This virus has come up twice in the college exams. Question 4 from the first paper of 2019 involved immunocompromised patients, whereas in  Question 21 from the second paper of 2010 the college had asked specifically about infection in the immunocompetent host. 

Oh's Manual only discusses cytomegalovirus in the context of organ or bone marrow transplants, as well as pediatric pneumonia and encephalitis. One my need to resort to Harrison's (Ch. 182) for more details.

A brief summary of what a CICM trainee is expected to know about CMV has been cobbled together from this chapter, as well as from articles such as this piece by Gandhi et al (2004).

CMV among the immunocompetent ICU population is also explored in this 2008 JAMA paper, andits complications are detailed in this excellent CID article from 1997.

Virology

• CMV is a β-herpesvirus.
• It has double-stranded DNA.
• Replicates in the nucleus.
• Able to replicate in a variety of cell types.
• May be present in breast milk, saliva, feces, and urine.
• Not readily spread by casual contact. Requires prolonged or intimate exposure.
• Once infected, it is carried for life.
• Reactivation occurs when T-cell immunity is impaired.
• In organ transplant recipients, the transplanted organ is often the reservoir.

Clinical manifestations of CMV

Disease causd by CMV is generally divided into "CMV syndrome" and "CMV end-organ disease". The actual distinction is probably more relevant to the stratification of patients in trials (Ljungman et al, 2016) rather than as a guide to therapy. For trial purposes, CMV syndrome is  appicabe only to solid organ transplant recipients, and is defined as viraemia plus any two of the following:

• Fever ≥38°C for at least 2 days.
• New or increased malaise or  fatigue 
• Leukopenia or neutropenia 
• >5% atypical lymphocytes
• Thrombocytopenia 
• Elevation of aminotransferases

A brief note on the pathology

• "Cytomegalovirus" is so called because the affected cells are two to four times the size of unaffected cells.
• These cells typically have large "owl's eye" inclusions

Diagnosis of CMV infection

• Positive CMV antibodies (IgM or IgG) - sensitive for recent or acute infection
  • However, this is obviously invalidated by IVIg infusion, plasmapheresis, and such other.
• Antigen assay (for pp65 antigen) is cheap and used to be the standard, but it requires a few neutrophils and cannot be performed in severe neutropenia. The blood needs to be fresh; one is better off sending this test during working hours.
• Qualitative PCR for CMV DNA- very sensitive for the presence of CMV, but they do not distingusih between active and latent infection.
• Quantitative PCR for CMV DNA - ideal test, as it provides a quantitative assessment of viral load, and allows the monitoring of therapy. The WHO have set up an international standard for this technique.
• Tissue histology: one cannot always have access to the tissue, but it does allow the identification of these characteristic cytomegalic cells.
• Viral cultures: These take some time; in vivo CMV is cultured in fibroblasts.

Management of CMV

There is little literature to guide your decisionmaking in the critically ill immunocompetent host. Fortunately, extensive literature exists concerning CMV in the immunocompromised.

• Ganciclovir is first-line: a selective inhibitor of CMV DNA polymerase. Much more activity against CMV than aciclovir.
  • Valganciclovir usually follows: it is the orally bioavailable pro-drug.
  • Ganciclovir-soaked pellets can be sutured into the eye to treat retinitis
• Foscarnet is second-line: it is sodium phosphonoformate, and it also inhibits CMV DNA polymerase.
  • Unfortunately, it is hideously nephrotoxic and neurotoxic.
• Cidofovir is an alternative second-line: a nucleotide analogue, it also selectively inhibits CMV DNA polymerase.
  • It is also nephrotoxic
  • The major approved indication is the management of CMV retinitis
• CMV immunoglobulin is specific antibodies to CMV from immune hosts.
  • The specific purified immunoglobulin may not be available, but experience suggests that non-specific immunoglobulin might be just as good. In spite of little evidence for efficacy or influence on mortality, this seems to be a well-accepted adjunctive strategy to throw at the dying CMV patient; given that CMV pneumonitis (for instance) has a mortality approaching 80%, one can be forgiven for straying out of one's EBM practice norms.