Cytomegalovirus (CMV) retinitis is the most common ocular opportunistic infection in patients with acquired immune deficiency syndrome (AIDS), affecting up to 30% of afflicted individuals.1'2 Rhegmatogenous retinal detachment has been reported to occur in up to 29% of CMV retinitis cases with a cumulative probability of up to 50% one year after the diagnosis of CMV retinitis.3'4 The extent of retinal involvement by CMV retinitis has been shown to be a risk factor for retinal detachment, leading to conclusions that a reduction in the extent of retinitis may result in fewer detachments.5 Recently, an intraocular sustained-release ganciclovir implant has been shown to be effective in delaying the median time to progression of CMV retinitis.6 When retinal detachment does occur, many reports have shown successful reattachment rates using pars plana vitrectomy with permanent silicone oil tamponade.4'7'13 Silicone oil infusion offers rapid visual rehabilitation, obviates the need for postoperative head positioning, and is compatible with good visual acuity.9·11'12 Perhaps more importantly, long-term tamponade appears necessary for multiple holes in atrophie or gliothic retina that are commonly found in detachments associated with CMV retinitis.12
Given the success of the ganciclovir implant in controlling CMV retinitis and the efficacy of pars plana vitrectomy with silicone oil tamponade in accomplishing retinal reattachment, it appears possible that a therapeutic benefit could be achieved by combining these 2 modalities in a single surgical procedure. In addition to anatomic reattachment with potential preservation or improvement of visual acuity, patients could benefit from sustained medical control of the size and activity of the CMV lesion(s). This may reduce the risks of further atrophy, retinal hole formation, and possible repeat detachment. Results of 1 0 patients who underwent siíicone oil tamponade combined with ganciclovir implantation for CMV retinitis-related retinal detachments are reported.
PATIENTS AND METHODS
The surgical results of 10 patients who presented to a large medical center over a 3-year period ( 1 996- 1 999) were retrospectively reviewed. All patients were diagnosed with CMV retinitis-related retinal detachments and treated with pars plana vitrectomy and silicone oil infusion combined with simultaneous placement of an 8-month sustained release Vitrasert ganciclovir implant (Chiron, Calif.). A standard 3-port pars plana vitrectomy technique was used. This included excision of the vitreous and detachment of the posterior hyaloid, followed by endodrainage through a nasal retinotomy, endolaser, and infusion of silicone oil to the posterior lenticular surface. During the procedure a ganciclovir implant was placed in the inferior temporal quadrant at a position 4.0 mm posterior to the surgical limbus using 8-0 nylon suture. The devices were implanted equally successfully prior to vitrectomy or after retinal reattachment via air/fluid exchange before infusion of silicone oil.
Parameters evaluated included location of retinal detachment, reattachment rate, preoperative and best postoperative Snellen visual acuity, and preoperative and postoperative CMV retinitis activity and location. Preoperative CMV retinitis was reported as active or inactive; active retinitis was further classified with respect to zone of involvement. Postoperative CMV retinitis was judged by the response to ganciclovir implant placement. In inactive eyes, success was measured by the ability to prevent recurrence over the duration of follow up. In eyes with active CMV retinitis preoperatively, success was measured by the ability to induce regression or prevent progression at an interval between 1 and 2 months of follow up. Partial regression or diminished activity was defined as a clinically evident decrease in the overall extent of active retinitis and retinitis-related hemorrhages, formation of a scar, and lack of progression along any border. This was judged by an experienced vitreoretínal specialist. Preoperative and postoperative adjunctive CMV antiviral regimens were documented. It should be noted that all patients received some form of systemic antiviral therapy posroperatively to protect the fellow eye and other susceptible organ systems. Finally, complications of the combined procedure were reviewed.
