- Email: [email protected]
Diplopia after glaucoma drainage device implantation
Diplopia after glaucoma drainage device implantation Amany Abdelaziz, MD, PhD, Hilda Cap o, MD, Michael R. Banitt, MD, Joyce Schiffman, MS, William J. Feuer, MS, Craig A. McKeown, MD, Nehemiah E. Spencer, MD, and Richard K. Parrish, MD PURPOSE
To determine incidence of diplopia after glaucoma drainage device (GDD) surgery and to report treatment outcomes.
METHODS
Financial claims data were used to identify patients who underwent GDD surgery (CPT [Current Procedural Terminology] 66180) at the Bascom Palmer Eye Institute from January 2, 1991, through December 31, 2005. After a second claims data search, the medical records of patients diagnosed with diplopia (International Classification of Diseases, 9th Revision code 368.2) and those who underwent extraocular muscle surgery (CPT-4 code 67311-67335) after GDD implantation were reviewed retrospectively.
RESULTS
A total of 2,661 patients underwent GDD surgery during the study period. Charges were submitted for 59 patients for strabismus surgery or office visits relating to diplopia. Of these, 27 patients were excluded because medical records did not document diplopia or included pre-existing diplopia, cranial nerve palsy, or diplopia attributed to another ocular procedure. The remaining 32 patients developed diplopia secondary to GDD. Superotemporal quadrant GDDs were identified in 23 eyes and inferonasal quadrant placement in 9 eyes. The 1-year cumulative incidence of diplopia was 1.4%. No patient developed diplopia after 1 year. The mean follow-up after diagnosis for patients with diplopia was 48 27 months (range, 1-124 months). The mean time of onset of diplopia after GDD implantation was 66 62 days, with median onset of 42 days (range, 8-278 days). Treatment of diplopia included prisms in 17 cases (53.1%), no treatment in 13 (40.6%), other therapies in 2 (6.3%), and surgery after prismatic treatment failed in 3 (9.4%).
CONCLUSIONS
The incidence of diplopia after GDD surgery is low, and most patients are treated with prisms. ( J AAPOS 2013;17:192-196)
G
laucoma drainage devices (GDDs) are commonly used to treat glaucoma that is refractory to medical therapy.1-12 Potential complications after the placement of a GDD include endophthalmitis, suprachoroidal hemorrhage, retinal detachment, hypotony, cataract, corneal edema, tube or plate erosion, hyphema, choroidal effusion,1-13 superior orbital vein occlusion,14 and diplopia.1-15 The development of diplopia has been reported in 1.4% to 37% of patients after GDD surgery.1-10 The incidence of strabismus after GDD placement ranges from 2.1% to 77%.1-10 In this study we estimated the incidence of diplopia after GDD implantation by using the financial claims data of a large university hospital-based Author affiliations: Bascom Palmer Eye Institute, University of Miami, Miller School of Medicine, Miami, Florida Supported in part by NIH-NEI grant EY014957 (to MRB). The Bascom Palmer Eye Institute is supported by an unrestricted grant from Research to Prevent Blindness, Inc. New York, New York, and National Institutes of Health Center Grant P30-EY014801. Submitted June 8, 2012. Revision accepted November 4, 2012. Correspondence: Hilda Capo, MD, Bascom Palmer Eye Institute, 900 NW 17th Street, Miami, FL 33136 (email: [email protected]). Copyright Ó 2013 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/$36.00 http://dx.doi.org/10.1016/j.jaapos.2012.11.017
192
glaucoma practice. The accuracy of the claims data was verified through a retrospective review of the medical records of the identified cases. In addition, we report treatment outcomes of diplopic patients.
Methods and Materials This study was approved by the Institutional Review Board of the University of Miami. Financial claims data were searched to identify patients .18 years who underwent GDD surgery (Current Procedural Terminology-4 66180) at the Bascom Palmer Eye Institute from January 2, 1991, through December 31, 2005. All patients received Molteno drainage implants (IOP Inc, Costa Mesa, CA), Baerveldt glaucoma implants (BGI: Abbott Medical Optics, Santa Ana, CA), Ahmed glaucoma drainage implants (New World Medical Inc, Rancho Cucamonga, CA), or Krupin implants (E. Benson Hood Lab Inc, Pembroke, MD). The specific type and number of implants were not identifiable in the financial claims data but were later documented by medical record review of patients who developed new-onset diplopia. A second search was performed to identify recipients of GDD who were diagnosed with diplopia (International Classification of Diseases, 9th Revision, code 368.2), who underwent extraocular muscle surgery (Current Procedural Terminology-4 codes 67311-67335), or both after the
Journal of AAPOS
Volume 17 Number 2 / April 2013 GDD surgery. Patients were excluded from the study if diplopia was not documented in the medical record, if diplopia existed before the GDD surgery, or if the diplopia was attributable to another ocular procedure or cranial nerve palsy. An Institutional Review Board–approved standard data collection sheet was used to record patient demographic information, systemic medical conditions, visual acuity, IOP, and the dates and laterality of all ocular procedures that were determined by medical records review. When corroborated by the record, the date of the first complaint of diplopia was noted. Other characteristics of the diplopia, including worsening of the symptoms at distance or near and persistence of the diplopia throughout the day, were also recorded. When available, ocular alignment measurements in the 9 standard positions of gaze were noted. Ocular alignment was measured by the cover and alternate cover tests, except in patients with poor visual acuity, in which cases the Krimsky test was used. Patients were classified as having received no treatment, prismatic correction in spectacles, extraocular muscle surgery, other surgical interventions, or a therapeutic device such as Bangerter foils and Min lenses. Ocular alignment measurements were recorded again approximately 6 months after diagnosis. Patients were classified as having responded to treatment if their symptoms resolved. Patients treated with prismatic correction who experienced residual diplopia were categorized as having significantly improved, partially improved, or not improved. Diplopia was considered improved when its frequency decreased or when there was reduction of diplopia only in certain positions of gaze or at certain distances. Patients were categorized as significantly improved if they reported only occasional diplopia and partially improved if they experienced diplopia on a daily basis.
