Bone Marrow Transplantation (2004) 33, 425–430 & 2004 Nature Publishing Group All rights reserved 0268-3369/04 $25.00 www.nature.com/bmt Viral infections Enterovirus infections following T-cell depleted allogeneic transplants in adults S Chakrabarti1, H Osman2, KE Collingham2, CD Fegan1 and DW Milligan1 1Departments of Haematology, Birmingham Heartlands Hospital, Birmingham, UK; and 2Public Health Laboratory, Birmingham Heartlands Hospital, Birmingham, UK Summary: kingdom, 40 366 confirmed enterovirus isolates were reported in a 20-year period (1975–1994), with 60% of Anecdotally, enteroviruses have been reported to cause the isolates being from children less than 5 years old.3 serious complications post BMT, but the exact impact of Given the fact that patients undergoing allogeneic stem cell these viruses in the post transplant period has not been transplantation are deficient in both cellular and humoral reported. We prospectively evaluated stool, urine and immunity for a prolonged period and are susceptible to a throat samples for enteroviruses by viral culture together number of endogenous and exogenous viral infections with relevant body fluids by RT-PCR in 64 allograft during this period, there is scanty data on the incidence of recipients receiving grafts T-cell depleted by Campath- enterovirus infections in this setting. Anecdotal reports 1H, following both conventional and reduced-intensity have implicated that enteroviruses might be responsible for conditioning. Seven patients (10.4%) developed nine serious complications in the post transplant period.4–8 Some episodes of enterovirus infections at a median of 146 days stool surveillance studies have reported on the isolation of post transplant. Four episodes were associated with these viruses in post transplant patients with diarrhoea, but symptomatic illnesses, which could be attributable to none have investigated the incidence and outcome of enteroviruses. There was no mortality directly related to enterovirus infections in a comprehensive manner.9–11 T- enteroviruses. There was no correlation between dose and cell depletion of the graft results in an increase in the mode of Campath-1H use, lymphocyte recovery, IgG and incidence of several viral infections, but whether the same is IgA levels and enterovirus isolation. Although entero- true for enterovirus infections is not known. viruses tended to be more frequent in TBI-based conven- We report the incidence and outcome of enterovirus tional conditioning recipients, the only significant risk infections with respect to the immune reconstitution in 64 factor for enterovirus infection was unrelated donor graft. adult patients receiving T-cell-depleted grafts following both The low incidence of the severe enterovirus infections conventional and reduced-intensity conditioning (RIC). could have been related to a low lymphocyte count in this cohort in the absence of GVHD, particularly CD4 þ count, which has been implicated in tissue damage in Patients and methods experimental animals. Further studies are needed to define its impact in different allograft settings. We evaluated 64 T-cell-depleted allograft recipients, who Bone Marrow Transplantation (2004) 33, 425–430. were treated in the BMT unit at Birmingham Heartlands doi:10.1038/sj.bmt.1704359 Hospital between May 1997 and September 2001, for Published online 22 December 2003 evidence of enterovirus infections. All the patients were Keywords: enterovirus; stem cell transplantation; T-cell nursed in single rooms with high-efficiency particulate air depletion (HEPA)filters. Conditioning treatment and T-cell depletion Enteroviruses are small RNA viruses belonging to the Conventional conditioning consisted of cyclophosphamide picornaviradae family with 67 distinct serotypes, causing a or etoposide and total body irradiation or busulphan. The wide range of clinical symptoms.1 The annual prevalence of RIC schedule consisted of fludarabine and melphalan. T- symptomatic enteroviral infections in United States is cell depletion was carried out with Campath (anti-CD52) estimated to be about 10–15 million.2 In the United antibodies (Therapeutic Antibody Centre, Oxford) in vivo or in vitro. All patients also received cyclosporin A as graft- versus-host disease (GVHD)prophylaxis from day À1. Correspondence: Dr S Chakrabarti, Department of Haematology, University Hospital Birmingham, Edgbaston, Birmingham B15 2TH, Supportive care UK; E-mail: [email protected] Received 20 August 2003; accepted 24 September 2003 Antimicrobial prophylaxis consisted of oral fluconazole, Published online 22 December 2003 ciprofloxacin and oral aciclovir from the beginning of Enterovirus infection in transplant recipients S Chakrabarti et al 426 conditioning treatment until engraftment. Oral