Anterior Nongranulomatous Uveitis after Intravitreal HPMPC (Cidofovir) for the Treatment of Cytomegalovirus Retinitis

Anterior N ongranulomatous Uveitis after lntravitreal HPMPC ( Cidofovir) for the Treatment of Cytomegalovirus Retinitis Analysis and Prevention

Eugenio Chavez-de Ia Paz, MD, 1 ]. Fernando Arevalo, MD/ Leonard S. Kirsch, MD/

David Munguia, BS/ Firas M. Rahhal, MD/ Erik DeClercq, MD,2

William R. Freeman, MD1

Background and Objective: The authors characterize and analyze the incidence of a previously reported mild anterior nongranulomatous uveitis associated with intravitreal injections of (S)-1-(3-hydroxy-2-phosphonylmethoxypropyl)cytosine (HPMPC), also termed cidofovir (Vistide, Gilead Sciences, Foster City, CA). This is an acyclic nucleoside phospho­nate analogue with a potent anticytomegalovirus effect. The authors also analyzed the effects of probenecid therapy, as well as prophylaxis with probenecid plus topical corticoste­roids and cycloplegics on the course and outcome of the uveitis.

Methods: Prospective case series from a tertiary referral center, which included 46 consecutive patients with acquired immune deficiency syndrome (AIDS) and cytomegalovi­rus (CMV) retinitis. There was a total of 130 injections in 69 eyes treated with 20 J..Lg of intravitreal HPMPC. Forty-one patients (119 injections) received oral probenecid, 5 patients (11 injections) did not, and 21 patients (53 injections) received topical corticosteroids and cycloplegics as an adjuvant to probenecid in the prophylaxis of iritis .

. Results: Mild to moderate nongranulomatous iritis was seen in 26% of patients after their first injection (n = 12). Patients receiving probenecid prophylaxis after first injection had a significantly lower frequency of iritis versus patients who did not receive probenecid at the time of first injection (P = 0.0089). In contrast, treatment with topical corticosteroid and cycloplegics after injection did not statistically significantly affect the frequency of iritis in patients (P = 0.44). The development of iritis after a second injection of HPMPC was more likely if it had occurred after the initial injection (P = 0.015; Fisher's exact test). All cases of iritis were treated with topical corticosteroids and cycloplegics, and there was no permanent impairment of vision secondary to iritis after HPMPC injection in any eyes.

Conclusions: Anterior uveitis was seen in 26% of patients after first-time HPMPC injection. Concomitant use of probenecid appears to decrease the frequency of the iritis from 71% to 18% in patients with AIDS and CMV retinitis after the first intravitreal injection of HPMPC. Topical corticosteroid administration after injection (before iritis) was ineffective in preventing iritis treatment with topical corticosteroids and cycloplegics resulted in resolu­tion of all iritis cases. Ophthalmology 1997; 104:539-544

Originally received: December 7, 1995. Revision accepted: November 25, 1996. 1 Department of Ophthalmology, Shiley Eye Center, University of Cali­fornia San Diego, La Jolla, California.

Supported in part by Research Grant EY07366 from the National Insti­tutes of Health and the University-Wide AIDS Research Program of the University of California (Dr. Freeman).

2 Rega Institute for Medical Research, Katholieke Universiteit Leuven, Leuven, Belgium.

Reprint requests to William R. Freeman, MD, Department of Oph­thalmology, Shiley Eye Center 0946, La Jolla, CA 92093-0946.

539

Ophthalmology Volume 104, Number 3, March 1997

Cytomegalovirus (CMV) retinitis is the most common cause of intraocular infection and vision loss in immuno­suppressed patients, particularly those with acquired im­mune deficiency syndrome. 1•

2 The only approved thera­pies currently available for CMV retinitis are daily intra­venous or oral administration of the antiviral agents ganciclovir or foscamet. 3-