Results A total of 2,661 patients who underwent GDD surgery during the study period were identified. Of these, 59 were coded as experiencing diplopia, undergoing extraocular muscle surgery, or both after a GDD procedure. After a review of 59 records, 32 patients were found to have diplopia for which the GDD surgery was determined to be a causative factor. Of the 59, 27 patients were excluded from the study for the following reasons: because neither a complaint or diagnosis of diplopia was evident in the medical record (4 patients); a single complaint of diplopia was followed by a normal motility examination (2 patients); the patients had a pre-existing motility disturbance (10 patients); diplopia occurred after other ocular procedures (8 patients); or the patient had a preexisting or new cranial nerve palsy (3 patients). The 1-year cumulative incidence of diplopia attributable to GDD was 1.4% (Kaplan-Meier analysis), indicating that, in our study, 1.4% of patients who underwent GDD surgery developed diplopia within 1 year of surgery. No patients developed diplopia due to GDD after 1 year. Patient demographics, including sex, age, and ethnicity, are summarized in Table 1. Glaucoma diagnoses and pre-
Journal of AAPOS
Abdelaziz et al
193
Table 1. Demographics of patients with diplopia after glaucoma drainage device surgery at the Bascom Palmer Eye Institute, 1991-2005 Characteristic Sex, n (%) Male Female Age in years, mean SD (range) Ethnicity, n (%) White Hispanic African American
14 (44) 18 (56) 69 15 (29-89) 25 (78) 6 (19) 1 (3)
Table 2. Type of glaucoma and previous ocular surgical procedures in patients with diplopia after glaucoma drainage device surgery Characteristic Type of glaucoma Primary open-angle glaucoma Chronic angle-closure glaucoma Pseudoexfoliation Pigmentary glaucoma Secondary open-angle Normal tension glaucoma Iridocorneal endothelial syndrome Uveitic Sturge-Weber syndrome Corticosteroid induced Unknown Previous ocular surgery (either eye)a Cataract extraction Penetrating keratoplasty Pars plana vitrectomy Scleral buckle Previous glaucoma surgery Trabeculectomy in study eye Glaucoma drainage device in study eyeb Glaucoma drainage device in fellow eye Trabeculectomy in fellow eye Operated (study) eye Right Left
Number (%) 12 (40) 3 (10) 3 (10) 3 (10) 1 (3) 2 (7) 2 (7) 2 (7) 1 (3) 1 (3) 2 28 (88) 5 (16) 5 (16) 3 (9) 16 (50) 1 (3) 4 (13) 4 (13) 12 (38) 20 (63)
a
More than one type of surgery possible for individual patients. If both eyes underwent glaucoma drainage device surgery, the study eye is the one most recently operated.
b
vious ocular surgical procedures are listed in Table 2, and the types and location(s) of the implanted glaucoma drainage device(s) are summarized in Table 3. The type of deviation, time to onset, characteristics of the diplopia, treatment(s) used, and the results of the treatment(s) are summarized in e-Supplement 1 (available at jaapos.org). The mean follow-up for the 32 patients with diplopia after diagnosis was 48 27 months (range, 1-124). The mean follow-up for all patients after GDD surgery was 44 44 months (range, 0-220). The mean time to the onset of symptomatic diplopia was 66 62 days, with a median onset of 42 days (range, 8-278). Of the 32 patients, 3 (9.4%) developed diplopia within the first 2 weeks after GDD surgery; 23 (71.9%), between 2 weeks and 3 months; 6 (18.8%), between 3 to 9 months (Figure 1). Symptoms
194
Abdelaziz et al
Volume 17 Number 2 / April 2013
Table 3. Type and location of GDD Type of implant Baerveldt Ahmed Molteno
Location
No. patients
Superotemporal Inferonasala Superotemporal Superotemporal
20 9 2 1
GDD, glaucoma drainage device. Eight placed after previous trabeculectomy or GDD in the superior quadrant.
a
FIG 1. Interval between date of GDD surgery and onset of diplopia.
were intermittent in 29 patients (91%); in 3 patients (9%) symptoms were constant. Twenty-five patients had ocular motility measurements in primary and eccentric positions of gaze. Of the 23 patients who underwent superotemporal GDD surgery, 18 (78%) developed exotropia and/or hypertropia (Figure 2). Of the 9 patients who had inferonasal tubes implanted (8 after a superotemporal GDD or trabeculectomy), 4 (44%) developed hypotropia (Figure 2). Of the 32 patients with diplopia, 17 (53.1%) were treated with prismatic correction in spectacles, of whom 12 had received superotemporal GDDs and 5 inferonasal GDDs. No treatment was prescribed in 13 patients (40.6%), 9 in the superotemporal group and 4 in the inferonasal group. A Bangerter foil and Min lens were used in 2 patients (6.3%), 1 patient each in the superotemporal and inferonasal groups (4.3% of patients). Three patients (9.5%) underwent surgical procedures after they did not respond to prismatic correction, 2 involving altering the drainage device or bleb; only 1 had strabismus surgery. The diplopia spontaneously resolved in 5 patients. The treatment outcomes are summarized in Figure 3. The results were similar for superotemporal and inferonasal implant location. In patients treated with prisms the diplopia was graded as significantly improved in 9 patients (53%), partially improved in 2 (12%), and not improved in 6 (35%). Three patients underwent surgery after they did not respond to prism therapy. Two patients had partial improvement of diplopic symptoms, after needling of the bleb in one, and, in the other, removal of the lateral wings of the BGI from under the belly of the lateral rectus muscle with
FIG 2. Direction and magnitude of ocular deviation in primary position in superotemporal (ST; A) and inferonasal (IN; B) groups after GDD in prism diopters (PD). Some points overlap. ET, esotropia; HoT, hypotropia; HT, hypertropia; PP, primary position; XT, exotropia.
