Cytomegalovirus Antigenemia in Acquired Immunodeficiency Syndrome Patients

With Untreated Cytomegalovirus Retinitis

CLAUDIO S. PANNUTI, MD, PHD, ESPER G. KALLAS, MD, CRISTINA MUCCIOLI, MD, RONALD K. ROLAND, MD,

ELIZABETH C. FERREIRA, BS, STELLA M. H. S. BUENO, MD CYNTHIA L. M. DO CANTO, BS, LUCY S. VILLAS BOAS, MS,

AND RUBENS BELFORT, JR, MD

• PURPOSE: To determine the frequency of cyto­megalovirus (CMV) viremia in patients with ac­quired immunodeficiency syndrome (AIDS) and untreated CMV retinitis using conventional cell culture isolation and the sensitive CMV antigene­mia assay. • METHODS: We examined 24 AIDS patients with ophthalmologic diagnosis of untreated CMV reti­nitis and 24 AIDS patients without present or past retinitis (control patients) from three medical centers between September 1992 and March 1994. Cytomegalovirus antigenemia was detected by an indirect peroxidase staining in 300,000 cytocentrifuged neutrophils, using a mixture of murine monoclonal antibodies directed against the pp65 lower matrix protein of CMV. • RESULTS: Positive antigenemia was demonstrat­ed in eight (33.3%) of the 24 retinitis patients and in none of the 24 control patients (P < .001).

Accepted for publication July 17, 1996. From the Division of Infectious Diseases, Hospital do Servidor Publico

Estadual de Säo Paulo (Drs Pannuti, Kallas, and Roland); the Department of Infectious Diseases, University of Sao Paulo School of Medicine, Institute de Medicina Tropical de Sâo Paulo, Laboratórios de lnvestigaçâo Medica do Hospital das Clinicas da Faculdade de Medicina da Universi-dade de Säo Paulo (LIMHCFMUSP) (Drs Pannuti, Ms Ferreira, Ms Do Canto, and Ms Villas Boas); the Department of Ophthalmology, Paulista School of Medicine (Drs Muccioli and Belfort); and Centro de Referência e Treinamento-AIDS, Secretarla Estadual da Saude de Sào Paulo (Drs Muccioli and Bueno), Sào Paulo, Brazil.

Reprint requests to Claudio S. Pannuti, MD, PhD, Virology Laboratory, Institute de Medicina Tropical de Säo Paulo, Av Dr Enéias de Carvalho Aguiar, 470, 05403-000 Säo Paulo, SP, Brazil; fax: 011-852-2174; e-mail: cpannuti@usp.br

Only two of the eight ahtigenemia-positive patients had a concurrent positive CMV isolation from blood leukocytes by conventional cell culture assay. • CONCLUSIONS: These results emphasize the risk of extraocular disease in AIDS patients with CMV retinitis because the virus is often present in peripheral blood leukocytes. The CMV antigene­mia assay may be a simple and rapid means of identifying those patients with unilateral retinitis at highest risk of developing CMV retinitis of the fellow eye or of visceral CMV disease if intravitre-al injections or implants are used as sole treatment for CMV retinitis.

A FTER THE ADVENT OF EFFECTIVE ANTIRETROV1-

ral therapy and routine prophylaxis against Pneumocystis carinii pneumonia, approximate­

ly 40% of acquired immunodeficiency syndrome (AIDS) patients will develop cytomegalovirus (CMV) disease,1 and 85% of those will develop retinitis.2

At present, two drugs, ganciclovir and foscarnet, are approved for the treatment of CMV retinitis. Cidofovir (hydroxyphosphonomethoxypropylcytosine dihydrate, or HPMPC), a cytosine-derived nucleotide analog with a long intracellular half-life and broad-spectrum activity against the herpesvirus family, including ganciclovir-resistant CMV,3 is a third drug recently approved for systemic administration.

