Treatment of Cytomegalovirus Retinitis with Ganciclovir

Treatment of Cytomegalovirus Retinitis with Ganciclovir

Treatment of Cytomegalovirus Retinitis with Ganciclovir DOUGLAS A. JABS, MD,* CHERYL NEWMAN, MD,t SERGE DE BUSTROS, MD,* B. FRANK POLK, MDt Abstract:...

1MB Sizes 4 Downloads 269 Views

Treatment of Cytomegalovirus Retinitis with Ganciclovir DOUGLAS A. JABS, MD,* CHERYL NEWMAN, MD,t SERGE DE BUSTROS, MD,* B. FRANK POLK, MDt

Abstract: Eighteen immunocompromised patients with cytomegalovirus (CMV) retinitis were treated with ganciclovir, an investigational antiviral drug. CMV retinitis in association with acquired immune deficiency syndrome (AIDS) developed in 17 patients; CMV retinitis developed in one patient after cardiac transplantation. Fourteen patients responded to ganciclovir treatment with improvement in CMV retinitis. In 11 patients, the response was classified as complete; in three patients, the response was partial. Continued improvement in the retinitis was often seen while the patient was on maintenance treatment. Maintenance ganciclovir therapy was required; relapse occurred in five of seven patients in whom ganciclovir treatment was interrupted. The major limiting toxic side effect of ganciclovir was neutropenia, which necessitated temporary discontinuation of ganciclovir in five patients but was reversible in all cases. Ganciclovir appears to be an effective therapy for CMV retinitis, but chronic maintenance therapy is required. [Key words: acquired immune deficiency syndrome (AIDS), cytomegalovirus retinitis, ganciclovir, immunocompromised host.] Ophthalmology 94: 824-830, 1987

Cytomegalovirus (CMV) infection of the retina is a well-recognized complication of immunosuppression or immunodeficiency. Originally described in immunosuppressed patients or patients undergoing transplantation, I-J it is now most frequently reported in patients with the acquired immune deficiency syndrome (AIDS). 4 - 7 Recently, an investigational anti-viral agent, ganciclovir, an acyclic nucleoside analog of guanine, has become available for the treatment of serious CMV infections. The drug is also known as 9-( 1,3-dihydroxy-2propoxymethyl) guanine (DHPG, Syntex Research, Palo Alto, CA) or BW B795U (Burroughs Wellcome, Research Triangle Park, NC). Ganciclovir inhibits viral

From the Department of Ophthalmology,* the Wilmer Ophthalmological Institute, the Department of Medicine, Division of Infectious Diseases,t the Johns Hopkins University School of Medicine, and the Department of Epidemiology,t the Johns Hopkins University School of Hygiene and Public Health, Baltimore. Supported in part by outpatient clinical research center grant 5M01 RR0072, from the Division of Research Resources (NIH). Reprint requests to Douglas A. Jabs, MD, Uveitis and Clinical Immunology Service, the Wilmer Ophthalmological Institute, Wilmer 300, the Johns Hopkins Hospital, 600 N. Wolfe St, Baltimore, MD 21205.

824

replication in vitro, 8 and preliminary studies suggest that it is effective in suppressing CMV retinitis in humans. 9- 18 In this article, we report our experience using ganciclovir to treat 18 patients with CMV retinitis.

SUBJECTS AND METHODS From July 1985 to January 1987, 23 patients with CMV retinitis were seen at the Johns Hopkins Hospital. Eighteen patients were treated with ganciclovir (DHPG, Syntex Research, Palo Alto, CA) under an open-labeled, compassionate-care protocol. Informed consent was obtained in all cases. Patients were treated for 2 to 3 weeks in-hospital with 5 mg/kg of ganciclovir intravenously (IV) every 12 hours. The dose was reduced for renal insufficiency according to the protocol in Table 1. Patients 1 to 3 were treated for a period of 2 weeks in-hospital, and then treatment was assessed for effectiveness. All subsequent patients were treated for up to 3 weeks in-hospital and then placed on maintenance outpatient DHPG. Maintenance treatment consisted of5 to 6 mg/kg IV daily, 5 of every 7 days, through a Hickman catheter. The maintenance dose also was adjusted for renal function.

