvenile glaucoma Association with Aphakia Pseudophakia 1° ACG
13
8
2 1 1 1 13
5 2 2 1 1 1 1 3 3
4 3 1 1
OAG = open-angle glaucoma; POAG = primary open-angle glaucoma; 2 ° ACG = secondary angle-closure glaucoma; ICE = iridocorneal endothelial syndrome; PK = penetrating keratoplasty; 1° ACG = primary angle-closure glaucoma.
915
Ophthalmology
Volume 100, Number 6, June 1993
achieve extremely low flow through the tube in the early postoperative period. A 23-gauge needle was used to form a tight tract into the anterior chamber, from a point 0.5 to 1 mm posterior to the surgical limbus. The tube was inserted through this tract and was secured to sclera with a 9-0 polypropylene mattress suture. The tube was covered with a scleral patch graft (alcohol-preserved and rehydrated in a gentamycinsaline bath), which was sutured to the sclera at its four corners with 8-0 polyglactin. Tenon's capsule and conjunctiva were closed with a running 8-0 polyglactin suture. The quadrant of implantation was chosen based on the condition of the conjunctiva and depth available for tube positioning in the anterior chamber. Nineteen implants were placed in the superotemporal quadrant, 14 in the superonasal quadrant, 3 in the inferonasal quadrant, and 1 in the inferotemporal quadrant. In addition to routine postsurgical glaucoma evaluation, as heterotropia and diplopia became apparent as a consistent complication, each patient available for followup was examined for evidence of a heterotropia and for restriction of extraocular motility in the surgical eye. Alternate cover testing with prism measurement of any heterotropia in primary gaze was performed on each followup examination. If the patient had poor fixation ability precluding accurate prism measurement, the heterotropia was estimated based on light reflex measurements in primary gaze. Versions and ductions were examined in the nine diagnostic gaze positions. Motility restriction was judged by the same observer on each visit, on a scale from 1 to 4 in each gaze position ( 1 = a trace restriction; 4 = severe restriction with inability to move the eye past the midline into the position in question). The data presented in the tables are the result of analysis of all postoperative examinations, with follow-up ranging from 1.25 to lO months. Heterotropias in primary gaze were quantified as above and recorded on each visit. All heterotropias worsened or remained unchanged over time.
Table 2. Superotemporal Implant* Heterotropia in Primary Gaze
No. of Patients
Average (11)
Range (a)
Hypertropia/XT H ypertropia/ET Hypotropia/XT XT Hypertropia Hypotropia Motility restriction, no heterotropia in primary gaze No apparent restriction
8
8/9 15/30 2/10
4-12/2-25
XT
=
exotropia; ET
=
1 1 1 1
6
Heterotropia in Primary Gaze
No. of Patients
Average (.:l)
Range (a)
Hypotropia/XT Hypotropia/ET Hypertropia/XT Hypertropia/ET Motility restriction, no heterotropia in primary gaze
5 2 1 1
9/9
3-15/2-18 4-10/9-12
2
esotropia.
7/11
14/20 4/12
2
XT = exotropia; ET = esotropia. * There were 11 eyes with motility follow-up.
Therefore, the final visit prism diopter measurement was obtained for each patient, and patients with similar heterotropias and same quadrant of implant were grouped together. Their heterotropias were averaged. These averages as well as the ranges for patients with superior implants are presented in Tables 2 and 3. The few patients with inferior implants are discussed in the Results section separately. Motility restriction was quantified as described above. All patients who had motility restriction were judged to have "3" to "4" plus restriction, which always increased to a maximum as the eye moved into the quadrant of the implant.
Results At a mean postoperative follow-up of 6. 75 months (range, 1.25-10 months), 23 (77%) of the 30 eyes with motility follow-up have significant heterotropia in primary gaze, with motility restriction into the quadrant of the implant (Table 4). Seventeen of these patients are binocular; 13 are functionally monocular. None had a previous complaint of diplopia. Eleven (65%) of 17 functionally binocular patients are diplopic in primary gaze postoperatively. Superotemporal implants most consistently resulted in hypertropia (average, 81:::,.) with exotropia (average, 91:::,.) in
Table 4. Postoperative Motility Complications Complications
*There were 16 eyes with motility follow-up (3 eyes with no motility follow-up).