Results of combined pars plana vitrectomy with silicone oïl infusion and ganciclovir implant placement are presented in the Table. All retinal detachments were repaired within 1 day of diagnosis, with the exception of 2 patients whose systemic illness precluded earlier surgery (patient 2 had a 9-day delay before surgery and patient 5 was delayed 5 days); in both cases, the macula was detached preoperatívely. Overall anatomic reattachment was achieved in all 10 patients. However, patient 3 redetached inferotemporally 6 weeks after initial repair and was successfully reattached the following day after repeat pars plana vitrectomy, silicone oil exchange, and endolaser. Four patients presented with macular involvement of their retinal detachments. Three of these patients experienced significant postoperative improvement in visual acuity, but patient 2 had significant visual deterioration even after macular reattachment This was thought to be caused by a combination of delay in surgical repair and extensive macular scarring from CMV retinitis. Surgery preserved visual acuity within 2 Snellen lines in the 6 patients who presented with macula attached. All 3 CMV retinitis patients with inactive retinitis preoperatívely remained free of retinitis for the duration of follow up (median 6 months; range 2 to 22 months). Of the remaining cases with active retinitis preoperatively, 5 became inactive and 2 diminished in activity at an interval between 1 and 2 months of follow up. No patients experienced progression of their retinitis postoperatively. Most patients were treated before the HAART (highly active antiretroviral therapy) era. No patients showed a significant change in CD4 count over the follow-up period that would account for improved retinitis control. One patient (patient 2) developed endophthalmitis 7 weeks after surgery. It appeared this was endogenous endophthalmitis because the patient presented late in the postoperative course with concurrent systemic symptoms. This patient died within several weeks from Pseudomonas sepsis and multiple organ failure.
The rationale for pars plana vitrectomy with silicone oil tamponade has already been established for the treatment of CMV retinitis- reía ted retinal detachments. These detachments are unique in that they may involve multiple breaks at the junction of normal and atrophie retina or within thin, gUotic retina.12 It has been suggested that minimal vitreous traction may produce detachment in necrotic retina that has a poor bond with its underlying degenerated retinal pigment epithelium.14 Pars plana vitrectomy with silicone oil infusion addresses the issue of vitreous traction while permanently tamponading multiple breaks that may not be detected at the time of surgery or may develop intra- or postoperatively.
In the treatment of active CMV retinitis, the ganciclovir implant provides highly active local therapy with a prolonged median time to progression.6 It achieves intravitreal ganciclovir levels approximately 4 times that achieved by intravenous administration6·15"19 and this may account for its efficacy in both newly diagnosed retinitis and reactivated retinitis. Intravenous ganciclovir, foscarnet, and cidofovir are all effective in treating CMV retinitis, but each yields a median time to progression that is less than that of the gancidovir implant.20'23 The ganciclovir implant is also effective for reactivated retinitis including treatment failure with or intolerance to intravenous ganciclovir.24
Given the efficacy of the ganciclovir implant in providing long-term control of active retinitis, implantation of this device was combined with repair of CMV retinitis- related retinal detachment in this series. This study corroborates prior reports of the success of silicone oil tamponade in achieving a high rate of retinal reattachment.4'7'12'25 However, past functional results have been less desirable and these prior studies indicate that control of CMV retinitis, particularly in the macula, is essential for an optimal visual outcome. The addition of the ganciclovir implant in the current series addresses this issue. Although limited by sample size, visual acuity was generally preserved from preoperative level in those presenting with the macula attached, and improved in those presenting with the macula detached. Best postoperative acuity was better than or equal to 20/100 in 7 of 10 patients. Freeman et al12 propose that CMV retinitis patients with macula off detachments may attain good postoperative acuity due to a wellformed vitreous. This may act to iimit the height of subretinal fluid and the exchange of vitreous fluid within the subretinal space. The 1 patient with macular detachment who experienced postoperative visual loss was delayed 9 days before repair. Delay in surgical treatment has also been reported as a risk factor for poor visual outcome.8
Table. Results of Patients Who Received Implant for Repair of CMV Retinitis-Related Retinal Detachment
It was initially unclear whether the ganciclovir implant could function properly in a silicone oil-filled vitreous cavity given the hydrophilic nature of ganciclovir. The present series, however, suggests the implant may function well in silicone oil. Of the 3 eyes that presented with inactive retinitis, there was no recurrence over the entire postoperative period. Eight of the 1 0 patients showed no active disease at last follow up and at no time during postoperative care did any patient experience progression or reactivation of retinitis. One suggested mechanism by which the ganciclovir implant may function in silicone oil is through diffusion into the peripheral vitreous. As movement occurs into the aqueous phase, the solution is redistributed along the retinal surface at the oil-retina interface. The risk of ocular toxicity may theoretically increase from higher concentrations of ganciclovir in the reduced aqueous phase of a silicone oil-filled eye. In this small series, however, no such adverse effects were noted.