JABS et a/



GANC/CLOVIR FOR CMV RETINITIS

Table 1. Protocol by which Ganciclovir Was Reduced for Renal Insufficiency Creatinine Clearance (ml/min/1. 73 m2)

Dose (mgjkg)

Dosing Interval (hrs)

>50 25-50 10-25 0-10

5.0 3.0 3.0 1.5

12 12 24 24

Patients received a complete ophthalmologic examination before beginning treatment and at least weekly during the inpatient phase of treatment. Outpatients on maintenance therapy were followed at regular intervals. Outpatients were seen 1 week after discharge; if the patient was stable, he was seen 2 weeks later, and then on a monthly basis. Fundus photographs were routinely taken to document and follow the status of the CMV retinitis. Hematologic monitoring was initially performed every other day, and subsequently performed weekly. If the absolute neutrophil count fell below 500 cells/mm 3, DHPG was temporarily discontinued and re-instituted when the white count had again risen. For the purposes of this study, a response was defined as arresting the progression of retinitis coupled with a decrease in the area of active retinal inflammation. A complete response was defined as greater than 90% resolution of all ophthalmoscopically visible active CMV lesions. 11 A complete response often left the retina with an atrophic scar with residual glial, lipid, or calcium deposition and increased pigmentation. A partial response was defined as a decrease in the ophthalmoscopically active retinitis but less than a complete response. The response to maintenance therapy was defined as complete if complete resolution of a partial response was observed while the patient was on maintenance therapy. Suppression was defined as maintenance of either a partial or complete response, but without progression of a partial response to a complete one. Relapse was defined as the recurrence of clinically evident CMV retinitis after the discontinuation or during the interruption of ganciclovir therapy. Breakthrough was defined as the extension of active lesions or the occurrence of new active lesions while a patient was receiving maintenance ganciclovir.

RESULTS PATIENTS

Ganciclovir treatment of 18 patients is outlined in Table 2. Seventeen of the patients had AIDS; one patient had undergone cardiac transplantation. Seventeen of the 18 patients were men; one was a woman. Eight patients had bilateral CMV retinitis and ten had unilateral disease. Nine patients had involvement of the paste-

rior pole and nine had disease peripheral to the temporal vascular arcades. Sixteen of the patients were treated with an initial induction dose of 5 mg/kg every 12 hours; two were treated at a reduced dose of 3 mg/kg every 12 hours because of renal disease. Eleven patients were placed on maintenance therapy; seven patients did not receive maintenance treatment. The visual acuity of case 1 dropped from 5/200 to no light perception in both eyes 1 day before admission to the hospital for ganciclovir treatment; after a 14-day course, he was declared a treatment failure and was not placed on maintenance ganciclovir. Case 3 had a partial response to 14 days of full-dose ganciclovir, but declined maintenance therapy. Case 15 received 5 days of ganciclovir, but treatment was interrupted for fever and confusion. He than elected to be treated with azidothymidine and not ganciclovir. Four patients died either during induction treatment with ganciclovir (cases 2 and 17) or shortly after the induction phase (cases 10 and 14) and were not treated with maintenance ganciclovir. Three patients (cases l, 6,and 12) suffered retinal detachments (RDs) associated with CMV retinitis. In two patients (cases I and 6), extensive retinal damage from CMV retinitis created large retinal tears and rhegmatogenous RDs, which were present before being treated with ganciclovir. Case 12 presented with an exudative RD and optic atrophy, and the degree of RD appeared to parallel the presence or absence of visible retinal necrosis (active CMV retinitis). RESPONSE TO TREATMENT

Fourteen patients responded to ganciclovir treatment. Of these 14 patients (cases 3-14, 16, and 18), 11 had a complete response to either the induction treatment or subsequently while on maintenance therapy; three had a partial response. During the initial in-hospital treatment, six patients had a complete response and eight had a partial response. In five patients, a partial response at the end of the initial 3 weeks of treatment converted to a complete response while on maintenance ganciclovir. All patients but one (case 18) had positive CMV cultures from either blood or urine before being treated with ganciclovir; one patient (case 1) had positive urine cultures only, four (cases 6, 9, 15, and 17) had positive blood cultures only, and the other 12 had both positive blood and urine cultures. In all cases, cultures became negative during ganciclovir treatment. It was common for the ophthalmoscopically active lesions to show minimal extension during the first week of treatment but then to stabilize and begin to fade at 10 to 14 days of ganciclovir treatment. Complete fading of all ophthalmoscopically active lesions often occurred by day 21 of treatment, although patients with more extensive disease often continued to show improvement on maintenance therapy over the following several weeks (Figs 1-5). The four patients who did not respond to ganciclovir received less than 3 weeks of treatment. Case 1 lost all 825

