Article Text


Discontinuing anticytomegalovirus therapy in patients with cytomegalovirus retinitis and AIDS
  1. Departments of Ophthalmology and Medicine, Johns Hopkins University School of Medicine, and the Department of Epidemiology, Johns Hopkins University School of Hygiene and Public Health
  1. The Wilmer Eye Institute, The Johns Hopkins Hospital, 550 North Broadway, Suite 700, Baltimore, MD 21205, USA djabs{at}

Statistics from

Cytomegalovirus (CMV) retinitis is the most common opportunistic ocular infection in patients with the acquired immune deficiency syndrome (AIDS).1 Before the advent of highly active antiretroviral therapy (HAART), CMV retinitis affected 30% of patients with AIDS at some time during the course of their disease.2 Cytomegalovirus retinitis is a late stage complication associated with low CD4+ T cell counts, typically less than 50 cells × 106/l.34 Cytomegalovirus retinitis was rare at CD4+ T cells >100 cells × 106/l.34 All of the available anti-CMV therapies suppress viral replication, but do not eliminate the virus. Unless immune reconstitution occurs, prolonged suppressive anti-CMV therapy (maintenance therapy) is required.15 Without immune reconstitution or maintenance therapy, CMV retinitis relapses within 3 weeks. As such, in the pre-HAART era, patients with CMV retinitis required lifetime maintenance anti-CMV therapy.

HAART consists of combination therapy for the human immunodeficiency virus (HIV), with at least three drugs, typically two nucleoside reverse transcriptase inhibitors and either a protease inhibitor or a non-nucleoside reverse transcriptase inhibitor. HAART can result in marked suppression of HIV replication, improvement in immune function, increases in CD4+ T cells, decreases in opportunistic infections, and improved survival.6 With HAART, there has been a 55%–95% reduction in the number of new cases of the CMV retinitis, and the decrease varies depending upon the population being served.6-8 However, CMV retinitis continues to occur, albeit at a reduced incidence, and there remains an increasing prevalent population of patients with AIDS and CMV retinitis who have experienced immune reconstitution as a consequence of HAART and are living for substantially longer times.

In addition to the increase in CD4+ T cell counts and the decrease in the incidence of CMV retinitis with HAART, studies have demonstrated the restoration of specific anti-CMV immunity in patients with CMV retinitis who have had immune reconstitution as a consequence of HAART.9 As such, several investigators have discontinued anti-CMV maintenance therapy in patients with immune reconstitution from HAART. These case series have reported that, as long as immune reconstitution is maintained, CMV retinitis does not relapse, and that the anti-CMV therapy can be discontinued safely.10-15

In this issue of the BJO (p 471), Curiet al have reported their experience with discontinuing anti-CMV therapy in 41 patients with CMV retinitis who had immune reconstitution. CD4+ T cell counts at the time of diagnosis of the CMV retinitis typically were low, with a median CD4 + T cell count of 42 cells × 106/l, and all of the patients experienced immune reconstitution, with a median CD4+ T cell count of 238 cells × 106/l at the time when anti-CMV therapy was discontinued. The lowest CD4+ T cell count at that time was 143 cells × 106/l. None of the patients suffered relapses of the retinitis, and immune reconstitution was maintained throughout. The median final CD4+ T cell count in this population was 427 cells × 106/l, and the lowest was 181 cells × 106/l. These results are in accord with other published studies, which have reported that as long as the CD4+ T cell count increases to over 100 cells × 106/l, and is maintained over 50 cells × 106/l, anti-CMV maintenance therapy can be discontinued safely.10-15 In the series by Curi et al the median follow up off anti-CMV therapy was nearly 2 years (21 months), suggesting that as long as immune reconstitution is maintained anti-CMV therapy can be withheld for prolonged periods of time.

Although it is clear that anti-CMV therapy may be discontinued safely in patients who experience immune reconstitution, there still are several issues. The first is the level of CD4+ T cell count to use for discontinuation of anti-CMV therapy. Most investigators have used a level of at least 100 cells × 106/l, although some have used 150 cells × 106/l. One centre used 50 cells × 106/l.15 The report by Curiet al does not enable us to better refine the estimate, as all but one patient had a CD4+ T cell count over 150 cells × 106/l at the time of discontinuation of anti-CMV therapy. However, because some series have reported safe discontinuation of anti-CMV therapy in patients with at least 100 cells × 106/l, it appears that this level is reasonable. Whether lower levels are as safe remains uncertain.

The second issue is the duration of immune reconstitution before discontinuation of anti-CMV therapy. Restoration of CD4+ T cell counts may occur before the restoration of specific CMV immunity, and cases of CMV retinitis have been reported to occur immediately after introduction of HAART.16 Although investigators have suggested a restoration of CD4+ T cell counts for at least 3–6 months, based on estimates of the time to restore specific anti-CMV immunity, most patients in the reported case series have been on HAART for longer time periods before discontinuing anti-CMV therapy. In the series by Curi et al the shortest time on HAART was 5 months, and the median time was 13 months. The third issue is the role of HIV viral load in monitoring patients of anti-CMV therapy.11-13 Although suppression of HIV replication to undetectable levels is the goal of antiretroviral therapy, several case series of patients with CMV retinitis have suggested that low levels of HIV replication, as long as the CD4+ T cell count has increased, are not associated with relapse. As such, it appears that immunological reconstitution is necessary for discontinuation of anti-CMV therapy, but that complete suppression of HIV replication may not be. Although ongoing low level HIV replication will probably result in loss of immune reconstitution long term, in the short term it appears that level of immune function is the superior way to follow patients when discontinuing anti-CMV therapy. The fourth issue is when to reinstitute anti-CMV therapy. Patients who have experienced an immune reconstitution and had successful discontinuation of anti-CMV therapy have been reported to relapse when immune reconstitution is lost and the CD4+ T cell count falls to <50 cells × 106/l.15 As such, it would appear to be prudent to consider reinstitution of anti-CMV therapy when the CD4+ T cell count falls to <50 cells × 106/l.

In conclusion, it appears that among patients who experience immune reconstitution as a consequence of HAART, anti-CMV therapy can be discontinued safely for prolonged periods of time. A threshold level of 100–150 cells × 106/l for a duration of 3–6 months appears to be a reasonable guideline for discontinuing anti-CMV therapy. Because of the occasional patient who will not recover specific CMV immunity despite an increase in CD4+ T cells, these patients will continue to need regular ophthalmological follow up. In addition, the CD4+ T cell count will need to be followed, as patients may relapse when the CD4+ T cell count falls below 50 cells × 106/l. However, as shown in the paper by Curiet al, prolonged immune reconstitution and prolonged periods off anti-CMV maintenance therapy are achievable.


Grant support: Supported in part by grants EY-08052, EY-10268, and EY-00405 from the National Eye Institute, the National Institutes of Health, Bethesda, MD, USA.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles