The diagnosis and management of cervical intraepithelial neoplasia in women with a transformation zone type 3

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)


Twenty percent of colposcopic assessments are inadequate due to a type 3 transformation zone (TZ3). Despite this, the literature relating to this finding is sparse. Management is guided by the referral screening test and, in this thesis, I have shown that the presence of a TZ3 is the strongest predictor of false positive cervical screening results. Analysis of colposcopists’ decision-making, both locally and nationally, identified heterogeneity of care in women with low grade cytology and a TZ3; total length and technique of cytological follow-up were affected by anxiety of missing a cancer and paucity of guidance, suggesting a need for a national consensus opinion. To date, the effectiveness of different cytological sampling techniques in a TZ3 assessment have not been evaluated. In the UK, routine cervical screening is completed by a Cervex-Brush alone. In my thesis, the addition of a cytobrush increased the yield of endocervical cells but this was not associated with increased predictability of CIN2+. This finding is relevant for resource allocation as I propose, that cytological follow-up with a Cervex-Brush alone can be safely undertaken in a primary care setting. My findings suggest women with high grade cytology and a TZ3 should be offered a LLETZ as 80% will have CIN2+. I have also reported, for the first time, that women with low grade cytology, high risk HPV and a TZ3 have double the risk of CIN2+ (36.7%) when compared to women where the TZ is visible. In these women I propose the use of HPV biomarkers (p16 and Ki-67) in combination with liquid based cytology; these biomarkers provide a >99% sensitivity for CIN2+ and improve the specificity of screening from 19.3% to 71.7%. When compared to dual-stained cytology, neither HPV 16/18 genotyping nor p16 & Ki67, in combination with endocervical curettings, demonstrated an equivocal sensitivity.
Date of Award6 Nov 2018
Original languageEnglish
Awarding Institution
  • University of Bristol
SupervisorRachna Bahl (Supervisor), Andrew Wills (Supervisor), Andres Lopez Bernal (Supervisor) & Amit Patel (Supervisor)

Cite this