OPHTHALMOLOGY

JULY 1987





VOLUME 94



NUMBER 7

Table 2. Ganciclovir Treatment for CMV Retinitis Dose of Ganciclovir (mgjkg)

Systemic Age Case No. (Yrs)/Sex Disease

Location CMV Retinitis

Initial Visual Acuity

Maintenance (5 of Initial Duration every Ganciclovir (every (wks)* 12 hrs) 7 days)

Treatment Response Induction

Maintenance

Final Visual Acuity

5 3

None None

2 2

None None

5

None

2

Partial

3

3

43

Partial

Complete

5

5

43

Partial

Complete

5

6

29

Complete

Suppression

5

6

25

Partial

Complete

20/20 OU

5

6

25

Complete

Suppression

LP OD 20/100 OS

5

6

7

Partial

Suppression

20/20 OU Peripheral OS 20/20 OU Peripheral OS Posterior OD 20/200 OD 20/20 OS 20/20 OU Peripheral

5

None

3

Complete

5

5

16

Complete

Suppression

20/20 OU

5

6

15

Partial

Complete

5

6

13

Partial

Complete

5

None

3

Partial

5

None

20/200 OD 20/20 OS 20/20 OD 20/25 OS 20/20 OD 20/100 OS 20/400 OD 20/20 OS

5

6

4

Partial

5

None

2

None

5

6

4

Complete

1 2

42/M 24/M

AIDS AIDS

3

35/M

AIDS

4

56/M

HT

5

60/M

AIDS

6

44/M

AIDS

7

34/M

AIDS

NLP OU HM OD 20/200 OS Posterior OD 20/400 OD 20/20 OS Posterior OU 20/20 OD 20/30 OS 20/30 OD Peripheral 20/25 OS OS 20/50 OD Peripheral 20/20 OS OD 20/20 OU Peripheral

8

43/M

AIDS

Peripheral

9

39/M

AIDS

Posterior OU

10

26/M

AIDS

11

41/M

AIDS

12

44/M

AIDS

13

31/M

AIDS

14

42/M

AIDS

15

40/M

AIDS

16

37/M

AIDS

17

35/F

AIDS

18

32/M

AIDS

Posterior OU Posterior OU

ou

ou

ou

20/20 OD Peripheral 20/25 OS OU Posterior OD 20/40 OD 20/20 OS

Peripheral OS Posterior OS

20/20 OU

20/100 OD 20/20 OS Posterior OD 20/60 OD 20/20 OS

NLP OU LP OD CF OS NLP OD 20/30 OD 20/200 OS 20/25 OU 6/200 20/20 20/25 20/20 20/20

Leukopeniat

+

+ +

OD OS OD OS OU

ou

20/20

Suppression

20/200 OD 20/30 OS 20/50 OD

Relapsed off ganciclovir; died Relapsed off ganciclovir Relapsed off ganciclovir; died RD, OD; died

Relapsed when noncompliant; complete response to reinduction; subsequent suppression Maintenance discontinued; Died Died

NLP OD 20/300 OS

Complete

RD, OU; died Died

Died

NA

None

Comment

+ RD, OD; breakthrough

+

Died Ganciclovir stopped for fever, confusion; treatment with AZT; progression CMV retinitis Died

CMV = cytomegalovirus; AIDS = acquired immune deficiency syndrome; HT = heart transplant; OU = both eyes; OD = right eye; OS = left eye; NLP = no light perception; HM = hand motions; LP = light perception; CF = count fingers; NA = not available; RD = retinal detachment; AZT = azidothymidine. * Total duration ganciclovir treatment including induction phase and maintenance treatment (if given). t Low leukocyte count is less than 500 granulocytes/mm 3.

