Since the widespread use of combination antiretroviral therapy (cART) in high – middle- and low-income settings, rates of death in people with HIV infection have continued to decline. In the Data collection on Adverse events of anti-HIV Drugs (D:A:D) study, a collaboration of 11 (9 ongoing) cohort studies conducted in Europe, Australia and the USA with just under 50 000 people with HIV infection enrolled, the rate of death has decreased from 17.5 per 1000 person-years (1999-2001) to 9.1 (2009-2011).1 The most common cause of death remains acquired immune deficiency syndrome (AIDS) related, but as a percentage of all deaths, AIDS-related deaths have also declined from 34% (1999-2001) to 22% (2009-2011), p < 0.0001.1 In those with profound immunodeﬁciency (CD4 T-lymphocyte cell (CD4) counts < 50/μL), i.e. those not treated with cART or who have failed to respond to cART, end-organ disease resulting from reactivated cytomegalovirus (CMV) infection is a signiﬁcant cause of mortality and morbidity. Other risks for CMV end-organ disease are other opportunistic infection (OI), high HIV plasma viraemia > 100 000 copies/mL and high plasma CMV viraemia. Although almost any organ system can be aﬀected, the most commonly aﬀected sites in individuals with HIV infection are the retina, colon, oesophagus and central nervous system. CMV retinitis is diagnosed clinically, but most other forms of CMV disease are diagnosed histologically in biopsy samples in association with viral polymerase chain reaction (PCR) assay and culture, or both. The availability of eﬀective oral agents for treating CMV infection, particularly the oral pro-drug, valganciclovir, has simpliﬁed the management of CMV end-organ disease in patients with advanced HIV infection.
13.9.1 Clinical presentations
CMV retinitis was the most common manifestation of reactivated CMV infection and the most common cause of blindness in patients with AIDS, before the use of cART. Up to 25% of patients with AIDS developed CMV retinitis and those with CD4 counts < 50 cells/μL were at greatest risk.2,3 The typical clinical scenario is for a patient with a previous AIDS diagnosis to present with unilateral visual disturbance: decreased visual acuity, ﬂoaters or unilateral visual ﬁeld loss. Initially the lesions are located peripherally. Untreated lesions spread centrally within the aﬀected eye towards the macula and optic disc, and haematogenously to the other eye. CMV retinitis may be asymptomatic. Patients with CD4 counts < 50 cells/μL and patients with non-retinitis CMV end-organ disease are advised to have regular ophthalmological checks to permit early detection and treatment of CMV retinitis. Pupillary dilatation and examination by indirect ophthalmoscopy is essential to visualise peripheral lesions. It is recommended that immunodeﬁcient patients (before and during immune recovery on cART) be educated about symptoms of CMV retinitis and advised to present promptly should they experience these symptoms.
Cytomegalovirus gastrointestinal disease
CMV may cause disease in all parts of the gastrointestinal tract;: ulcerative disease of the colon and oesophagus are the most common sites and occur in 5-10% of patients with AIDS and CMV end-organ disease. Less common sites include the biliary tree, mouth and rectum. Patients with CMV colitis present with chronic diarrhoea, anorexia, weight loss, abdominal pain, nausea, vomiting and fever. CT scanning may reveal non-specific colonic mucosal thickening. Colonoscopy reveals widespread submucosal haemorrhages and diﬀuse mucosal ulcerations. Life-threatening haemorrhage and bowel perforation may occur, the latter especially if the caecum is involved and the patient may present with an acute abdomen.
CMV is the aetiological agent in 10 to 20% of cases of oesophageal disease in patients with HIV. The clinical presentation of CMV oesophagitis is indistinguishable from Candida oesophagitis. Patients who fail to respond to empirical antifungal therapy within 72 hours after presenting with dysphagia and fever should be investigated for the possibility of CMV oesophagitis with endoscopy and biopsy.
Cytomegalovirus neurological disease
CMV typically causes two broad types of neurological dysfunction in immunodeﬁcient individuals: ascending polyradiculopathy and ventriculoencephalitis. CMV polyradiculopathy presents with urinary retention and bilateral leg weakness (a Guillain-Barré-like syndrome) with progressive deterioration over weeks leading to a ﬂaccid paraplegia. CMV ventriculoencephalitis usually presents in the setting of symptomatic CMV disease at another site. Typically, patients present with lethargy, mental slowing, confusion and fever; the onset may be acute with rapid deterioration and death. Cranial nerve defects can occur. Spastic myelopathy and sacral paraesthesia have been reported.
