Proffered papers 9.30–10.00 Monday 15 September 2003

Both the original Bethesda system and the current UK classifications of cervical cytology have proved robust but each has a major weakness in the area of abnormalities of uncertain significance. Cytologists recognize that sometimes it is simply impossible to differentiate between reactive and dyskaryotic material. For this reason, the Australian version of the Bethesda system introduced a new category of 'high grade inconclusive' with a recommendation for referral to colposcopy. Approximately 60% of such cases are found to have high grade lesions at colposcopy (Schoolland M, Sterrett G, Knowles S et al .). The present UK system even with the proposed changes requires of the pathologist, a decision as to whether such cases are probably high grade (=a report of moderate dyskaryosis) or not (= a report of borderline). This continues to ignore the fact that sometimes you just cannot tell, even on review. We have taken a consecutive series of 50 referral smears, reported as moderate dyskaryosis, where the histological outcome (by loop cone) is known. These cases were rescreened and then reviewed blind by a pathologist with extensive experience of the Australian NH & MRC modified Bethesda system. On review, the material was reclassified along NH & MRC lines. The results were compared with the biopsy findings in order to determine whether the category of 'inconclusive' might be of value in the context of the NHSCSP.     

Phenotypic differentiation of bifidobacteria of human and animal origins.

The phenotypes of 153 strains belonging or related to the genus Bifidobacterium were studied. These organisms included 38 collection strains and 115 wild strains (41 strains of human origin, 56 strains of animal origin, and 18 strains obtained from rivers or sewage). Our phenotypic analysis revealed seven main groups that were subdivided into 20 subgroups. Seven subgroups contained no type or collection strain. Among the human strains, the type strains of Bifidobacterium pseudocatenulatum and B. catenulatum fell into group I, which contained the type strains of B. adolescentis (subgroup Ib), B. dentium (subgroup Ic), and B. angulatum (ungrouped). The type strain of B. breve belonged to subgroup IIIa1, and the type strains of B. infantis and B. longum fell into subgroup IIIb1. Group VII comprised only wild strains that were isolated from human infant feces. Among the animal strains, group II consisted mainly of bifidobacteria that were isolated from pig feces and contained the type strains of B. suis (subgroup IIb), B. thermophilum (subgroup IIf), B. choerinum, and B. boum (ungrouped). Wild strains belonging to group V were isolated from pig, calf, cow, and chicken feces; this included the type strains of B. animalis (subgroup Va), B. magnum (subgroup Vb), B. pseudolongum, and B. globosum (subgroup Vc). The strains of human origin (groups I, III, and VII) were well separated from the animal strains (groups II, IV, and V). It was not surprising that the wild strains isolated from surface water or sewage were distributed in the animal groups as well as the human groups. Thus, bifidobacteria can be considered to be successful indicators of human or animal fecal pollution when they are correctly classified. The acidification patterns were not adequate to differentiate Bifidobacterium species, as determined previously by Mitsuoka (Bifidobacteria Microflora 3:11-28, 1984) and Scardovi (p. 1418-1434, in P. H. A. Sneath, N. S. Mair, M. E. Sharpe, and J. G. Holt, ed., Bergey's Manual of Systematic Bacteriology, vol. 2, 1986). However, enzymatic tests furnished new taxonomic criteria for the genus.