916
Table 3. Superonasal Implant*
Heterotropia in primary gaze and motility restriction Motility restriction, no heterotropia in primary gaze No restriction *There were 30 eyes with motility follow-up.
No. of Patients (%)
23 (77) 5 (17) 2 (6)
Smith et al · Clinical Experience with the Baerveldt Glaucoma Implant the surgical eye (Table 2). Motility was consistently re-stricted into the quadrant of the implant. Of the nine binocular patients with superotemporal implants, all had measurable heterotropia with motility restriction into the quadrant of the implant. Seven (78%) of these were diplopic as a result of their heterotropia in primary gaze. One of the remaining two patients returned to another physician and had no motility follow-up. The other became functionally monocular due to an exposurerelated corneal ulcer after_ eyelid surgery. Of the seven patients initially found to be functionally monocular with superotemporal implants, six (86%) had motility restriction into the quadrant of the implant, and four (57%) of these had sufficient fixation ability to exhibit measurable heterotropia. Superonasal implants most consistently resulted in hypotropia (average, 96.) with exotropia (average, 96.) in the surgical eye (Table 3). Motility was consistently restricted into the quadrant of the implant, with a pattern similar to a Brown syndrome with little or no elevation in adduction, presumably due to a tight superior oblique tendon. Of the eight binocular patients with superonasal implants, seven (88%) had measurable heterotropia with motility restriction into the quadrant of the implant. Four (50%) ofthese patients were diplopic. Two ofthe remaining four patients returned to other physicians and had no motility follow-up. The third patient has consistently measured hypertropia of 46. with esotropia of 126. with no complaint of diplopia, and this patient is probably functionally monocular. The final patient is a deaf, nonverbal child who does not express a complaint of diplopia. Of the three patients initially found to be functionally monocular with superonasal implants, two (67%) had motility restriction into the quadrant of the implant and sufficient fixation ability to exhibit measurable heterotropia. Inferonasal implants were placed in two functionally monocular patients with motility follow-up, one of whom experienced motility restriction into the quadrant of the implant. An inferotemporal implant was placed in one functionally monocular patient with sufficient fixation ability to measure hypertropia of 1OL., with motility restriction into the quadrant of the implant. Various other postoperative complications expected with aqueous drainage devices were noted (Table 5). The mid-sized Baerveldt 350-mm 2 implant does seem comparable with the double-plate Molteno implant in its ability to decrease intraocular pressure (lOP). The average preoperative lOP was 29 mmHg (range, 3-58 mmHg). Low preoperative lOPs were seen in two patients with leaking, unsalvageable filtering blebs. The average postoperative lOP was 15 mmHg (range, 4-50 mmHg). High postoperative lOPs were seen in two patients with tubes retracted out of the anterior chamber. Postoperative lOPs were achieved with 58% of the eyes requiring an average of one topical glaucoma medication per eye; and 38% of these requiring an additional carbonic anhydrase inhibitor.
Table
5. Postoperative Surgical Complications
Complications
No. of Eyes (%)
Choroidal effusion Hyphema Severe inflammation Conjuntiva melt over SPG Shallow AC Tube retraction/occlusion
RD Intubation stitch prolapse Iris occluding tube Comeal decompensation over tube CME Presumed fibrous downgrowth SPG = scleral patch graft; AC = anterior chamber; RD tachment; CME = cystoid macular edema.
