Fungal infections of the central nervous system and paranasal sinuses in onco-haematologic patients. Epidemiological study reporting the diagnostic-therapeutic approach and outcome in 89 cases.

Invasive fungal infections (IFI) of the Central Nervous System (IFI-CNS) and Paranasal Sinuses (IFI-PS) are rare, life-threatening infections in haematologic patients, and their management remains a challenge despite the availability of new diagnostic techniques and novel antifungal agents. In addition, analyses of large cohorts of patients focusing on these rare IFI are still lacking. Between January 2010 and December 2016, 89 consecutive cases of Proven (53) or Probable (36) IFI-CNS (71/89) and IFI-PS (18/89) were collected in 34 haematological centres. The median age was 40 years (range 5-79); acute leukaemia was the most common underlying disease (69%) and 29% of cases received a previous allogeneic stem cell transplant. Aspergillus spp. were the most common pathogens (69%), followed by mucormycetes (22%), Cryptococcus spp. (4%) and Fusarium spp. (2%). The lung was the primary focus of fungal infection (48% of cases). The nervous system biopsy was performed in 10% of IFI-CNS, and a sinus biopsy was performed in 56% of IFI-PS (P = 0.03). The Galactomannan test on cerebrospinal fluid has been performed in 42% of IFI-CNS (30/71), and it was positive in 67%. Eighty-four pts received a first-line antifungal therapy with Amphotericine B in 58% of cases, Voriconazole in 31% and both in 11%. Moreover, 58% of patients received 2 or more lines of therapy and 38% were treated with a combination of 2 or more antifungal drugs. The median duration of antifungal therapy was 60 days (range 5-835). A surgical intervention was performed in 26% of cases but only 10% of IFI-CNS underwent neurosurgical intervention. The overall response rate to antifungal therapy (complete or partial response) was 57%, and 1-year overall survival was 32% without significant differences between IFI-CNS and IFI-PS. The overall mortality was 69% but the IFI attributable mortality was 33%. Mortality of IFI-CNS/PS remains high but, compared to previous historical data, it seems to be reduced probably due to the availability of newer antifungal drugs. The results arising from this large contemporary cohort of cases may allow a more effective diagnostic and therapeutic management of these very rare IFI complications in haematologic patients.

Bloodstream infections in haematological cancer patients colonized by multidrug-resistant bacteria.

Infections by multidrug-resistant (MDR) bacteria are a worrisome phenomenon in hematological patients. Data on the incidence of MDR colonization and related bloodstream infections (BSIs) in haematological patients are scarce. A multicentric prospective observational study was planned in 18 haematological institutions during a 6-month period. All patients showing MDR rectal colonization as well as occurrence of BSI at admission were recorded. One-hundred forty-four patients with MDR colonization were observed (6.5% of 2226 admissions). Extended spectrum beta-lactamase (ESBL)-producing (ESBL-P) enterobacteria were observed in 64/144 patients, carbapenem-resistant (CR) Gram-negative bacteria in 85/144 and vancomycin-resistant enterococci (VREs) in 9/144. Overall, 37 MDR-colonized patients (25.7%) developed at least one BSI; 23 of them (62.2%, 16% of the whole series) developed BSI by the same pathogen (MDRrel BSI), with a rate of 15.6% (10/64) for ESBL-P enterobacteria, 14.1% (12/85) for CR Gram-negative bacteria and 11.1% (1/9) for VRE. In 20/23 cases, MDRrel BSI occurred during neutropenia. After a median follow-up of 80 days, 18 patients died (12.5%). The 3-month overall survival was significantly lower for patients colonized with CR Gram-negative bacteria (83.6%) and VRE (77.8%) in comparison with those colonized with ESBL-P enterobacteria (96.8%). CR-rel BSI and the presence of a urinary catheter were independent predictors of mortality. MDR rectal colonization occurs in 6.5% of haematological inpatients and predicts a 16% probability of MDRrel BSI, particularly during neutropenia, as well as a higher probability of unfavourable outcomes in CR-rel BSIs. Tailored empiric antibiotic treatment should be decided on the basis of colonization.

PMR studies of phosphatidylcholine-cholesterol vesicles interacting with lucensomycin.

Spin-lattice relaxation times (T1) were measured above the phase transition temperature on sonicated vesicles of egg- or dipalmitoyl-phosphatidylcholine containing cholesterol and/or the polyenic antibiotic, lucensomycin. T1 values of only the terminal methyl groups of the fatty acyl chains were significantly reduced by cholesterol. Lucensomycin caused, more markedly in cholesterol-containing vesicles, a selective reduction of the T1 values of the N-methyl groups. An even more conspicuous decrease, occurring only in cholesterol-containing vesicles, was observed for the transverse relaxation times (T2) of the N-methyl signals upon addition of lucensomycin. The polyene failed to remove the well-known broadening effect of cholesterol on phosphatidylcholine methylene signals. These results indicate that as lucensomycin binds to cholesterol-containing membranes, there is a detectable perturbation of the dynamic structure of the N-methyl groups with an increase in the degree of motional anisotropy. But the non-polar region of the bilayer seems not significantly perturbed by the polyene.