Gastrointestinal varicella zoster infection. Dissemination or reactivation of a latent virus in the gut?

Varicella zoster virus (VZV) is an exclusively human, highly neurotropic herpes virus that can present as two different clinical entities. The primary infection causes varicella, or chicken pox, while reactivation of the latent VZV causes herpes zoster, also known as shingles. It is less well known that the gastrointestinal system may be commonly affected by VZV infection and produces a variety of clinical symptoms. The pathogenesis of gastrointestinal involvement remains controversial. Recently, we read with interest two articles published in Endoscopy. One was written by Remmerswaal et al. and the other by Krones et al. [1] [2]. Both articles described cases of “disseminated” VZV infection of the gastrointestinal system. We would like to comment on a few aspects regarding the pathogenesis of gastrointestinal VZV infection and these two cases in particular.

The authors of the first case presented an elderly, immunocompromised man with gastric ulcers, hepatitis, and pancreatitis in association with a disseminated maculovesicular rash on the face, trunk, and abdomen, while the authors of the second case presented an elderly, immunocompetent man with multiple esophageal, gastric, and duodenal ulcers associated with a few vesicular eruptions over his head and trunk. Both cases were determined to be VZV. However, it was unclear from the description in either case whether these were cases of primary VZV infection or whether they were due to reactivation of latent VZV. It is very important to make the distinction between the two, as they may present differently. In primary infection, after the contraction of the virus and its initial replication, there is viremia with spread of the virus to the skin as well as to internal organs, such as the liver and spleen. This may produce a diffuse skin rash and hepatitis. Subsequently, the virus establishes latency in the ganglia along the entire neuraxis. It is unclear from the current literature whether the gastrointestinal mucosa is commonly affected by primary VZV infection similarly to the skin. In secondary, or reactivated VZV infection, the virus travels along the axon causing a dermatomal skin rash. Viremia has also been detected in patients with lytic, localized herpes zoster, which does not exclude it as another route of spreading the virus, especially in immunocompromised patients [3]. Furthermore, disseminated cutaneous herpes zoster is usually defined as a generalized eruption of more than 20 extradermatomal vesicles occurring within a week of classic dermatomal herpes zoster. The second case by Krones et al. does not meet this description.

Second, both authors state that in most cases of herpes zoster, latent VZV is reactivated in the dorsal root ganglia and spreads to the viscera. During the past decade, evidence has suggested that the involvement of the gastrointestinal system may be the rule rather than the exception, and that the enteric nervous system of the gut can be a primary site of reactivation [4] [5] [6]. Chen et al. studied samples of the human gut from three groups of children, which included one group who had previously been infected with VZV, one group who had received VZV vaccination, and one group who had neither had varicella nor received the vaccine. VZV DNA and transcripts encoding gene products were identified in 6/6 children with previous varicella infection and in 6 /7 of the children who had been vaccinated. However, neither DNA nor transcripts were detected in any of the seven children who had been neither infected nor vaccinated. These recent studies provide evidence of the ability of VZV to establish latency in the enteric neurons of the gut. It is quite possible that cases of gastrointestinal zoster with no cutaneous involvement represent local reactivation of the VZV in the gastrointestinal system.

Third, viremia has been detected during primary varicella infection as well as in herpes zoster, which makes hematogenous spread to the gastrointestinal system possible [3]. We believe that viremia during herpes zoster in the immunocompetent patient is subclinical and would not cause disseminated VZV infection, whereas in immunocompromised patients, viremia may cause disseminated infection due to the decreased immune status of the body.

In conclusion, we believe that the term “disseminated” should not be used when referring to zoster of the gastrointestinal system. Instead, the term “gastrointestinal zoster” should be used until further confirmatory evidence of the exact pathogenic mechanism is available. In addition, we encourage physicians to routinely search for gastrointestinal complications in patients with herpes zoster, as they may be more common than previously thought.

• References

• 1 Remmerswaal RG, de Vries AC, Ramsoekh D et al. Varicella zoster-associated gastric ulcers, hepatitis and pancreatitis in an immunocompromised patient. Endoscopy 2012; 44: E140
• 2 Krones E, Petritsch W, Valentin T et al. Visceral dissemination of herpes zoster with multiple ulcers in the upper gastrointestinal tract of an apparently immunocompetent patient. Endoscopy 2012; 44: E302-303
• 3 de Jong MD, Weel JF, Schuurman T et al. Quantitation of varicella-zoster virus DNA in whole blood, plasma, and serum by PCR and electrochemiluminescence. J Clin Microbiol 2000; 38: 2568-2573
• 4 Chen JJ, Gershon AA, Li Z et al. Varicella zoster virus (VZV) infects and establishes latency in enteric neurons. J Neurovirol 2011; 17: 578-589
• 5 Gershon AA, Chen J, Gershon MD. A model of lytic, latent, and reactivating varicella-zoster virus infections in isolated enteric neurons. J Infect Dis 2008; 197: 61-5
• 6 Gershon AA, Chen J, Davis L et al. Latency of varicella zoster virus in dorsal root, cranial, and enteric ganglia in vaccinated children. Trans Am Clin Climatol Assoc 2012; 123: 17-33 ; discussion 33–35