FIG 3. Treatment outcomes. NA, not available; No Tx, no treatment; Others, Min lens and Bangerter foil.
excision of fibrotic tissue. A third patient had strabismus surgery with adjustable sutures, with no improvement of diplopia. Treatment effects were not documented in 2 patients (6%) because of loss to follow-up, 1 with no treatment and 1 treated with a Min lens. Most patients were treated with prisms, but treatment may be ineffective.
Discussion The acute onset of diplopia between 2 weeks and 3 months after GDD surgery in the majority of patients (71.9%)
Journal of AAPOS
Volume 17 Number 2 / April 2013 suggests that mass effect or scarring associated with the device or bleb may be a cause of the restrictive strabismus and resultant diplopia. This finding is consistent with a previous report in which the authors found the mean interval between implantation of GDDs and the onset of diplopia to be 23.5 days (range, 1-90 days).16 Patients with superotemporal GDDs who developed diplopia in our study had a greater incidence of exotropia and/or hypertropia (78%) compared with those with inferonasal tubes (33%). These results reflect those previously reported by Mu~ noz and Parrish12 in patients with Molteno and BGI,13 Smith and colleagues8 in patients with BGIs, Prata and colleagues,14 Dobler-Dixon and colleagues9 in patients given only Molteno implants, and the Tube versus Trabeculectomy (TVT) study1 for BGI recipients. This pattern of ocular deviation may be explained by displacement of the muscles adjacent to the implant away from the sclera, altering the length–tension relationship and affecting motility in the same way as a resection.13,17 Our results differ from studies and case reports by authors8,16,18-22 who report hypotropia in eyes with superiorly placed GDDs in 14 patients, in 9 of whom, however, GDDs were placed in the superonasal quadrant. Inferonasally positioned GDDs implanted after a superonasal GDD or trabeculectomy in our study were associated with hypotropia in 44% cases, with either esotropia or exotropia. In contrast, Wilson-Holt23 and colleagues reported hypertropia after inferonasal Molteno plate surgery, and C ¸ ardakli and Perkins24 reported hypertropia after inferonasal placement of a Krupin valve with disc implantation. When identified by claims data and verified by medical record review, we determined a 1.4% cumulative annual incidence of diplopia after GDD implantation, which is less than that noted in previous studies. The results of previous studies in which authors have evaluated the incidence of diplopia and strabismus after GDD implantation are summarized in Table 4. More recent studies, as compared with previous reports, show decreased occurrence of postoperative diplopia, most likely resulting from the use of GDDs modified to include fenestrations.2 The present study is limited by both retrospective nature and financial claims reporting design. The true incidence of diplopia was likely underestimated. Patients with postoperative diplopia for whom the diagnosis of diplopia was not included in submitted financial charges would not have been identified through a search of the claims database. Furthermore, a lack of complaints from monocular patients and those with poor visual acuity or extensive visual field defects would lead to a missed diagnosis of induced strabismus. The absence of a full ocular motility examination before GDD surgery in most patients limits the recognition of pre-existing phorias or intermittent tropias, which complicates the analysis of the postoperative strabismus. For these reasons, the true incidence of diplopia or induced strabismus after GDD implantation would most accurately be provided by a prospective study. This
Journal of AAPOS
Abdelaziz et al
195
Table 4. Summary of studies reporting incidence of diplopia or strabismus after GDD implantation
Authors (date) Ayyala et al (1998)10 Britt et al (1999)7 Dobler-Dixon et al (1999)9 Frank et al (1995)21 Harbick et al (2006)2 Huang et al (1999)11 Krishna et al (2001)4 Lloyd et al (1994)6 Rauscher et al (2009)1 Roy et al (2001)5 Smith et al (1993)8 Tsai et al (2003)3 Wilson-Holt et al (1992)23
Implant type
Incidence Study No. diplopia/ type patients strabismus, %
Ahmed RCS BGI RCT Molteno PCS
85 107 24
4.7/NR 14.9/14.9 NR/46
Krupin BGI Ahmed BGI BGI BGI BGI BGI BGI Ahmed Molteno
7 182 144 65 73 101 51 36 70 48 16
70/NR 2/NR NR/2.7 NR/3 8/18 5/9.9 NR/2.1 30/77 1.4/NR 0/NR 31.2/56.2
RCS RCS RCS RCS RCT RCT RCS RCS RCS RCS
BGI, Baerveldt glaucoma implant; GDD, glaucoma drainage device; NR, not reported; PCS, prospective clinical series; RCS, retrospective case series; RCT, randomized clinical trial.