vision 1 day before beginning ganciclovir treatment. He had extensive bilateral RDs and was declared a treatment failure after 2 weeks of ganciclovir treatment. Case 2 had renal disease requiring a reduced induction dosage, and leukopenia developed, which required frequent interruption of ganciclovir treatment. His CMV retinitis did not respond to induction therapy and he died while

826

on ganciclovir. Case 15 received only 5 days of ganciclovir, had treatment stopped for fever and confusion, and elected to go on azidothymidine rather than resume ganciclovir treatment. Case 17 had extensive optic nerve involvement and died from recurrent Pneumocystis carinii pneumonia while receiving induction treatment. Patients with involvement of the posterior pole by

JABS et al



GANCICLOVIR FOR CMV RETINITIS

Fig 1. Case 16. December 1986, left eye, showing active CMV retinitis between superior arcades.

Fig 3. Case 5. March 1986, left eye, showing active CMV retinitis along inferior nasal arcade.

Fig 2. Case 16. Four weeks later, left eye, showing resolution of active CMV retinitis.

Fig 4. Case 5. April 1986, left eye, showing partial resolution active CMV lesion.

CMV retinitis did not appear to do as well as those with retinitis peripheral to the arcades. Of the nine patients with posterior pole involvement, four (44%) had no response and only three (33%) had a complete response. Conversely, the nine patients with peripheral involve-

ment had a response to ganciclovir treatmen t and eight (89%) had a complete response. Not surprisingly, the visual outcome was better with peripheral than with posterior involvement. Of the 13 eyes with involvement of the posterior pole, 11 (85%) had a final visual acuity of

827

OPHTHALMOLOGY



JULY 1987 •

VOLUME 94



NUMBER 7

less than 20/200. Of the II eyes with peripheral involvement in whom follow-up data were available, only one (9%) had a final visual acuity of less than 20/200. COMPLICATIONS

Ganciclovir treatment was interrupted in seven patients; five had treatment interrupted for severe leukopenia (<500 granulocytes/mm 3). In all cases, ganciclovir-induced neutropenia was reversible upon discontinuation of ganciclovir. In addition, in case 5, ganciclovir treatment was held for 2 weeks due to intense pruritus. Withholding ganciclovir did not alter his pruritus, which was therefore not considered a side effect of the drug. Ten of the I8 patients have died. Four patients died either during the induction phase of treatment (cases 2 and I7) or shortly (within I week) thereafter (cases 10 and 14). Three patients died 1 to 10 months after discontinuation of maintenance ganciclovir (cases 1, 3, and 9). Three patients died while on maintenance ganciclovir (cases 5, 6, and 7). In all cases, death was due to other infectious complications of AIDS; and in no case was it believed that ganciclovir contributed to death. RELAPSE AND BREAKTHROUGH

Five patients suffered a relapse of active CMV retinitis while off therapy. Of the five patients in whom a relapse occurred, three responded to a second induction course of ganciclovir. Case 3 declined maintenance therapy, and his vision decreased from 20/400 to no light perception in his only affected eye over a 3-month period after treatment was halted. Case 4 had a partial response to induction ganciclovir treatment, but significant leukopenia developed and he was not placed on maintenance therapy for a period of 18 days. The CMV retinitis recurred and involved the macula of his left eye, resulting in a loss of visual acuity from 20/30 to 20/200. The CMV retinitis partially responded to a second course of induction therapy, and a complete response was ultimately achieved on maintenance therapy. Case 5 had maintenance ganciclovir discontinued for 14 days due to intense pruritus, and CMV retinitis recurred at the border of the previously atrophic scar. The retinitis responded completely to a second course of induction therapy and remained suppressed on maintenance treatment. Case 8 was noncompliant with outpatient maintenance therapy, had a relapse of CMV retinitis, and responded to reinduction. Case 11 had a complete response to 3 weeks of induction therapy, but significant leukopenia developed after 3 weeks of therapy. His granulocyte count normalized over the next 3 weeks, and maintenance therapy was reinstituted while the CMV retinitis was still in remission. He did not have a relapse. Case 15 had ganciclovir treatment stopped after 5 days for fever and confusion, elected not to resume ganciclovir, and was placed on azidothymidine; CMV retinitis progressed. One patient had breakthrough of active retinitis while on maintenance therapy. Case 12 presented with an ex828

Fig 5. Case 5. May 1986, left eye, showing atrophic scar with no active CMV retinitis.