9 (25) 5 (14) 3 (8) 3 (8) 2 (5) 2 (5) 2 (5) 2 (5) 1 (3) 1 (3) 1 (3) 1 (3) =
retinal de-
Discussion Among the various available modern aqueous drainage devices used for treatment of glaucoma, translimbalequatorial shunts have proven most effective. Molteno, Schocket, and Krupin implants have been the most popular. These devices have in common an anterior chamber tube shunt to an episcleral, equatorial, bleb-promoting device or plate. A fibrous "cocoon" develops over the plate to accommodate aqueous humor drainage postoperatively. The surface area of the filtration cocoon, and thus the surface area of the plate, may be one of the key factors in long-term lOP control postoperatively. 5•8•9 The authors have had success with and have favored the Molteno implant, particularly the double-plate implant over the single-plate, due to its increased surface area for filtration. Literature regarding induced heterotropia with the Molteno implant is sparse. Hypertropia has been noted with inferior quadrant implants; hypotropia with superior quadrant implants. Mass effect, as well as a Faden, or posterior fixation suture effect, have been proposed as etiologic factors in these cases. 10 Hypertropia also has been seen with a superotemporal implant, and the fat adherence syndrome was proposed as a restrictive etiology. 11 An acquired Brown's superior oblique tendon syndrome was encountered frequently with superonasal placement of the Baervelt implant. This resulted in an inability to elevate the eye in the adducted position due to a tight superior oblique tendon. Based on this experience and other reported cases, implantation superonasally should be avoided. 12 Among muscle imbalances induced by all varieties of scleral explants, those associated with retinal detachment surgery are the most common. The most recurrent themes of etiology are large size of the implant, and its location or extension under a muscle belly. Certainly, these factors, as well as displacement or tightening of the superior
917
Ophthalmology
Volume 100, Number 6, June 1993
oblique tendon, and possible muscle insertion damage, come -1nto play with the Baerveldt implant, which has a large surface area and extends beneath two adjacent rectus muscles. The authors do not believe that significant muscle sleeve damage or violation of Tenon's capsule occurs with placement of this implant. With aqueous flow to the plate, as the fibrous cocoon surrounding the implant expands beneath the rectus muscles, the mass effect stretches and tightens them, inducing heterotropia and motility restriction into the quadrant ofthe implant. The authors' surgical technique differed from Baerveldt's, in that the tube ligatures were adjusted so that minimal flow was allowed to the plate in the immediate postoperative period, most likely before the development of the fibrous cocoon. Early flow may have been a factor in the development of prominent heterotropias. A mass effect also may be exerted simply by the presence of the plate itself beneath the muscles, excluding aqueous flow to the plate, and the development of the fibrous cocoon. Also, the fibrous cocoon may have a direct effect on the muscles, and there may be a window of opportunity for removal of the implant during which a good result would be seen. Since initial presentation of these results, the authors have received communication from the manufacturer of the Baerveldt implant describing product improvement aimed at decreasing the height of the bleb surrounding the implant. The implant now has several "bleb control fenestrations" in the plate's surface which allow growth of fibrous "rivets" through the plate, anchoring it to sclera posteriorly. The manufacturer believes that this modification will decrease the incidence of induced diplopia. In this series, Baerveldt implants have been removed from five patients due to induced heterotropia and motility restriction. In four patients, Baerveldt implants have been replaced with double-plate Molteno implants. The fifth patient refused a second aqueous drainage device and has undergone trabeculectomy revision with rapid failure. The earliest removal of implant was at I month postoperatively; the latest was at nine months. All have resulted in significant improvement, some to near resolution of the heterotropia. One patient with hypertropia and exotropia was referred to a strabismologist, and received botulinum injections to the superior and lateral rectus muscles, with subsequent reversal to hypotropia and esotropia. Spectacle prisms have been prescribed for the remaining patients with diplopia as a temporizing measure for relief in primary gaze and/or for reading. Patients with remaining implants and heterotropia have not improved over time. No patient with diplopia has become orthophoric without removal of the implant.
large surface area. The disadvantage of this implant is that, in this experience, it causes an unacceptably high incidence of heterotropia and ocular motility restriction, with diplopia in binocular patients, possibly due to multiple factors, most likely large implant size, and its extension beneath adjacent rectus muscles. Perhaps these recurrent etiologic themes in the literature regarding similar problems after retinal detachment surgery should serve as warnings against the design of an implant that imposes a mass effect beneath rectus muscles. The authors have ceased using the Baerveldt 350-mm2 implant in binocular patients, and that hesitation extends to include functionally monocular patients, due to induced motility restriction and, in some cases, unacceptable cosmesis.