Generally, patients receiving ganciclovir implants also receive oral ganciclovir to limit CMV infection systemically and in the fellow eye. This combined procedure can facilitate potent combination antiviral therapy in which patients receive systemic antiviral therapy (such as foscarnet or cidofovir) in addition to the local ganciclovir implant for refractory retinitis. One study showed a benefit of combination intravenous foscarnet and ganciclovir versus monotherapy for relapsed retinitis, but the combination antiviral therapy group also experienced the greatest number of drug-related toxicides.26 The ganciclovir implant could provide the treatment benefits associated with combination therapy while obviating the additional time-consuming daily administration of systemic agents along with their associated toxicities.
This study has several limitations including its retrospective design and small sample size. However, results indicate that patients benefit from excellent anatomic reattachment rates, preservation or improvement of visual acuity in most cases, and extended control of their CMV retinitis. The 2 procedures appear compatible without significant prohibitive complications. This technique, with its high intravitreal antiviral levels as well as its potential to facilitate potent combination antiviral therapy, may be particularly well-suited for those cases of refractor CMV retinitis that progress to retinal detachment. Futther study is warranted to clarify anatomic and functional outcomes in a larger treatment group.
1. Holland ON, Pepose IS, Pettit TH, et al. Acquired immune deficiency syndrome. Ocular manifestations. Ofhthalmology. 1983;90:859-873.
2. Jabs DA, Enger C, Bárdete JG. Cytomegalovirus retinitis and acquired immune deficiency syndrome. Arch Ophthalmoi 1989;107:75-80.
3. Freeman WR, Henderly DE, Wan WL, et al. Prevalence, pathophysiology, and treatment of rhegmatogenous retinal detachment in treated cytomegalovirus retinitis. Am ] Ophthalmoi 1987;103:527-536.
4. Jabs DA, Enger C, Haller J, et al. Retinal detachments in patients with cytomegalovirus retinitis. Arch Ophthalmoi. 199I;109:794-799.
5. Studies of Ocular Complications of AIDS Research Group in Collaboration with the AIDS Clinical Trials Group. Rhegmatogenous retinal detachment in patients with cytomegalovirus retinitis: the foscarnet-ganciclovir cytomegalovirus retinitis trial. Am J Ophthalmoi. 1997;124:61-70.
6. Martin DF, Parks DJ, Mellow SD, et al. Treatment of cytomegalovirus retinitis with an intraocular sustained release ganciclovir implant: a randomized controlled clinical trial. Arch Ophthalmoi \ 994; 112:1531-1539.
7. Kuppermann BD, Flores-AguÜar M, Ouiceno JI, et al. A masked prospective evaluation of outcome parameters for cytomegalovi rus- related retinal detachment surgery in patents with acquired immune deficiency syndrome. Ophthalmology. 1994;101:46-55.
8. Um JL, Enger C, Haller JA, et al. Improved visual results after surgical repair of cytomegalovirus-related retinal detachments. Ophthalmology. 1994;101:264-269.
9. Garcia RF, Flores-AguÜar M, Quiceno JL, et al. Results of rhegmatogenous retinal detachment repair in cytomegalovirus retinitis with and without scierai buckling. Ophthalmology. 1995;102:236-245.
