Prolonged varicella-zoster virus reinfection in an adult after unrelated cord blood transplantation.

Most varicella-zoster virus (VZV) infections after cord blood transplantation (CBT) present as localized herpes zoster. Here, we report a case of VZV reinfection in an adult patient after CBT that appeared clinically to be varicella. A 50-year-old Japanese man underwent CBT for the management of acute lymphoblastic leukemia. Seventeen months later, he developed a small number of vesicles with umbilicated centers. A skin biopsy showed an intraepidermal blister containing degenerated balloon cells. Subsequently, the skin eruption developed over his entire body. The patient was treated with intravenous acyclovir for 5 days, followed by oral valacyclovir for 9 days. It took more than 3 weeks for most of the skin lesions to scab. Serum levels of anti-VZV IgG on days 3 and 33 after the onset of the skin eruption were negative and 260 mIU/ml, respectively. Serum anti-VZV IgM on days 3 and 33 was not detected. Our patient was diagnosed with VZV reinfection.

Introduction

The frequency of umbilical cord blood transplantation (CBT) from unrelated donors has increased among adult patients who require allogeneic hematopoietic stem cell transplantation but have no suitable bone marrow or peripheral blood stem cell donors.[1] Varicella-zoster virus (VZV) infection is a major complication of CBT.[23] Most VZV infections after CBT are cases of localized herpes zoster.[3] Although very rare, there are reports of patients who have undergone CBT and subsequently developed cutaneous disseminated VZV infection.[3] There is a lack of detailed information on the clinical course and serological findings in such cases. Here, we report in detail the clinical course and serological findings of a case of VZV reinfection in an adult post CBT, which appeared clinically to be varicella and followed a severe and prolonged course.

Case Report

A 50-year-old Japanese man with a history of CBT for acute lymphoblastic leukemia in March 2009 developed remittent fever and fatigue on 25th August 2010. Subsequently, he noticed an itchy skin eruption and visited our dermatology department on 1st September, 2010. The patient had a childhood history of varicella. Physical examination revealed sparsely distributed vesicles, some with umbilication, and small erythematous erosions on the face, anterior chest, upper back, and upper extremities [Figure 1a, b]. A tense bulla was also present on the left thigh. A total of seven lesions were observed. Histological analysis of a vesicle showed an intraepidermal blister containing degenerated balloon cells [Figure 1c, d]. Within a few days, there was widespread eruption of vesicles and bullae over the body [Figure 1e, f] and the fever became continuous. Zosteriform lesions were not observed.

Figure 1

Clinical appearance of the skin lesions at day 3: (a) erythematous erosion in the region of the left eyebrow and (b) an umbilicated vesicle. Histological findings in a vesicle on the right forearm: (c) acanthosis with intraepidermal blisters and a mild perivascular infiltration of lymphocytes in the upper dermis (hematoxylin and eosin; original magnification ×40) and (d) an intraepidermal blister containing degenerated balloon cells (hematoxylin and eosin; original magnification ×200). Clinical appearance of the skin lesion at day 10: scattered vesicles and bulla, some of which became erosions on the chest, abdomen (e), and back (f)

He was admitted for intravenous acyclovir (10 mg/kg/day) for 5 days, with a reduced dose as the patient had renal dysfunction of an unknown cause. The skin lesions responded promptly to therapy: no new skin lesions developed and the bullous lesions began to scab, though some lesions remained active. In addition, the fever decreased. The patient was discharged after 5 days of intravenous acyclovir, with oral valacyclovir (20 mg/kg/day) advised for a further 9 days. By September 21, most of the skin lesions had scabbed.

We measured the serum levels of anti-VZV IgG and anti-VZV IgM by enzyme-linked immunosorbent assay kits (Enzygnost Anti-VZV/IgG and Enzygnost Anti-VZV/IgM, Dade Behring, Deerfield, IL, USA) on days 3 and 33 after the appearance of the skin eruption. The values of anti-VZV IgG were expressed in international units (i.e., mIU/ml, with values <57.0 mIU/ml being considered negative); the measurement of anti-VZV IgM were expressed as a cut-off index (i.e., <1.0: negative, 1.0–2.0: indeterminate, and >2.0: positive). Serum anti-VZV IgG was negative on day 3 and 260 mIU/ml on day 33. Serum anti-VZV IgM was negative on both days 3 and 33. Based on the clinical and serological findings we diagnosed VZV reinfection.

Discussion

Both cell-mediated and humoral immunity are important for host defense against VZV. The former is considered to play a more important role in the host defense against VZV, since VZV is cell-associated during active infection. [4] As cord blood contains naive T-cells but no antigen-specific memory T-cells,[5] primary VZV infections by exogenous VZV are expected to occur frequently in CBT recipients. [4] However, localized herpes zoster is the most frequent clinical presentation in adult patients after CBT.[3] In addition to localized herpes zoster, a rare occurrence of VZV reinfection manifesting as a generalized vesicular eruption without herpes zoster lesions, as was seen in our patient, has been previously observed in patients after CBT.[24] This type of VZV reinfection can be caused by either reactivation of latent virus or exogenous reinfection with VZV. Genome analysis is necessary to differentiate between exogenous reinfection and reactivation of latent virus.[67] Since we did not compare VZV DNA between the primary and present infection, it remains unknown whether the infection in our patient was of exogenous or endogenous origin.

Anti-VZV IgG has anti-VZV activity and remains in the body for a long time.[4] Considering that our patient had a childhood history of varicella and that the serum anti-VZV IgG at day 3 was negative, it appears that humoral immunity against VZV had been lost after CBT, resulting in susceptibility to VZV reinfection.

In general, VZV reinfection that appears clinically to be varicella follows a mild clinical course in normal elderly people and immunocompromised hosts.[7] In contrast, the infection in our patient followed a prolonged clinical course. This may be partly explained by the lack of anti-VZV IgG at the onset of the skin eruption. However, Aisa et al.[4] have reported a case of post-CBT varicella with a relatively severe clinical course despite the presence of a high anti-VZV IgG titer. Further investigations are necessary to determine why certain post-CBT patients develop varicella, and to clarify why VZV reinfection in post-CBT patients may be more severe than that in normal aged people and immunocompromised patients.