5 In patients with acquired im­mune deficiency syndrome, treatment with ganciclovir or foscarnet must be maintained for the duration of the pa­tient's life because both drugs are virostatic, and discon­tinuation of therapy quickly leads to progressive retinitis, resulting in severe visual loss and blindness.6 There may be difficulty with intravenous antiviral therapy because of toxic side effects or intravenous catheter complications, which may lead to discontinuation of the antiviral drug and progression of the disease? Even with daily intrave­nous infusions of ganciclovir or foscamet, progression of retinitis occurs after a median time of 8 weeks.8 The time to progression of retinitis decreases after each successive reactivation.9 The reasons for the difficulty in controlling retinitis are unclear, but they may be related to low intra­vitreal levels of foscarnet and ganciclovir after intrave­nous administration. Ganciclovir- and foscamet-resistant strains of CMV also have been isolated. 10 In addition, previous reports from our laboratory have demonstrated that intravitreal levels of ganciclovir or foscamet after intravenous treatment are at or below those required to inhibit 50% of the CMV viral growth. 11

Recently, the use of intravitreal antiviral agents in the treatment of CMV retinitis has become more prevalent. 12

Treatment with intravitreal injections of ganciclovir or foscamet at different doses have been reported, but fre­quent injections are required to prevent reactivation. 13

-22

Even with weekly injections of 400 J.Lg of ganciclovir, the median time to reactivation is 8 weeks.23 A newly approved intravitreal sustained-release ganciclovir im­plant is available for local therapy; however, surgery, with its attendant potential complications, is required and experience with the use of the device is still limited. 24

-26

(S)-l-(3-hydroxy-2-phosphonylmethoxypropyl)cytosine (HPMPC), also termed Cidofovir, is a potent anti-CMV acyclic nucleoside phosphonate analogue.27

-35 When ad­

ministered intravenously, nephrotoxicity is a dose-limiting toxic effect of HPMPC. Concomitant administration of pro­benecid orally appears to ameliorate this problem; however, its ability to control CMV retinitis is still under investiga­tion.

We recently reported dose-ranging and efficacy studies of intravitreal HPMPC in the local treatment of CMV retinitis in humans.36

•37 Intravitreal HPMPC therapy alone

was effective in healing the retinitis for a prolonged time; the median time to the retinitis progression (determined by masked fundus photograph reading) after a single 20-J.Lg HPMPC dose was 49 days (Kaplan-Meier analysis).36

The long duration of antiviral effect may allow infrequent dosing (every 6-8 weeks) and better control of retinitis. In our initial series of patients, no retinal toxicity was observed clinically. The most significant side effects were a dose-related hypotony and vitritis noted in eyes treated with HPMPC at doses of l 00 J.Lg, and in one of three eyes treated with 40 J.Lg. Conversely, the incidence of

540

hypotony causing vision loss-even after repeated injec­tions of 20 J.Lg-was extremely small.36

"37 The indications

for treatment were progressive active retinitis despite ade­quate intravenous therapy with ganciclovir or foscamet, intolerance to intravenous therapy, or noncompliance with or refusal of intravenous ganciclovir or foscamet.

The objective of this study was to determine the inci­dence and characteristics of anterior nongranulomatous uveitis in the first 46 consecutive patients treated with HPMPC. In addition, we hoped to determine whether the inflammation recurs after repeated injections of HPMPC, as well as the possible relationship of probenecid therapy and prophylaxis with probenecid plus topical corticoste­roids and cycloplegic on the course and outcome of the uveitis.

Patients and Methods

The 46 consecutive patients included in this report belong to a study protocol conducted as a prospective consecutive case series, open label clinical trial, to determine the effi­cacy of intravitreal HPMPC as sole therapy for CMV retinitis. These were consecutive patients treated with 20 J.Lg of HPMPC from April 19, 1993, to August 16, 1994. This report excludes patients treated with 10-, 40-, and 100-J.Lg injections as a part of a preliminary dose-ranging study?6 The inclusion criteria for patients in this study were patients with acquired immune deficiency syndrome, patients with progressive active CMV, retinitis despite adequate intravenous therapy with ganciclovir or foscar­net, patients with intolerance to current therapies, and patients' refusal of intravenous therapy. Patients were ex­cluded from the study if they had a history of ocular inflammation or infection not attributable to CMV, active extraocular CMV, previous ocular surgery, or retinal de­tachment in the study eye.