type of information was generated by the TVT study, which demonstrated an incidence of 5% new-onset and persistent diplopia (defined as the new onset of diplopia postoperatively with continued presence at follow-up of 6 months or later) and 9.9% development or worsening of ocular motility disturbances.1 Other limitations of our financial claims data methodology are the inability to accurately determine the number of patients who were lost to follow-up and the inability to predict the risk of diplopia for each type of GDD. If patients returned to referring physicians for postoperative care, the diplopia may not have been included in submitted claims and thus not identified through our search of the claims database, consequently underestimating the rate of postoperative diplopia. The lack of data on the specific type of GDD, location of implantation, and primary or sequential surgery prevents an analysis of diplopia for each type of GDD as the denominator is not known. Prismatic spectacle correction was the most common form of treatment in our study and resulted in improvement in 65% of treated patients. Prisms did not always provide satisfactory results because of the incomitant nature of some deviations. The ability to interpret diplopia treatment results is limited in our study by the lack of a detailed sensorimotor examination in some patients after the GDD surgery. Surgery was performed for diplopia in 3 patients, with improvement of symptoms in 67%. Conservative nonsurgical treatment may have been chosen in most cases because of advanced optic nerve damage and greater operative risk in performing strabismus surgery adjacent to drainage devices and filtering blebs. Our estimate of the incidence of diplopia after GDD surgery identified by financial data is low, but the methodology has its weaknesses. Although patients can be
196
Abdelaziz et al
managed with prisms or surgery, treatment may be ineffective. Patients and ophthalmologists who chose GDD surgery as a therapeutic option should be aware of this complication. References 1. Rauscher FM, Gedde SJ, Schiffman JC, Feuer WJ, Barton K, Lee RK, Tube versus Trabeculectomy Study Group. Motility disturbances in the Tube versus Trabeculectomy Study during the first year of follow-up. Am J Ophthalmol 2009;147:458-66. 2. Harbick KH, Sidoti PA, Budenz DL, et al. Outcomes of inferonasal Baerveldt glaucoma drainage implant surgery. J Glaucoma 2006;15:7-12. 3. Tsai JC, Johnson CC, Dietrich MS. The Ahmed shunt versus the Baerveldt shunt for refractory glaucoma: A single-surgeon comparison of outcome. Ophthalmology 2003;110:1814-21. 4. Krishna R, Godfrey DG, Budenz DL, et al. Intermediate-term outcomes of 350-mm2 Baerveldt glaucoma implants. Ophthalmology 2001;108:621-6. 5. Roy S, Ravinet E, Mermoud A. Baerveldt implant in refractory glaucoma: Long-term results and factors influencing outcome. Int Ophthalmol 2001;24:93-100. 6. Lloyd MA, Baerveldt G, Heuer DK, Minckler DS, Martone JF. Initial clinical experience with the Baerveldt implant in complicated glaucomas. Ophthalmology 1994;101:640-50. 7. Britt MT, LaBree LD, Lloyd MA, et al. Randomized clinical trial of the 350-mm2 versus the 500-mm2 Baerveldt implant: Longer term results: Is bigger better? Ophthalmology 1999;106:2312-18. 8. Smith SL, Starita RJ, Fellman RL, Lynn JR. Early clinical experience with the Baerveldt 350-mm2 glaucoma implant and associated extraocular muscle imbalance. Ophthalmology 1993;100:914-18. 9. Dobler-Dixon AA, Cantor LB, Sondhi N, Ku WS, Hoop J. Prospective evaluation of extraocular motility following double-plate Molteno implantation. Arch Ophthalmol 1999;117:1155-60. 10. Ayyala RS, Zurakowski D, Smith JA, et al. A clinical study of the Ahmed glaucoma valve implant in advanced glaucoma. Ophthalmology 1998;105:1968-76.
Volume 17 Number 2 / April 2013 11. Huang MC, Netland PA, Coleman AL, Siegner SW, Moster MR, Hill RA. Intermediate term clinical experience with the Ahmed glaucoma valve implant. Am J Ophthalmol 1999;127:27-33. 12. Mu~ noz M, Parrish R II. Hypertropia after implantation of a Molteno drainage device. Am J Ophthalmol 1992;113:98-100. 13. Mu~ noz M, Parrish R II. Strabismus following implantation of Baerveldt drainage devices. Arch Ophthalmol 1993;111:1096-9. 14. Prata JA, Minckler DS, Green RL. Pseudo Brown’s syndrome as a complication of glaucoma drainage implant surgery. Ophthalmic Surg 1993;24:608-11. 15. Coats DK, Paysee EA, Orenga-Nania S. Acquired Pseudo-Brown’s syndrome immediately following Ahmed valve glaucoma implant. Ophthalmic Surg Lasers 1999;30:396-7. 16. Roizen A, Ela-Dalman N, Velez FG, Coleman AL, Rosenbaum AL. Surgical treatment of strabismus secondary to glaucoma drainage device. Arch Ophthalmol 2008;126:480-86. 17. Rosenbaum A, Santiago P. Strabismus after glaucoma implant procedures. In: Rosenbaum A, Santiago P, eds. Clinical strabismus management. Philadelphia, PA: WB Saunders Co; 1999:304-7. 18. Ball SF, Ellis GS Jr, Herrington RG, Liang K. Brown’s superior oblique tendon syndrome after Baerveldt glaucoma implant. Arch Ophthalmol 1992;110:1368. 19. Christmann LM, Wilson ME. Motility disturbances after Molteno implants. J Pediatr Ophthalmol Stabismus 1992;29:44-8. 20. Dobler AA, Sondhi N, Cantor LB, Ku S. Acquired Brown’s Syndrome after a double plate Molteno implant. Am J Ophthalmol 1993;116: 641-2. 21. Frank JW, Perkins TW, Kushner BJ. Ocular motility defects in patients with the Krupin valve implant. Ophthalmic Surg 1995;26: 228-32. 22. Ventura MP, Vianna RN, Souza Filho JP, Solari HP, Curi RL. Acquired Brown’s syndrome secondary to Ahmed valve implant for neovascular glaucoma. Eye 2005;19:230-32. 23. Wilson-Holt N, Franks W, Nourredin B, Hitchings R. Hypertropia following insertion of inferiorly sited double plate Molteno tubes. Eye 1992;6:515-20. 24. C ¸ ardakli UF, Perkins TW. Recalcitrant diplopia after implantation of a Krupin valve with disc. Ophthalmic Surg 1994;24:256-8.