Other manifestations of CMV end-organ disease include pneumonitis, but this is much rarer in HIV infection than it is in the transplant setting. The diagnosis of CMV pneumonia in HIV is one of exclusion, multiple other pathogens including Pneumocystis jirovecii can present with a similar pneumonitis. These other pathogens must be excluded before CMV is deemed the causative pathogen as opposed to a ‘bystander’ in the clinical presentation.
Cytomegalovirus immune reconstitution inflammatory syndrome (IRIS)
A CMV immune recovery uveitis or vitritis is a well recognised complication following the initiation of cART in human immunodeficiency virus (HIV) patients with previous CMV retinitis. In one historical control study, immediate versus delayed cART was associated with a much higher risk of immune recovery uveitis (IRU) i.e. 71% vs 31%.4 Clinically, this condition presents in a similar way to CMV retinitis, with which it was initially confused.5 The prime distinguishing feature is inﬂammation of the anterior chamber and vitreous humour, which is unusual in CMV retinitis. Other distinguishing characteristics are recent onset of cART (within the prior 3 months) and a CD4 count > 200 cells/μL at the time of diagnosis. Attempts to isolate CMV from the vitreous humour are unsuccessful. IRU can occur in patients with previously treated CMV retinitis or in patients with no history of CMV retinitis, the latter representing an unmasking of occult infection through restoration of immune and inflammatory responses against CMV. The inﬂammatory reaction in most cases is self-limited and does not require speciﬁc anti-CMV therapy. However, sight-threatening IRU can occur and may require intervention with corticosteroids.6
The risk of CMV IRU mandates thorough ophthalmological review of patients with profound immunodeﬁciency before initiation of antiretroviral therapy as specific anti-CMV treatment of asymptomatic CMV retinitis is warranted in these individuals. A brief delay (≈ 2 weeks) in the initiation of cART until maintenance CMV therapy is established is recommended, but delaying cART for much longer periods should be avoided. Overall, the small risk of CMV IRU has to be balanced against the risk of other OIs (and death) that might occur during a delay period before the start of cART.7 The exact timing of cART initiation has not been deﬁned (See Section 22.2) but as CMV replication is controlled within 1-2 weeks of specific CMV therapy being introduced, and as the current rates of IRU are low i.e. approx. 0.04 per person-year,8 a ≈ 2 week delay before cART seems reasonable.
Cytomegalovirus antibodies and diagnostic utility
In a large CMV seroprevalence study in Australia, seroprevalence was 57% (95% Confidence Interval [CI], 55.2 to 58.6%), with an association between CMV antibody and increasing age. For children aged 1-2 years of age and 3-4 years of age, seroprevalence was 38% and 39% respectively. In the 20-24-year olds seroprevalence was 50% and in the 50-59-years age group, seroprevalence was 79% (95% CI: 72.7 to 84.7%).9 Rates of CMV antibody seropositivity are reported to be much higher at a younger age in men who have sex with men. CMV is shed in semen, cervical secretions as well as saliva and sexual transmission is a well described acquisition pathway. As CMV antibody seropositivity is so common, detection of CMV IgG antibodies is a near useless test in the diagnosis of CMV end-organ disease. The only place in which this test might have some utility is when a patient is CMV IgG antibody negative, which would make CMV end-organ disease far less likely unless the end-organ disease represented a clinical manifestation of acute CMV infection.
