Carcinoma of the cervical stump. The Radiumhemmet series 1959-1987. Treatment and prognosis

Background. The purpose of this retrospective study is to evaluate the long term prognosis for cervical stump cancer compared to matched controls with cancer in an intact uterus. Methods. From 1959 to 1987, 145 patients were treated for an infiltrating c arcinoma of the cervical stump at Radiumhemmet representing 2.2% of all cer vical cancers. Three control cases to each case were selected from the coho rt of cervical carcinoma cases - matched to year of treatment, stage, histo logy and age (plus, minus 2 years). Actuarial survival was calculated for c ases and controls. Survival differences were analyzed with the Kaplan-Meier technique. The age distribution for cases ranged between 36 and 84 years w ith a mean age of 60.6 years. The mean age for the control series is 9 year s of age (range 35-86 years). Among the cases 87.6% were squamous cell carc inoma and 12.4% were adenocarcinomas. Treatment of carcinoma of the uterine stump at Radiumhemmet followed the same modality as was practised for ordi nary cervical cancer cases i.e. two brachyradium applications with 3 weeks interval followed by external irradiation. The dose of irradiation from the intracavitary application given to the stump cancers was lower than that g iven to comparable cases of the common cervical cases. Results. No evidence was found of poorer longterm prognosis for radiologica lly treated squamous cell carcinoma of the uterine stump compared to that o f the ordinary cervical carcinomas. Stump cancers of the adenocarcinoma typ e had a worse prognosis than adenocarcinomas in an intact uterus (p<0.07) a nd also compared with stump cancers of the squamous epithelial type (p=0.05 ). The complication rate was higher for the stump cancer cases compared wit h that for cervical cancers in intact uterus. The mean time interval from s ubtotal hysterectomy to the stump cancer diagnosis was 17.6 years with a ra nge from 1 to 46 years. Conclusions. Recent discussions argue for a better sexual function after su btotal hysterectomy. Our study gave no convincing argument in terms of poor er prognosis for radiologically treated carcinoma of the uterine stump comp ared to that of the total cervical cancer series. It is thus necessary to w eigh the possible gains with subtotal hysterectomy against the relatively l ow risk to fall victim of a stump cancer. Complications following surgery, as well as possible physiologic and sexual functions of the cervix, should be taken into account.