Evaluation of a screening test for detecting urinary tract infection in newborns and infants

A total of 31 cervical biopsy specimens were taken from 29 women attending a genitourinary medicine clinic, nine women (11 biopsy specimens) were known to have Chlamydia trachomatis cervicitis and 20 women were known to be free of chlamydial infection. The specimens were routinely processed to paraffin wax and stained by an antiChlamydia immunoperoxidase technique to localise the organisms. Of the 11 positive biopsy specimens three showed positive staining of elementary/reticulate bodies. In one case the surface endocervical cells showed large inclusions which were packed with chlamydial bodies. The diagnosis of chlamydial infection is difficult to make clinically and in routine cytological and histological specimens but immunoperoxidase staining can clearly identify C trachomatis inclusions in cervical biopsy specimens provided infection is severe. Department of Pathology, Royal Preston Hospital J M Edwards A R Campbell Department of Genitourinary Medicine, University of Liverpool A Tait Department of Virology, University of Manchester M Lusher Correspondence to: Mr A R Campbell, Department of Pathology, Royal Preston Hospital, Preston, Lancashire PR2 4HG. Accepted for publication 29 May 1991 Infections caused by Chlamydia trachomatis are now recognised as the most prevalent and among the most damaging of all sexually transmitted diseases.' Infection is often low grade and chronic but is an important cause of pelvic inflammatory disease and its sequelae. Treatment is simple and effective but signs and symptoms are often subtle. Papanicolou stained cervical smear cell changes are now regarded as too non-specific to allow Chlamydia to be differentiated from other cervical infections.2 Histological features associated with chlamydial infection in cervical biopsy specimens have recently been described3 and their specificity awaits further confirmation. Use of electron microscopy to search for inclusions has a detection rate of only 4%.4 Fluorescent antibody staining of cervical smears is nearly as sensitive as culture provided the smear is taken from the endocervix,' but fluorescent antibody staining of cervical biopsy specimens is less sensitive.3 A peroxidase-antiperoxidase method has been used on cervical smears with some success in localising inclusions.6 We attempted to stain for C trachomatis inclusions in paraffin wax embedded cervical biopsy specimens using an immunoperoxidase adaptation of the fluorescence technique to give a more convenient and more permanent result. Methods Endocervical swabs were taken from nulliparous, adult women attending a genitourinary medicine clinic. The swabs were immediately transferred to the laboratory in Chlamydia culture medium and an enzyme linked immunosorbent assay (ELISA) test was performed to detect the group specific lipopolysaccharide antigen (LPS) (Antigenz, Northumbria Biologicals Ltd, Newcastle). All positive ELISA cases were confirmed with direct fluorescence (Microtrak, Syva). Eleven known positive and 20 known negative cases were brought back to the clinic one week later and a colposcopically directed cervical biopsy specimen was taken from the squamocolumnar junction. The specimens were fixed in formalin and processed to paraffin wax as normal. A routine haematoxylin and eosin section was examined to check that all biopsy specimens included the squamocolumnar junction. Two monoclonal antibodies were used on the 31 cervical biopsy specimens plus several infected and non-infected McCoy cell monolayer cultures. One method used the genus specific monoclonal antibody LPS (NovoBiolabs) which is the same antigen as is detected in the ELISA technique performed in our laboratory. The other used the species specific monoclonal antibody MOMP (Syva Company, USA) which recognises the major outer membrane protein of C trachomatis elementary bodies.7 Both the commercially available LPS and MOMP monoclonal mouse antibodies are labelled with fluorocien isothiocyanate (FITC). We used two separate immunocytochemistry systems to convert to an immunoperoxidase end point, a 1 in 100 rabbit anti-mouse secondary antibody and a 1 in 100 mouse anti-FITC monoclonal secondary. An avidin-biotin-complex (ABC) system was then used to bypass the FITC signal and visualised using peroxidase and diaminobenzidine (DAB). Both these systems gave identical results. All immunocytochemical experiments included positive and negative control slides. To begin with the positive control consisted of infected McCoy cell monolayers. After positive results were achieved on paraffin wax sections these cases were used as positive controls. The negative control was performed by omitting the primary antibody from the protocol. Results Both monoclonal antibodies LPS and MOMP 1027 group.bmj.com on November 7, 2017 Published by http://jcp.bmj.com/ Downloaded from