Cervical Cone Measurements and Residual Disease in LLETZ Conisation for Cervical Intraepithelial Neoplasia

Discussion

This study has shown that the higher the grade of lesion, the greater the percentage of women with conisation depth >10 mm and cone volume > 4ml. Despite the greater amount of cervical tissue removed, as intraepithelial lesions progressed from LGSIL to HGSIL and early invasive cancer, the percentage of incomplete removal also increased. Irrespective of conisation depth and cone volume, the possibility of incomplete lesion removal and thus residual disease was 25% for a woman in the general cohort, ranging from 12.7% in the LGSIL group to 34.3% and 71.5% in the HGSIL and invasive cancer group respectively.

Concerning the oncological limits for safe procedure, it can be seen in Figure 1 that conisation depth has a significant association with the complete/incomplete removal of lesions. In particular, for cones measuring <10 mm in length there was a significantly higher rate of incompletely removed lesions. If the colposcopist were to perform cervical conisation and wanted to significantly reduce the rate of residual disease, then they would have to choose the depth limit of 10 mm. If the colposcopist were to provide a HGSIL patient less than 40 years of age with a 95% chance of no residual disease, then they would have to opt for a deeper cone with depth of 18 mm.

Cone volume was unfortunately not significantly associated with complete/incomplete removal of lesions. This could be due to the method of cone volume calculation applied. Measurements of conisation depth were accurate and unequivocal in all cases with the use of a ruler prior to fixation, whereas volumes were mathematically calculated with the use of the geometric formula for an ellipse. It has been shown in phantom studies (5) that the accuracy of volume measurements with the use of this formula lies within 5-20% of the actual volume. Perhaps the lack of accuracy accounts for the non-significant findings concerning the association between free/non-free excision margins with cone volume. A far more accurate method of cone volume assessment would have been the fluid displacement technique based on Archimedes' principle (6). In the study of Rubio et al. (6) in 1978, volumetric determinations were performed for 247 cone specimens with the use of water displacement technique and the authors demonstrated that a cone volume of <2.5 ml had a significantly higher proportion of residual disease than those cones with greater volume. This is the only report found in literature suggesting that in terms of cone volume, the limit for oncologically safe procedure may probably be 2.5 ml.

This is a retrospective observational study that aims to provide an analysis of cone measurements and to investigate the oncologically safe limits in cervical conisation. There are certain limitations to be considered about our study. First of all, the retrospective nature of our case-series data unfortunately does not warrant a very high level of evidence. Secondly, the surprisingly large proportion of cases with positive or non-free cone margins suggests that LLETZ conisation technique was not optimal, despite being performed by presumably experienced and well-trained colposcopists in our department. Moreover, although cone margins in pathology reports may be positive, the patient may in fact have no residual disease. This can be attributed to the thermal effect of the loop during conisation at the margins of the remaining cervix and also to the use of diathermy for hemostasis, which both eradicate any remaining dysplastic cells. This means that less than 25% of women in our study have residual disease. The precise number is unknown and it is difficult to determine accurately, even with close follow-up with colposcopy/cytology.

Our study has shown that conisation depth <10 mm increases significantly the rate of residual disease and may therefore be a risk factor predicting positive resection margins. Moreover, if cone depth exceeds 18-20 mm, this may lead to no residual disease with 95% probability. However, this would be an unacceptable routine practice as such deep cones (18-20 mm) would obstetrically compromise future pregnancies in women of reproductive age. There are literature reports that large cones expose the patient to a higher risk of future adverse obstetric outcomes, especially when conisation depth is >10 mm (7-9) and cone volume is >4 ml (10). Additionally, previous reports have correlated the amount of cervical tissue removed and the occurrence of adverse obstetric outcomes (11-13). Specifically, if cone depth is >10 mm, there is a significantly increased risk of preterm delivery (pooled relative risk=2.6, 95% CI: 1.3-5.3) (4, 14). Other research has highlighted the association between cone volume and adverse obstetric outcome (9, 10). In these studies, when the excised cone volume was >4 ml, the risk of preterm delivery was 31.7% versus 3.2% (10). Another report found that increased cone depth was associated with a significant increase in the risk of preterm delivery, with an estimated 6% increase in risk per each additional millimeter of tissue excised (15).

In conclusion, conisation depth of <10 mm may be a risk factor predicting positive resection margins, while cone volume is an inappropriate clinical marker. These key points should be taken into consideration by the colposcopist prior to performing LLETZ conisation. In the end, therapy should be undertaken by experienced colposcopists who can achieve the best balance between maximum eradication rates of intraepithelial lesions and minimum disruption of cervical anatomy and function.