Medical Dosimetry, Vol. 28, No. 1, pp.

7–11, 2003
Copyright © 2003 American Association of Medical Dosimetrists
Printed in the USA. All rights reserved
0958-3947/03/$–see front matter

PII: S0958-3947(02)00136-X

PATIENT TRAINING IN RESPIRATORY-GATED RADIOTHERAPY

VIJAY R. KINI, M.D., SUBRAHMANYA S. VEDAM, M.S., PAUL J. KEALL, PH.D.,
SUMUKH PATIL, B.S., CLAYTON CHEN, B.A., and RADHE MOHAN, PH.D.
Department of Radiation Oncology, Medical College of Virginia and the Hunter Holmes McGuire Veterans Medical
Center, Richmond, VA
( Accepted 15 January 2002)

Abstract— Respiratory gating is used to counter the effects of organ motion during radiotherapy for chest
tumors. The effects of variations in patient breathing patterns during a single treatment and from day to day are
unknown. We evaluated the feasibility of using patient training tools and their effect on the breathing cycle
regularity and reproducibility during respiratory-gated radiotherapy. To monitor respiratory patterns, we used
a component of a commercially available respiratory-gated radiotherapy system (Real Time Position Manage-
ment (RPM) System, Varian Oncology Systems, Palo Alto, CA 94304). This passive marker video tracking system
consists of reflective markers placed on the patient’s chest or abdomen, which are detected by a wall-mounted
video camera. Software installed on a PC interfaced to this camera detects the marker motion digitally and
records it. The marker position as a function of time serves as the motion signal that may be used to trigger
imaging or treatment. The training tools used were audio prompting and visual feedback, with free breathing as
a control. The audio prompting method used instructions to “breathe in” or “breathe out” at periodic intervals
deduced from patients’ own breathing patterns. In the visual feedback method, patients were shown a real-time
trace of their abdominal wall motion due to breathing. Using this, they were asked to maintain a constant
amplitude of motion. Motion traces of the abdominal wall were recorded for each patient for various maneuvers.
Free breathing showed a variable amplitude and frequency. Audio prompting resulted in a reproducible
frequency; however, the variability and the magnitude of amplitude increased. Visual feedback gave a better
control over the amplitude but showed minor variations in frequency. We concluded that training improves the
reproducibility of amplitude and frequency of patient breathing cycles. This may increase the accuracy of
respiratory-gated radiation therapy. © 2003 American Association of Medical Dosimetrists.

Key Words: Respiratory gating, Radiotherapy, Chest tumors, Patient training.