Because many patients will have to discontinue systemic treatment because of severe toxicity,4 there is

VOL.122, No. 6 © AMERICAN JOURNAL OF OPHTHALMOLOGY 1996;122:847-852 847

growing interest in local alternative therapies, such as intravitreal injections or implants (references 5 through 11 and Chiron Ganciclovir Implant Study Group, unpublished data, Second National Confer­ence on Human Retroviruses and Related Infections, Washington, DC, January 29 to February 2, 1995). However, the detection of CMV viremia in patients with untreated CMV retinitis,12'14 the documentation of systemic involvement with CMV in autopsy studies of AIDS patients with CMV ocular infection,15 and the frequent bilateral presentation12 suggest that pa­tients with CMV retinitis may have nonocular CMV infection. Additionally, in a randomized controlled clinical trial to assess the safety and efficacy of a ganciclovir implant for treatment of CMV retinitis, cytomegalovirus retinitis developed in 67% (14/21) of the patients who entered the study with unilateral retinitis, and biopsy-proven symptomatic visceral CMV disease developed in 31% (8/26) of the study patients. Hence, the development and standardiza­tion of methods are desired for identifying those patients at highest risk of developing CMV retinitis in the fellow eye or visceral CMV disease when treated exclusively with local intraocular therapeutic ap­proach.

In organ-transplanted patients, the direct demon­stration of a tegument phosphoprotein (pp65) of CMV in peripheral blood leukocytes using monoclo­nal antibodies (antigenemia assay) has been shown to be a specific, rapid, and sensitive method for CMV detection.1617

The purpose of this study was to determine the frequency of CMV in peripheral blood leukocytes of patients with untreated CMV retinitis, using conven­tional cell culture isolation and the CMV antigene­mia assay.

METHODS

PATIENTS WITH AIDS AND PREVIOUSLY UNTREATED CMV retinitis from three medical centers of Sâo Paulo, Brazil (Escola Paulista de Medicina, Hospital do Servidor Publico Estadual de Sâo Paulo, and Centro de Referência e Treinamento em AIDS) were enrolled in the study between September 1992 and March 1994.

848 AMERICAN JOURNAL OF

A group of AIDS patients without present or past retinitis, admitted for treatment of opportunistic infections other than CMV disease, was included as a control. The control patients were matched with patients for gender and age, Karnofsky performance, and peripheral blood lymphocyte count. Neither study patients nor control patients were using any antiviral drug active against CMV before enrollment in the study. Toxoplasmic encephalitis (n = 7), disseminated tuberculosis (n = 5), and other forms of Mycobacterium tuberculosis disease (n = 4) were the most common opportunistic infections in the control group. CD4 lymphocyte counts could not be obtained from a substantial number of patients at the time of inclusion in the study, thus preventing the use of this marker of immunodeficiency for matching patients and controls.

The diagnosis of AIDS was based on the surveil­lance case definition adopted by the Centers for Disease Control and the presence of antibodies to the human immunodeficiency virus (HIV) by enzyme-linked immunosorbent assay, western blot analysis, or both. The study protocol was reviewed and approved by the Ethics Committee of the three medical centers.

Evaluation of patients included a history taking and a physical examination. The Karnofsky score was determined in all patients, and a score of less than 70 was considered a low performance. The ocular exami­nation consisted of biomicroscopy and indirect oph-thalmoscopy after pupillary dilation. Cytomegalovirus retinitis was diagnosed based on the presence of the characteristic white, progressively enlarging, necrotiz-ing retinitis with minimal inflammation. In all pa­tients with CMV retinitis, an induction treatment with ganciclovir was given immediately after the collection of a blood sample for virologie studies. The induction treatment consisted of a 10-mg/kg daily dose of ganciclovir (5 mg/kg/12 h) for 14 or 21 days and was followed by a maintenance dose of 5 mg/kg/d 7 days a week or 6.5 mg/kg/d 5 days a week.