The patients received baseline and follow-up examina­tions of both eyes, including prospective visual acuity and periodic refraction measured with Early Treatment of Diabetic Retinopathy Study (ETDRS) charts,38 medical history including medications, slit-lamp examination, in­traocular pressure measurement by applanation tonome­try, and dilated indirect ophthalmoscopy with fundus drawings on a standardized form. In addition, fundus pho­tography of both eyes was performed at each visit. Pa­tients were examined at weekly or biweekly intervals until death, lost to follow-up, or conclusion of the study.

On the day of HPMPC injection, patients received oral probenecid-2 g 3 hours before the injection and 1 g 2 and 8 hours after the procedure. Five patients (11 injec­tions) who had a history of sulfa allergy or who were noncompliant did not receive oral probenecid at any time during the study. (There was a total of 17 injections given without probenecid.)

HPMPC was injected via the pars plana into the vitre­ous cavity according to a standard protocol. In brief, for the intravitreal injections, the eye was exposed with a lid speculum, a topical anesthetic (proparacaine 0.5%) was instilled, and the injection site was prepared with povi­done iodine applicators and then irrigated with sterile

Chavez~de laPaz · Anterior Nongranulomatous Uveitis after lntravitreal HPMPC

Figure 1. Patient 1. Ocular injection associated with active anterior cham­ber inflammatory reaction with posterior synechiae formation in the left eye, 5 days after 20 j.1g of intravitreal HPMPC (Cidofovir).

normal saline solution. After subconjunctival injection of 2% lidocaine with epinephrine, a 0.1-ml solution con­taining 20 f.Lg of sterile HPMPC was injected intravitreally using a 30-gauge needle attached to a tuberculin syringe. The diluent for the HPMPC was normal saline. The infer­otemporal or superotemporal quadrants 4 mm posterior to the limbus were used for the injection. Immediately after the injection, a cotton applicator was placed over the injection site to prevent intraocular fluid efflux. If there was no perfusion of the central retinal artery for more than 60 seconds after injection, an anterior chamber paracentesis was performed. Patients received antibiotic ointment at the conclusion of the procedure.

Patients in whom iritis developed received topical cyclopentolate 1% or atropine 1% three times daily and topical prednisolone acetate 1% every 2 to 3 hours. All iritis cases were symptomatic and had at least 2+ cell and flare, which was an increase from baseline. Grading of cell and flare was performed under high-power magni­fication by focusing into the anterior chamber with a small bright beam from the slit lamp. The rating system has

d . 1 19 been reporte previOus y. · . A second phase prospective study of 53 sequential

injections was performed in which all eyes received topi­cal cycloplentolate l% twice daily and prednisolone ace­tate 1% four times daily, between days 5 and 14 after the injection, to determine if steroid treatment after HPMPC injection would help prevent iritis. Therefore, in this pr?­spective substudy, all patients receiv~d oral probenec~d before each injection and topical stermds and cycloplegia after injection to attempt to prevent iritis.

The statistical software used for this study was JMP 3.0 for the Macintosh (SAS Institute, Cary, NC). Data were entered into a database before transfer to the statisti­cal program.

Results

Forty-six patients were included in the study. These were consecutive patients treated with 20 J.Lg of HPMPC be-

Figure 2. Same patient 15 days postinjection. Iritis resolved when treated with mydriatic/cycloplegic agents and topical corticosteroids; these medi­cations then were tapered over 2 weeks. Posterior synechiae and keratic precipitates were still present.

tween April 19, 1993, and August 16, 1994. There was a total of 130 injections in 69 eyes. Of the 130 injections, there were 30 cases of iritis (23% ). Of the 46 patients (46 first injected eyes), the total incidence of iritis was 12 (26%).

