Cytomegalovirus (CMV) Workup

Updated: May 05, 2018
  • Author: Kauser Akhter, MD; Chief Editor: Michael Stuart Bronze, MD  more...
  • Print

Laboratory Studies

Cytomegalovirus (CMV) has been detected via culture (human fibroblast), serologies, antigen assays, PCR, and cytopathology. The IgM level is elevated in patients with recent CMV infection, or there is a 4-fold increase in IgG titers. False-positive CMV IgM results may be seen in patients with EBV or HHV-6 infections, as well as in patients with increased rheumatoid factor levels. [8]

Some tests are sensitive enough to detect anti-CMV IgM antibody early in the course of the illness (CMV early [nuclear] antigen, CMV viral capsid antigen) and during CMV reactivation. As with EBV infection, observing reactivation of the virus with a positive IgM result in the presence of IgG antibody is not uncommon. This is most commonly observed during intercurrent infection in immunocompromised patients.

An anti-CMV immediate early antigen monoclonal antibody assay is available. [44] This reacts with an early protein and can detect CMV infection 3 hours into the infection. Intense coarse granular intranuclear inclusion staining is noted. No other nuclear staining or cytoplasmic staining is visualized. [44]

In the transplant population, antigen assays or PCR is used (sometimes in conjunction with cytopathology) for diagnosis and treatment determinations, with the choice of test varying among institutions.

Antigen testing

Antigenemia is defined as the detection of the CMV pp65 antigen in leukocytes. [7]

The pp65 assay is used to detect messenger matrix proteins on the CMV virus, with either immunofluorescence assay or messenger RNA amplification. These proteins are typically expressed only during viral replication.

Antigen tests are often the basis for institution of antiviral therapy in transplant recipients and may allow for the detection of subclinical disease in high-risk patients. The assay is sensitive and specific yields results quickly.

Antigen assays cannot be used in patients with leukopenia, as these tests detect antigen within neutrophils.

In immunocompromised patients, low or moderate CMV antigenemia may indicate reactivation or infection. [8]

It has been reported that the pp65 antigen assay and quantitative CMV PCR (COBAS Amplicor Monitor Test; see Quantitative polymerase chain reaction) yield similar effectiveness in diagnosing and monitoring patients with active CMV infection. [45]

Qualitative polymerase chain reaction

Qualitative PCR is used to detect CMV in blood and tissue samples.

PCR depends on the multiplication of primers specific for a portion of a CMV gene. The primers usually bind to the area of virus that codes for early antigen.

Qualitative PCR is extremely sensitive, but, because CMV DNA can be detected in patients with or without active disease, the clinical utility of qualitative PCR is limited. [46, 47, 48] Serial PCR may be more helpful clinically.

It yields a positive result before the antigenemia test in transplant recipients with viremia.

Results are typically negative in patients without CMV viremia.

In transplant recipients, a negative CMV PCR result goes against reactivation, but not infection. [44]

Commercially available qualitative PCR testing can be performed using the NucliSens CMV Test, a nucleic acid sequence-based amplification assay (NASBA, Organon Teknika Corporation, Durham, North Carolina).

Quantitative polymerase chain reaction

Quantitative PCR has been used to detect plasma CMV. The advantage of quantitative PCR over regular PCR is unknown. Ideally, quantitative PCR is as sensitive as qualitative PCR and provides an estimate of the number of CMV genomes present in plasma.

A study of newborns compared real-time PCR assays of liquid-saliva and dried-saliva specimens with rapid culture of saliva specimens obtained at birth. Both PCR assays showed high sensitivity and specificity for detecting CMV infection. [49]

A study of more than 3400 blood specimens from organ transplant recipients tested with CMV PCR and pp65 antigenemia found that quantitative real-time PCR for CMV DNA could be used in lieu of antigenemia for monitoring CMV infection and determining when to initiate preemptive treatment. [50]

In theory, the CMV viral load would indicate whether therapy is necessary because patients whose viral load is below a certain cutoff would not develop CMV disease. However, the level of viremia necessary for CMV disease to occur may vary depending on host factors and the type of organ transplant, and this may need to be determined empirically. For example, in CMV retinitis, the viral load has a poor positive predictive value, meaning its clinical utility is limited. A detectable CMV viral load at the time of CMV retinitis diagnosis was shown in one study to correlate with increased mortality (P = 0.007). [51] CMV involvement of the GI tract also has a poor correlation with CMV viremia.

Commercially available quantitative PCR assays include the COBAS Amplicor CMV Monitor test (Roche Diagnostics, Indianapolis, IN) and various laboratory-developed PCR assays. The COBAS Amplicor CMV Monitor test measures viremia in the range of 600-100,000 copies/mL. [52]

In the quest to standardize viral load testing for CMV, the FDA approved marketing in July 2012 of a fully automated assay, the COBAS AmpliPrep/COBAS TaqMan CMV test (CAP/CTM CMV test, Roche Molecular Diagnostics, Pleasanton, CA), which uses an international standard to quantitate plasma CMV load. This assay is available in Europe, but not yet in the United States. It measures viremia in a range of 150-10,000,000 copies/mL. [53]

Because viral loads are not comparable among different assays, it is important to use the same test and same sample type (whole blood or plasma) when monitoring patients over time. [54]

Shell vial assay

The shell vial assay is performed by adding the clinical specimen to a vial that contains a permissive cell line for CMV. The shell vials are centrifuged at a low speed and placed in an incubator. After 24 and 48 hours, the tissue culture medium is removed and the cells are stained using a fluorescein-labeled anti-CMV antibody. The cells are read using a fluorescent microscope. Alternatively, the cells are stained with an antibody against CMV, followed by a fluorescein-labeled anti–immune globulin.