Journal of AAPOS
To determine incidence of diplopia after glaucoma drainage device (GDD) surgery and to report treatment outcomes.
METHODS
Financial claims data were used to identify patients who underwent GDD surgery (CPT [Current Procedural Terminology] 66180) at the Bascom Palmer Eye Institute from January 2, 1991, through December 31, 2005. After a second claims data search, the medical records of patients diagnosed with diplopia (International Classification of Diseases, 9th Revision code 368.2) and those who underwent extraocular muscle surgery (CPT-4 code 67311-67335) after GDD implantation were reviewed retrospectively.
RESULTS
A total of 2,661 patients underwent GDD surgery during the study period. Charges were submitted for 59 patients for strabismus surgery or office visits relating to diplopia. Of these, 27 patients were excluded because medical records did not document diplopia or included pre-existing diplopia, cranial nerve palsy, or diplopia attributed to another ocular procedure. The remaining 32 patients developed diplopia secondary to GDD. Superotemporal quadrant GDDs were identified in 23 eyes and inferonasal quadrant placement in 9 eyes. The 1-year cumulative incidence of diplopia was 1.4%. No patient developed diplopia after 1 year. The mean follow-up after diagnosis for patients with diplopia was 48 27 months (range, 1-124 months). The mean time of onset of diplopia after GDD implantation was 66 62 days, with median onset of 42 days (range, 8-278 days). Treatment of diplopia included prisms in 17 cases (53.1%), no treatment in 13 (40.6%), other therapies in 2 (6.3%), and surgery after prismatic treatment failed in 3 (9.4%).
CONCLUSIONS
The incidence of diplopia after GDD surgery is low, and most patients are treated with prisms. ( J AAPOS 2013;17:192-196)
G
laucoma drainage devices (GDDs) are commonly used to treat glaucoma that is refractory to medical therapy.1-12 Potential complications after the placement of a GDD include endophthalmitis, suprachoroidal hemorrhage, retinal detachment, hypotony, cataract, corneal edema, tube or plate erosion, hyphema, choroidal effusion,1-13 superior orbital vein occlusion,14 and diplopia.1-15 The development of diplopia has been reported in 1.4% to 37% of patients after GDD surgery.1-10 The incidence of strabismus after GDD placement ranges from 2.1% to 77%.1-10 In this study we estimated the incidence of diplopia after GDD implantation by using the financial claims data of a large university hospital-based Author affiliations: Bascom Palmer Eye Institute, University of Miami, Miller School of Medicine, Miami, Florida Supported in part by NIH-NEI grant EY014957 (to MRB). The Bascom Palmer Eye Institute is supported by an unrestricted grant from Research to Prevent Blindness, Inc. New York, New York, and National Institutes of Health Center Grant P30-EY014801. Submitted June 8, 2012. Revision accepted November 4, 2012. Correspondence: Hilda Capo, MD, Bascom Palmer Eye Institute, 900 NW 17th Street, Miami, FL 33136 (email: [email protected]). Copyright Ó 2013 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/$36.00 http://dx.doi.org/10.1016/j.jaapos.2012.11.017
192
glaucoma practice. The accuracy of the claims data was verified through a retrospective review of the medical records of the identified cases. In addition, we report treatment outcomes of diplopic patients.
Methods and Materials This study was approved by the Institutional Review Board of the University of Miami. Financial claims data were searched to identify patients .18 years who underwent GDD surgery (Current Procedural Terminology-4 66180) at the Bascom Palmer Eye Institute from January 2, 1991, through December 31, 2005. All patients received Molteno drainage implants (IOP Inc, Costa Mesa, CA), Baerveldt glaucoma implants (BGI: Abbott Medical Optics, Santa Ana, CA), Ahmed glaucoma drainage implants (New World Medical Inc, Rancho Cucamonga, CA), or Krupin implants (E. Benson Hood Lab Inc, Pembroke, MD). The specific type and number of implants were not identifiable in the financial claims data but were later documented by medical record review of patients who developed new-onset diplopia. A second search was performed to identify recipients of GDD who were diagnosed with diplopia (International Classification of Diseases, 9th Revision, code 368.2), who underwent extraocular muscle surgery (Current Procedural Terminology-4 codes 67311-67335), or both after the
Journal of AAPOS
Volume 17 Number 2 / April 2013 GDD surgery. Patients were excluded from the study if diplopia was not documented in the medical record, if diplopia existed before the GDD surgery, or if the diplopia was attributable to another ocular procedure or cranial nerve palsy. An Institutional Review Board–approved standard data collection sheet was used to record patient demographic information, systemic medical conditions, visual acuity, IOP, and the dates and laterality of all ocular procedures that were determined by medical records review. When corroborated by the record, the date of the first complaint of diplopia was noted. Other characteristics of the diplopia, including worsening of the symptoms at distance or near and persistence of the diplopia throughout the day, were also recorded. When available, ocular alignment measurements in the 9 standard positions of gaze were noted. Ocular alignment was measured by the cover and alternate cover tests, except in patients with poor visual acuity, in which cases the Krimsky test was used. Patients were classified as having received no treatment, prismatic correction in spectacles, extraocular muscle surgery, other surgical interventions, or a therapeutic device such as Bangerter foils and Min lenses. Ocular alignment measurements were recorded again approximately 6 months after diagnosis. Patients were classified as having responded to treatment if their symptoms resolved. Patients treated with prismatic correction who experienced residual diplopia were categorized as having significantly improved, partially improved, or not improved. Diplopia was considered improved when its frequency decreased or when there was reduction of diplopia only in certain positions of gaze or at certain distances. Patients were categorized as significantly improved if they reported only occasional diplopia and partially improved if they experienced diplopia on a daily basis.