An edetic acid-treated blood sample for detection of CMV viremia and antigenemia was taken when the patients were admitted into the study, before starting ganciclovir therapy. A second sample was taken, when possible, at the end of the induction treatment from patients with CMV retinitis and positive antigenemia.

OPHTHALMOLOGY DECEMBER 1996

For cell cultures, burly-coat cells of edetic acid-treated blood samples were obtained by dextran sedimentation. Briefly, 2 ml of 4% dextran in phos­phate-buffered saline were added to 4 ml of blood. The suspension was incubated for 20 minutes at 37 C. The leukocyte-rich upper layer was removed, and the cells were pelleted through centrifugation at 1,000 rpm for 7 minutes. The leukocytes were washed twice with Hank's medium without calcium or magnesium and resuspended in 1 ml of Eagles minimal essential medium; then 0.3 ml of the suspension was inoculat­ed in a tube containing a monolayer of human foreskin fibroblast cells. After 30 minutes, 0.7 ml of culture medium (Eagle's minimal essential medium, supplemented with 10% fetal bovine serum) was added. Cell cultures were maintained for 4 weeks and were examined twice a week. Cytomegalovirus was identified by the production of characteristic cyto-pathic effect.

The antigenemia assay was performed according to van der Bij and associates16 and Halwachs and associates17 with minor modifications. Briefly, leuko­cytes were separated by sedimentation of 3 to 5 ml of edetic acid-treated blood in a 5% dextran solution at a 2:1 ratio at 37 C for 20 minutes. Using a plastic Pasteur pipette, the polymorphonuclear leukocyte-rich plasma was transferred to a 15-ml tube and centrifuged for 7 minutes at 1,500 rpm. The pelleted cells were suspended in 10 ml of cold ammonium chloride solution to lyse the contaminating erythro-cytes and placed on ice for 10 minutes. After washing twice in phosphate-buffered saline, the polymorpho­nuclear leukocytes were suspended in phosphate-buffered saline and counted. The polymorphonuclear leukocyte spots were prepared by cytocentrifuging 3 X 105 cells in 100 μΐ for 5 minutes at 900 rpm onto a glass slide, using a cytocentrifuge. The CMV antigen detection in polymorphonuclear leukocytes (antigen­emia) was accomplished by incubating the polymor­phonuclear leukocyte spots with 25 μΐ of a 1:10 dilution of a mixture of C10 and C l l monoclonal antibodies to CMV antigens for 45 minutes at room temperature. These antibodies recognize the 65-kd lower matrix phosphoprotein (pp65) of CMV.

After washing the slide three times with phos­phate-buffered saline, the slide was incubated for 30 minutes at room temperature with 25 μΐ of a peroxi-

VOL. 122, No. 6 CYTOMECALOVIRUS ANTIGENEMIA IN ,

dase-labeled anti-mouse conjugate. After washing the slide three times as before, the enzyme reaction was performed with a solution of 3-amino-9-ethylcar-bazole in 0.1 M acetate buffer (pH 4.9) supplemented with 50 μΐ of hydrogen peroxide (30%) per 50 ml of acetate buffer immediately before use (10 minutes at room temperature). The slides were then washed with acetate buffer for 5 minutes, rinsed with distilled water, counterstained with Mayer's hematoxylin for approximately 1 minute, rinsed with tap water, and mounted in glycerol. Cells positive for the CMV antigen showed a red-brown stain in the nuclei of polymorphonuclear leukocytes. The nuclear stain was homogeneous or perinuclear. Preparations known to contain CMV antigen-positive cells were used as positive controls in every reaction. Cytomegalovirus antigen-positive polymorphonuclear leukocytes were counted and recorded.

We defined positive antigenemia as the presence of one or more positive staining cell per slide. Antigene­mia was defined as low when the number of positive cells was lower than 10 per slide and as high when the number of positive cells was more than 10 per slide.

