The correlation between size of colposcopic lesions and the histological grade, using various colposcopy methods

Poster presentation at IFCPC (London, 2014)

Background

Since its evolution in 1924, colposcopy has developed through the decades. Along with the use of directed biopsies, it has become an established tool in the diagnosis and management of cervical pathology. Over time, various grading systems have been employed to aid description of cervical lesions found at colposcopy; and furthermore, to try and predict the subsequent histological findings. However as demonstrated in a previous meta-analysis, the sensitivity and specificity of predicting the presence of high grade lesions confirmed on histology from their colposcopic findings, can be wide ranged (87-99% and 23-87% respectively). Previous studies have also demonstrated an association between the size of the cervical lesion and the histological grade. However recent research surrounding this topic is lacking. Therefore, we decided to conduct a small audit at a local level, to examine the association between colposcopic findings, in particular the size of the lesion and the subsequent histological grade. We employed three different colposcopy methods, to determine if there was any difference between the methods at determining high grade lesions; these included Standard Colpscopy, Standard DYSIS, and DYSIS using the colourmap. A fourth cohort where the lesion size is measured directly from the colpophotograph stored electronically on Viewpoint software is being analysed. The results will be available later.

Methods

A total of 69 patients underwent one of the three colposcopic methods. Women were examined by qualified colposcopists and a standardised form completed for each patient. They were then grouped according to the size of the lesion found on colposcopy; small < 26%, medium 26-50% and large > 50% and the histology grades were compared.

Results

The results showed the larger the lesion (medium and large), the higher the grade on histology. However, with smaller lesions, this relationship was less defined and high grade lesions were still present. Using Standard Colposcopy and DYSIS colourmap produced similar results with regards to the colposcopists’ impression and the subsequent histology. However, using the standard DYSIS alone appeared to be superior, with 100% correlation between the two. The audit suggests there is an association between lesion size and histological grading. Furthermore, use of the newer DYSIS colposcopy machine may provide better accuracy when determining lesions. Admittedly larger studies are needed to examine this further.