The universal symptom of iritis was photophobia that appeared between 3 to 12 days after intravitreal injection (median = 4 days). In all cases, the iritis severity was 2 to 3+ cells plus 2 to 3+ flare. No eyes had 4+ cells or hypopyon. Fine nongranulomatous keratic precipitates were seen in all iritis episodes. The duration of iritis ranged from 5 to 10 days. Posterior synechiae were seen in 37% of the total number of cases of iritis (11 of 30). In more severe cases (n = 5), there was miosis, low intraocular pressure, and injection of the perilimbal blood vessels (Figs 1-3). Symptoms were pain, red eye, photo-

Figure 3. Patient 2. Active inflammatory reaction with formation of a fibrin plaque over the lens and posterior synechiae. After 14 days of topical cycloplegic agents and topical corticosteroids, the acute process resolved with no visual threatening sequelae.

541

Ophthalmology Volume 104, Number 3, March 1997

phobia, mildly decreased vision, and tearing. Therapy with topical prednisolone acetate 1% every 2 to 3 hours four to six times daily, and cycloplegia resulted in resolu­tion of iritis in all cases within 2 weeks.

We evaluated the incidence of iritis in relation to oral probenecid use and topical steroid treatment after injec­tion analyzing only the first injection of HPMPC. The unit of analysis was the patient. The total incidence of iritis by first injected eye (by patient) was 26%. Further, in five of the seven patients who did not receive probene­cid, iritis developed (71% ). Of the 39 patients that did receive probenecid at the time of the first injection, iritis developed in only 7 (18%). The association between ad­ministration of probenecid and the lower incidence of iritis was statistically significant (P = 0.0089; Fisher's exact test).

Analysis of the incidence of iritis by injection showed similar results. Of the 130 total injections, iritis developed in 30 (23% ). Probenecid was administered before HPMPC injection in 113 cases, with the rate of iritis in this group being 19% (n = 21). Of the 17 patients (injec­tions) that did not receive probenecid, iritis developed in 9 (53%). The association between administration of probenecid and the absence of iritis was statistically sig­nificant (P = 0.0039; Fisher's exact test).

An analysis of the relationship between administered topical steroids/cycloplegia (after injection and before de­velopment of iritis) and the development of iritis also was performed for first injections in patients (analysis by first eye injected by patient). Of 46 first injected eyes, 11 received steroid/cycloplegia treatment after injection; iri­tis developed in 4 of these 11 eyes. Thirty-five eyes did not receive steroid/cycloplegic treatment; iritis developed in 8 of these. Steroid treatment given after HPMPC injec­tions did not affect the incidence of iritis (P = 0.44, Fisher's exact test). Analysis by injection revealed similar results. Of the 130 total injections, iritis developed in 30 (23% ). Fifty-three injected eyes received steroid/cyclo­plegia treatment after injection, with a total incidence of iritis of 25% (n = 13), whereas the respective incidence of iritis for those eyes not receiving this treatment was 22% (17 of77). The association between steroid treatment and absence of iritis was not statistically significant (P = 0.74; chi square).

In two patients, an anterior chamber paracentesis was required after 0.1 ml of HPMPC was injected. There was no relationship between the iritis and the paracentesis.

Next, we investigated the relationship between the de­velopment of iritis after a second injection and before positive history of iritis after the first injection in the same eye. The unit of analysis was the patient. There was a total of 24 patients who received a second injection in the initially treated eye and who had sufficient follow-up (visible for 4 weeks after injection) for analysis. Of these 24 patients, iritis developed in 7 immediately after the first injection, with a second occurrence of iritis developing in that same eye in 4 (57%) of that subgroup. Conversely, of those patients in whom iritis did not develop after a first injection and who received a second injection in the same eye (n = 17), iritis subsequent to the second injec­tion developed in one (6%). This association between

542

development of a second episode of iritis and the first episode of iritis in the same eye was statistically signifi­cant (P = 0.015; Fisher's exact test).