10. Sandy CJ, Bloom PA, Graham EM, et al. Retinal detachment in AIDS-related cytomegalovirus retinids. Eye. 1995:9:277-281.
11. Regillo CD, Vander JF, Duker JS, et al. Repair of retinitis-related retinal detachments with silicone oil in patients with acquired immunodeficiency syndrome. Am J Ophthalmol. 1992;113:21-27.
12. Freeman WR, Qunceno JI, Crapotta JA, et al. Surgical repair of rhegmatogenous retinal detachment in immunosuppressed patients with cytomegalovirus retinids. Ophthalmology. 1992;99:466-474.
13. Dugel PU, Liggett PE, Lee MB, et al. Repair of retinal detachment caused by cytomegalovirus retinids in patients with the acquired immunodeficiency syndrome. Am J Ophthalmol 1991;112:235-242.
14. OrellanaJ,Teich SA, Lieberman RM, et al. Treatment of retinal detachments in patients with the acquired immune deficiency syndrome. Ophthalmology. 1001; 98:939-943.
15. Sanborn G, Anond R, Turn RE, et al. Sustained release ganciclovir therapy for treatment of cytomegalovirus retinids: use of an intravitreal device. Arch Ophthalmol. 1992;110:188-195.
16. Kupperman BD, Quiceno JI, Flores-Aguilar M, et al. Intravitreal ganciclovir concentration after intravenous administration in AIDS patients with cytomegalovirus retinitis: implications of therapy. J Infect Dis. 1993;168:1506-1509.
17. Jabs DA, Newman C, deBustros S, Polk BF. Treatment of cytomegalovirus retinitis with ganciclovir. Ophthalmology. 1987:94:824-830.
18. Jabs DA, Wingard JR, deBustros S, et al. BW5759U for cytomegalovirus retinids: intraocular drug penetration. Arch OphthalmoL 1986;104:1436-1437.
19. Mursch DC, Mardn DF, Gordon JF, et al. Treatment of cytomegalovirus retinitis with a sustained release ganciclovir implant. N EnglJ Med. 337:83-90.
20. Studies of Ocular Complications of AIDS Research Group in Collaboration with the AIDS Clinical Trials Group. Foscarnet-ganciclovir cytomegalovirus retinitis trial, 4: visual outcomes. Ophthalmology. 1994;101: 1250-1261.
21. Studies of Ocular Complications of AIDS Research Group in Collaboration with the AIDS Clinical Trials Group. Mortality in patients with die acquired immunodeficiency syndrome treated with either foscarnet or ganciclovir for cytomegalovirus retinitis. N EnglJ Mea. 1992;326:213-220.
22. Lalezari JP, Stagg RJ, Kupperman ED, et al. Intravenous cidovovir for peripheral cytomegalovirus retinitis in patients with AIDS: a randomized, controlled trial. Ann intern Med. 1997;126:257-263.
23. Studies of Ocular Complications of AIDS Research Group in Collaboration with the AIDS Clinical Trials Group. Parenteral cidofovir for cytomegalovirus retinitis in patients with AIDS: the HPMPC peripheral cytomegalovirus retinitis trial: a randomized, controlled trial. Ann Intern Mea. 1997;126:264-274.
24. Kempen JH, Chuang EL. Use of a sustained-release ganciclovir implant for management of previously treated CMV retinitis. Investig Ophthalmol Vis Sci. 1996,37(3):S371.
25. Davis JL, Serfass MS, Lai MY, et al. Silicone oil in repair of retinal detachments caused by necrotizing retinitis in HIV infection. Arch Ophthalmol. 1995;113:1401-1409.
26. Studies of Ocular Complications of AIDS Research Group in Collaboration with the AIDS Clinical Trials Group. Combination foscarnet and ganciclovir thetapy versus monotherapy for the treatment of relapsed cytomegalovirus retinitis in patients with AIDS. Arch Ophthalmol. 1996; 114:23-33.
Table. Results of Patients Who Received Implant for Repair of CMV Retinitis-Related Retinal Detachment