The authors of this article1 are to be congratulated on their excellent treatment results. Their conclusion that margin status at the endocervix is an independent prognostic factor is similar to that of our findings.2 They also note that the majority of post-treatment disease is diagnosed more than 2 years after treatment. In their study, in the absence of data on the location of the squamo-columnar junction, the authors use age as a surrogate, assuming that younger women have ectocervical lesions, whereas older women have lesions involving the canal. Unfortunately, as our data have shown, age is not a good indicator of the location of the squamo-columnar junction. The authors’ advice to base the depth of treatment on the woman’s age obscures the important, time-honoured and sensible criterion to tailor the treatment to the extent of endocervical involvement, which was probably the basis of their own highly successful treatment policy. I suggest that we should use demonstrable clinical criteria that individualise every woman’s treatment rather than using an inaccurate surrogate.
It is not appropriate to compare the depth of treatment with the risk of recurrence without knowing the location of the squamo-columnar junction because the surgeon will usually have tailored the depth of treatment on the basis of this characteristic. In Figure 2A, excision to 5 mm in older women seems to have given the best results. This is probably a result of the surgeon using shallower treatment in women with disease only on the ectocervix. We recognised, as a limitation of our study, the fact that our treatment policy was to excise to 10 mm if the lesion was on the ectocervix, 15 mm if the upper limit of the lesion was easily visible in the lower end of the canal and 15–25 mm if the upper limit was not clearly visible.
It is not particularly useful to use histologically derived measurements of the maximum depth of cervical intraepithelial neoplasia, because these are all measured from the surface of the epithelium and take no account of the need to treat disease in the canal. Finally, I was surprised to see the authors stating that a study from their own department, ‘also showed no increased risk of disease recurrence with involved margins’. Indeed, with only a 3-month follow up, Murdoch et al.3 showed that incomplete excision was associated with an increased risk of recurrent disease. With longer follow-up, this finding was confirmed by another study from the department.4 Although the rates are not stated specifically in the present article, it is also clear that incomplete excision is associated with an increased risk of recurrent disease.
The conclusion of this study, as in ours, is that the treating surgeon should aim for complete excision of cervical intraepithelial neoplasia, preferably in a single fragment. Failure to achieve this goal will increase the risk of post-treatment disease requiring a second treatment and potentially increase the risk of an adverse outcome in any subsequent pregnancy.