IgG antibodies to CMV late antigens were mea­sured by indirect immunofluorescence assay using antigens prepared in our laboratory, as previously described.18

RESULTS

A TOTAL OF 48 PATIENTS WERE ENROLLED IN THIS STUDY.

Of the 48, 24 (50%) had CMV retinitis (retinitis group) and 24 had a normal ophthalmologic exami­nation (control group).

All patients in the control group had IgG CMV antibodies, were in category IV Cl definition criteria by the Centers for Disease Control, and had been admitted for treatment of opportunistic infections other than CMV disease. Both groups were similar in gender (four women in the retinitis group and three women in the control group), age (median, retinitis patients, 35.0 years; control patients, 34.5 years), low Karnofsky performance (retinitis group, eight; control group, 11), and peripheral blood lymphocyte count (average, retinitis group, 812.5 cells/mm3; control group, 951.2 cells/mm3).

ACQUIRED IMMUNODEFICIENCY SYNDROME 849

Antigenemia was detected in eight (33.3%) of the 24 retinitis patients and in none of the 24 control AIDS patients (P < .001, Fisher's exact test). One additional patient with retinitis had negative antigen­emia results when admitted to the study, but a second sample, obtained 14 days after ganciclovir treatment, tested positive. Because statistical comparisons be­tween patients were based upon the first sample only, this patient was included in the negative-antigenemia group.

Bilateral retinal involvement occurred in 50.0% of patients (4/8) with a positive CMV antigenemia and in 31.2% of the retinitis patients (5/16) with negative antigenemia (P = .32, Fisher's exact test). A low Karnofsky performance was observed in 12.5% of the CMV antigenemia-positive patients (1/8) and in 43.7% of antigenemia-negative patients (7/16; P = .14, Fisher's exact test).

Only three of the patients with positive CMV antigenemia had more than 10 positive leukocytes per slide (70, 87, and 33 positive cells), two of them also having CMV isolated from peripheral blood leuko­cytes. All the remaining positive patients had low-grade antigenemia (two, two, three, three, and nine positive cells per slide), and viremia was not detected in any of them.

A second blood sample could be obtained 14 to 25 days after the beginning of ganciclovir therapy from five of the eight antigenemia-positive retinitis pa­tients. Four of them had either a decrease in the number of polymorphonuclear leukocyte antigen-positive cells (two patients) or no positive cells detected in the second sample (two patients). The remaining patient, who showed an increase in the number of polymorphonuclear leukocyte antigen-positive cells, could not be treated with ganciclovir or foscarnet because of severe side effects and died with disseminated CMV disease 17 days after admission into the study.

DISCUSSION

ITS HIGH DIAGNOSTIC ACCURACY, RAPIDITY, QUANTITA-tive nature, and technical simplicity have made the CMV antigenemia assay one of the cornerstone methods for diagnosis and management of CMV

850 AMERICAN JOURNAL

infection in immunocompromised patients.19 In a series of 130 consecutive patients receiving a renal allograft, van den Berg and associates20 demonstrated that the antigenemia assay had a sensitivity of 89% and a specificity of 93% in the diagnosis of CMV infection. In 45 liver transplant recipients, van den Berg and associates compared the value of weekly monitoring of CMV antigenemia with serology and rapid viral isolation. The sensitivity of the antigene­mia assay was 96% compared with 57% for viral isolation of blood.21 In 33 patients receiving allogeneic bone marrow transplantation for malignant hémato­logie diseases, we found (Machado CM, Villas Boas LS, Do Canto CLM, Ostronoff M, Dulley F, Pannuti CS, unpublished data, Eighth International Symposi­um on Infections in the Immunocompromised Host, Davos, Switzerland, June 19-22, 1994) that positive antigenemia had a positive predictive value for CMV disease (pneumonitis, enteritis, or both) of 80% and a negative predictive value of 95.6%.