Additional analysis was performed to determine if de­velopment of a first event of iritis in one eye in patients who received bilateral treatment with HPMPC was associ­ated with the incidence of iritis in the fellow eye. There were 23 patients who had bilateral treatment and who had sufficient follow-up examinations. Of the seven patients who had bilateral treatment and in whom iritis developed after a first injection, iritis developed in the fellow eye of four (57%) after that eye's initial injection. Of the patients in whom iritis did not develop after the first injec­tion (n = 16), iritis developed in the fellow eye in one (6%; P = 0.017; Fisher's exact test). Thus, there is a strong association with the development of iritis in the fellow eye after its first injection if iritis developed in the contralateral eye after that eye's first injection.

Finally, we evaluated the incidence of iritis, synechiae, and treatment with corticosteroids and cycloplegia after injection. When considering first-time injected eyes of those patients in whom iritis developed, synechiae devel­oped in 3 of 12 (25%). Synechiae developed in two of the four patients (50%) injected at first visit and treated with steroids/cycloplegia after the injection, whereas sy­nechiae developed in only one of the eight (13%) patients not treated prophylactically. The difference was not sig­nificant (P = 0.24; Fisher's exact test). When pooling all injections (n = 130), of the 30 (23%) patients in whom iritis developed, synechiae subsequently developed in 11 (37%). Synechiae developed in 7 of the 13 (54%) patients who received corticosteroid cycloplegic postinjection, whereas synechiae developed in 4 of 17 (24%) of those eyes not treated after injection with steroids and cyclople­gia. There was no association between synechiae and lack of steroid/cycloplegic treatment (P = 0.1322; Fisher's exact test).

Discussion

In this study, we report the occurrence of anterior non­granulomatous uveitis in 24% of patients after HPMPC intravitreal injections. This appears to be the most com­mon side effect of this promising medication, which ap­pears to prevent CMV retinitis progression for a median of 49 days after a single 20-J-Lg injection?6

•37

Topical and systemic medications have been reported previously as a cause of an anterior uveitis. Topical agents such as mitomycin-C40, and systemic agents such as co­caine,41 BCG,42 rifabutin,43 and sulfonamides,44 are some examples. In all cases, the acute process of iritis was associated with the concurrent use of the respective ther­apy. Discontinuation of the medication was associated with a decrease in the inflammation, and reinstitution was related to a reactivation of the process.

The symptoms of our patients with uveitis associated with HPMPC included pain, red eye, photophobia, and tearing. This uveitis usually appeared 5 to 12 days after intravitreal injection of 20 J-Lg of HPMPC. The clinical signs were cells and flare in the anterior chamber and fine

Chavez-de laPaz · Anterior Nongranulomatous Uveitis after lntravitreal HPMPC

keratic precipitates. In a few cases, posterior synechiae developed and generally were associated with a low intra­ocular pressure. Usually the duration of the process was 5 to 10 days, with a good response to topical cycloplegics ( cyclopentolate 1%) and steroids (prednisolone acetate 1% ). The treatment doses were higher than the doses used to attempt to prevent iritis.

The development of a more severe inflammation or posterior synechiae was seen in eyes that had a preceding episode of mild anterior uveitis after a previous HPMPC injection. It may be that more severe inflammation was a result of a cumulative dose of HPMPC. Alternatively, the eye may have become "sensitized" after the first injection, with subsequent HPMPC administration elic­iting an increased inflammatory reaction. The iritis re­solved when treated with mydriatic/cycloplegic agents and topical corticosteroids; these medications then were tapered over 2 to 3 weeks.

There was a significant difference in the incidence of iritis in patients who had received oral probenecid versus those who were not treated with oral probenecid. Con­comitant use of probenecid appears to dramatically de­crease the incidence of the iritis in patients with acquired immune deficiency syndrome and CMV retinitis being treated with intravitreal injections ofHPMPC. Probenecid is a highly lipid-soluble benzoic acid derivative that acts as a competitive anion transport inhibitor in epithelial barriers. We postulate that probenecid may suppress or decrease the absorption of HPMPC (which is an organic acid) or metabolite derivatives into the ciliary body, thus decreasing the possibility of a toxic effect.45 Alternatively, patients who did not receive probenecid (in our series usually because of a sulfa allergy) may in some way be more sensitive to the uveitogenic effect of any agent. We found no cases of severe adverse reactions to probenecid in our series, even among patients with a history of sulfa allergy; this may be because of the infrequency of admin­istration of probenecid in our protocol. The incidence of the inflammation appeared to be less in those patients who had received oral probenecid; therefore, we continue to recommend its use. Rifabutin may cause iritis; how­ever, it was not widely available in our community during this study. Because the incidence of rifabutin uveitis is low (less than 1% ), it is unlikely that it could cause the uveitis seen in our cases.43 However, it is possible that Cidofovir can potentiate rifabutin uveitis.