udative RD, active CMV retinitis, and optic atrophy. There was a partial response of CMV retinitis to 3 weeks of induction therapy and a complete response after 4 weeks of maintenance ganciclovir. He was continued on maintenance ganciclovir at 6 mg/kg/day, 5 of every 7 days. One month later, he had a breakthrough of active CMV retinitis, and was switched to daily maintenance ganciclovir (5 mg/kg/day). Within 3 weeks, the active CMV retinitis was again quiet, and he has not had any further episodes of breakthrough while on daily maintenance. INTRAOCULAR DRUG LEVELS

Intraocular ganciclovir levels were measured using high-pressure liquid chromatography on the aqueous, vitreous, and subretinal fluid specimens from one patient (case 6) who underwent two retinal reattachment procedures. The first surgery was a conventional retinal reattachment procedure. Aqueous and subretinal fluid specimens were obtained 2112 hours after a 1-hour infusion of 6 mg/kg of maintenance ganciclovir with the following results: serum level, 6 JLm; subretinal fluid level, 3.6 JLm; and aqueous fluid level, 2.4 JLm. The second surgery was a vitrectomy 21 hours after a 1-hour infusion of 6 mg/kg of ganciclovir, and the following results were obtained: serum level, 0 JLm (undetectable); and vitreous fluid level, 0.8 JLm.

DISCUSSION Ganciclovir appears to be an effective drug for the treatment of CMV retinitis. Fourteen of 18 patients had

JABS et al



GANCICLOVIR FOR CMV RETINITIS

either a partial or complete response to treatment, and all patients who received greater than 2 weeks of ganciclovir responded. Although ganciclovir was effective in suppressing CMV infection, it did not eradicate the virus. This result has been found by other investigators using ganciclovir; and in AIDS patients, relapse is universal without continued maintenance therapy. 9 - 14 •16 •17 This experience has led to the recommendation that maintenance ganciclovir be given as a once-daily IV infusion through a permanent indwelling central IV catheter 5 of every 7 days. Interruption of maintenance ganciclovir treatment for as short as 2 weeks can result in recurrence of the active CMV retinitis, which occurred in five of our patients. Although several studies have documented the efficacy of ganciclovir for CMV retinitis, the optimum dose and duration of both induction and maintenance therapy have yet to be determined. The dose of ganciclovir used during the initial induction phase has ranged from 7.5 to 15.0 mg/kg daily, with most investigators using 2.5 mg/kg every 8 hours or 5.0 mg/kg every 12 hours. Similarly, the duration of induction therapy has ranged from 10 to 36 days, with most investigators using a 10-to-20-day induction phase. The dose of maintenance drug has been escalated up to its current recommendation of 6 mg/kg/day, 5 of 7 days, due to the frequent occurrence of breakthrough at lower doses or doses given less frequently. The recommended maintenance therapy has remained at 5 of 7 days rather than daily therapy due to concerns that leukopenia might be more likely to develop with daily therapy. We attempted to use a standardized protocol to treat patients with CMV retinitis. This protocol consisted of the use of an induction dose of 5 mg/kg every 12 hours for 21 days, with the dosage adjusted for renal failure if necessary. The duration of induction treatment was shortened if a complete response occurred before 21 days of treatment. Patients were then placed on maintenance therapy (6 mgjkg/day, 5 of every 7 days), again adjusted for renal function. The dosage and duration of induction treatment used in this study are among the highest reported. One result of our treatment protocol was the recognition that continued clinical improvement of the CMV retinitis occurred while patients were on maintenance therapy because five patients converted from a partial to a complete response on maintenance treatment. With this regimen, we saw breakthrough in only 1 of 11 patients (9%) placed on maintenance therapy, and the retinitis responded to a change in maintenance dosage from 5 of 7 days per week to daily ganciclovir. Although the numbers are small and may not be significantly different, this result contrasts with that in the study by Holland et al 17 where four of ten patients (40%) receiving continuous maintenance therapy for longer than 3 weeks suffered breakthrough re-activation of CMV retinitis. Whether our apparent lower frequency of breakthrough is due to a higher induction dosage, a higher maintenance dosage, or whether our frequency of

