High-dose rate intracavitary therapy for carcinoma of the uterine cervix: II. Risk factors for rectal complication.

From August 1978 through December 1982, a total of 267 patients with carcinoma of the uterine cervix were treated using remote afterloading high-dose rate intracavitary therapy (RALS) with non-rigid applicator at our department. The data from 199 previously untreated patients with standard application out of 267 were available for this analysis of rectal complication. The incidence of moderate to severe rectal complication (Kottmeier's grade 2 and 3) was 7% (13/199). Cox's regression model was used for the analysis of risk factors for rectal complication in which even minor injuries (grade 1) were included. By using this method, it was clear that the first significant risk factor was z-coordinates of weighted geometric center (WGC-z) (p = 0.0007) and the second corresponding factor was rectal TDF (p = 0.0082), the sum of the rectal dose measured by semiconductor dosimeter (ICD-5) and the dose of external whole pelvic irradiation. From the analysis of application pattern of intracavitary sources, WGC-z indicated the approximation of intracavitary sources to the anterior wall of rectum quantitatively and three-dimensionally. The significance of rectal TDF also implied that the monitoring of rectal dose by ICD-5 had an important role for the prediction and prevention of rectal complication after RALS.