The incidence of iritis increased after a second HPMPC injection in eyes in which iritis developed after a first injection. Similarly, contralateral eyes of patients in which iritis developed were more likely to be affected. This information will be of value in determining who is at risk for this complication.

HPMPC is a highly effective intravitreal therapy for CMV retinitis in patients with acquired immune defi­ciency syndrome because of its effectiveness in healing and impeding the advancement of retinitis, as well as its long duration of therapeutic action?7

•38 The use of oral

probenecid does reduce the incidence of iritis from 71% to 18% of first-time injections. The use of topical cyclop1egics and steroids for 5 to 12 days after the injec­tion did not reduce the incidence of iritis. The prompt

administration of higher doses of topical cycloplegics and steroids at the onset of the iritis appears to control the iritis and prevent vision-threatening complications in all cases. Further analyses of the long-term effects of iritis in a larger cohort of patients is warranted particularly to determine whether posterior synechiae worsen over time and whether cataract development in iritis cases becomes a problem. It should be noted that HPMPC became com­mercially available as an intravenous formulation after completion of our study. Our HPMPC was not synthe­sized by Gilead Sciences (Foster City, CA), the manufac­turer of intravenous HPMPC. However, we have assayed our preparation and found that it contains approximately 100% less active HPMPC than an equivalent amount of commercially available HPMPC.46 Thus, the 20-f.Lg dose we used probably is equivalent to 18 f.Lg of intravenous HPMPC, but we have not performed any formal equiva­lent studies.

References

1. Freeman WR, Lerner CW, Mines JA, et al. A prospective study of the ophthalmologic findings in the acquired im­mune deficiency syndrome. Am J Ophthalmol 1984; 97:133-42.

2. Holland GN, Pepose JS, Pettit TH, et al. Acquired immune deficiency syndrome: ocular manifestations. Ophthalmol­ogy 1983;90:859-73.

3. Henderly DE, Freeman WR, Causey DM, Rao NA. Cyto­megalovirus retinitis and response to therapy with gan­ciclovir. Ophthalmology 1988;94:425-34.

4. Palestine AG, Stevens G Jr, Lane HC. Treatment of cyto­megalovirus retinitis with dihydroxy propoxymethyl gua­nine. Am J Ophthalmol 1986; 101:95-101.

5. Fanning MM, Read SE, Benson M, et al. Foscamet therapy of cytomegalovirus retinitis in AIDS. J Acquir Immune Defic Synd 1990;3:472-9.

6. Erice A, Chou S, Biron KK, et al. Progressive disease due to ganciclovir-resistant cytomegalovirus in immunocom­promised patients. N Engl J Med 1989;320:389-93.

7. Holland GN, Sakamoto MJ, Hardy D, et al. Treatment of cytomegalovirus retinopathy in patients with acquired im­munodeficiency syndrome. Arch Ophthalmol 1986; 104:1794-800.

8. Studies of Ocular Complications of AIDS Research group in collaborations with the AIDS Clinical Trials Group: mor­tality in patients with the acquired inmunodeficiency syn­drome treated with either foscamet or ganciclovir for cyto­megalovirus retinitis. N Engl J Med 1992;326:213-20.

9. Studies of Ocular Complications of AIDS Research group in collaborations with the AIDS Clinical Trials Group: fosc­amet-ganciclovir cytomegalovirus retinitis trial. Ophthal­mology 1994; 101:1250-61.

10. Knox K, Drobyski W, Carrigan D. Cytomegalovirus isolate resistant to ganciclovir and foscamet from a marrow trans­plant patient. Lancet 1991;255:1292-3.

