Role of image guided radiation therapy in obese patients with gynecologic malignancies

Tracy S. Bray, Adelaide Kaczynski, Kevin Albuquerque, Francesca Cozzi, John C. Roeske

Research output: Contribution to journalArticlepeer-review

10 Scopus citations


Purpose: We investigated the effect of body mass index on setup errors by analyzing daily shifts required in treating patients undergoing image guided radiation therapy (IGRT) for gynecologic malignancies. Methods and Materials: Forty successive patients treated with daily kV-based IGRT for gynecologic malignancies between April 2009 and June 2012 were identified. Directional setup corrections were analyzed according to patient body mass index. Random and systematic setup errors were calculated. Image acquisition dose was estimated by performing ionization chamber measurements in a phantom. Results: Obese patients had larger random setup errors, particularly in the right-left (R-L) direction, with a setup error of 7.6 mm, versus 3.9 mm for nonobese patients. The range of individual patient random errors in the R-L direction was 1.5 to 7.6 mm among nonobese patients versus 2.0 to 17.0 mm among obese patients (P = .03, F-test). For obese patients, daily IGRT prevented treating outside the planning target volume in 33% of fractions, versus 16% in the nonobese group (P = .001). The mean total image acquisition dose from daily kV-IGRT was approximately 3 cGy, versus 150 cGy if daily megavoltage portal imaging were used to correct for erratic setup errors. Conclusions: Daily kV-based IGRT in obese patients allows for correction of erratic setup error and minimizes excess dose from portal imaging.

Original languageEnglish (US)
Pages (from-to)249-255
Number of pages7
JournalPractical Radiation Oncology
Issue number4
StatePublished - Oct 2013

ASJC Scopus subject areas

  • Oncology
  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Role of image guided radiation therapy in obese patients with gynecologic malignancies'. Together they form a unique fingerprint.

Cite this