breakthrough will increase with longer follow-up remains to be determined. We emphasize that our study was open labeled and not a randomized, controlled study. However, given the natural history of CMV retinitis (particularly in AIDS patients, where a progressive disease results in profound visual loss) and that patients on ganciclovir relapse when maintenance therapy is interrupted, it appears that ganciclovir is an effective treatment for CMV retinitis. Carefully designed future studies will be needed to determine the optimum induction dosage, duration of induction therapy, and dosage and frequency of maintenance therapy. Leukopenia is the limiting toxic side effect of ganciclovir therapy and occurred to a variable degree in most of our patients treated with the drug. Severe leukopenia with absolute neutrophil counts less than 500 cellsjmm 3 occurred in 5 of 18 (28%) patients and necessitated interruption of therapy. Fortunately, neutropenia was reversible with discontinuation of treatment, and ganciclovir therapy could be resumed. RDs occurred in 3 of our 18 patients ( 17% ); in two cases, the detachment appeared to be rhegmatogenous and due to retinal destruction from the CMV infection. In the third case, the RD appeared to be exudative and the extent and severity of RD parallel the activity of the CMV infection. All three patients presented with RD before being treated with ganciclovir. It has been suggested that ganciclovir may contribute to rhegmatogenous RDs by inhibiting cellular proliferation. 15 Although we cannot exclude this possibility, our experience suggests that the rhegmatogenous RDs are a consequence of CMV retinitis with its attendant retinal destruction, similar to the acute retinal necrosis syndrome, 19 which also appears to be caused by a herpesvirus, either herpes simplex or Varicella zoster virus.Z 0 •21 The occurrence of RDs requiring surgical intervention has enabled us to obtain specimens of intraocular fluids for drug levels. In case 6, at 2! hours after ganciclovir infusion, both aqueous (2.4 J,Lm) and subretinal fluid (3.6 J,Lm) levels were similar to, but slightly less than that in a simultaneous serum specimen (6 J,Lm). Conversely, at 21 hours after ganciclovir infusion, the serum level was undetectable, whereas the vitreous level was 0.8 J,Lm. We have previously reported a patient treated for CMV retinitis after bone marrow transplantation who suffered an RD, and in whom subretinal fluid ganciclovir levels were-obtained at two separate surgeries. 14 At 51/z hours after the infusion, the serum and subretinal levels were very close ( 8.16 versus 7.16 J,Lm, respectively), whereas at 8 hours after the infusion the subretinal fluid level actually exceeded the serum level (2.5 versus 1.28 J,Lm, respectively). Because of the different fluids analyzed and the small number of samples, these data must be interpreted with caution. However, they suggest that the intraocular levels parallel those in the serum, but that penetration of ganciclovir into and clearance of ganciclovir from the eye are delayed relative to serum. While the exact relationships among the ret829

OPHTHALMOLOGY •

JULY 1987

inal and subretinal fluids, vitreous and aqueous drug levels are unclear, these data suggest that ganciclovir penetration into the infected eye is good and that therapeutic levels are achieved, 8 a fact mirrored by the biologic response of the CMV retinitis to ganciclovir treatment. Furthermore, the presence of detectable ganciclovir in the eye 21 hours after an infusion of maintenance drug provides rationale for the use of once daily maintenance therapy. That breakthrough was seen in a patient given 5- of7-day maintenance therapy suggests that intraocular drug levels may fall below the therapeutic level during the "off" days with 5- of 7-day maintenance therapy and allow for re-activation of CMV retinitis. Daily maintenance therapy may be necessary in some patients. The authors have no commercial or proprietary interest in the drug, DHPG.



8.

9.

10.

11.

12.

13.

ACKNOWLEDGMENT 14.

The authors wish to thank Jean-Pierre Sommadossi, PhD, PharO, for measuring the ganciclovir levels.

15.