Results A total of 2,661 patients who underwent GDD surgery during the study period were identified. Of these, 59 were coded as experiencing diplopia, undergoing extraocular muscle surgery, or both after a GDD procedure. After a review of 59 records, 32 patients were found to have diplopia for which the GDD surgery was determined to be a causative factor. Of the 59, 27 patients were excluded from the study for the following reasons: because neither a complaint or diagnosis of diplopia was evident in the medical record (4 patients); a single complaint of diplopia was followed by a normal motility examination (2 patients); the patients had a pre-existing motility disturbance (10 patients); diplopia occurred after other ocular procedures (8 patients); or the patient had a preexisting or new cranial nerve palsy (3 patients). The 1-year cumulative incidence of diplopia attributable to GDD was 1.4% (Kaplan-Meier analysis), indicating that, in our study, 1.4% of patients who underwent GDD surgery developed diplopia within 1 year of surgery. No patients developed diplopia due to GDD after 1 year. Patient demographics, including sex, age, and ethnicity, are summarized in Table 1. Glaucoma diagnoses and pre-
Journal of AAPOS
Abdelaziz et al
193
Table 1. Demographics of patients with diplopia after glaucoma drainage device surgery at the Bascom Palmer Eye Institute, 1991-2005 Characteristic Sex, n (%) Male Female Age in years, mean SD (range) Ethnicity, n (%) White Hispanic African American
14 (44) 18 (56) 69 15 (29-89) 25 (78) 6 (19) 1 (3)
Table 2. Type of glaucoma and previous ocular surgical procedures in patients with diplopia after glaucoma drainage device surgery Characteristic Type of glaucoma Primary open-angle glaucoma Chronic angle-closure glaucoma Pseudoexfoliation Pigmentary glaucoma Secondary open-angle Normal tension glaucoma Iridocorneal endothelial syndrome Uveitic Sturge-Weber syndrome Corticosteroid induced Unknown Previous ocular surgery (either eye)a Cataract extraction Penetrating keratoplasty Pars plana vitrectomy Scleral buckle Previous glaucoma surgery Trabeculectomy in study eye Glaucoma drainage device in study eyeb Glaucoma drainage device in fellow eye Trabeculectomy in fellow eye Operated (study) eye Right Left
Number (%) 12 (40) 3 (10) 3 (10) 3 (10) 1 (3) 2 (7) 2 (7) 2 (7) 1 (3) 1 (3) 2 28 (88) 5 (16) 5 (16) 3 (9) 16 (50) 1 (3) 4 (13) 4 (13) 12 (38) 20 (63)
a
More than one type of surgery possible for individual patients. If both eyes underwent glaucoma drainage device surgery, the study eye is the one most recently operated.
b
vious ocular surgical procedures are listed in Table 2, and the types and location(s) of the implanted glaucoma drainage device(s) are summarized in Table 3. The type of deviation, time to onset, characteristics of the diplopia, treatment(s) used, and the results of the treatment(s) are summarized in e-Supplement 1 (available at jaapos.org). The mean follow-up for the 32 patients with diplopia after diagnosis was 48 27 months (range, 1-124). The mean follow-up for all patients after GDD surgery was 44 44 months (range, 0-220). The mean time to the onset of symptomatic diplopia was 66 62 days, with a median onset of 42 days (range, 8-278). Of the 32 patients, 3 (9.4%) developed diplopia within the first 2 weeks after GDD surgery; 23 (71.9%), between 2 weeks and 3 months; 6 (18.8%), between 3 to 9 months (Figure 1). Symptoms
194
Abdelaziz et al
Volume 17 Number 2 / April 2013
Table 3. Type and location of GDD Type of implant Baerveldt Ahmed Molteno
Location
No. patients
Superotemporal Inferonasala Superotemporal Superotemporal
20 9 2 1
GDD, glaucoma drainage device. Eight placed after previous trabeculectomy or GDD in the superior quadrant.
a
FIG 1. Interval between date of GDD surgery and onset of diplopia.
were intermittent in 29 patients (91%); in 3 patients (9%) symptoms were constant. Twenty-five patients had ocular motility measurements in primary and eccentric positions of gaze. Of the 23 patients who underwent superotemporal GDD surgery, 18 (78%) developed exotropia and/or hypertropia (Figure 2). Of the 9 patients who had inferonasal tubes implanted (8 after a superotemporal GDD or trabeculectomy), 4 (44%) developed hypotropia (Figure 2). Of the 32 patients with diplopia, 17 (53.1%) were treated with prismatic correction in spectacles, of whom 12 had received superotemporal GDDs and 5 inferonasal GDDs. No treatment was prescribed in 13 patients (40.6%), 9 in the superotemporal group and 4 in the inferonasal group. A Bangerter foil and Min lens were used in 2 patients (6.3%), 1 patient each in the superotemporal and inferonasal groups (4.3% of patients). Three patients (9.5%) underwent surgical procedures after they did not respond to prismatic correction, 2 involving altering the drainage device or bleb; only 1 had strabismus surgery. The diplopia spontaneously resolved in 5 patients. The treatment outcomes are summarized in Figure 3. The results were similar for superotemporal and inferonasal implant location. In patients treated with prisms the diplopia was graded as significantly improved in 9 patients (53%), partially improved in 2 (12%), and not improved in 6 (35%). Three patients underwent surgery after they did not respond to prism therapy. Two patients had partial improvement of diplopic symptoms, after needling of the bleb in one, and, in the other, removal of the lateral wings of the BGI from under the belly of the lateral rectus muscle with
FIG 2. Direction and magnitude of ocular deviation in primary position in superotemporal (ST; A) and inferonasal (IN; B) groups after GDD in prism diopters (PD). Some points overlap. ET, esotropia; HoT, hypotropia; HT, hypertropia; PP, primary position; XT, exotropia.
