Reactivation of varicella zoster virus (VZV) is a common event in patients undergoing allogeneic bone marrow transplantation (BMT) and may lead to life-threatening complications. We retrospectively analyzed the incidence, clinical outcome, and risk factors for VZV infections occurring within the first 5 years of transplantation in 100 consecutive adults undergoing allogeneic BMT between 1992 and 1997. Forty-one patients (41%) developed VZV reactivation a median of 227 days (range 45-346 days) post-transplantation. Twelve percent of VZV reactivation occurred in the first 100 days and 88% within the first 24 months. Among those who survived for 2 or more years after transplantation (n = 47), 59% developed VZV infection. Forty percent of patients with VZV reactivation required admission with a mean hospital stay of 7.2 days. Two patients developed encephalitis, and 1 died despite antiviral therapy. The most frequent complications were post-herpetic neuralgia and peripheral neuropathy (68%). Thoracic dermatomal zoster represented 41% of the infections; disseminated cutaneous involvement was observed in 17% of patients. No clinical or epidemiologic risk factors were associated with recurrence. Administration of ganciclovir for prevention of cytomegalovirus infection delayed the onset of VZV infection beyond 4 months (P = .06). In a further subset analysis, patients with a limited chronic graft-versus-host disease (GVHD) had a lower estimated incidence of VZV reactivation compared with those with extensive chronic GVHD (P = .11). We conclude that complications from reactivation of VZV infection are common and associated with considerable morbidity and mortality in patients undergoing allogeneic BMT.

A potential association between human herpesvirus 6 (HHV-6) and cytomegalovirus (CMV) following kidney transplantation was explored by retrospectively testing serial serum specimens for HHV-6 IgG and IgM antibody. HHV-6 reactivation occurred in 35 (66%) of 53 transplant recipients. Fungal or parasitic opportunistic infections, graft rejection or loss, and mortality were not associated with HHV-6 reactivation. HHV-6 reactivation was associated with primary CMV infection (P=.001) and CMV syndrome (P=.003) and with trends for CMV-related hepatitis (P=.095), CMV-related neutropenia (P=.104), and serious CMV disease (P=.085). After controlling for CMV immune globulin (CMVIG) prophylaxis, the association between HHV-6 reactivation and primary CMV infection and syndrome remained significant (P=.002 and 0.006, respectively). The reduction in CMV syndrome among those receiving CMVIG prophylaxis remained significant (P=.007) after controlling for HHV-6 reactivation. HHV-6 reactivation in kidney transplant recipients at risk for primary CMV infection is associated with CMV infection and CMV-related disease, and these effects are independent of CMVIG prophylaxis.

A systematic evaluation of the attributable mortality of Stenotrophomonas maltophilia bacteremia was undertaken in a matched, retrospective, case-control study. We determined the attributable mortality rate (26.7%) and mortality risk ratio (an 8-fold increase) of S. maltophilia bacteremia. The attributable mortality rate for S. maltophilia bacteremia is similar to the attributable mortality rate for other nosocomial bloodstream infections.