The diagnosis of CMV retinitis is made on clinical grounds, preferably by an experienced ophthalmologist. Typical lesions are yellow white and granular in appearance and follow a vascular distribution. The lesions are associated with perivascular exudates and haemorrhages. The diﬀerential diagnosis includes HIV retinopathy, progressive outer retinal necrosis and other ocular opportunistic infections, such as toxoplasmosis, tuberculosis, syphilis and herpes simplex virus or herpes zoster virus disease. CMV retinitis lesions are referred to as ‘cottage cheese and ketchup’ to distinguish them from the typical ‘cotton wool’ appearance of HIV retinopathy. There should be caution regarding the use of CMV PCR in this setting. CMV PCR in peripheral blood is positive in only ≈ 70% of patients with CMV eye disease, so a negative PCR does not preclude the diagnosis. CMV PCR in vitreous humour is positive in ≈ 80% of patients with CMV eye disease and can be useful in supporting the clinical diagnosis, however, a negative PCR does not exclude the diagnosis.10,11
Other cytomegalovirus end-organ disease
The diagnosis of CMV disease in other organs cannot be made on clinical grounds alone. Isolation of the virus or detection of CMV antigens or DNA from appropriate clinical specimens is required. Viral culture of appropriate specimens on human ﬁbroblast monolayers is the gold standard, and characteristic cytopathic eﬀects may be detected within a few days (however, this is unavailable in many laboratories today, which prefer to use molecular techniques in its place). A rapid technique involving the detection of an immediate-early antigen following overnight culture is also available. Molecular techniques involve CMV PCR assay or techniques to detect CMV antigens. The detection of CMV antigens or CMV DNA in blood, cerebrospinal ﬂuid (CSF) or tissues may yield positive results before a positive culture for CMV. CMV inclusions (Owl’s eye), neutrophilic inﬁltration, non-speciﬁc inﬂammatory changes and ulceration are seen on histological examination of CMV-infected tissue. Importantly, histopathological evidence of inflammation is important in putting the finding of CMV inclusion bodies into context; the presence of inclusion bodies alone might cast some doubt on CMV being the aetiological agent and a search for other pathogens would be mandatory. CMV viraemia (qualitative and quantitative) in blood has poor predictive value for the presence or absence of end-organ disease. In a similar way, there is still uncertainty regarding the best way to use quantitative CMV PCR to monitor response to therapy in those with proven end-organ disease, or to initiate therapy in the absence of clinically apparent end-organ disease.12-14
The diagnosis of CMV neurological disease is often one of exclusion. Magnetic resonance imaging (MRI) generally demonstrates periventricular enhancement. The CSF ﬁndings are non-speciﬁc and generally reveal a normal-to-elevated protein level and a normal-to-low CSF glucose level. There may be a lymphocytic or neutrophilic pleocytosis. CSF CMV culture is usually negative. A positive CSF PCR coupled with typical changes on the MRI add greatly to the likelihood of CMV neurological disease. Biopsy demonstrates periventriculitis with ependymal necrosis and CMV intranuclear inclusion bodies.15
The choice of initial anti-CMV therapy depends on the location and severity of the end-organ disease. The approach to the management of CMV end-organ disease is frequently one of induction-maintenance with specific antiviral agents coupled with the use of cART to effect immune restoration. Choices for induction therapy (usually of 2 to 3 weeks duration) include valganciclovir, intravenous ganciclovir and intravenous foscarnet or cidofovir, with the latter therapies usually used for recurrent or relapsed disease. Combination therapy is sometimes used, in particular for CNS disease, although there is a paucity of data to support this approach. The appropriate timing of cART initiation in patients with CMV end-organ disease is not known, see above in Clinical presentations: cytomegalovirus immune reconstitution inflammatory syndrome (IRIS).
In a paper by Jabs and colleagues from the Studies of Ocular Complications of AIDS Research Group, the effects of systemic therapy (vs intraocular therapy only) on systemic outcomes and the effect of intraocular therapies (ganciclovir implants, intravitreal injections) on ocular outcomes were evaluated.16 Importantly, regimens containing systemic anti-CMV therapy were associated with a 50% reduction in mortality (adjusted hazard ratio [HR], 0.5; 95% CI, 0.3-0.7; p = 0.006), a 90% reduction in new visceral CMV disease (adjusted HR, 0.1; 95% CI, 0.04-0.4; p = 0.004), and among those with unilateral CMV retinitis at presentation, an 80% reduction in second eye disease (adjusted HR, 0.2; 95% CI, 0.1-0.5; p = 0.0005) vs those using only intraocular therapy (implants or injections).
Currently available anti-CMV agents
is a guanosine derivative with anti-CMV activity; the drug also has activity against other herpes group viruses including herpes simplex. After intracellular conversion by a viral phosphotransferase, encoded by CMV gene region UL97, ganciclovir triphosphate is a selective inhibitor of CMV DNA polymerase. The following formulations are available: intravenous (IV), intravitreal injection (2 mg per injection for 1-4 doses over a 7-10 day period). The intravitreal implant has now been discontinued.