INTRODUCTION These techniques are most effective when the patient’s

Normal tissue toxicity encountered in lung cancer radio-
therapy is frequently due to large treatment margins.1
Respiratory motion necessitates that large margins be
added to the clinical target volume to ensure adequate
dose coverage during the entire range of motion of the
tumor. This intrafraction organ motion forms a critical
component of treatment margins in chest radiotherapy.
Decreasing such motion may lead to significant gains in
chest radiotherapy. Some reports have analyzed the po-
tential for dose escalation if the respiratory component of
the PTV (planning target volume) were eliminated.2 Sec-
ondly, intensity-modulated radiation therapy (IMRT) at-
tempts to reduce normal tissue toxicity while simulta-
neously allowing potential dose escalation.3 However,
the motion artifacts resulting from breathing motion may
compromise the effectiveness of IMRT.4 Various devel-
opments that aim to reduce the negative effects of respi-
ratory organ motion are actively being investigated as
adjuncts to the use of 3DCRT or IMRT for chest tumors.5– 8
Reprint requests to: Vijay R. Kini, M.D., Department of Radia-
tion Oncology, Medical College of Virginia, 401 College Street, PO
Box 980058, Richmond, VA 23298-0058. E-mail: vrkini@hsc.vcu.edu
Presented in part at the International Congress on Radiation
Oncology, Melbourne, Australia, January 2001.
7
breathing pattern during imaging and treatment is reproduc-
ible. However, without any external intervention, patients
cannot voluntarily maintain a stable breathing pattern and
reproduce this pattern from day to day. Some reports have
attempted to characterize organ motion using various im-
aging modalities.9 –12 However, many analyses do not ad-
dress the critical aspect of temporal variation in breathing
patterns and the effect on treatment. To further study this
very important issue related to most modern techniques of
chest radiotherapy, we evaluated the technical feasibility
and effect of using training tools on respiratory patterns.
METHODS AND MATERIALS
The respiratory motion monitoring system em-
ployed in our investigations is a component of a com-
mercially available respiratory-gated radiotherapy sys-
tem (Real Time Position Management (RPM) System,
Varian Oncology Systems, Palo Alto, CA 94304). A
small lightweight hollow plastic box of 6.5 ⫻ 3.5 ⫻ 4
cm3 with 2 or more retroreflective marker dots, each
approximately 5 mm in diameter, is used. This box is
placed on the patient at a spot where the amplitude of
respiratory motion is high, usually the midpoint between
the xyphoid process and the umbilicus. Tattoos, perma-
nent ink, or measurements may be used to precisely
8 Medical Dosimetry
Fig. 1. The gating apparatus. Volunteer in simulation position
showing the infrared reflective marker placed 3 cm inferiorly to
the xyphoid process. The camera with an infrared illuminator
surrounding the lens is inset (left). Headphones used for audio
training and LCD monitor used for visual training are also
shown. Patient’s view of the LCD monitor is inset (right).
reposition this marker on a daily basis. A combined
camera and infrared illuminator unit is mounted on one
of the walls of the simulator computed tomography (CT)/
treatment room, as shown in Fig. 1. The illuminator
floods the marker with infrared light, and the camera,
which is interfaced to a personal computer (PC) records
the reflection of the dots on the marker. The software
installed on the PC digitally records the position of the
reflective dots from the camera image at 30 Hz. This
position trace, as a function of time, serves as the respi-
ratory signal.
Patients were trained to regulate their breathing
patterns. Graphical traces of respiratory patterns were
recorded using “free breathing,” “audio prompting,” and
“visual feedback” modes. These concepts are defined
below.
Free breathing
The patient is instructed to breathe as he/she usually
does at rest. No modifications to his/her breathing pattern
are suggested.
Audio prompting
This technique involves using a computer-generated
prerecorded message that instructs the patient to “breathe
in” or “breathe out” at periodic intervals. This tool is
pre-installed in the commercially available version of the
motion monitoring system. The PC processes the feed-
back from the respiratory motion trace during the pa-
tient’s first few breathing cycles. Based on this informa-
tion, the PC determines the rate at which the patient can
comfortably breathe and “instructs” the patient to breathe
in and breathe out.
Volume 28, Number 1, 2003
Fig. 2. Variation in respiratory patterns in a single patient on 2
days with free breathing. Note the variation in both frequency
and amplitude of motion.
Visual feedback
At our institution, we have developed and incorpo-
rated a visual feedback training tool. This tool consists of
an LCD monitor positioned near the patient (Fig. 1). The
patient can see a continuous trace of his or her own
breathing pattern. He/she is then instructed to perform
various breathing maneuvers while looking at the breath-
ing trace on the LCD monitor. This arrangement allowed
us to set an upper and a lower “threshold line” on the
real-time graphic trace of the abdominal wall motion.
The patient is then instructed to limit the amplitude of
breathing and hence that of the abdominal motion be-
tween these thresholds.