References 1. Sewell JJ, Knobloch WH, Eifrig DE. Extraocular muscle imbalance after surgical treatment for retinal detachment. Am J Ophthalmol 1974;78:321-3. 2. Mets MB, Wendell ME, Gieser RG. Ocular deviation after retinal detachment surgery. Am J Ophthalmol 1985;99:66772. 3. Smiddy WE, Loupe D, Michels RG, eta!. Extraocular muscle imbalance after scleral buckling surgery. Ophthalmology 1989;96: 1485-90. 4. Fison PN, Chignell AH. Diplopia after retinal detachment surgery. Br J Ophthalmol 1987;71:521-5. 5. Lloyd MA, Sedlak T, Heuer DK, et a!. Clinical experience with the single-plate Molteno implant in complicated glaucomas. Update of a pilot study. Ophthalmology 1992;99: 679-87. 6. Wilson RP, Cantor L, Katz U, eta!. Aqueous shunts: Molteno versus Schocket. Ophthalmology 1992;99:672-8. 7. Latina MA. Single stage Molteno implant with combination internal occlusion and external ligature. Ophthalmic Surg 1990;21:444-6. 8. Heuer DK, Lloyd MA, Abrams DA, eta!. Which is better? One or two? A randomized clinical trial of single-plate versus double-plate Molteno implantation for glaucomas in aphakia and pseudophakia. Ophthalmology 1992; 99: 151219. 9. Loeffler KU, Jay JL. Tissue response to aqueous drainage in a functioning Molteno implant. Br J Ophthalmol1988;72: 29-35. 10. Christmann LM, Wilson ME. Motility disturbances after Molteno implants. J Pediatr Ophthalmol Strabismus 1992;29:44-8.
Summary
11. Munoz M, Parrish R. Hypertropia after implantation of a Molteno drainage device [letter]. Am J Ophthalmol 1992; 113:98-100.
In summary, the authors were motivated to use the Baerveldt 350-mm 2 implant due to its ease of insertion and
12. Ball SF, Ellis GS, Herrington RG, eta!. Brown's superior oblique tendon syndrome after Baerveldt glaucoma implant [letter]. Arch Ophthalmol 1992;110:1368.
918
Extraocular muscle imbalance has been well-documented after scleral buckling procedures, with the most significant factors being the size of the material used and its location under a rectus muscle. 1- 4 Muscle imbalance after glau-
Originally received: September 24, 1992. Revision accepted: December 28, 1992. 1 Lehmann Eye Center, Nacogdoches, Texas. 2 Glaucoma Associates of Texas, Dallas. 3 Department of Ophthalmology, University of Texas Health Science Center at Dallas, Southwestern Medical School, Dallas. Presented in part at the Association for Research in Vision and Ophthalmology Annual Meeting, Sarasota, May 1992. Reprint requests to Shannon L. Smith, MD, Lehmann Eye Center, 4848 NE Stallings, Suite 102, Nacogdoches, TX 75961.
914
coma drainage implants is thought to be rare when the device is placed superiorly. Among the available modernday ocular drainage devices, the Molteno implant has proven to be more popular. It comes in single ( 130-160 mm 2 ), and double (265-320 mm 2 ) polypropylene plate varieties. The calculated surface area of the plates varies, depending on the method used for calculation. 5 •6 The Molteno plates respect the extraocular muscles in their positioning on the globe. The plates are positioned over the equatorial sclera in one or two adjacent quadrants, between the rectus muscles, with the interplate tube straddling a muscle. The space beneath the muscle remains undisturbed (Fig 1). The actual incidence of muscle imbalance after placement of the Molteno implant is unreported. Our retrospective review of 100 of our patients who underwent
Smith et al · Clinical Experience with the Baerveldt Glaucoma Implant
Figure 1. Molteno implant.