In AIDS patients, the likelihood of CMV viremia increases as CD4+ T-lymphocyte count decreases.22

In the present study, even though CD4+ cell counts were not feasible, the fact that patients from both the retinitis and control groups had low peripheral blood lymphocyte counts and opportunistic diseases sug­gests that they were comparable in terms of advanced HIV-induced immunosuppression. Therefore, the ab­sence of any marker of active CMV infection in the control group was somewhat unexpected because recent reports in AIDS patients without CMV dis­ease and CD4+ cell counts below 100/μΙ have found frequencies of CMV viremia or antigenemia ranging from 2.1% to 12% (references 23 and 24 and Sebastiani G, Magno MS, Mastroianni C, et al, unpublished data, Fifth International Cytomegalovi-rus Conference, Stockolm, Sweden, May 21-24, 1995). However, not having used CD4+ cell counts for matching cases and controls might have affected our results because, although CMV retinitis occurs almost always in AIDS patients with CD4+ cell counts of less than 50/μΙ,25'27 other opportunistic infections can take place in less advanced stages of HIV disease.

On the other hand, in AIDS patients with estab­lished CMV disease, high rates of CMV viremia have been observed. Jabs and associates12 reported positive

OPHTHALMOLOGY DECEMBER 1996

blood cultures in 74% of AIDS patients (29/39) with CMV retinitis, and Mazzulli and associates,14 examin­ing multiple blood samples from nine patients with CMV retinitis, reported a 100% positivity of the antigenemia assay. Bek and associates23 detected posi­tive antigenemia in 96.7% of patients with proven CMV disease and reported that 42% of patients with retinitis had fewer than five positive cells per slide, suggesting that viral loads can be low in AIDS patients with retinitis, as previously reported.28 If this is the case, minor differences in antigenemia and virus isolation techniques could have affected the proportion of viremic patients in the present series, and the rate of 33.3% may be underestimated, espe­cially because in this study, only a single pretreatment blood sample was obtained from each patient. The number of polymorphonuclear granulocytes cytocen-trifuged onto microscopic slides, type of fixation, type of monoclonal antibodies, or the timing for sample collection and processing are some of the technical details that must be considered when comparing antigenemia results from different studies.29 Converse­ly, in organ transplant recipients, comparison of CMV isolation from blood with the antigenemia assay has shown that current viral isolation methods can detect only high viral loads.21 Also, increasing the frequency and volume of blood sampling increases the detection of CMV viremia by shell vial assay in transplantation recipients, suggesting that CMV vi­remia can occur with a small number of viral particles and can be intermittent.30 More frequent sampling could therefore have increased the number of viremic patients in this series and might be an important step in establishing the true frequency of viremia in AIDS patients with retinitis. In this study, the observation that a patient had positive antigenemia results only in a second sample corroborates this hypothesis. The optimal number of samples, volume of blood, and crucial test conditions that have to be met when comparing results of different laboratories for diagnos­ing CMV viremia by CMV isolation or antigenemia remains unknown and must be addressed by con­trolled studies.19

The fact that no correlation was found between the Karnofsky score or the unilateral or bilateral retinal involvement and positive CMV antigenemia discloses the difficulty in finding a consistent clinical

VOL.122, No. 6 CYTOMECALOVIRUS ANTIGENEMIA IN

marker to distinguish localized from systemic CMV infection. However, because of small sample size, the present study had low power to detect statistically significant differences within subgroups of patients.

In this study, by examining blood samples collected at one time, just before ganciclovir therapy was started, we could demonstrate the presence of CMV in peripheral blood leukocytes in at least 33.3% of AIDS patients with CMV retinitis. The antigenemia assay may thus be a simple and rapid means of identifying those patients at highest risk of developing CMV retinitis in the fellow eye or visceral CMV disease when sustained-release intraocular ganciclo­vir implants or intravitreal injections of ganciclovir, foscarnet, or cidofovir are used as the only treatment for CMV retinitis.

