Comparison between in vivo dosimetry and barium contrast technique for prediction of rectal complications in high-dose-rate intracavitary radiotherapy in cervix cancer patients.

PURPOSE: To investigate the correlation between late rectal complications and rectal dose in cervix cancer patients treated with high-dose-rate intracavitary radiotherapy (HDR ICR) and to analyze factors reducing rectal complications. PATIENTS AND METHODS: A total of 136 patients with cervix cancer who were treated with external beam radiotherapy (EBRT) and HDR ICR from 1995 to 1999 were retrospectively analyzed. Radiotherapy (RT) consisted of EBRT plus HDR ICR. The median EBRT dose was 50.4 Gy, and midline block was done after 30-50 Gy of EBRT. A total of six fractions of HDR ICR with 4 Gy fraction size each were applied twice per week to the A point. The rectal dose was calculated at the rectal reference point using the barium contrast criteria. In vivo measurement of the rectal dose was performed with thermoluminescent dosimeter (TLD) during HDR ICR. The median follow-up period was 26 months (range 6-60 months).
RESULTS: A total of 16 patients (12%) experienced rectal bleeding, which occurred 4-33 months (median 11 months) after the completion of RT. The calculated rectal doses did not differ in patients with rectal bleeding and those without, but the measured rectal doses were higher in affected patients. The differences of the measured ICR fractional rectal dose, ICR total rectal dose, and total rectal biologically equivalent dose (BED) were statistically significant. When the measured ICR total rectal dose exceeded 16 Gy, the ratio of the measured rectal dose to A point dose was > 70%; when the measured rectal BED exceeded 110 Gy(3), a high possibility of late rectal complications could be found.
CONCLUSION: In vivo dosimetry using TLD during HDR ICR was a good predictor of late rectal complications. Hence, if data from in vivo dosimetry shows any possibility of rectal bleeding, efforts should be made to reduce the rectal dose.