P‐11THE IMPACT OF INTRODUCING LIQUID BASED CYTOLOGY INTO A ROUTINE SCREENING LABORATORY

Introduction: A glacial acetic acid wash performed retrospectively or prospectively on visibly bloodstained cervical ThinPrep® specimens can reduce the unsatisfactory rate and increase the number of diagnostic epithelial cells. This study was undertaken to determine which specimens are most likely to benefit from a prospective glacial acetic acid wash. Methods: Bloodstained ThinPrep® specimens selected for routine lysing prior to processing were macroscopically assessed and scored based on the level of blood present (+ to +++). Both unlysed and lysed slides were prepared from each specimen and microscopically examined. Results: Fifty‐eight specimens (32 scored +, 12 ++ and 14+++) were assessed. Three unlysed slides prepared from the ++ specimens and 12 from +++ specimens were evaluated as unsatisfactory due to excessive blood and inadequate numbers of squamous cells. In contrast, only one of the unlysed slides from the 32 + specimens was considered to be unsatisfactory. Almost all the lysed slides were satisfactory and generally more cellular than the unlysed slides. Abnormal cells were found in four cases (both unlysed and lysed paired slides). Discussion: Although the acetic acid wash increases cellularity of bloodstained ThinPrep® specimens, prospectively lysing all bloodstained specimens is an unnecessary procedure. Lysing only very heavily bloodstained specimens prior to processing will reduce laboratory workload, costs and the possibility of specimen mix up. Occasionally a retrospective wash may be required but screening staff should be aware that although blood may be present on an unlysed ThinPrep® slide, a lysed slide should not be requested unless there are insufficient numbers of squamous cells present.     

Evolutionary relationships among hares from North Africa (Lepus sp. or Lepus spp.), cape hares (L. capensis) from South Africa, and brown hares (L. europaeus), as inferred from mtDNA PCR-RFLP and allozyme data

Systematics and taxonomy of hares of the genus Lepus (Lagomorpha) are under contentious debate, and phylogenetic relationships among many taxa are not well understood. Here we study genetic differentiation and evolutionary relationships among North African hares, currently considered subspecies of Lepus capensis , cape hares ( L. capensis ) from the Cape province in South Africa, and brown hares ( L. europeaus ) from Europe and Anatolia, using maternally (mtDNA) and biparentally (allozymes) inherited markers. A polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) analysis of a c. 1.8 kb long segment of the mitochondrial control region using eight hexanucleotide-recognizing restriction endonucleases yielded 28 haplotypes, and horizontal starch gel electrophoresis of proteins encoded by 25 structural gene loci revealed 52 alleles at 18 polymorphic loci. Diverse phylogenetic analyses (neighbor joining dendrogram, median joining network, multidimensional scaling of pairwise distances, AMOVA, F -statistics, hierarchical F -statistics) of genetic variants revealed marked substructuring of mtDNA into three phylogeographic groups, namely an African, a central European, and an Anatolian, but a somewhat less pronounced overall differentiation of the nuclear genome, despite a relatively high number of population-specific (private) alleles. However, all our results are not incongruent with Petter's (1959: Mammalia 23 , 41; 1961: Z. f. Säugetierkunde 26 , 30; 1972 : Société Des Sciences Naturelles et Physiques du Maroc 52 , 122) hypothesis that North African hares generally belong to L. capensis and that brown hares should be included in this species as well.     

Conditional Maximum Likelihood Estimation Following a Group Sequential Test

We consider estimation after a group sequential test. An estimator that is unbiased or has small bias may have substantial conditional bias (Troendle and Yu, 1999, Coburger and Wassmer, 2001). In this paper we derive the conditional maximum likelihood estimators of both the primary parameter and a secondary parameter, and investigate their properties within a conditional inference framework. The method applies to both the usual and adaptive group sequential test designs. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)