FIG 3. Treatment outcomes. NA, not available; No Tx, no treatment; Others, Min lens and Bangerter foil.
excision of fibrotic tissue. A third patient had strabismus surgery with adjustable sutures, with no improvement of diplopia. Treatment effects were not documented in 2 patients (6%) because of loss to follow-up, 1 with no treatment and 1 treated with a Min lens. Most patients were treated with prisms, but treatment may be ineffective.
Discussion The acute onset of diplopia between 2 weeks and 3 months after GDD surgery in the majority of patients (71.9%)
Journal of AAPOS
Volume 17 Number 2 / April 2013 suggests that mass effect or scarring associated with the device or bleb may be a cause of the restrictive strabismus and resultant diplopia. This finding is consistent with a previous report in which the authors found the mean interval between implantation of GDDs and the onset of diplopia to be 23.5 days (range, 1-90 days).16 Patients with superotemporal GDDs who developed diplopia in our study had a greater incidence of exotropia and/or hypertropia (78%) compared with those with inferonasal tubes (33%). These results reflect those previously reported by Mu~ noz and Parrish12 in patients with Molteno and BGI,13 Smith and colleagues8 in patients with BGIs, Prata and colleagues,14 Dobler-Dixon and colleagues9 in patients given only Molteno implants, and the Tube versus Trabeculectomy (TVT) study1 for BGI recipients. This pattern of ocular deviation may be explained by displacement of the muscles adjacent to the implant away from the sclera, altering the length–tension relationship and affecting motility in the same way as a resection.13,17 Our results differ from studies and case reports by authors8,16,18-22 who report hypotropia in eyes with superiorly placed GDDs in 14 patients, in 9 of whom, however, GDDs were placed in the superonasal quadrant. Inferonasally positioned GDDs implanted after a superonasal GDD or trabeculectomy in our study were associated with hypotropia in 44% cases, with either esotropia or exotropia. In contrast, Wilson-Holt23 and colleagues reported hypertropia after inferonasal Molteno plate surgery, and C ¸ ardakli and Perkins24 reported hypertropia after inferonasal placement of a Krupin valve with disc implantation. When identified by claims data and verified by medical record review, we determined a 1.4% cumulative annual incidence of diplopia after GDD implantation, which is less than that noted in previous studies. The results of previous studies in which authors have evaluated the incidence of diplopia and strabismus after GDD implantation are summarized in Table 4. More recent studies, as compared with previous reports, show decreased occurrence of postoperative diplopia, most likely resulting from the use of GDDs modified to include fenestrations.2 The present study is limited by both retrospective nature and financial claims reporting design. The true incidence of diplopia was likely underestimated. Patients with postoperative diplopia for whom the diagnosis of diplopia was not included in submitted financial charges would not have been identified through a search of the claims database. Furthermore, a lack of complaints from monocular patients and those with poor visual acuity or extensive visual field defects would lead to a missed diagnosis of induced strabismus. The absence of a full ocular motility examination before GDD surgery in most patients limits the recognition of pre-existing phorias or intermittent tropias, which complicates the analysis of the postoperative strabismus. For these reasons, the true incidence of diplopia or induced strabismus after GDD implantation would most accurately be provided by a prospective study. This
Journal of AAPOS
Abdelaziz et al
195
Table 4. Summary of studies reporting incidence of diplopia or strabismus after GDD implantation
Authors (date) Ayyala et al (1998)10 Britt et al (1999)7 Dobler-Dixon et al (1999)9 Frank et al (1995)21 Harbick et al (2006)2 Huang et al (1999)11 Krishna et al (2001)4 Lloyd et al (1994)6 Rauscher et al (2009)1 Roy et al (2001)5 Smith et al (1993)8 Tsai et al (2003)3 Wilson-Holt et al (1992)23
Implant type
Incidence Study No. diplopia/ type patients strabismus, %
Ahmed RCS BGI RCT Molteno PCS
85 107 24
4.7/NR 14.9/14.9 NR/46
Krupin BGI Ahmed BGI BGI BGI BGI BGI BGI Ahmed Molteno
7 182 144 65 73 101 51 36 70 48 16
70/NR 2/NR NR/2.7 NR/3 8/18 5/9.9 NR/2.1 30/77 1.4/NR 0/NR 31.2/56.2
RCS RCS RCS RCS RCT RCT RCS RCS RCS RCS
BGI, Baerveldt glaucoma implant; GDD, glaucoma drainage device; NR, not reported; PCS, prospective clinical series; RCS, retrospective case series; RCT, randomized clinical trial.