Figure 2. Baerveldt implant.
placement of double-plate Molteno implants indicates that it may be as high as 6%, and is much lower if inferior implants are excluded. The Baerveldt single-plate implant (Iovision, Inc, Irvine, CA) is manufactured in three sizes, based on the presumed surface area of the collecting reservoir: 200, 350, and 500 mm 2 • The surface area of the mid-sized Baerveldt 350-mm2 implant compares well with that of the double-plate Molteno implant. The Baerveldt singleplate implant lies in one scleral quadrant. It extends beneath two adjacent rectus muscles (Fig 2). We were highly motivated to use the Baerveldt 350-mm2 implant due to its potential advantage over the double-plate Molteno implant of relative ease of insertion (requiring only one quadrant of dissection). Also, the plate is silicone, which is less inflammatory than polypropylene, 6 and more flexible, lending ease of manipulation into the quadrant between and under the muscles.
propylene, 8 mm from the surgical limbus, with the plate extensions lying 2 mm posterior to the rectus muscle insertions. The plate was checked to ensure its placement 2 mm posterior to the muscle insertions. The tube was cut to an appropriate length, and was intubated with a 50 or 6-0 nylon suture, which was buried subconjunctivally in the inferior fornix, to be pulled at a later date. One or two 6-0 and/or 7-0 polyglactin ligatures were tied around the tube as an additional safeguard against hypotony, as described by Latina. 7 These ligatures were adjusted to
Methods After obtaining informed written consent, the Baerveldt 350-mm 2 implant was placed in 37 eyes of 36 patients with complicated glaucoma, having previously failed conventional medical, laser, and surgical management. Five patients had evidence of preexisting heterotropia or heterophoria; four were functionally monocular and one was binocular. None had preoperative complaints of diplopia or poor cosmesis. Each patient had an average of three prior intraocular surgeries. The diagnostic categories of these patients were varied (Table 1). The patients' mean age was 57 years (range, 8-86 years). The standard surgical technique was as follows: a conjunctival incision was made 6 mm posterior to the surgical limbus, in the planned quadrant of implantation. Dissection was carried anteriorly to the limbus. Two adjacent rectus muscles were isolated, and the Baerveldt 350-mm2 implant was inserted to extend beneath the muscles. Tenon's capsule was not violated, and the muscle sleeves were not stripped in the course of the procedure. The anterior lip of the plate was sutured to the sclera with 9-0 poly-
Table 1. Diagnostic Categories Diagnosis
No. of Eyes
OAG Association with Pseudophakia Uveitis Aphakia Juvenile glaucoma POAG 2° ACG Association with Rubeosis Fibrous ingrowth Pseudophakia ICE Aphakia Aphakia, PK Juvenile glaucoma Traumatic glaucoma Congenital glaucoma Juvenile glaucoma Association with Aphakia Pseudophakia 1° ACG
13
8
2 1 1 1 13
5 2 2 1 1 1 1 3 3
4 3 1 1
OAG = open-angle glaucoma; POAG = primary open-angle glaucoma; 2 ° ACG = secondary angle-closure glaucoma; ICE = iridocorneal endothelial syndrome; PK = penetrating keratoplasty; 1° ACG = primary angle-closure glaucoma.
915
Ophthalmology
Volume 100, Number 6, June 1993
achieve extremely low flow through the tube in the early postoperative period. A 23-gauge needle was used to form a tight tract into the anterior chamber, from a point 0.5 to 1 mm posterior to the surgical limbus. The tube was inserted through this tract and was secured to sclera with a 9-0 polypropylene mattress suture. The tube was covered with a scleral patch graft (alcohol-preserved and rehydrated in a gentamycinsaline bath), which was sutured to the sclera at its four corners with 8-0 polyglactin. Tenon's capsule and conjunctiva were closed with a running 8-0 polyglactin suture. The quadrant of implantation was chosen based on the condition of the conjunctiva and depth available for tube positioning in the anterior chamber. Nineteen implants were placed in the superotemporal quadrant, 14 in the superonasal quadrant, 3 in the inferonasal quadrant, and 1 in the inferotemporal quadrant. In addition to routine postsurgical glaucoma evaluation, as heterotropia and diplopia became apparent as a consistent complication, each patient available for followup was examined for evidence of a heterotropia and for restriction of extraocular motility in the surgical eye. Alternate cover testing with prism measurement of any heterotropia in primary gaze was performed on each followup examination. If the patient had poor fixation ability precluding accurate prism measurement, the heterotropia was estimated based on light reflex measurements in primary gaze. Versions and ductions were examined in the nine diagnostic gaze positions. Motility restriction was judged by the same observer on each visit, on a scale from 1 to 4 in each gaze position ( 1 = a trace restriction; 4 = severe restriction with inability to move the eye past the midline into the position in question). The data presented in the tables are the result of analysis of all postoperative examinations, with follow-up ranging from 1.25 to lO months. Heterotropias in primary gaze were quantified as above and recorded on each visit. All heterotropias worsened or remained unchanged over time.