Five patients who were on treatment with conven-
tional radiotherapy for chest malignancies underwent
monitoring. Each patient performed the following
breathing maneuvers at least once: free breathing, breath-
ing using audio prompting, and breathing using visual
feedback. The motion traces were then compared to each
other with respect to the range of motion and the effect
of training on the breathing patterns. An implicit assump-
tion with the gating system used is that the abdominal
marker motion correlates to the tumor motion. For the
purpose of this study, we accepted this assumption.
RESULTS
Figures 2, 3, and 4 show a sample comparison of
breathing traces for free breathing, breathing using audio
prompting, and breathing using visual feedback, respec-
tively, over different sessions for patient no. 5. As shown
in Fig. 2, during free breathing, i.e., without any form of
coaching, breathing patterns showed poor reproducibility
in frequency and amplitude. This was also evident on
consecutive monitoring sessions on the same patient.
As shown in Fig. 3, with audio prompting, patients
 Patient training in respiratory-gated radiotherapy ● V. R. KINI et al.
Fig. 3. Comparison of respiratory patterns in a single patient on
2 days with audio prompts. Note the reproducibility of fre-
quency as well as variation in amplitude.
were able to reproduce the frequency of respiration, and
more regular pattern of respiration was seen. However,
we noticed an increase in the amplitude of motion of the
anterior abdominal wall. Furthermore, the amplitude was
not always the same during the same session, and varied
even more markedly from one session to the next. Pa-
tients were able to achieve a good control over the
amplitude of motion of the anterior abdominal wall with
visual feedback (Fig. 4); however, the frequency of
breathing cycles was variable with this technique.
The bar chart shown in Fig. 5 provides a quantita-
tive analysis of the differences between untrained and
trained respiratory patterns in patient no. 5. This figure
Fig. 4. Comparison of respiratory patterns in a single patient on
2 days with visual feedback. Note the reproducibility of the
amplitude of motion as well as the variation in frequency of
respiration.
9
Fig. 5. Variation in the average duration of a breathing cycle in
a single patient on 4 days with and without training.
shows the average breathing period (time elapsed be-
tween consecutive minima, averaged over one session)
for the same patient for 4 different sessions. It is clear
that the average breathing period for audio prompting
varies much less compared to visual feedback mode.
Free breathing shows the most variation in time period.
The data for all 5 patients is summarized in Figs. 6
through 8. Figure 6 shows that the range of motion is
generally significantly smaller for visual feedback mode
as compared to audio prompting mode. Figures 7 and 8
show intra- and inter-session variability of range of mo-
tion and breathing period in terms of their standard
deviations (composite of all sessions), respectively. The
data for the first patient was acquired while developing
the procedures and likely represents an effect of our
learning curve. For visual feedback, the variability in the
range of motion for patient nos. 2 through 5 is small. For
audio prompting, it can be more variable, indicating the
reduced control over breathing amplitude in this mode.
Similarly, except for the first patient, the breathing pe-
riod variability is small for audio prompting but can be
more variable for the visual feedback.
DISCUSSION
We have evaluated the feasibility of using various
training tools and their effect on respiratory patterns.
This may allow us to fashion gated radiotherapy with a
higher level of confidence. Patients were able to tolerate
these maneuvers well and follow instructions when audio
prompting or visual feedback techniques were utilized.
Three-dimensional conformal radiation therapy
planning typically takes into account the range of motion
of the tumor during a breathing cycle by assigning a
portion of the margin to this motion. (See, for example,
RTOG 93-11 protocol for non-small cell lung cancers.)
10 Medical Dosimetry
Fig. 6. Range of motion for all patients with and without
training. (ROM ⫽ range of motion)
A fluoroscopic simulation provides a reasonable magni-
tude of this range of motion. However, we found that this
range may vary during a single treatment fraction (Fig. 2)
as well as between treatment fractions, i.e., from day to
day, in the same patient. This could conceivably alter the
coverage of the tumor if the PTV is based on a single
fluoroscopy session. More commonly, however, CT-
based planning uses published numbers to assign a value
for the intrafraction range of motion of the tumor. This
could prove to be equally inaccurate. Currently, it is not
clear what role respiratory motion plays on dose-volume
histogram (DVHs) and isodose coverage of the target.
Newer treatment techniques such as IMRT and respira-
tory-gated therapy attempt to decrease treatment mar-
gins. If daily variations in patient breathing patterns are
Fig. 7. Variability in the range of motion in all patients with and
without training. (ROMSD ⫽ standard deviation of the range of
motion).
Volume 28, Number 1, 2003
Fig. 8. Variation in the average duration of a respiratory cycle
in all patients with and without training. (Period SD ⫽ standard