type of information was generated by the TVT study, which demonstrated an incidence of 5% new-onset and persistent diplopia (defined as the new onset of diplopia postoperatively with continued presence at follow-up of 6 months or later) and 9.9% development or worsening of ocular motility disturbances.1 Other limitations of our financial claims data methodology are the inability to accurately determine the number of patients who were lost to follow-up and the inability to predict the risk of diplopia for each type of GDD. If patients returned to referring physicians for postoperative care, the diplopia may not have been included in submitted claims and thus not identified through our search of the claims database, consequently underestimating the rate of postoperative diplopia. The lack of data on the specific type of GDD, location of implantation, and primary or sequential surgery prevents an analysis of diplopia for each type of GDD as the denominator is not known. Prismatic spectacle correction was the most common form of treatment in our study and resulted in improvement in 65% of treated patients. Prisms did not always provide satisfactory results because of the incomitant nature of some deviations. The ability to interpret diplopia treatment results is limited in our study by the lack of a detailed sensorimotor examination in some patients after the GDD surgery. Surgery was performed for diplopia in 3 patients, with improvement of symptoms in 67%. Conservative nonsurgical treatment may have been chosen in most cases because of advanced optic nerve damage and greater operative risk in performing strabismus surgery adjacent to drainage devices and filtering blebs. Our estimate of the incidence of diplopia after GDD surgery identified by financial data is low, but the methodology has its weaknesses. Although patients can be
196
Abdelaziz et al
managed with prisms or surgery, treatment may be ineffective. Patients and ophthalmologists who chose GDD surgery as a therapeutic option should be aware of this complication. References 1. Rauscher FM, Gedde SJ, Schiffman JC, Feuer WJ, Barton K, Lee RK, Tube versus Trabeculectomy Study Group. Motility disturbances in the Tube versus Trabeculectomy Study during the first year of follow-up. Am J Ophthalmol 2009;147:458-66. 2. Harbick KH, Sidoti PA, Budenz DL, et al. Outcomes of inferonasal Baerveldt glaucoma drainage implant surgery. J Glaucoma 2006;15:7-12. 3. Tsai JC, Johnson CC, Dietrich MS. The Ahmed shunt versus the Baerveldt shunt for refractory glaucoma: A single-surgeon comparison of outcome. Ophthalmology 2003;110:1814-21. 4. Krishna R, Godfrey DG, Budenz DL, et al. Intermediate-term outcomes of 350-mm2 Baerveldt glaucoma implants. Ophthalmology 2001;108:621-6. 5. Roy S, Ravinet E, Mermoud A. Baerveldt implant in refractory glaucoma: Long-term results and factors influencing outcome. Int Ophthalmol 2001;24:93-100. 6. Lloyd MA, Baerveldt G, Heuer DK, Minckler DS, Martone JF. Initial clinical experience with the Baerveldt implant in complicated glaucomas. Ophthalmology 1994;101:640-50. 7. Britt MT, LaBree LD, Lloyd MA, et al. Randomized clinical trial of the 350-mm2 versus the 500-mm2 Baerveldt implant: Longer term results: Is bigger better? Ophthalmology 1999;106:2312-18. 8. Smith SL, Starita RJ, Fellman RL, Lynn JR. Early clinical experience with the Baerveldt 350-mm2 glaucoma implant and associated extraocular muscle imbalance. Ophthalmology 1993;100:914-18. 9. Dobler-Dixon AA, Cantor LB, Sondhi N, Ku WS, Hoop J. Prospective evaluation of extraocular motility following double-plate Molteno implantation. Arch Ophthalmol 1999;117:1155-60. 10. Ayyala RS, Zurakowski D, Smith JA, et al. A clinical study of the Ahmed glaucoma valve implant in advanced glaucoma. Ophthalmology 1998;105:1968-76.
Volume 17 Number 2 / April 2013 11. Huang MC, Netland PA, Coleman AL, Siegner SW, Moster MR, Hill RA. Intermediate term clinical experience with the Ahmed glaucoma valve implant. Am J Ophthalmol 1999;127:27-33. 12. Mu~ noz M, Parrish R II. Hypertropia after implantation of a Molteno drainage device. Am J Ophthalmol 1992;113:98-100. 13. Mu~ noz M, Parrish R II. Strabismus following implantation of Baerveldt drainage devices. Arch Ophthalmol 1993;111:1096-9. 14. Prata JA, Minckler DS, Green RL. Pseudo Brown’s syndrome as a complication of glaucoma drainage implant surgery. Ophthalmic Surg 1993;24:608-11. 15. Coats DK, Paysee EA, Orenga-Nania S. Acquired Pseudo-Brown’s syndrome immediately following Ahmed valve glaucoma implant. Ophthalmic Surg Lasers 1999;30:396-7. 16. Roizen A, Ela-Dalman N, Velez FG, Coleman AL, Rosenbaum AL. Surgical treatment of strabismus secondary to glaucoma drainage device. Arch Ophthalmol 2008;126:480-86. 17. Rosenbaum A, Santiago P. Strabismus after glaucoma implant procedures. In: Rosenbaum A, Santiago P, eds. Clinical strabismus management. Philadelphia, PA: WB Saunders Co; 1999:304-7. 18. Ball SF, Ellis GS Jr, Herrington RG, Liang K. Brown’s superior oblique tendon syndrome after Baerveldt glaucoma implant. Arch Ophthalmol 1992;110:1368. 19. Christmann LM, Wilson ME. Motility disturbances after Molteno implants. J Pediatr Ophthalmol Stabismus 1992;29:44-8. 20. Dobler AA, Sondhi N, Cantor LB, Ku S. Acquired Brown’s Syndrome after a double plate Molteno implant. Am J Ophthalmol 1993;116: 641-2. 21. Frank JW, Perkins TW, Kushner BJ. Ocular motility defects in patients with the Krupin valve implant. Ophthalmic Surg 1995;26: 228-32. 22. Ventura MP, Vianna RN, Souza Filho JP, Solari HP, Curi RL. Acquired Brown’s syndrome secondary to Ahmed valve implant for neovascular glaucoma. Eye 2005;19:230-32. 23. Wilson-Holt N, Franks W, Nourredin B, Hitchings R. Hypertropia following insertion of inferiorly sited double plate Molteno tubes. Eye 1992;6:515-20. 24. C ¸ ardakli UF, Perkins TW. Recalcitrant diplopia after implantation of a Krupin valve with disc. Ophthalmic Surg 1994;24:256-8.
Journal of AAPOS