Table 2. Superotemporal Implant* Heterotropia in Primary Gaze
No. of Patients
Average (11)
Range (a)
Hypertropia/XT H ypertropia/ET Hypotropia/XT XT Hypertropia Hypotropia Motility restriction, no heterotropia in primary gaze No apparent restriction
8
8/9 15/30 2/10
4-12/2-25
XT
=
exotropia; ET
=
1 1 1 1
6
Heterotropia in Primary Gaze
No. of Patients
Average (.:l)
Range (a)
Hypotropia/XT Hypotropia/ET Hypertropia/XT Hypertropia/ET Motility restriction, no heterotropia in primary gaze
5 2 1 1
9/9
3-15/2-18 4-10/9-12
2
esotropia.
7/11
14/20 4/12
2
XT = exotropia; ET = esotropia. * There were 11 eyes with motility follow-up.
Therefore, the final visit prism diopter measurement was obtained for each patient, and patients with similar heterotropias and same quadrant of implant were grouped together. Their heterotropias were averaged. These averages as well as the ranges for patients with superior implants are presented in Tables 2 and 3. The few patients with inferior implants are discussed in the Results section separately. Motility restriction was quantified as described above. All patients who had motility restriction were judged to have "3" to "4" plus restriction, which always increased to a maximum as the eye moved into the quadrant of the implant.
Results At a mean postoperative follow-up of 6. 75 months (range, 1.25-10 months), 23 (77%) of the 30 eyes with motility follow-up have significant heterotropia in primary gaze, with motility restriction into the quadrant of the implant (Table 4). Seventeen of these patients are binocular; 13 are functionally monocular. None had a previous complaint of diplopia. Eleven (65%) of 17 functionally binocular patients are diplopic in primary gaze postoperatively. Superotemporal implants most consistently resulted in hypertropia (average, 81:::,.) with exotropia (average, 91:::,.) in
Table 4. Postoperative Motility Complications Complications
*There were 16 eyes with motility follow-up (3 eyes with no motility follow-up).
916
Table 3. Superonasal Implant*
Heterotropia in primary gaze and motility restriction Motility restriction, no heterotropia in primary gaze No restriction *There were 30 eyes with motility follow-up.
No. of Patients (%)
23 (77) 5 (17) 2 (6)
Smith et al · Clinical Experience with the Baerveldt Glaucoma Implant the surgical eye (Table 2). Motility was consistently re-stricted into the quadrant of the implant. Of the nine binocular patients with superotemporal implants, all had measurable heterotropia with motility restriction into the quadrant of the implant. Seven (78%) of these were diplopic as a result of their heterotropia in primary gaze. One of the remaining two patients returned to another physician and had no motility follow-up. The other became functionally monocular due to an exposurerelated corneal ulcer after_ eyelid surgery. Of the seven patients initially found to be functionally monocular with superotemporal implants, six (86%) had motility restriction into the quadrant of the implant, and four (57%) of these had sufficient fixation ability to exhibit measurable heterotropia. Superonasal implants most consistently resulted in hypotropia (average, 96.) with exotropia (average, 96.) in the surgical eye (Table 3). Motility was consistently restricted into the quadrant of the implant, with a pattern similar to a Brown syndrome with little or no elevation in adduction, presumably due to a tight superior oblique tendon. Of the eight binocular patients with superonasal implants, seven (88%) had measurable heterotropia with motility restriction into the quadrant of the implant. Four (50%) ofthese patients were diplopic. Two ofthe remaining four patients returned to other physicians and had no motility follow-up. The third patient has consistently measured hypertropia of 46. with esotropia of 126. with no complaint of diplopia, and this patient is probably functionally monocular. The final patient is a deaf, nonverbal child who does not express a complaint of diplopia. Of the three patients initially found to be functionally monocular with superonasal implants, two (67%) had motility restriction into the quadrant of the implant and sufficient fixation ability to exhibit measurable heterotropia. Inferonasal implants were placed in two functionally monocular patients with motility follow-up, one of whom experienced motility restriction into the quadrant of the implant. An inferotemporal implant was placed in one functionally monocular patient with sufficient fixation ability to measure hypertropia of 1OL., with motility restriction into the quadrant of the implant. Various other postoperative complications expected with aqueous drainage devices were noted (Table 5). The mid-sized Baerveldt 350-mm 2 implant does seem comparable with the double-plate Molteno implant in its ability to decrease intraocular pressure (lOP). The average preoperative lOP was 29 mmHg (range, 3-58 mmHg). Low preoperative lOPs were seen in two patients with leaking, unsalvageable filtering blebs. The average postoperative lOP was 15 mmHg (range, 4-50 mmHg). High postoperative lOPs were seen in two patients with tubes retracted out of the anterior chamber. Postoperative lOPs were achieved with 58% of the eyes requiring an average of one topical glaucoma medication per eye; and 38% of these requiring an additional carbonic anhydrase inhibitor.