The oral formulation has been superseded by the oral pro-drug, valganciclovir, and is no longer available. The intravitreal injection oﬀers no protection or treatment for extraocular CMV disease. The main side-eﬀects of IV ganciclovir, are haematological toxicity (neutropenia and thrombocytopenia), renal dysfunction, nausea and diarrhoea. In some cases, adjunctive use of granulocyte-colony stimulating factor (G-CSF) may be needed in order to avoid or treat neutropenia. The ganciclovir product information (www.tga.gov.au) provides detailed information on dosing, usually, 5 mg per kg twice a day (every 12 hours) as induction, with 5 mg per kg daily as maintenance (if IV maintenance rather than oral maintenance with valganciclovir – see below), as well as guidance on dose reduction for those with renal impairment.
Valganciclovir is a pro-drug (a valine ester) of ganciclovir and has significantly enhanced oral absorption compared with oral ganciclovir with 60% bioavailability. Blood levels obtained are equivalent to those achieved with intravenous ganciclovir (i.e. 900 mg valganciclovir = 5 mg/kg intravenous ganciclovir) although the peak level is higher with the intravenous formulation of the drug. Valganciclovir has similar efficacy and tolerability to intravenous ganciclovir in the induction and maintenance of CMV retinitis.17 However, there is a paucity of data on the use of valganciclovir in the treatment of CMV neurological or gut disease in patients with HIV infection. There are considerably more data on its use both prophylactically and for treatment of CMV disease in the organ transplant setting.18-21 The valganciclovir product information (www.tga.gov.au) provides detailed information on dosing, usually, 900 mg twice a day orally (every 12 hours) as induction, with 900 mg daily orally as maintenance, as well as guidance on dose reduction for those with renal impairment.
Foscarnet, a pyrophosphate analogue, inhibits CMV DNA polymerase. Because this agent does not require phosphorylation to be active, it is also eﬀective against most ganciclovir-resistant CMV isolates. Foscarnet has similar eﬃcacy to intravenous ganciclovir in treating acute CMV retinitis and preventing relapses.22 Foscarnet needs to be administered intravenously. The predominant dose-limiting toxicity is nephrotoxicity. Foscarnet is considerably less well tolerated than ganciclovir and therefore is not recommended as ﬁrst-line therapy. Foscarnet-resistant mutants emerge with continued treatment. Adverse eﬀects include: anaemia, nephrotoxicity, electrolyte disturbance and neurological dysfunction including seizures. The foscarnet product information (www.tga.gov.au), provides detailed information on dosing, usually 60 mg per kg three times a day IV (every 8 hours) or 90 mg per kg twice daily (every 12 hours). Foscarnet has some anti-HIV activity and before the availability of the potent anti-HIV drugs available in our armamentarium and, even today, in those with multi-drug resistant virus, has found some utility in the induction-maintenance approach to treatment.23
Cidofovir is a cytosine nucleotide analogue that has widespread antiviral activity against the herpes group of viruses. Cidofovir has the theoretical advantage of not requiring viral activation. It is active against some ganciclovir-resistant strains. The drug’s prolonged intracellular half-life permits it to be administered once every 2 weeks as maintenance therapy. Cidofovir has similar eﬃcacy to intravenous ganciclovir in previously untreated patients with CMV retinitis.24 Furthermore, cidofovir is associated with improved outcomes in patients with relapsed CMV retinitis who had previously been treated with either ganciclovir, foscarnet or both.25 Renal toxicity, secondary to a dose-dependent proximal tubular eﬀect, is signiﬁcant with cidofovir but can be alleviated, to some extent by co-administration with intravenous hydration with normal saline pre and post dosing plus oral probenecid. The cidofovir product information (www.tga.gov.au) provides detailed information on dosing, usually, 5 mg per kg weekly for 2 weeks (induction) followed by dosing every other week with IV saline (pre and post dose) and2 g of oral probenecid 3 hours before each cidofovir dose and 1 g at 2 hours and again at 8 hours after completion of 1 hour cidofovir infusion (for a total of 4 g).
Pregnancy and breastfeeding
Ganciclovir and valganciclovir are Category D drugs in pregnancy. While there are limited data in pregnancy, animal data indicate the drug is embryotoxic and teratogenic, causing cleft palate, anophthalmia, aplastic kidney and pancreas and hyrocephalus. There have been case reports of safe use in pregnant women. It is unknown if ganciclovir enters breast milk, but this possibility exists as does the potential to cause serious adverse events in the breast-fed infant; breast-feeding should be discontinued. Foscarnet is associated with skeletal anomalies in rats and rabbits. There is no experience with the drug used in the first trimester in humans. Cidofovir is embryotoxic and teratogenic in rats and rabbits. Use in pregnancy for both these agents is not advised.