REFERENCES

1. Hoover DR, Saah AJ, Bacellar H, et al. Clinical manifestations of AIDS in the era of Pneumocystis prophylaxis. N Engl J Med 1993;329:1922-1926.

2. Gallant JE, Moore RD, Richman DD, Keruly J, Chaisson RE, Zidovudine Epidemiology Study Group. Incidence and natural history of cytomegalovirus disease in patients with advanced human immunodeficiency virus disease treated with zidovudine. J Infect Dis 1992;166:1223-1227.

3. Jacobson MA. Update (1995) on clinical trials of antiviral therapy and prophylaxis for AIDS-related cytomegalovirus disease. Scand J Infect Dis 1995;99(suppl):96-99.

4. Studies of Ocular Complications of AIDS Research Group, AIDS Clinical Trials Group. Mortality in patients with the acquired immunodeficiency syndrome treated with either foscarnet or ganciclovir for cytomegalovirus retinitis. N Engl J Med 1992;326:213-220.

5. Henry K, Cantrill H, Fletcher C, Chinnock BJ, Balfour HH Jr. Use of intravitreal ganciclovir (dihydroxy propoxymethyl gua-nine) for cytomegalovirus retinitis in a patient with AIDS. Am J Ophthalmol 1987;103:17-23.

6. Cochereau-Massin I, Le Hoang P, Lautier-Frau M, et al. Retinite à cytomegalovirus au cours du SIDA: traitement par injections intravitréennes de ganciclovir. Presse Med 1990;19:1313-1316.

7. Young SH, Morlet N, Heery S, Hollows FC, Coroneo MT. High dose intravitreal ganciclovir in the treatment of cyto­megalovirus retinitis. Med J Aust 1992;157:370-373.

8. Diaz-Llopis M, Chipont E, Sanchez S, et al. Intravitreal foscarnet for cytomegalovirus retinitis in a patient with ac­quired immunodeficiency syndrome. Am J Ophthalmol 1992;114:742-747.

9. Kirsch LS, Arevalo JF, de Clercq E, et al. Phase I/II study of intravitreal cidofovir for the treatment of cytomegalovirus retinitis in patients with the acquired immunodeficiency syndrome. Am J Ophthalmol 1995;119:466-476.

10. Kirsch LS, Arevalo JF, de la Paz EC, et al. Intravitreal

ACQUIRED IMMUNODEFICIENCY SYNDROME 851

cidofovir (HPMPC) treatment of cytomegalovirus retinitis in patients with acquired immune deficiency syndrome. Oph­thalmology 1995;102:533-543.

11. 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 Ophthalmol 1994;112:1531-1539.

12. Jabs DA, Enger C, Bartlett JG. Cytomegalovirus retinitis and acquired immunodeficiency syndrome. Arch Ophthalmol 1989;107:75-80.

13. Palestine AG, Polis MA, De Smet MD, et al. A randomized, controlled trial of foscarnet in the treatment of cytomegalovi­rus retinitis in patients with AIDS. Ann Intern Med 1991;115:665-673.

14· Mazzulli T, Rubin RH, Ferraro MJ, et al. Cytomegalovirus antigenemia: clinical correlations in transplant recipients and in persons with AIDS. J Clin Microbiol 1993;31:2824-2827.

15. Pepose JS, Holland GN, Nestor MS, Cochran AJ, Foos RY. Acquired immune deficiency syndrome: pathogenic mecha­nisms of ocular disease. Ophthalmology 1985;92:472-484.

16. van der Bij W, Schirm J, Torensma R, van Son WJ, Tegzess AM, The TH. Comparison between viremia and antigene­mia for detection of cytomegalovirus in blood. J Clin Microbiol 1988;26:2531-2535.

17. Halwachs G, Zach R, Pogglitsch H, et al. A rapid immunocy-tochemical assay for CMV detection in peripheral blood of organ-transplanted patients in clinical practice. Transplanta­tion 1993;56:338-342.