Table
5. Postoperative Surgical Complications
Complications
No. of Eyes (%)
Choroidal effusion Hyphema Severe inflammation Conjuntiva melt over SPG Shallow AC Tube retraction/occlusion
RD Intubation stitch prolapse Iris occluding tube Comeal decompensation over tube CME Presumed fibrous downgrowth SPG = scleral patch graft; AC = anterior chamber; RD tachment; CME = cystoid macular edema.
9 (25) 5 (14) 3 (8) 3 (8) 2 (5) 2 (5) 2 (5) 2 (5) 1 (3) 1 (3) 1 (3) 1 (3) =
retinal de-
Discussion Among the various available modern aqueous drainage devices used for treatment of glaucoma, translimbalequatorial shunts have proven most effective. Molteno, Schocket, and Krupin implants have been the most popular. These devices have in common an anterior chamber tube shunt to an episcleral, equatorial, bleb-promoting device or plate. A fibrous "cocoon" develops over the plate to accommodate aqueous humor drainage postoperatively. The surface area of the filtration cocoon, and thus the surface area of the plate, may be one of the key factors in long-term lOP control postoperatively. 5•8•9 The authors have had success with and have favored the Molteno implant, particularly the double-plate implant over the single-plate, due to its increased surface area for filtration. Literature regarding induced heterotropia with the Molteno implant is sparse. Hypertropia has been noted with inferior quadrant implants; hypotropia with superior quadrant implants. Mass effect, as well as a Faden, or posterior fixation suture effect, have been proposed as etiologic factors in these cases. 10 Hypertropia also has been seen with a superotemporal implant, and the fat adherence syndrome was proposed as a restrictive etiology. 11 An acquired Brown's superior oblique tendon syndrome was encountered frequently with superonasal placement of the Baervelt implant. This resulted in an inability to elevate the eye in the adducted position due to a tight superior oblique tendon. Based on this experience and other reported cases, implantation superonasally should be avoided. 12 Among muscle imbalances induced by all varieties of scleral explants, those associated with retinal detachment surgery are the most common. The most recurrent themes of etiology are large size of the implant, and its location or extension under a muscle belly. Certainly, these factors, as well as displacement or tightening of the superior
917
Ophthalmology
Volume 100, Number 6, June 1993
oblique tendon, and possible muscle insertion damage, come -1nto play with the Baerveldt implant, which has a large surface area and extends beneath two adjacent rectus muscles. The authors do not believe that significant muscle sleeve damage or violation of Tenon's capsule occurs with placement of this implant. With aqueous flow to the plate, as the fibrous cocoon surrounding the implant expands beneath the rectus muscles, the mass effect stretches and tightens them, inducing heterotropia and motility restriction into the quadrant ofthe implant. The authors' surgical technique differed from Baerveldt's, in that the tube ligatures were adjusted so that minimal flow was allowed to the plate in the immediate postoperative period, most likely before the development of the fibrous cocoon. Early flow may have been a factor in the development of prominent heterotropias. A mass effect also may be exerted simply by the presence of the plate itself beneath the muscles, excluding aqueous flow to the plate, and the development of the fibrous cocoon. Also, the fibrous cocoon may have a direct effect on the muscles, and there may be a window of opportunity for removal of the implant during which a good result would be seen. Since initial presentation of these results, the authors have received communication from the manufacturer of the Baerveldt implant describing product improvement aimed at decreasing the height of the bleb surrounding the implant. The