Use in cytomegalovirus retinitis
Oral valganciclovir, intravenous ganciclovir, intravenous ganciclovir followed by oral valganciclovir, intravenous foscarnet, intravenous cidofovir and the ganciclovir intraocular implant coupled with valganciclovir have all been shown to be eﬀective treatments for CMV retinitis. Slow-release intravitreal ganciclovir is more eﬀective in controlling CMV retinitis within the aﬀected eye than intravenous ganciclovir.26 In severe, sight-threatening CMV retinitis intravitreal injections of either ganciclovir or cidofovir can be undertaken to provide very high local levels of drug and faster control of the infection until steady state intraocular levels of ganciclovir are achieved through IV administration. Co-administration of systemic therapy, usually oral valganciclovir,27,28 is required to prevent CMV disease outside the treated eye. Ganciclovir implants are no longer available, since manufacture was discontinued, but when they were used in the past, remained active for up to 6 months. Part of the reason for their discontinuation is that their use became increasingly rare due to declining incidence of CMV retinitis. Even in those patients who developed disease, the use of highly bioavailable and effective oral agents coupled with the expectation that cART-induced immune recovery were expected to protect against disease relapse even in the absence of a long-acting locally applied anti-CMV drug. Intravitreal injection of ganciclovir remains an option in very severe CMV retinitis as described above.
Use in cytomegalovirus gastrointestinal and neurological disease
Most specialists use parenteral therapy with ganciclovir for 2 to 4 weeks (note: the US guidelines state 2- 6 weeks) 27 until symptoms and signs have resolved. In severe CNS disease, which is often associated with poor outcome including death, combination intravenous ganciclovir and foscarnet may be preferred to stabilise disease despite being associated with high rates of toxicity. Optimal duration of parenteral treatment for CNS disease is unknown. Moreover, the role of valganciclovir in both induction and maintenance of gastrointestinal and CNS disease is unknown. Clinicians are likely more comfortable using oral valganciclovir in the maintenance phase but in some cases may choose to use oral valganciclovir to treat mild gastrointestinal disease. There are no randomised controlled trials to support this approach in the setting of HIV.
In those patients with cART-induced immune recovery, relapse is much less likely. Overall, ganciclovir resistance is less common in the cART era. The 2-year incidence of ganciclovir resistance decreased from 28% before to 9% after 1996.29 Ganciclovir resistance is associated with progression of retinal disease in the cART era.30 Risk factors for resistance include long-term therapy. Low-level resistance to ganciclovir occurs with mutations of the CMV UL97 gene, high-level resistance with mutations of the UL97 plus UL54 genes. Resistance to foscarnet and cidofovir occurs with mutations of the UL54 gene. Cross-resistance exists between ganciclovir and cidofovir and sometimes foscarnet, especially when high-level resistance to ganciclovir exists.27 Resistance testing may be useful (if available) in patients with treatment failure. Importantly, as there is often homology (≈ 90%) between virus in the eye and in blood, sequencing CMV derived from blood is sufficient to confirm the presence of resistance mutations in the end-organ(s) involved.10 Patients with treatment failure may respond to re-induction with the agent used initially for treatment. Alternatively, the use of combination therapy may be tried, or if the relapse is in the eye, placement of a ganciclovir implant in order to achieve very high levels of the drug locally may be warranted.
In the absence of cART-induced immune recovery, maintenance therapy is required. Time to relapse is prolonged from 37 to 145 days in patients who receive maintenance therapy.31 Relapses of non-retinitis CMV disease are less common and maintenance therapy is not recommended for these patients.
The most eﬀective strategy in the prevention of CMV end-organ disease is to avoid immunodeficiency that increases the risk of CMV reactivation, by earlier use of cART. Despite differences in the international guidelines in regards to the CD4 cell count threshold for the introduction of cART, all patients with HIV should be starting cART at CD4 cell counts well above the values at which clinical disease from CMV is a risk. Patients presenting with CMV end-organ disease may therefore represent a failure of the health-care system, either that they were unaware of their HIV diagnosis or were aware of their HIV diagnosis but had not accessed medical care for whatever reason, or that medical care including cART was not available to them, for example, in developing world settings. Another scenario where patients are at increased risk of CMV end-organ disease is where immune recovery is suboptimal. The optimal management of immune failure on virologically suppressive cART is unknown.