11. Kuppermann BD, Quinceno JI, Flores-Aguilar M, et al. Intravitreal ganciclovir concentration after intravenous ad­ministration in AIDS patients with cytomegalovirus retini­tis: implications for therapy. J Infect Dis 1993; 168:1506-9.

12. Freeman WR. Intraocular antiviral therapy. Arch Ophthal­mol1989;107:1737-9.

543

Ophthalmology Volume 104, Number 3, March 1997

13. Heinemann MH. Long-term intravitreal ganciclovir therapy for cytomegalovirus retinopathy. Arch Ophthalmol 1989; 107:1767-72.

14. Harris ML, Mathalone MBR. Intravitreal ganciclovir in CMV retinitis: case report. Br J Ophthalmol 1989;73:382-4.

15. Cantril H, Henry K, Melroe H, eta!. Treatment of cytomeg­alovirus retinitis with intravitreal ganciclovir: long-term re­sults. Ophthalmology 1989, 96:367-74.

16. Maurice DM. Use of intravitreal ganciclovir (propoxy­methyl guanine) for cytomegalovirus retinitis in a patient with AIDS. Am J Ophthalmol 1987; 103:842-4.

17. Ussery FM, Gibson SR, Conklin RH, et a!. Intravitreal ganciclovir in the treatment of AIDS-associated cytomega­lovirus retinitis. Ophthalmology 1988; 95:640-8.

18. Henry K, Cantrill H, Fletcher C, et a!. Use of intravitreal ganciclovir (dihydroxy propoxymethyl guanine) for cyto­megalovirus retinitis in a patient with AIDS. Am J Ophthal­mol 1987; 103:17-23.

19. Daikos GL, Pulido J, Kathapalia SB, Jackson GG. Intrave­nous and intraocular ganciclovir for CMV retinitis in pa­tients with AIDS or chemotherapeutic immunosuppression. Br J Ophthalmol 1988;72:521-4.

20. Schulman J, Peyman GA, Horton MB, eta!. Intraocular 9-([2-hydroxy-1-{hydroxymethyl }ethoxy ]methyl)guanine levels after intravitreal and subconjunctival administration. Ophthalmic Surg 1986; 17:429-32.

21. Cochereau-Massin I, Lehoang P, Lautier-Frau M, eta!. Ef­ficacy and tolerance of intravitreal ganciclovir in cytomega­lovirus retinitis in acquired immune deficiency syndrome. Ophthalmology 1991;98:1348-55.

22. Biichi ER, Fiitting PL, Michel AE. Long-term intravitreal ganciclovir for cytomegalovirus retinitis in a patient with AIDS. Arch Ophthalmol 1989; 106:1349-50.

23. Cochereau-Massin I, Lehoang P, Lautier-Frau M, eta!. Ef­ficacy and tolerance of intravitreal ganciclovir in cytomega­lovirus retinitis in acquired immune deficiency syndrome. Ophthalmology 1991;98:1348-53 [discussion 1353-5].

24. Smith TJ, Pearson PA, Blandford DL, et a!. Intravitreal sustained-release ganciclovir. Arch Ophthalmol 1992; 110:255-258.

25. Sanborn GE, Anand R, Torti RE, et a!. Sustained-release ganciclovir therapy for treatment of cytomegalovirus retini­tis: use of an intravitreal device. Arch Ophthalmol 1992; 110:188-95.

26. Martin DF, Parks DJ, Mellow SD, et a!. Treatment of cyto­megalovirus retinitis with an intraocular sustained-release gancilcovir implant: a randomized controlled clinical trial. Arch Ophthalmol 1994; 112:1531-9.

27. DeClercq E, Walker RT. Chemotherapeutic agents for her­pes virus infections. Prog Med Chern 1986;23:187-218.

28. Votruba I, Bemaerts R, Sakuma T, eta!. Intracellular phos­phorylation of broad-spectrum anti-DNA virus agent (S)-9-(3-hydroxy-2-phosphonylmethoxypropyl)adenine and in­hibition of viral DNA synthesis. Mol Pharmacal 1987; 32:524-9.

