Intra-operative frozen section analysis for suspected early-stage ovarian cancer - Twigg - 2012 - BJOG: An International Journal of Obstetrics & Gynaecology
Correspondence
Sir,
We read with interest the article by Cross et al.1
on the use of intra-operative frozen section for suspected early
ovarian cancer. We would like to commend the authors for their work in
providing these data and we recognise the need for mechanisms that can
be used to address the National Institute for Health and Clinical
Excellence (NICE) Guidelines CG122, which recommend assessment of the
para-aortic lymph nodes in women with early ovarian cancer.2
However,
there are a number of areas of practice that we feel need to be
examined further before frozen section procedures can be used to alter
the management of women with suspected early-stage ovarian cancer.
First,
we are perplexed that the authors deemed it necessary to undertake
para-aortic lymphadenectomy for women with borderline ovarian tumours.
These are by nature an unpredictable class of tumour with mostly good
outcomes and little in the way of nonsurgical treatment options when
there is disseminated disease. Further, they are usually early-stage
tumours and so the utility of a para-aortic lymph node dissection is
questionable. If the authors had described the rate of disease in lymph
nodes and the difference in outcome this provided for the woman with
positive nodes their data would lend stronger support for more
widespread implementation.
Accepting this
and examining the authors data for ‘all comers’ (Table 1) we calculate
that 28.8% (415) of women had an appropriate para-aortic lymph node
dissection on the basis of the frozen section prediction, which
represents the real-world scenario for the gynaecological oncology
surgeon waiting in theatre for a frozen section analysis to be phoned
back.
If the authors changed their
protocol to only using dissection in women with malignancy on frozen
section, 63.8% (918) of women would appropriately not undergo a
para-aortic dissection. The total number of women correctly triaged by
frozen section analysis would be 92.6%. Of the remainder, 7% would not
undergo a para-aortic dissection that should and 0.35% would have a
dissection they do not need. Such a protocol change compares with the
authors’ figures who, when including a policy of para-aortic dissection
for borderline tumours on frozen section, overtreated 8% of the women
and undertreated 1.3%.
The answer to
deciding which strategy one would wish to take up must come down to the
differences in outcome for these women, defined by morbidity and
mortality comparisons from overtreatment or undertreatment by surgery or
chemotherapy, respectively, and any subsequent influence this has on
overall survival. Unfortunately the authors do not provide this
information, and only allude to data in preparation that indicate their
ability to increase the stage of a woman’s disease. However, this figure
can be calculated from their data in Table 1 to equate to 82 women
(5.7%) who had a frozen section showing borderline disease but whose
final paraffin section report showed a malignancy. Until other centres
can validate their techniques and such practice can be shown to
translate into a survival benefit for women, it is unlikely that their
data will change surgical practice in women with early ovarian cancer.
References
- 1, , , , , , et al. Intra-operative frozen section analysis for suspected early-stage ovarian cancer: 11 years of Gateshead Cancer Centre experience. BJOG 2012;119:194–201.
Direct Link: - 2National Institute for Health and Clinical Excellence. The recognition and initial management of ovarian cancer. [http://www.nice.org.uk/CG122]. Accessed 20 January 2012.