This test has been found to be as sensitive as traditional tissue culture.


Intracellular inclusions surrounded by a clear halo may be demonstrated with various stains (Giemsa, Wright, hematoxylin-eosin, Papanicolaou). This gives the appearance of an "owl's eye" (see Pathophysiology).

Hematoxylin-eosin–stained lung section showing typ Hematoxylin-eosin–stained lung section showing typical owl-eye inclusions (480X). Courtesy of Danny L Wiedbrauk, PhD, Scientific Director, Virology & Molecular Biology, Warde Medical Laboratory, Ann Arbor, Michigan.

Imaging Studies

The diagnosis of CMV pneumonia can be suggested by chest radiography findings, but these findings cannot be used to differentiate between other common causes of pneumonia in immunocompromised hosts. A chest radiograph finding consistent with pneumonia and a BAL result that is CMV positive is a common method for diagnosis.

CT scan is more sensitive for the identification of infiltrate. It has been valuable in patients who present with hypoxia and no infiltrate visible on chest roentgenography.


Other Tests

Cytomegalovirus resistance testing

CMV infection continues to pose a major problem in transplant recipients, and antiviral resistance is encountered in all forms of transplantation. In solid-organ transplant recipients, ganciclovir resistance is found mainly among donor-positive, recipient-negative lung, kidney, and kidney/pancreas transplant recipients. Among stem cell transplant recipients, resistance primarily affects the donor-negative, recipient-positive group. Other risk factors include T-cell depletion, more than 3 months of antiviral therapy, very high viral loads, recurrent episodes of CMV disease, increased levels of immunosuppression, and suboptimal antiviral drug concentrations due to noncompliance or decreased absorption. [55] Resistance to foscarnet and cidofovir has also been reported in solid-organ and stem cell transplant recipients.

Resistance typically takes weeks to months to develop. In fact, among patients with HIV infection, a 10% ganciclovir resistance rate has been reported at 3 months. [55] Resistance should be suspected in patients who initially respond to CMV therapy but who subsequently develop an increasing viral load despite drug compliance. It should also be considered in patients who are clinically deteriorating.

Only two CMV resistance genes have been reported to date: UL-97 and UL-54. UL-97 (a phosphotransferase gene), encodes ganciclovir resistance, while UL-54 (viral DNA polymerase) mutations confer resistance to ganciclovir, foscarnet, and cidofovir. In approximately 90% of patients, ganciclovir resistance initially results from UL-97 mutations. To date, proven ganciclovir resistance mutations in UL-97 are found only in codons 460, 520, and 590-607. Mutations in codons 696-850 mediate foscarnet resistance, and mutations in these sites do not usually mediate cross-resistance to the other anti-CMV drugs. If a patient develops resistance while taking cidofovir, it is caused by a UL-54 mutation, which will encode cross-resistance to ganciclovir. [55]

Specialized assays can be used to test resistance. The most widely used of these is a genotypic assay using fluid samples (eg, CSF, blood) that contain CMV DNA or samples with cultures positive for CMV. Genotype assay results can be performed and results received in a matter of days. Unfortunately, the assay is expensive and may pick up irrelevant mutations. Hence, familiarity in interpreting the results is key.

Other resistance assays include those used to measure viral load via antigenemia or quantitative DNA, as well as a phenotypic plaque reduction assay. [56] The former is not well standardized, and interpretation may vary from one institution to the next. In addition, in certain CMV diseases (eg, retinitis), viral load testing yields a low positive predictive value. [51] The plaque reduction assay takes at least 1 month to complete, is poorly standardized, and is not routinely performed in the laboratory. [56]


Histologic Findings

The hallmark of CMV infection is the finding of intranuclear inclusions consistent with herpesvirus infection. CMV infection may be confirmed using in situ hybridization or direct or indirect staining of intranuclear inclusions using CMV-specific antibodies linked to an indicator system (eg, horseradish peroxidase, fluorescein).

Hematoxylin-eosin–stained lung section showing typ Hematoxylin-eosin–stained lung section showing typical owl-eye inclusions (480X). Courtesy of Danny L Wiedbrauk, PhD, Scientific Director, Virology & Molecular Biology, Warde Medical Laboratory, Ann Arbor, Michigan.
Here, using immunofluorescent technique, a specime Here, using immunofluorescent technique, a specimen of human embryonic lung (25X) reveals the presence of cytomegalovirus. Courtesy of the CDC and Dr Craig Lyerla.