Ka-Hing WONG
Cytomegalovirus (CMV) disease is a common and potentially debilitating condition, occurring in 21-44% of AIDS patients in western countries before the era of highly active antiretroviral therapy (HAART). The most frequent presentation is retinitis which accounts for 75-85% of the diseases1 while gastrointestinal (colitis, oesophagitis), respiratory (pneumonitis), and neurological (encephalitis, polyradiculopathy, neuropathy) diseases can also occur. In Hong Kong, CMV diseases, primarily retinitis, contributed to some 4-5% of the primary AIDS-defining illnesses (ADI) reported up to 2005 and was the commonest subsequent ADI in the pre-HAART era.2
However, the incidence of CMV retinitis has substantially declined after the institution of HAART. In a multicentre prospective study, 1.8% of HAART-treated patients with CD4 <200/μL developed CMV diseases over 2 years, as compared to 11.1% on dual therapy.3 In addition, survival of AIDS patients with CMV retinitis has significantly improved. Although more data is present for protease inhibitor (PI) being associated with markedly prolonged time to relapse and improved survival in AIDS patients with CMV retinitis,4 the benefits conceivably also occurs for non-PI containing HAART. The decline in CMV incidence as a primary ADI was also seen in Hong Kong, despite the low overall occurrence.
CMV retinitis primarily occurs in patients with CD4 count <50-100/μL. In Hong Kong, over 75% of patients with CMV retinitis reported as primary ADI has CD4 <50/μL at diagnosis, with a median count of 19/μL.5 The common clinical presentations include floaters, loss of visual fields, flashing and blurring of vision. If left untreated, CMV retinitis invariably progresses rapidly to blindness due to macular involvement or retinal detachment from extensive peripheral retinitis. In pre-HAART era, prognosis of CMV retinitis was invariably poor despite treatment, with a mean survival of less than 1 year. Even with HAART, CMV retinitis is still associated with significant visual impairment in both newly diagnosed patients and those with long standing disease. The detection of CMV viraemia with positive plasma CMV DNA PCR is a predictor independent of CD4 and viral load for mortality in AIDS patients.6
CMV infection can also involve gastrointestinal tract, respiratory tract and central nervous system. CMV colitis typically presents with bloody diarrhoea while oesophagitis presents with dysphagia and solitary ulcer on endoscopy (See Chapter 19: Gastrointestinal system). CMV pneumonitis may occur alone or, more commonly, co-exist with Pneumocystis jiroveci pneumonia. CMV encephalitis is typically characterised by diffuse cerebral involvement but cerebral space-occupying lesion has been reported.7 Spinal cord involvement of CMV infection may lead to transverse myelitis or polyradiculopathy. CMV polyradiculopathy often presents with subacute lower limb weakness, paraesthesia and urinary retention.
Owing to the potential serious impact on vision, prompt diagnosis and treatment is essential, especially for patients with near macular involvement. Diagnosis of CMV retinitis is primarily based on presence of retinal lesion, which is characterised by creamy yellow white, haemorrhagic, full thickness opacification, usually at zone 2 or 3. Health care providers should be vigilant especially to patients with visual complaints or the CD4 count is <50-100/μL. For CMV infections with gastrointestinal and respiratory involvement, diagnosis can be determined by the presence of CMV inclusion bodies and/or tissue damage in endoscopic biopsy specimens. Detection of CMV DNA in the cerebrospinal fluid is the definitive diagnostic tool of CNS involvement. Patients with CMV infection of other organs are often at high risk of CMV retinitis and ophthalmologic assessment must be performed even when there are no visual symptoms. Primary prophylaxis against CMV is currently unavailable. All HIV-infected patients should be educated about the symptoms and importance of early treatment, particularly when their CD4 count falls below 100/μL. In this case, regular indirect ophthalmoscopy by an ophthalmologist shall be considered.
When CMV retinitis is suspected, urgent referral for ophthalmologist assessment and confirmation is warranted before initiating anti-CMV treatment. The objective is to establish a baseline for monitoring of treatment and to exclude other less common but critical ophthalmologic conditions, which include herpetic retinal necrosis, toxoplasma retinochoroiditis, cryptococcal choroiditis, large cell lymphoma, candidiasis, syphilitic uveitis and Pneumocystis jiroveci choroiditis.
The advent of HAART has revolutionised the management as well as prognosis of CMV infection in HIV-infected patients. The objectives of management are to minimise morbidity, mortality and quality of life impairment (especially blindness and impaired vision). This could be achieved through early diagnosis, timely treatment, appropriate maintenance therapy and monitoring for relapse of disease.
The benefit of drug treatment of CMV retinitis has to be balanced against its side effects and tolerance. It should be individualised to meet patient's specific needs. Treatment regimen should be discussed with the patient and chosen according to characteristics of the retinitis, patient's work and living conditions, family support, concomitant drugs, and contraindications. Management of CMV retinitis should follow a multi-disciplinary approach. The diagnosis, treatment and monitoring should involve both HIV physician and ophthalmologist. Health counselling and psychological support should be provided by nursing team and mobilisation of community resources can be arranged by medical social worker. Close monitoring for relapse of retinitis or involvement of the other eye by the ophthalmologist is essential. The sequelae of retinitis and impact on daily activities should be assessed and managed accordingly.