Primary prophylaxis for CMV infection is not generally undertaken in Australia because of concerns about cost of therapy, pill burden and the potential to develop resistance. Trials of oral ganciclovir for primary prophylaxis against CMV disease in patients with advanced HIV disease have demonstrated variable eﬃcacy.32,33 Valganciclovir has not been evaluated in this context in the HIV setting. However, data from randomised controlled trials in the transplant setting have led to the production of guidelines on Caring for Australians with Renal Impairment (CARI) for the prevention of CMV disease, in which valganciclovir is one of the anti-CMV drugs recommended as primary prophylaxis.34 In the HIV setting, primary prophylaxis of patients based upon monitoring of CMV PCR levels in blood is not currently recommended. However, there are some data to support such an approach.35,36
Discontinuing maintenance anti- cytomegalovirus therapy
CMV maintenance therapy can be discontinued in patients who experience immune recovery following the introduction of cART and who have a CD4 count > 100 cells/μL for 3 to 6 months. These recommendations are based on several studies that show that individuals with sustained immune recovery remain free of active CMV disease for up to 18 months after ceasing CMV prophylaxis.6,37,38
The safety of this approach for patients with sight-threatening lesions has not been formally evaluated. The decision to cease anti-CMV therapy prophylaxis should be undertaken only with ophthalmological consultation and followed by regular ophthalmological review. However, the risk of reactivation of CMV retinitis in persons who discontinue maintenance therapy is estimated to be low at 0.016 per person-years follow-up.39
Congenital cytomegalovirus infection
Primary infection and reactivation of CMV disease have been reported during pregnancy and these scenarios can lead to congenital CMV infection. The management of CMV in HIV-positive pregnant women is beyond the scope of this chapter and readers should refer to specific guidelines.27
 Smith CJ, Ryom L, Weber R, Morlat P, Pradier C, Reiss P,et al; D:A:D Study Group. Trends in underlying causes of death in people with HIV from 1999 to 2011 (D:A:D): a multicohort collaboration. Lancet 2014;384:241-8.
 Hoover DR, Saah AJ, Bacellar H, Phair J, Detels R, Anderson R, et al. Clinical manifestations of AIDS in the era of Pneumocystis prophylaxis. Multicenter AIDS Cohort Study. N Engl J Med 1993;329:1922-6.
 Kuppermann BD, Petty JG, Richman DD, Mathews WC, Fullerton SC, Rickman LS, et al. Correlation between CD4 counts and prevalence of cytomegalovirus retinitis and human immunodeﬁciency virus-related noninfectious retinal vasculopathy in patients with AIDS. Am J Ophthalmol 1993;115:575-82.
 Ortega-Larrocea G, Espinosa E, Reyes-Terán G. Lower incidence and severity of cytomegalovirus-associated immune recovery uveitis in HIV-infected patients with delayed highly active antiretroviral therapy. AIDS 2005;19:735-8.
 Jacobson MA, Zegans M, Ravan PR, Pavan PR, O’Donnell JJ, Sattler F, et al. Cytomegalovirus retinitis after initiation of highly active antiretroviral therapy. Lancet 1997;349:1443-5.
 Wohl DA, Kendall MA, Owens S, Holland G, Nokta M, Spector SA, et al; ACTG 379 Study Team. The safety of discontinuation of maintenance therapy for cytomegalovirus (CMV) retinitis and incidence of immune recovery uveitis following potent antiretroviral therapy. HIV Clin Trials 2005;6:136-46.
 Hoffmann CJ, Lewis JJ, Dowdy DW, Fielding KL, Grant AD, Martinson NA, et al. Mortality associated with delays between clinic entry and ART initiation in resource-limited settings: results of a transition-state model. J Acquir Immune Defic Syndr 2013;63:105-11
 Jabs DA, Ahuja A, Van Natta M, Lyon A, Srivastava S, Gangaputra S; Studies of the Ocular Complications of AIDS Research Group. Course of cytomegalovirus retinitis in the era of highly active antiretroviral therapy: five-year outcomes. Ophthalmology 2010;117:2152-61.
