Various techniques of contouring the rectum and their impact on rectal dose-volume histograms.

Late rectal bleeding resulting from radiotherapy has been correlated with rectal dose-volume histograms (DVHs). The techniques of contouring the rectum have been inconsistent within the literature, making interpretations of DVHs difficult. This study was conducted to investigate the impact on rectal DVHs when using different ways of contouring. Ten prostate cancer patients were treated with a 4-field box-technique and received 70 Gy. Six different ways of contouring the rectum were implemented by using 3 different cross-sections and 2 different lengths. The 3 different cross sections were (1) anterior rectal wall (arw): only the anterior half of the rectal wall was contoured; (2) whole rectal wall (wrw): the entire rectal wall was contoured but excluding the rectal contents; and (3) the rectum (rec): including the rectal contents. Two different lengths were used for the above 3 volumes: (1) long (Lg): cranial border starting at where the rectum turned horizontally into the sigmoid and the caudal border 2 cm below the prostatic apex; (2) short (Sh): from 2 cm above to 2 cm below the prostate. Therefore, a total of 6 different volumes (Sh arw, Lg arw, Sh wrw, Lg wrw, Sh rec, and Lg rec) were generated. DVHs of all 6 volumes were compared with the y-axis being percentage volume as well as absolute volume (cc). When using percentage volume as the y-axis, Sh arw gave an impression that a large portion of rectum (median of 41.8%) received high dose (greater than 90% prescribed dose), while the Lg wrw and Lg rec revealed a smaller portion of rectum (median of 17.1% and 14.7%, respectively) received high dose. The other contours were somewhere in-between. When using absolute volume as the y-axis, the DVHs of the 4 rectal volumes, excluding the rectal content (Sh and Lg arw, Sh and Lg wrw), merged at doses greater than 80% to 85% prescription, therefore providing similar information within these high-dose regions. Configurations of rectal DVHs varied drastically with different techniques of contouring and may lead to different interpretations. By using absolute volume (cc) as the y-axis, the shape of the 4 rectal DVHs, excluding the rectal content, were similar in the high-dose region. Reporting rectal toxicities in relations to DVHs using absolute volume, as well as percentage volume, may eliminate inconsistencies secondary to different methods of contouring.