Several effective drugs are available for treatment and prevention of relapse of CMV retinitis and they are listed in Box 23.1. A 2-3 week course of induction treatment should be given and assessed for response. It halts progression and induces resolution of the retinitis. It aims at preserving vision and preventing the complication of retinitis, in particular retinal detachment. Without effective immune recovery, life-long maintenance therapy is mandatory to prevent further retinal necrosis, which can occur in 2-6 weeks if therapy is stopped.8 Oral valganciclovir, IV ganciclovir, IV foscarnet and IV cidofovir are all appropriate initial choices for induction and maintenance therapy but valganciclovir is the drug of choice nowadays.
Cidofovir obviates the need of long-term intravenous access as is the case with IV ganciclovir or foscarnet. If IV foscarnet is used for maintenance, the patient should have access to an infusion pump to avoid dangerous electrolyte disturbance. Valganciclovir , a pro-drug of ganciclovir, is the only effective oral form of antiviral drugs that is active against CMV. It overcomes the limitations of oral ganciclovir, and also obviates the need of invasive intravenous or intraocular injection. Due to its excellent bioavailability, the maximum drug level achievable is similar to intravenous ganciclovir. Valganciclovir has also been shown to have similar efficacy as IV ganciclovir for both induction and maintenance treatment in newly diagnosed CMV AIDS patients.9 The rate of emergence of resistant strains is similar for both IV ganciclovir and oral valganciclovir.
Ganciclovir implant is a useful alternative for those who cannot tolerate systemic treatment. Through a high local concentration, ganciclovir implant controls CMV retinitis with little toxicity. It is especially indicated in patients with zone one (within 3000 μm from the center of the fovea or 1500 μm from the optic disc) disease. The drawbacks are:
(a) Implantation may be complicated by retinal detachment, vitreous haemorrhage or infection,
(b) The implant has a lifespan of 6-8 months, requiring regular replacement, and
(c) It offers no systemic protection against CMV, and does not protect contralateral ocular and extraocular diseases.
In a retrospective cohort study in a US managed-care setting, use of oral and intraocular CMV treatment saves cost compared with intravenous treatment, against a background of decreased incidence in the HAART era.10
Prior to the era of HAART, almost all patients had relapse of CMV retinitis despite long term suppressive therapy. Relapse commonly occurs at the border of an inactive lesion, and is related to inadequate drug level rather than resistance. Others might have disease refractory to initial CMV treatment. Management of relapsing or refractory retinitis is difficult and should be individualised according to initial/previous response to therapy, concomitant medical conditions, characteristics of retinitis, immunologic and virologic status, and expertise of specialized treatment available. Adherence to maintenance therapy should be reassessed for relapsed cases.
For relapse while on IV ganciclovir or foscarnet maintenance, re-induction with the maintenance drug or switch to the other drug has similar efficacy.10 Re-induction with cidofovir is also effective for relapse on maintenance with IV ganciclovir, IV foscarnet or cidofovir maintenance.
Combination therapy may be considered for refractory disease during initial treatment or relapse with monotherapy. Combination of IV ganciclovir and foscarnet has been shown to be superior to using either alone in treating relapse.8In vitro data suggests synergistic inhibition of CMV by cidofovir in combination with ganciclovir or foscarnet.8 Ganciclovir ocular implant or intravitreal ganciclovir/foscarnet injection can be considered if expertise for such treatment is available. Either of these two treatments can be given alone or as adjuvant to IV therapy.
Induction treatment with IV ganciclovir or foscarnet for 3-6 weeks is often indicated. A combination of the two drugs may be considered for CMV encephalitis or polyradiculopathy and in patients who have received prior anti-CMV therapy. Maintenance therapy is generally not necessary for non-retinal CMV diseases unless there is relapse.
Response to anti-CMV treatment of extra-ocular diseases varies. CMV gastrointestinal conditions generally respond to anti-CMV treatment. Ganciclovir and foscarnet have also improved the prognosis of CMV polyradioculopathy and neuropathy but response of encephalitis remains poor.
The occurrence of CMV retinitis in advanced AIDS patients signifies severe immunosuppression. Restoration of immune function is crucial for disease control with CMV-specific treatment. The role of HAART in this regard is clear, and provides impetus to study the virologic and immunologic correlates of protection from CMV diseases.11 Furthermore, mounting evidence show that maintenance therapy can be stopped in patients who respond to HAART with good CD4 response (>100/μL).12,13 This could be linked to the increase in CMV-specific immune response together with the decrease in CMV reactivation.