implant now has several "bleb control fenestrations" in the plate's surface which allow growth of fibrous "rivets" through the plate, anchoring it to sclera posteriorly. The manufacturer believes that this modification will decrease the incidence of induced diplopia. In this series, Baerveldt implants have been removed from five patients due to induced heterotropia and motility restriction. In four patients, Baerveldt implants have been replaced with double-plate Molteno implants. The fifth patient refused a second aqueous drainage device and has undergone trabeculectomy revision with rapid failure. The earliest removal of implant was at I month postoperatively; the latest was at nine months. All have resulted in significant improvement, some to near resolution of the heterotropia. One patient with hypertropia and exotropia was referred to a strabismologist, and received botulinum injections to the superior and lateral rectus muscles, with subsequent reversal to hypotropia and esotropia. Spectacle prisms have been prescribed for the remaining patients with diplopia as a temporizing measure for relief in primary gaze and/or for reading. Patients with remaining implants and heterotropia have not improved over time. No patient with diplopia has become orthophoric without removal of the implant.
large surface area. The disadvantage of this implant is that, in this experience, it causes an unacceptably high incidence of heterotropia and ocular motility restriction, with diplopia in binocular patients, possibly due to multiple factors, most likely large implant size, and its extension beneath adjacent rectus muscles. Perhaps these recurrent etiologic themes in the literature regarding similar problems after retinal detachment surgery should serve as warnings against the design of an implant that imposes a mass effect beneath rectus muscles. The authors have ceased using the Baerveldt 350-mm2 implant in binocular patients, and that hesitation extends to include functionally monocular patients, due to induced motility restriction and, in some cases, unacceptable cosmesis.
References 1. Sewell JJ, Knobloch WH, Eifrig DE. Extraocular muscle imbalance after surgical treatment for retinal detachment. Am J Ophthalmol 1974;78:321-3. 2. Mets MB, Wendell ME, Gieser RG. Ocular deviation after retinal detachment surgery. Am J Ophthalmol 1985;99:66772. 3. Smiddy WE, Loupe D, Michels RG, eta!. Extraocular muscle imbalance after scleral buckling surgery. Ophthalmology 1989;96: 1485-90. 4. Fison PN, Chignell AH. Diplopia after retinal detachment surgery. Br J Ophthalmol 1987;71:521-5. 5. Lloyd MA, Sedlak T, Heuer DK, et a!. Clinical experience with the single-plate Molteno implant in complicated glaucomas. Update of a pilot study. Ophthalmology 1992;99: 679-87. 6. Wilson RP, Cantor L, Katz U, eta!. Aqueous shunts: Molteno versus Schocket. Ophthalmology 1992;99:672-8. 7. Latina MA. Single stage Molteno implant with combination internal occlusion and external ligature. Ophthalmic Surg 1990;21:444-6. 8. Heuer DK, Lloyd MA, Abrams DA, eta!. Which is better? One or two? A randomized clinical trial of single-plate versus double-plate Molteno implantation for glaucomas in aphakia and pseudophakia. Ophthalmology 1992; 99: 151219. 9. Loeffler KU, Jay JL. Tissue response to aqueous drainage in a functioning Molteno implant. Br J Ophthalmol1988;72: 29-35. 10. Christmann LM, Wilson ME. Motility disturbances after Molteno implants. J Pediatr Ophthalmol Strabismus 1992;29:44-8.
Summary
11. Munoz M, Parrish R. Hypertropia after implantation of a Molteno drainage device [letter]. Am J Ophthalmol 1992; 113:98-100.
In summary, the authors were motivated to use the Baerveldt 350-mm 2 implant due to its ease of insertion and
12. Ball SF, Ellis GS, Herrington RG, eta!. Brown's superior oblique tendon syndrome after Baerveldt glaucoma implant [letter]. Arch Ophthalmol 1992;110:1368.
918