REFERENCES 16. 1. Murray HW, Knox DL, Green WR, Susel RM. Cy1omegalovirus retinitis in adults: a manifestation of disseminated viral infection. Am J Med 1977; 63:574-84. 2. Pollard RB, Egbert PR, Gallagher JG, Merigan TC. Cy1omegalovirus retinitis in immunosuppressed hosts: I. Natural history and effects of treatment with adenine arabinoside. Ann Intern Med 1980; 93:65564. 3. Egbert PR, Pollard RB, Gallagher JG, Merigan TC. Cy1omegalovirus retinitis in immunosuppressed hosts: II. Ocular manifestions. Ann Intern Med 1980; 93:664-70. 4. Holland GN, Pepose JS, Pettit TH, et al. Acquired immune deficiency syndrome: ocular manifestations. Ophthalmology 1983; 90:859-73. 5. Khadern M, Kalish SB, Goldsmith J, et al. Ophthalmologic findings in acquired immune deficiency syndrome (AIDS). Arch Ophthalmol 1984; 102:201-6. 6. Freeman WR, Lerner CW, Mines JA, et al. A prospective study of the ophthalmologic findings in the acquired immune deficiency syndrome. Am J Ophthalmol1984; 97:133-42. 7. Newsome DA, Green WR, Miller ED, et al. Microvascular aspects of

830

17.

18.

19.

20.

21.

VOLUME 94



NUMBER 7

acquired immune deficiency syndrome retinopathy. Am J Ophthalmol1984; 98:590-601. Mar EC, Cheng YC, Huang ES. Effect of 9-(1 ,3-dihydroxy-2-propoxymethyl) guanine on human cy1omegalovirus replication in vitro. Antimicrob Agents Chemother 1983; 24:518-21. Bach MC, Bagwell SP, Knapp NP, et al. 9-(1 ,3-dihydroxy-2-propoxymethyl) guanine for cy1omegalovirus infections in patients with the acquired immunodeficiency syndrome. Ann Intern Med 1985; 103:381-2. Felsenstein D, D'Amico OJ, Hirsch MS, et al. Treatment of cy1omegalovirus retinitis with 9-[2-hydroxy-1-(hydroxymethyl) ethoxymethyl] guanine. Ann Intern Med 1985; 103:377-80. Palestine AG, Stevens G Jr, Lane HC, et al. Treatment of cy1omegalovirus retinitis with dihydroxy propoxymethyl guanine. Am J Ophthalmol1986; 101:95-101. Collaborative DHPG Treatment Study Group. Treatment of serious cy1omegalovirus infections with 9-(1 ,3-dihydroxy-2-propoxymethyl) guanine in patients with AIDS and other immunodeficiencies. N Engl J Med 1986; 314:801-5. Rosecan LR, Stahi-Bayliss CM, Kalman CM, Laskin OL. Antiviral therapy for cy1omegalovirus retinitis in AIDS with dihydroxy propoxymethyl guanine. Am J Ophthalmol1986; 101:405-18. Jabs DA, Wingard JR, de Bustros S, et al. BW B759U for cy1omegalovirus retinitis: intraocular drug penetration. Arch Ophthalmol 1986; 104:1436-7. Freeman WR, Henderly DE, Rao NA, et al. Rhegmatogenous retinal detachment in treated cy1omegalovirus retinitis: prevalence, pathophysiology, and treatment. ARVO Abstracts. Invest Ophthalmol Vis Sci 1986; 27(Suppl):260. D'Amico OJ, Talamo JH, Felsenstein D, et al. Ophthalmoscopic and histologic findings in cytomegalovirus retinitis treated with BW-B759U. Arch Ophthalmol1986; 104:1788-93. Holland GN, Sakamoto MJ, Hardy D, et al. Treatment of cy1omegalovirus retinopathy in patients with acquired immunodeficiency syndrome: use of the experimental drug 9-[2-hydroxy-1-(hydroxymethyl) ethoxymethyl] guanine. Arch Ophthalmol1986; 104:1794-800. Rosecan LR, Laskin OL, Kalman CM, et al. Antiviral therapy with ganciclovir for cy1omegalovirus retinitis and bilateral exudative retinal detachments in an immunocompromised child. Ophthalmology 1986; 93:1401-7. Fisher JP, Lewis ML, Blumenkranz M, et al. The acute retinal necrosis syndrome: Part 1. Clinical manifestations. Ophthalmology 1982; 89:1309-16. Culbertson WW, Blumenkranz MS, Pepose JS, et al. Varicella zoster virus is a cause of the acute retinal necrosis syndrome. Ophthalmology 1986; 93:559-69. Willig JL, Kaplan HJ, Holland GN, et al. Herpes zoster virus may be only one cause of ARN (Abstract). Ophthalmology 1986; 93(Suppl):83.