The use of active breathing control (ABC) to minimize breathing motion during radiation therapy

Purpose. Reducing the treatment margin for organ motion during breathing reduces the volume of irradiated normal tissues. This may allow a higher dose of radiation to be delivered to the target volume for thoracic and abdominal tumors. However, such margin reduction must not increase the risk of marginal misses which may lead to local failure. In this study, we investigate the feasibility of using Active Breathing Control (ABC) to temporarily immobilize the patient's breathing. Planning CT scans and radiation delivery can then be performed at identical ABC conditions such that a minimal margin for breathing motion can be prescribed safely. Methods and Materials. An active breathing control (ABC) apparatus was constructed consisting of two pairs of flow monitor and scissors valve; one each to control the inhalation and exhalation paths to the patient. The patient breathed through a mouth-piece or face mask connected to the ABC apparatus. A personal computer was used to process the respiratory signal and to display the changing lung volume in real-time. At some time after the patient achieved a stable breathing pattern, the operator activated ABC at a pre-selected point in the breathing cycle. Both valves were then closed to immobilize breathing motion. The period of active breath-hold was that which could be comfortably and repeatedly tolerated by each individual patient, as determined during a training session. The feasibility of the ABC procedure was studied by acquiring volumetric CT scans of a patient during active breath-hold. A helical CT scanner was used. These ABC scans were acquired at one-half to one-third the dose delivered with routine CT scanning. Nine patients with tumors in the thorax and abdomen were studied. Contiguous CT slices were obtained for a region which encompassed the target volume. At least 4 sets of volumetric scans were obtained; one with the patient breathing normally; two ABC scans at the same point near the end of normal inspiration or during deep inspiration; and one ABC scan near the end of normal expiration. For two patients, the ABC scans were repeated 1 week later. Contours of the lungs, liver, and kidneys were outlined when applicable. The positions and volumes of the organs in the different scans were quantified and compared. Results. The ABC procedure was tolerated well by all patients. When ABC was applied near the end of normal expiration, the minimal duration of active breath-hold was 15 seconds for a lung patient, and 20 seconds or more for all other patients. Two patients maintained 50 seconds of breath-hold when ABC was applied during deep inspiration. The usual scan artifacts associated with breathing motion were not observed in the ABC scans. The volumes of soft tissue organs were similar for the ABC scans. The positions of the liver and kidney varied by about 1.5 cm and 0.5 cm respectively between the end of normal inspiration and expiration. Organ positions were highly reproducible for identical points in the breathing cycle. Lung volumes varied by less than 5% in ABC scans that were acquired during the same CT session and when they were separated by 1 week. Figure 1 shows an example of the almost identical diaphragm positions of a normal subject in 2 ABC 'scout-view' profiles taken 5 minutes apart. Conclusions. The results are promising. ABC provides a simple means to reproducibly minimize breathing motion. When applied for treatment, the ABC procedure requires no more than normal operation of the treatment machine. The minimal variation in organ position with repeated application of ABC indicates that it is possible to complete, piece-meal, a treatment that might require extended beam-on time. Research is being conducted to examine the variation of breathing motion during the course of treatment and to incorporate the information in treatment planning