deviation of the average duration of the breathing cycle)
not acknowledged while using such techniques, PTVs
based on a single imaging study could possibly under-
dose the target and/or overdose normal tissues.
Audio prompting improved the frequency of breath-
ing, but resulted in a variation in the amplitude of motion
(Figs. 3, 5, and 8). The impact of an increase in the range
of motion of the tumor depends on the treatment tech-
nique. Most new techniques aim to decrease the portion
of the PTV that accounts for the intrafraction motion. If
the target has a greater range of motion, a larger volume
of normal tissue has to be included in the PTV to account
for this motion. This would increase the toxicity of
conventional (non-gated) radiotherapy. This increase in
amplitidue could also have a negative impact on the
logistics involved in administering gated radiotherapy.
The Varian respiratory-gating system is based on the
principle of turning the radiotherapy beam on and off
depending upon the position of the target. As such, any
portion of the breathing cycle may be used to treat the
target, based on the amplitude of motion of the target. If
the amplitude of motion is large, a smaller portion of a
single breathing cycle may be used to treat in a gated
fashion to reduce the volume of normal tissue irradiated.
Conceivably, a larger number of breathing cycles would
then be required to deliver the same number of monitor
units using gating. Thus, the increase in amplitude with
audio prompting could prove to be detrimental to patient
throughput by increasing treatment time.
Showing patients a trace of their respiratory excur-
sion seemed to be the logical method to accomplish
control over the variation in the amplitude of motion.
Visual feedback was more successful in controlling the
amplitude of breathing motion in most patients. While
the concept of using video training has previously been
mentioned, quantitative studies of this method are not
available. This could perhaps be attributed to overcom-
 Patient training in respiratory-gated radiotherapy ● V. R. KINI et al.
pensation by patients when they saw the trace approach-
ing the “thresholds” set by the operator. We should note,
however, that constancy of amplitude is much more
important for gated treatments than constancy of fre-
quency. Furthermore, visual feedback may also allow the
patient to hold his or her breath at a predefined repro-
ducible level, i.e., a “gated breath-hold” mode of imag-
ing and treatments. Because audio prompting can repro-
duce the frequency of the patient’s breathing cycle and
visual feedback can limit the amplitude of motion, all
aspects of the patient’s respiratory pattern could, in prin-
ciple, be made reproducible with adequate coaching.
CONCLUSIONS
We concluded that training improves the reproduc-
ibility of amplitude and frequency of patient breathing
cycles. This may increase the accuracy of respiratory-
gated radiation therapy.
REFERENCES
1. Graham, M.V.; Purdy, J.A.; Emami, B.; et al. Clinical dose-
volume histogram analysis for pneumonitis after 3D treatment for
non-small cell lung. Int. J. Radiat. Oncol. Biol. Phys. 45:323–9;
1999.
2. Ten Haken, R.K.; Balter, J.M.; Marsh, L.H.; Robertson, J.M.;
Lawrence, T.S. Potential benefits of eliminating target volume
11
expansions for patient breathing in the treatment of liver tumors.
Int. J. Radiat. Oncol. Biol. Phys. 38:613–7; 1997.
3. Mohan, R.; Liebel, S.A. Radhe. Intensity modulation of the radi-
ation beam. In: DeVita, V.S., Hellman, S., Rosenberg, S.A., edi-
tors. Cancer: Principles and Practice of Oncology. 5th ed. Phila-
delphia: Lippincott Raven; 1997: 3093–3106.
4. Yu, C.X.; Jaffray, D.A.; Wong, J.W. The effects of intra-fraction
organ motion on the delivery of dynamic intensity modulation.
Phys. Med. Biol. 43:91–104; 1998.
5. Hanley, J.; Debois, M.M.; Mah, D.; et al. Deep inspiration breath-
hold technique for lung tumors: The potential value of target
immobilization and reduced lung density in dose escalation. Int. J.
Radiat. Oncol. Biol. Phys. 45:603–11; 1999.
6. Kubo, D.H.; Len, P.M.; Minohara, S.; Mostafavi, H. Breathing-
synchronized radiotherapy program at the University of California
Davis Cancer Center. Med. Phys. 27:346 –53; 2000.
7. Kubo, D.H.; Hill, B.C. Respiration gated radiotherapy treatment: A
technical study. Phys. Med. Biol. 41:83–91; 1996.
8. Wong, J.W.; Sharpe, M.B.; Jaffray, D.A.; et al. The use of active
breathing control (ABC) to reduce margin for breathing motion.
Int. J. Radiat. Oncol. Biol. Phys. 44:911–9; 1999.
9. Davies, S.C.; Hill, A.L.; Holmes, R.B.; Halliwell, M.; Jackson,
P.C. Ultrasound quantitation of respiratory organ motion in the
upper abdomen. Br J Radiol. 67:1096 –1102; 1994.
10. Ross, C.S.; Hussey, D.H.; Pennington, E.C.; Stanford, W.; Doorn-
bos, J.F. Analysis of movement of intrathoracic neoplasms using
ultrafast computerized tomography. Int. J. Radiat. Oncol. Biol.
Phys. 18:671–7; 1990.
11. Balter, J.; Lam, K.; McGinn, C.; Lawrence, T.; Ten Haken, R.
Improvement of CT-based treatment-planning models of abdomi-
nal targets using static exhale imaging. Int. J. Radiat. Oncol. Biol.
Phys. 41:939 – 43; 1998.
12. Shimizu, S.; Shirato, H.; Kagei, K.; et al. Impact of respiratory
movement on the computed tomographic images of small lung
tumors in three-dimensional radiotherapy. Int. J. Radiat. Oncol.
Biol. Phys. 46:1127–33; 2000.