 Seale H, MacIntyre CR, Gidding HF, Backhouse JL, Dwyer DE, Gilbert L. National serosurvey of cytomegalovirus in Australia. Clin Vaccine Immunol 2006;13:1181-4.
 Hu H, Jabs DA, Forman MS, Martin BK, Dunn JP, Weinberg DV, et al; Cytomegalovirus Retinitis and Viral Resistance Study Group. Comparison of cytomegalovirus (CMV) UL97 gene sequences in the blood and vitreous of patients with acquired immunodeficiency syndrome and CMV retinitis. J Infect Dis 2002;185:861-7.
 Jabs DA, Martin BK, Forman MS, Ricks MO; Cytomegalovirus Retinitis and Viral Resistance Research Group. Cytomegalovirus (CMV) blood DNA load, CMV retinitis progression, and occurrence of resistant CMV in patients with CMV retinitis. J Infect Dis 2005;192:640-9. Erratum in: J Infect Dis 2005;192:1310.
 Yoshida A, Hitomi S, Fukui T, Endo H, Morisawa Y, Kazuyama Y, et al. Diagnosis and monitoring of human cytomegalovirus diseases in patients with human immunodeﬁciency virus infection by use of a real-time PCR assay. Clin Infect Dis 2001;33:1756-61.
 Edwards SG, Grover D, Scott C, Tedder RS, Pillay D, Copas A, et al. Cytomegalovirus viral load testing of blood using quantitative polymerase chain reaction in acutely unwell HIV-1-positive patients lacks diagnostic utility. Int J STD AIDS 2007;18:321-3.
 Wohl DA, Zeng D, Stewart P, Glomb N, Alcorn T, Jones S, 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.
 Salazar A, Podzamczer D, Rene R, Santin M, Perez JL, Ferrer I, et al. Cytomegalovirus ventriculoencephalitis in AIDS patients. Scand J Infect Dis 1995;27:165-9.
 Jabs DA, Ahuja A, Van Natta M, Dunn JP, Yeh S; Studies of the Ocular Complications of AIDS Research Group. Comparison of treatment regimens for cytomegalovirus retinitis in patients with AIDS in the era of highly active antiretroviral therapy. Prospective cohort study, the Longitudinal Study of the Ocular Complications of AIDS. Ophthalmology 2013;120:1262-70.
 Martin DF, Sierra-Madero J, Walmsley S, Wolitz RA, Macey K, Georgiou P, et al. A controlled trial of valganciclovir as induction therapy for cytomegalovirus retinitis. N Engl J Med 2002;346:1119-26.
 Åsberg A, Humar A, Rollag H, Jardine AG, Mouas H, Pescovitz MD, et al; VICTOR Study Group. Oral valganciclovir is noninferior to intravenous ganciclovir for the treatment of cytomegalovirus disease in solid organ transplant recipients. Am J Transplant 2007;7:2106-13.
 Fishman JA, Emery V, Freeman R, Pascual M, Rostaing L, Schlitt HJ, et al.Cytomegalovirus in transplantation - challenging the status quo. Clin Transplant 2007;21:149-58.
 Len O, Gavaldà J, Aguado JM, Borrell N, Cervera C, Cisneros JM, et al; RESITRA. Valganciclovir as treatment for cytomegalovirus disease in solid organ transplant recipients. Clin Infect Dis 2008;46:20-7.
 Snydman DR. Editorial commentary: use of valganciclovir for prevention and treatment of cytomegalovirus disease. Clin Infect Dis 2008;46:28-9.
 Foscarnet-ganciclovir cytomegalovirus retinitis trial: 4. Visual outcomes. Studies of ocular complications of AIDS Research Group, in collaboration with AIDS Clinical Trials Group. Ophthalmology 1994;101:1250-61.
 Leporrier J, Etienne M, Chapuzet C, Peytavin G, Bord S, Borsa-Lebas F, et al. Association of dolutegravir and rilpivirine, enhanced by foscarnet induction, in effective salvage antiretroviral therapy. J Clin Virol 2014;60:428-30.
 Lalezari JP, Stagg RJ, Kuppermann BD, Holland GN, Kramer F, Ives DV, et al. Intravenous cidofovir for peripheral cytomegalovirus retinitis in patients with AIDS. A randomized controlled trial. Ann Intern Med 1997;126:257-63.
