Prediction of cervical infiltration in Stage II endometrial cancer by different preoperative evaluation techniques (D&C, US, CT, MRI).

OBJECTIVE

Our clinical practice for FIGO Stage II endometrial cancer consists of Wertheim's radical hysterectomy as first choice of treatment. The evaluation of patients is based on D&C. The accuracy of this preoperative staging method is examined here.

METHODS

Twenty-nine patients with endometrial cancer with suspected cervical involvement (FIGO Stage II) based on endocervical curettage underwent Wertheim's radical hysterectomy between January 1, 1989 and December 31, 2001 at the Gynaecological Department of the National Institute of Cancer, Budapest, Hungary. In all cases surgico-pathological staging was performed to examine the accuracy of preoperative D&C and to find out whether radical surgery was necessary in all patients and how the preoperative evaluation of patients should be improved.

RESULTS

Out of 29 patients who underwent Wertheim's hysterectomy the pathological examination found primary cervical cancer in two patients. These two patients were eliminated from further evaluation. Out of the remaining 27 patients only eight (29.6%) had cervical involvement of endometrial cancer documented by a pathologic review on the hysterectomy specimen. Extrauterine disease was documented in one of the patients with cervical infiltration (1/8) and in one in the cervix-negative group (1/19). Ovarian spread was found in the first case and ovarian infiltration with penetration of the tumour into the parametric tissue in the second case. According to the FIGO classification 18 (66.6%) patients had less extensive disease and two (7.4%) had more extensive disease. Only 26% of the patients (7/27) had surgical findings consistent with the Stage II disease.

CONCLUSION

We can conclude that "overtreatment" seems to have occurred in 19 patients, whose cervical infiltration by endometrial cancer could not be proved by pathological staging. It can also be assessed that understaging took place in two cases, which can be explained by two reasons; we did not make use of preoperative imaging techniques since US was applied in six patients, CT in 16 and the most accurate, MRI, on three patients only. The other possible reason, which can point out the bad efficacy of the imaging techniques as well, could be that a major part of the patients received preoperative AL treatment, which could also have influenced the cervical progression. This is possible, but has not been proved. The difference in the number of cervical infiltrations in the group of patients who received preoperative radiotherapy and in the group where they did not, is not significant (p = 0.9742), and infiltration of the endometrium was present in all cases. In the future, proper selection of imaging modalities can improve the staging of gynaecological disorders and preclude unnecessary procedures. In endometrial cancer cases US, especially with the use of TVUS, is often considered to be the primary imaging approach. However, in patients where ultrasound is suboptimal, where there is a large tumour present or the result of imaging studies will directly influence the choice of therapy and guide therapy planning then the higher accuracy of MRI warrants its use. CT is of use in the later stages of disease; differentiation between Stage I and II is difficult and CT is limited in the assessment of myometrial invasion.