29. Snoeck R, Sakuma T, DeClercq E, eta!. (S)-1-(3-hydroxy-2-phosphonylmethoxypropyl)cytosine, a potent and selec-

544

tive inhibitor of human cytomegalovirus replication. Anti­microb Agents Chemother 1988;32:1839-44.

30. De Clercq E, Sakuma T, Baba M, et al. Antiviral activity of phosphonylmethoxyalkyl derivatives of purine and py­rimidines. Antiviral Res 1987;8:261-72.

31. Shigeta S, Konno K, Baba M, et a!. Comparative inhibitory effects of nucleoside analogues on different clinical isolates of human cytomegalovirus in vitro. J Infect Dis 1991; 63:270-5.

32. Li SB, Yang ZH, Feng JS, et a!. Activity of (S)-1-(3-hydroxy-2-phosphonylmethoxypropyl)cytosine (HPMPC) against guinea pig cytomegalovirus infection in cultured cells and in guinea pigs. Antiviral Res 1990; 13:237-52.

33. Neyts J, Snoeck R, Balzarini J, De Clercq E. Particular characteristics of the anti-human cytomegalovirus activity of (S)-1-(3-hydroxy-2-phosphonylmethoxypropyl)cytosine (HPMPC) in vitro. Antiviral Res 1991; 16:41-52.

34. Neyts J, Snoeck R, Schols D, et a!. Selective inhibition of human cytomegalovirus DNA synthesis by (S)-1-(3-hy­droxy-2-phosphony lmethoxypropy !)cytosine [ ( S)-HPMPC] and 9-(1,3-dihydroxy-2-propoxymethyl)guanine (DHPG). Virology 1990; 179:41-50.

35. Matthews T, Boehme R. Antiviral activity and mechanism of action of ganciclovir. Rev Infect Dis 1988; 10:S490-S494.

36. Kirsch LS, Arevalo JF, De Clercq E. A prospective clinical trial of intravitreal (S)-1-(3-hydroxy-2-phosphonylmethox­ypropyl)cytosine (HPMPC) in the treatment of cytomegalo­virus retinitis in patients with the acquired immune defi­ciency syndrome. Am J Ophthalmol 1995; 119:466-76.

37. Kirsch LS, Arevalo JF, Chavez-delaPaz E, et al. Intravi­treal cidofovir treatment of cytomegalovirus retinitis in pa­tients with the acquired immune deficiency syndrome. Oph­thalmology 1995; 102:533-43.

38. Ferris FL III, Kassoff A, Bresnick GH, Bailey I. New visual acuity charts for clinical research. Am J Ophthalmol 1982; 94:91.

39. Hogan MH, Kimura SJ, Thygeson P. Signs and symptoms of uveitis. I. Anterior uveitis. Am J Ophthalmol 1959;47: 155.

40. Rubinfeld RS, Pfister RR, Stein RM, et a!. Serious compli­cations of topical mitomycin-C after pterygium surgery. Ophthalmology 1992;99:1647-54.

41. Wang ES. Cocaine-induced iritis. Ann Emerg Med 1991; 20:192-3.

42. Price GE. Arthritis and iritis after BCG therapy for bladder cancer. J Rheumatoll994;21:564-5.

43. Saran BR, Maguire AM, Nichols C, eta!. Hypopyon uveitis in patients with acquired immunodeficiency syndrome treated for systemic mycobacterium avium complex infec­tion with rifabutin. Arch Ophthalmol 1994; 112:1159-65.

44. Tilden ME, Rosenbaum JT, Fraunfelder FT. Systemic sul­fonamides as a cause of bilateral, anterior uveitis. Arch Ophthalmol 1991; 109:67-9.

45. Lesar T, Fiscella RG. Antimicrobial drug delivery to the eye. Drug Intel! Clin Pharm 1985; 19:642-54.

46. Freeman WR. New developments in the treatment of CMV retinitis [editorial]. Ophthalmology 1996; 103:999-1000.