Norovirus (NV) infections are a frequent cause of gastroenteritis (GE), but data on this disease in immunocompromised patients are limited. We analyzed an NV outbreak, which affected immunosuppressed patients in the context of chemotherapy or HSCT. On recognition, 7 days after admission of the index patient, preventive measures were implemented. Attack rates were only 3% (11/334) and 10% (11/105) among patients and staff members, respectively. The median duration of symptoms was 7 days in patients compared with only 3 days in staff members (P = .02). Three patients died of the NV infection. Commonly used clinical diagnostic criteria (Kaplan-criteria) were unsuitable because they applied to 11 patients with proven NV-GE but also to 15 patients without NV-GE. With respect to the therapeutic management, it is important to differentiate intestinal GVHD from NV-GE. Therefore, we analyzed the histopathologic patterns in duodenal biopsies, which were distinctive in both conditions. Stool specimens in patients remained positive for NV-RNA for a median of 30 days, but no transmission was observed beyond an asymptomatic interval of 48 hours. NV-GE is a major threat to patients with chemotherapy or HSCT, and meticulous measures are warranted to prevent transmission of NV to these patients.

Background: We performed a prospective, randomized, open-label trial to evaluate the efficacy of low-dose liposomal amphotericin B (L-AmB) to reduce the incidence of invasive fungal infections (IFI) in patients with hematological malignancies and prolonged neutropenia (>10 days) following intensive chemotherapy. Patients and methods: In 219 neutropenic episodes (NE) of 132 patients randomization was performed. Patients received either 50 mg L-AmB every other day (arm A) or no systemic antifungal prophylaxis (arm B). Results: In the first NE of each patient the incidence of proven or probable IFI (primary end point) was five of 75 patients (6.7%) in arm A and 20 of 57 patients (35%) in arm B (P = 0.001). Invasive aspergillosis occurred less frequently in patients receiving L-AmB-prophylaxis (P = 0.0057), whereas the reduction of invasive candidiasis did not reach statistical significance (P = 0.0655). In all NE the incidence of IFI was five of 110 NE (4.6%) in arm A versus 22 of 109 NE (20.2%) in arm B (P <0.01). Adverse events, possibly related to L-AmB, were observed in five NE (4.6%) and L-AmB was discontinued in three NE (2.8%). No grade 3 or 4 toxicities were observed. Conclusions: Antifungal prophylaxis with low-dose L-AmB proved to be feasible and effective in our trial.

Invasive aspergillosis (IA) is a life-threatening infection in severely immunocompromised patients, especially those receiving intensive chemotherapy or undergoing haematopoietic stem cell transplantation. As the clinical diagnosis of IA is mostly based on biomarkers (galactomannan, β-d-glucan, PCR assays) indicating Aspergillus as the underlying pathogen, the effect of antifungal treatment on the performance of these parameters is still controversial. We evaluated the effect of antifungal treatment on the performance of an Aspergillus-specific PCR assay in bronchoalveolar lavage (BAL) samples. Two-hundred-and-twenty-six BAL samples from 226 patients with haematological malignancies at high risk for IA classified according to the 2008 European Organization for the Research and Treatment of Cancer criteria were analysed retrospectively for the diagnostic performance of a nested Aspergillus PCR assay in relation to the number and type of mould-active antifungals received prior to BAL sampling. Sensitivity of BAL PCR for patients without antifungal treatment prior to BAL sampling was 0.69, whereas specificity was 0.87. While no significant change in diagnostic performance by the addition of one antifungal was observed, receiving two or more antifungals prior to BAL sampling led to a significant decrease in the diagnostic performance of BAL PCR testing (P < 0.009). Treatment with mould-active antifungals prior to BAL sampling significantly decreases the performance of the Aspergillus PCR assay in haematological patients if BAL was performed after administration of more than one antifungal agent. Performing BAL sampling for Aspergillus PCR diagnostic despite pre-treatment with one antifungal or while on prophylaxis is feasible.

Central nervous system (CNS) invasive aspergillosis (IA) is a fatal complication in immunocompromised patients. Confirming the diagnosis is rarely accomplished as invasive procedures are impaired by neutropenia and low platelet count. Cerebrospinal fluid (CSF) cultures or galactomannan (GM) regularly yield negative results thus suggesting the need for improving diagnostic procedures. Therefore the performance of an established Aspergillus-specific nested polymerase chain reaction assay (PCR) in CSF samples of immunocompromised patients with suspicion of CNS IA was evaluated. We identified 113 CSF samples from 55 immunocompromised patients for whom CNS aspergillosis was suspected. Of these patients 8/55 were identified as having proven/probable CNS IA while the remaining 47 patients were classified as having either possible (n = 22) or no CNS IA (n = 25). PCR positivity in CSF was observed for 8/8 proven/probable, in 4/22 possible CNS IA patients and in 2/25 NoIA patients yielding sensitivity and specificity values of 1.0 (95% CI 0.68–1) and 0.93 (95% CI 0.77–0.98) and a positive likelihood ratio of 14 and negative likelihood ratio of 0.0, respectively, thus resulting in a diagnostic odds ratio of ∞. The retrospective analysis of CSF samples from patients with suspected CNS IA yielded a high sensitivity of the nested PCR assay. PCR testing of CSF samples is recommended for patients for whom CNS IA is suspected, especially for those whose clinical condition does not allow invasive procedures as a positive PCR result makes the presence of CNS IA in that patient population highly likely.