cotrimox- The absolute lymphocyte count (ALC)was noted at least azole was initiated when the neutrophil count was greater once in every 4 weeks following discharge for the first 6 than 1.0 Â 109/l. Metronidazole 400 mg thrice daily was months. ALC was also recorded at the first detection of an started from the day of transplant to engraftment for enterovirus isolate. The CD4 þ and CD8 þ T-cell counts anaerobic decontamination, as a part of GVHD prophy- (since August 1999)were measured every 6–8 weeks post laxis. transplant. Immunoglobulin levels (IgG and IgA)were also Patients at risk of CMV disease (recipient or donor measured every 3 months for the first year following the positive for CMV IgG)received pre-emptive therapy transplantation. For statistical analysis, the median with ganciclovir and/or foscarnet, based on a PCR assay.9 CD4 þ , ALC, IgG and IgA levels between 100 and 180 CMV sero-negative patients received blood products days post transplant were calculated, if more than one set from CMV sero-negative donors. None of the patients of values were available during that period. received prophylactic intravenous immunoglobulins (IVIg). Statistical methods Study design Univariate P-values and odds ratios were calculated from 2 Â 2 contingency tables using Epi info version 6 (CDC, Stool, urine and throat samples were examined by virus Atlanta). The continuous variables were compared using culture from all patients prior to transplant, and weekly to Mann–Whitney’s nonparametric method. The probability of fortnightly thereafter to 180 days post transplant. In various events was examined by the Kaplan–Meier method addition, the stool and urine samples were also examined and the groups were compared using the log-rank test. by electron microscopy. After 180 days post transplant, only relevant samples from symptomatic patients were examined. Tissue samples from liver, lungs or gut obtained Results ante or postmortem were examined by EM and culture. Incidence of enterovirus infection Laboratory methods. For EM, a 50% v/v suspension of each faecal sample was prepared in phosphate-buffered Enteroviruses were isolated in seven out of 64 patients saline, and placed on EM grids with strengthened formvar (10.4%)at a median of 146 days (range 25–796),with an membranes. The grids were stained with 2.5% phospho- actuarial probability of 14% (95% CI 4–24). Two patients tungstic acid and examined using a Jeol 100CX electron had multiple episodes of enterovirus infection. Coxsackie B microscope with an accelerating voltage of 80 kV. For was isolated in three episodes with poliovirus type 1 and an culture, 0.2 ml of the supernatant prepared from centrifu- echovirus in one episode each. The enteroviruses isolated gation of a 10% v/v suspension of faeces or urine in virus from the other four episodes were not serotyped. Entero- transport medium was added to monolayers of Rhesus viruses were isolated from the throat sample in five episodes monkey kidney and Hep 2 and human fibroblast cell lines. and from stool in the other four. The median duration of The cells were incubated at 331C for 14 days and examined virus excretion was 2 weeks (range 1–4 weeks). The thrice weekly for cytopathic effects. Enteroviruses were characteristics of the patients developing enterovirus typed using pooled fluorescein-conjugated monoclonal infections have been detailed in Table 1. antibodies (Chemicon International Inc., Temecula, CA, USA). In addition, cerebrospinal fluid and other relevant Outcome of enterovirus infections sterile body fluid samples were examined by an RT-PCR- based assay for the presence of enterovirus, as previously Symptomatic illness attributable to enteroviruses was described.12 documented in four patients (Table 2). All had upper Specimens for bacteriological and fungal investigations respiratory illness with systemic symptoms. One patient were processed using standard methods of microscopy, (patient 6)had protracted myalgia, weakness and fatigue culture and sensitivity testing. following an enterovirus infection for which no other cause Isolation of enteroviruses from any surveillance site was was identified. Patient 3 had coxsackie B virus isolated termed as enterovirus infection. Enterovirus disease was from a sputum sample during a respiratory illness, which defined as definite on demonstration of enteroviruses from progressed to severe pneumonia, and CMV was subse- a tissue site, and/or detection in the CSF or other sterile quently detected in a bronchial lavage sample by PCR. body fluids with clinical illness attributable to enteroviruses CMV was not detected in blood samples and tissue in the absence of other pathogens. If the viruses were specimens were not available. This patient improved isolated from surveillance samples only,