Abstract
OBJECTIVE
To determine which patient related variables, available at the point of referral, predict the presence of high grade intraepithelial neoplasia when the smear result is mild dyskaryosis with or without co-existent koilocytosis.
DESIGN
Multivariate analysis of prospective programme trial.
SETTING
Academic Department Colposcopy clinics at Dudley Road Hospital, Birmingham, UK.
SUBJECTS
One hundred and sixty-seven women, whose worst ever cervical smear was mild dyskaryosis with or without koilocytosis, referred to colposcopy clinics. All the women had complete data sets, as determined by the format of an ongoing programme, and all were treated by diathermy loop excision of the cervical transformation zone.
MAIN OUTCOME MEASURES
Age, parity, contraceptive practice, smoking habit, duration of abnormal cytology and the grade of histology in the excised transformation zone.
RESULTS
Forty-seven out of 78 (60%) smokers, compared with 18 out of 73 (25%) nonsmokers, had high grade disease. The other variables considered in the analysis were not significant or very weakly associated with histological grade when analysed in a univariate analysis. Stepwise logistic regression identified cigarette smoking as a powerful independent predictor of high grade disease.
CONCLUSION
These data suggest a strong association between smoking and high grade intraepithelial neoplasia in a population of women whose worst ever smear report was mild dyskaryosis. Smoking is a variable that could be built into models to facilitate referral for colposcopic assessment.
Collapse