Twenty patients with irritable bowel syndrome due to food intolerance were randomized to either oral sodium cromoglycate or placebo in a double-blind cross-over trial. The study consisted of treatment with either sodium cromoglycate or placebo for 8 weeks, followed by the cross-over treatment for 8 further weeks. Patients were allowed to eat the offending foods during the study. Eighteen patients completed the study. Analysis of patients' diary card scores showed a statistically significant difference in favour of sodium cromoglycate. There was a long carry-over effect in the active-placebo order group. Therefore oral sodium cromoglycate seems to be a useful treatment in patients with irritable bowel syndrome and proven food intolerance.

To analyze national prevalence, genomovar distribution, and epidemiology of the Burkholderia cepacia complex in Italy, 225 putative B. cepacia complex isolates were obtained from 225 cystic fibrosis (CF) patients attending 18 CF centers. The genomovar status of these isolates was determined by a polyphasic approach, which included whole-cell protein electrophoresis and recA restriction fragment length polymorphism (RFLP) analysis. Two approaches were used to genotype B. cepacia complex isolates: BOX-PCR fingerprinting and pulsedfield gel electrophoresis (PFGE) of genomic macrorestriction fragments. A total of 208 (92%) of 225 isolates belonged to the B. cepacia complex, with Burkholderia cenocepacia as the most prevalent species (61.1%). Clones delineated by PFGE were predominantly linked to a single center; in contrast, BOX-PCR clones were composed of isolates collected either from the same center or from different CF centers and comprised multiple PFGE clusters. Three BOX-PCR clones appeared of special interest. One clone was composed of 17 B. cenocepacia isolates belonging to recA RFLP type H. These isolates were collected from six centers and represented three PFGE clusters. The presence of insertion sequence IS1363 in all isolates and the comparison with PHDC reference isolates identified this clone as PHDC, an epidemic clone prominent in North American CF patients. The second clone included 22 isolates from eight centers and belonged to recA RFLP type AT. The genomovar status of strains with the latter RFLP type is not known. Most of these isolates belonged to four different PFGE clusters. Finally, a third clone comprised nine B. pyrrocinia isolates belonging to recA RFLP type Se13. They represented three PFGE clusters and were collected in three CF centers.In the late 1970s and 1980s, reports on the recovery of Burkholderia cepacia from cystic fibrosis (CF) specimens began to appear (29), and the emergence of this pathogen was subsequently reviewed (25,35,36). Polyphasic taxonomic studies identified bacteria tentatively classified as B. cepacia as a complex of at least nine closely related species (genomovars). This B. cepacia complex consists of B. cepacia,

Known MRSA epidemic clones such as USA600, USA800, USA1100, and UK EMRSA-3 were described for the first time in Italy. The diffusion of MRSA strains with high pathogenic potential in the CF population suggests that analysis of the MRSA strains involved in pulmonary infections of these patients is needed.

A total of 1,830 specimens (75.7% respiratory and 24.3% nonrespiratory) were cultured in parallel with the MB/BacT and BACTEC 460 TB systems and on Lowenstein-Jensen (LJ) medium. Mycobacteria were identified from 173 (6.5%) specimens. The most common species recovered were Mycobacterium tuberculosis complex (65.9%),Mycobacterium avium complex (22.5%), andMycobacterium chelonae (9.2%). The recovery rates by individual systems were 96.5, 99.4, and 95.9% for MB/BacT, BACTEC 460 TB, and LJ medium, respectively, for all mycobacteria; the recovery rates were 99.1, 100, and 98.2%, respectively, for M. tuberculosis complex alone. The difference among the recovery rates for all mycobacteria and those for individual species was not significant. The BACTEC 460 TB system detected M. tuberculosis isolates more rapidly than the MB/BacT system (8 versus 11.8 days for smear-positive specimens [P < 0.01] and 18 versus 21 days for smear-negative specimens [P < 0.05]), whereas the MB/BacT system more rapidly detected the nontuberculous mycobacteria (17.1 versus 12.7 days [P < 0.01]). These results indicate that the nonradiometric MB/BacT system is a rapid, sensitive, and efficient method for the recovery ofM. tuberculosis and nontuberculous mycobacteria from both pulmonary and extrapulmonary clinical specimens.

Three immunoenzymatic techniques for specific IgE detection (Pharmacia CAP System, Kallestad Allercoat System, Neo Abellò Hamlet-IgE) and the classical Phadebas RAST were compared using 34 sera from patients with a clinical diagnosis of allergic disease and 19 sera from healthy non-atopic controls. IgE antibodies to 9 aeroallergens and 6 food antigens were assessed and 399 tests were run with each method. All techniques showed a high specificity (92%-100%) and satisfactory efficiency (82%-98%), while the sensitivity for RAST, CAP, Allercoat and Hamlet was 89%, 91%, 83% and 53%, respectively, with the lowest values for food allergens. There was a good overall correlation of the four techniques, except when the Hamlet method was compared with the other methods for food-specific IgE detection (correlation coefficient < 0.3). These data indicate that CAP, Allercoat and RAST are satisfactory techniques for specific IgE determination, either for inhalants or for food allergens; CAP, however, offers the highest sensitivity without loss of specificity.