Clinical investigation
Cardiac chamber and coronary artery doses associated with postmastectomy radiotherapy techniques to the chest wall and regional nodes

https://doi.org/10.1016/j.ijrobp.2004.04.026Get rights and content

Abstract

Purpose

To compare the estimated radiation doses delivered to the cardiac chambers (CC) and coronary arteries (CA) for 5 postmastectomy radiotherapy (PMRT) techniques.

Methods and materials

A dosimetry study of 20 left-sided PMRT cases was conducted. Cardiac chambers (left and right atria [LA/RA] and left and right ventricles [LV/RV]) and coronary vessels (left main [LM], left anterior descending [LAD], left circumflex [LCX], right coronary [RCA] and posterior descending [PDA] arteries) were contoured on contrast-enhanced CT scans and verified by a cardiologist (T.K.). Five PMRT techniques were applied to each case; 50 Gy in 2 Gy fractions was delivered to the chest wall ± internal mammary node targets. The techniques were: (1) standard tangents (TAN); (2) cobalt (Co); (3) reverse hockey stick (RHS); (4) mixed photon/electron beam (20/80); and (5) partially wide tangent fields (PWTF). Three-dimensional dose calculations for 100 plans were performed for all structures. Plans were compared by using the mean dose (Dmean) and the volume that received more than 30 Gy (V30) and 45 Gy (V45) for each structure.

Results

Cobalt and 20/80 techniques delivered higher Dmeans to the whole heart and individual cardiac chambers (RA, RV, LA, and LV) as compared with the other three techniques. The heart received a Dmean of 21.03 ± 3.5 Gy from Co and 11.87 ± 5.22 Gy from 20/80. The remaining techniques delivered heart Dmeans of 2.90–4.94 Gy. When V30 was used as a metric, all techniques had comparably low V30 to the heart, except for Co, which resulted in a significantly higher irradiated volume of right-sided cardiac chambers (59.06% ± 30.7 for RA F-test < 0.0001; and 61.46% ± 22.13 for RV, F-test < 0.0001). Dmean to the proximal LAD (LAD_p) was significantly higher for RHS (17.64 ± 7.43 Gy) and 20/80 (20.52 ± 8.36 Gy) and lowest for PWTF (9.5 ± 4.16 Gy). The Dmean for the distal LAD (LAD_d) was significantly lower with PWTF (11.02 ± 7.34 Gy) than with all other techniques, including TAN (p < 0.0001). Similar results for PWTF and TAN were observed when V30 and V45 were used.

Conclusions

Cardiac substructures receive the most radiation exposure after PMRT with CO, 20/80 or both and least exposure with PWTF. Although TAN resulted in significant sparing of the majority of the cardiac structures, a significantly higher dose and volume of LAD was exposed when compared with PWTF. Although the clinical relevance of these dose differences is not clearly understood, these dosimetric estimates can serve as a baseline in the development of new techniques for locoregional treatment that will further reduce cardiac exposure.

Introduction

Although contemporary radiotherapy has substantially reduced the volume of heart exposed to radiation, which has resulted in a decrease in ischemic events (1), CT-based studies of breast-only radiation treatment still show delivery of high doses to smaller regions of the left anterior heart, with the left coronary vasculature at particularly high risk 2, 3. Furthermore, a meta-analysis of breast cancer patients treated in randomized trials demonstrated a significant excess of cardiac deaths in patients randomized to radiation therapy (4).

The intentional inclusion of the internal mammary node (IMN) region increases the complexity of radiation delivery because the nodal region lies in close proximity to the heart. Recent studies have demonstrated that certain postmastectomy radiotherapy (PMRT) techniques, in fact, increase cardiac exposure and complication probabilities 5, 6. Although these complication probabilities were generally greater when the internal mammary nodes were treated, some complication probabilities were also increased by the use of standard tangents, which do not intentionally treat the IMN region. To our knowledge, no detailed comparison of cardiac chamber or coronary vessel exposures using CT-based PMRT planning has been done. Therefore, the purpose of this study was to compare the estimated RT dose delivered to the entire heart, left and right atria (LA and RA) and left and right ventricles (LV and RV), and proximal (_p) and distal (_d) segments of each coronary artery for 5 PMRT techniques using three-dimensional–based (3D-based) calculations. We hypothesized that specific PMRT techniques deliver significantly higher mean doses (Dmeans) to the heart and cardiac substructures compared with other techniques, which could increase the risk of radiation-induced damage. To test this hypothesis, we conducted a dosimetry study utilizing CT data and sophisticated 3D tools for treatment planning to compare five common RT techniques.

Section snippets

Methods and materials

Twenty patients who had been treated consecutively for left-sided breast cancer in the Department of Radiation Oncology at the University of Michigan were identified for this study. Each patient had undergone modified radical mastectomy and lymph node dissection for node-positive breast cancer. As part of radiation planning, contrast-enhanced CT scans were obtained in 3-mm slices. During simulation, each patient was immobilized, with the ipsilateral arm above her head. Radiopaque catheters were

Mean dose

Cobalt and 20/80 delivered higher Dmeans to the heart and cardiac chambers (RA, RV, LA, and LV) as compared with the other three techniques (Table 1). The other three techniques had comparably low Dmeans to the heart. Similarly, TAN, PWTF, and RHS delivered low Dmeans to the RA (TAN = 0.60 ± 0.17 Gy, PWTF = 1.92 ± 0.57 Gy, RHS = 1.92 ± 0.96 Gy), RV (TAN = 3.58 ± 3.80 Gy, PWTF = 4.88 ± 1.90 Gy, RHS = 6.30 ± 3.66 Gy), LA (TAN = 0.86 ± 0.22 Gy, PWTF = 2.09 ± 0.49 Gy, RHS = 3.04 ± 1.27 Gy), and LV

Discussion

To our knowledge, this study is the first comparison of PMRT techniques using 3D data and dose–volume analyses of cardiac substructures. We hypothesized that different PMRT techniques would result in variable whole-heart and cardiac-substructure doses, with some techniques resulting in significantly greater dose and potential clinical relevance. Our data show that the Co technique, the 20/80 technique, or both resulted in the highest Dmean to the heart, cardiac chambers, and coronary vessels,

Conclusions

Our data show that the Co technique and the 20/80 technique resulted in the highest exposure to cardiac chambers and coronary vessels, with exception of the LAD, among the PMRT techniques studied here. PWTF resulted in the lowest exposure to partial volumes of the heart. Although TAN resulted in significant sparing of the majority of the cardiac structures, a significantly higher Dmean was delivered to the LAD compared with PWTF. Currently, whether the observed dose differences are clinically

References (26)

  • H.M. Lu et al.

    Reduction of cardiac volume in left-breast treatment fields by respiratory maneuversA CT study

    Int J Radiat Oncol Biol Phys

    (2000)
  • A.J. Nixon et al.

    No long-term increase in cardiac-related mortality after breast-conserving surgery and radiation therapy using modern techniques

    J Clin Oncol

    (1998)
  • Janjan NA. Cardiac doses in post-operative breast irradiation by Fuller et al. (Radiother Oncol. 25(1):19–24, 1992)....

    • Cited by (0)

    View full text
    Copyright © 2004 Elsevier Inc. All rights reserved.