 Lalezari JP, Holland GN, Kramer F, McKinley GF, Kemper CA, Ives DV, et al. Randomised, controlled study of the safety and eﬃcacy of intravenous cidofovir for the treatment of relapsing cytomegalovirus retinitis in patients with AIDS. J Acquir Immune Deﬁc Syndr Hum Retrovirol 1998;17:339-44.
 Musch DC, Martin DF, Gordon JF, Davis MD, Kupperman BD. Treatment of cytomegalovirus retinitis with a sustained-release ganciclovir implant N Engl J Med 1997;337:83-90.
 Panel on Opportunistic Infections in HIV-Infected Adults and Adolescents. Guidelines for the prevention and treatment of opportunistic infections in HIV-infected adults and adolescents. Recommendations from the Centers for Disease Control and Prevention, the National Institutes of Health, and the HIV Medicine Association of the Infectious Diseases Society of America. Available at: http://aidsinfo.nih.gov/contentfiles/lvguidelines/adult_oi.pdf (last accessed 27 June 2015).
 Martin DF, Kuppermann BD, Wolitz RA, Palestine AG, Li H, Robinson CA. Oral ganciclovir for patients with cytomegalovirus retinitis treated with a ganciclovir implant. N Engl J Med 1999;340:1063-70.
 Martin BK, Ricks MO, Forman MS, Jabs DA; Cytomegalovirus Retinitis and Viral Resistance Study Group. Change over time in incidence of ganciclovir resistance in patients with cytomegalovirus retinitis. Clin Infect Dis 2007;44:1001-8.
 Jabs DA, Martin BK, Ricks MO, Forman MS; Cytomegalovirus Retinitis and Viral Resistance Study Group. Detection of ganciclovir resistance in patients with AIDS and cytomegalovirus retinitis: correlation of genotypic methods with viral phenotype and clinical outcome. J Infect Dis 2006;193:1728-37.
 Drew WL, Buhles W, Erlich KS. Management of herpes infections (CMV, HSV VZV). In: Sande MA, Volberding PA, editors. The Medical Management of AIDS. 3rd ed. Philadelphia: Saunders; 1992:359-82.
 Spector SA, McKinley GF, Lalezari JP, Samo T, Andruczk R, Follansbee S, et al. Oral ganciclovir for the prevention of CMV disease in persons with AIDS. Roche Cooperative Oral Ganciclovir Study Group. N Engl J Med 1996;334:1491-7.
 Brosgart CL, Louis TA, Hillman DW, Craig CP, Alston B, Fisher E, et al. A randomized, placebo-controlled trial of the safety and eﬃcacy of oral ganciclovir for the prophylaxis of cytomegalovirus disease in HIV-infected individuals. AIDS 1998;12:269-77.
 Pilmore H. Pre-emptive treatment of cytomegalovirus. The KHA-CARI Guidelines – Caring for Australasians with Renal Impairment. March 2011. Available at: http://www.cari.org.au/Transplantation/transplantation%20cmv/Pre-emptive_treatment_CMV_Oct_2011.pdf (last accessed 26 June 2015).
 Spector SA, Wong R, Hsia K, Pilcher M, Stempien MJ. Plasma cytomegalovirus (CMV) DNA load predicts CMV disease and survival in AIDS patients. J Clin Invest 1998;101:497-502.
 Deayton JR, Sabin CA, Johnson MA, Emery VC, Wilson P, Griﬃths PD. Importance of cytomegalovirus viraemia in risk of disease progression and death in HIV-infected patients receiving highly active antiretroviral therapy. Lancet 2004;363:2116-21.
 Tural C, Romeu J, Sirera G, Andreu D, Conejero M, Ruiz S, et al. Long-lasting remission of cytomegalovirus retinitis without maintenance therapy in human immunodeﬁciency virus- infected patients. J Infect Dis 1998;177:1080-3.
 Jouan M, Saves M, Tubiana R, Carcelain G, Cassoux N, Aubron- Olivier C, et al; RESTIMOP study team. Discontinuation of maintenance therapy for cytomegalovirus retinitis in HIV- infected patients receiving highly active antiretroviral therapy. AIDS 2001;15:23-31.
 Walmsley SL, Raboud J, Angel JB, Mazzulli T, Shen S, Casciaro L, et al. Long-term follow-up of a cohort of HIV-infected patients who discontinued maintenance therapy for cytomegalovirus retinitis. HIV Clin Trials 2006;7:1-9.