18. Turchi MD, Pannuti CS, Sumita LM, et al. Infecçâo pelo citomegalovirus em pacientes com a sindrome da imunode-ficiência adquirida (AIDS): relaçôes clinico-virológicas e anatomopatológicas. Revista do Institute de Medicina Tropi­cal de Sào Paulo 1991;33:243-250.

19. The TH, van der Berg AP, Harmsen MC, van der Bij W, van Son WJ. The cytomegalovirus antigenemia assay: a plea for standardization. Scand J Infect Dis 1995;99(suppl):25-29.

20. van den Berg AP, van der Bij W, van Son WJ, et al. Cytomegalovirus antigenemia as a useful marker of sympto­matic cytomegalovirus infection after renal transplantation: a report of 130 consecutive patients. Transplantation 1989;48: 991-995.

21. van den Berg AP, Klompmaker IJ, Haagsma EB, et al.

Antigenemia in the diagnosis and monitoring of active cytomegalovirus infection after liver transplantation. J Infect Dis 1991;164:265-270.

22. Gema G, Parea M, Percivalle E, et al. Human cytomegalovi­rus viraemia in HIV-1-seropositive patients at various stages of infection. AIDS 1990;4:1027-1031.

23. Bek B, Boeckh M, Lepenies J, et al. High-level sensitivity of quantitative pp65 cytomegalovirus (CMV) antigenemia assay for diagnosis of CMV disease in AIDS patients and follow-up. J Clin Microbiol 1996;34:457-459.

24. MacGregor RR, Pakola SJ, Graziani AL, et al. Evidence of active cytomegalovirus infection in clinically stable HIV-infected individuals with CD4+ lymphocyte counts below 100/μ1 of blood: features and relation to risk of subse­quent CMV retinitis. J Acquir Immune Defic Syndr 1995;10: 324-330.

25. Studies of Ocular Complications of AIDS Research Group, AIDS Clinical Trials Group. Mortality in patients with the acquired immunodeficiency syndrome treated with either foscamet or ganciclovir for cytomegalovirus retinitis. N Engl J Med 1992;326:213-220.

26. Kuppermann BD, Petty JG, Richman DD, et al. Correlation between CD4+ counts and prevalence of cytomegalovirus retinitis and human immunodeficiency virus-related nonin-fectious retinal vasculopathy in patients with acquired im­munodeficiency syndrome. Am J Ophthalmol 1993; 115: 575-582.

27. Spector SA, Weigeist T, Pollard RB, et al. A randomized, controlled study of intravenous ganciclovir therapy for cyto­megalovirus peripheral retinitis in patients with AIDS. J Infect Dis 1993;168:557-563.

28. Drew WL, Ives D, Lalezari JP, et al. Oral ganciclovir as maintenance treatment for cytomegalovirus retinitis in pa­tients with AIDS. N Engl J Med 1995;333:615-620.

29. Saltzman RL, Quirk MR, Jordan MC. High levels of circulat­ing cytomegalovirus DNA reflect visceral organ disease in viremic immunosuppressed patients other than marrow recip­ients. J Clin Invest 1992;90:1832-1838.

30. Patel R, Klein DW, Espy MJ, et al. Optimization of detection of cytomegalovirus viremia in transplantation recipients by shell vial assay. J Clin Microbiol 1995;33:2984-2986.

Authors Interactive™ Comments and questions of 100 words or less regarding this article may be addressed to authors via THE JOURNAL'S Web Site on the Internet. Selected questions and comments along with the authors' statements and responses will be posted on the Authors Interactive™ Bulletin Board, a feature of THE JOURNAL'S home page. Readers may access THE JOURNAL'S home page and the Authors Interactive™ Bulletin Board at http://www.ajo.com/

852 AMERICAN JOURNAL OF OPHTHALMOLOGY DECEMBER i996