There have been reports of CMV retinitis presenting soon after the initiation of HAART, which may be due to the unmasking of subclinical CMV retinal infection by a HAART-induced immune inflammatory response, also known as immune reconstitution disease (refer to chapter 15).14 In addition, sight-threatening immune recovery vitritis or uveitis has been described in HAART responders.15 This may occur in up to two-thirds of patients with inactive CMV retinitis and elevated CD4 counts; vision loss can ensue from epiretinal membrane, cataract and cystoid macular oedema.16 Other inflammatory reactions can present as vitreous haze and optic disc oedema. Large CMV lesions of >=25% retinal areas and use of intravenous cidofovir are the risk factors.17 It is thus necessary to be vigilant of the potential "precipitation" of CMV retinitis after initiation of HAART in advanced HIV disease patients.
Withdrawal of maintenance therapy is considered safe in patients with regressed CMV retinitis on HAART if all of the following criteria are met: (a) CD4 count >100-150/μL for 3-6 months, (b) vision adequate in contralateral eye, (c) undetectable plasma viral load for >3-6 months, (d) patient can attend regular ophthalmic examination. The CD4 and plasma viral load of the patients should be regularly monitored. Failure of viral suppression and unsustainable rise in CD4 is a presage of recurrence.18 Risks and benefits of withdrawing secondary prophylaxis need to be explained to the patient. Some doctors may prefer a higher threshold of CD4, e.g. >200/μL for >6 months, and an adequate vision (after treatment) of both eyes before withdrawal of maintenance therapy is contemplated.
Gallant JE, Moore RD, Richman DD, Keruly J, Chaisson RE. Incidence and natural history of cytomegalovirus disease in patients with advanced human immunodeficiency virus disease treated with zidovudine. The Zidovudine Epidemiology Study Group. J Infect Dis 1992;166:1223-7.
Wong KH, Lee SS. Comparing the first and second hundred AIDS cases in Hong Kong. Singapore Med J 1998;39:236-40.
Yust I, Fox Z, Burke M, et al. Retinal and extraocular cytomegalovirus end-organ disease in HIV-infected patients in Europe: a EuroSIDA study, 1994-2001. Eur J Clin Microbiol Infect Dis 2004;23:550-9.
Skiest DJ, Chiller T, Chiller K, Park A, Keiser P. Protease inhibitor therapy is associated with markedly prolonged time to relapse and improved survival in AIDS patients with cytomegalovirus retinitis. Int J STD AIDS 2001;12:659-64.
Holland GN, Tufail A, Jordan MC. Cytomegalovirus diseases. In: Pepose JS, Holland GN, Wilhelmus KKR, eds. Ocular infection and immunity. St. Louis: Mosby-Year Book, 1996:1088-129.
Wohl DA, Zeng D, Stewart P, et al. Cytomegalovirus viremia, mortality, and end-organ disease among patients with AIDS receiving potent antiretroviral therapies. J Acquir Immune Defic Syndr 2005;38:538-44.
Bassil HF, William DC. Cytomegalovirus encephalitis in an HIV positive patient presenting with a cerebral mass lesion. AIDS Patient Care STDS 1997;11:319-21.
Jacobson MA, O'Donnell JJ, Brodie HR, Wofsy C, Mills J. Randomized prospective trial of ganciclovir maintenance therapy for cytomegalovirus retinitis. J Med Virol 1988;25:339-49.
Martin DF, Sierra-Madero J, Walmsley S, et al. A controlled trial of valganciclovir as induction therapy for cytomegalovirus retinitis. N Engl J Med 2002;346:1119-26.
Mahadevia PJ, Gebo KA, Pettit K, Dunn JP, Covington MT. The epidemiology, treatment patterns, and costs of cytomegalovirus retinitis in the post-haart era among a national managed-care population. J Acquir Immune Defic Syndr 2004;36:972-7.
Springer KL, Weinberg A. Cytomegalovirus infection in the era of HAART: fewer reactivations and more immunity. J Antimicrob Chemother 2004;54:582-6.
Tural C, Romeu J, Sirera G, et al. Long-lasting remission of cytomegalovirus retinitis without maintenance therapy in human immunodeficiency virus-infected patients. J Infect Dis 1998;177:1080-3.
Whitcup SM, Fortin E, Lindblad AS, et al. Discontinuation of anticytomegalovirus therapy in patients with HIV infection and cytomegalovirus retinitis. JAMA 1999;282:1633-7.
Jacobson MA, Zegans M, Pavan PR, et al. Cytomegalovirus retinitis after initiation of highly active antiretroviral therapy. Lancet 1997;349:1443-5.
Karavellas MP, Lowder CY, Macdonald C, Avila CP Jr, Freeman WR. Immune recovery vitritis associated with inactive cytomegalovirus retinitis: a new syndrome. Arch Ophthalmol 1998;116:169-75.
Goldberg DE, Smithen LM, Angelilli A, Freeman WR. HIV-associated retinopathy in the HAART era. Retina 2005;25:633-49.
Kempen JH, Min YI, Freeman WR, et al. Risk of immune recovery uveitis in patients with AIDS and cytomegalovirus retinitis. Ophthalmology 2006;113:684-94.
Reed JB, Briggs JW, McDonald JC, Freeman WR, Morse LS. Highly active antiretroviral therapy-associated regression of cytomegalovirus retinitis: long-Term results in a small case series. Retina 2001;21:339-43.
