A Practical Method to Prolong Expiratory Breath Holds for Abdominal Stereotactic Body Radiation Therapy

Introduction

Stereotactic body radiation therapy (SBRT) is an increasingly utilized radiation technique that enables accurate delivery of ablative radiation doses with a steep dose fall-off to surrounding tissues. However, the utilization of SBRT for moving targets can be a significant challenge. Failure to account for respiratory motion can lead to underdosing targets and overdosing normal tissues.1,2 Motion management is especially important for abdominal SBRT cases given the high dose per fraction and steep dose gradients between the tumor target and nearby gastrointestinal viscera (such as the stomach and the bowel). Various strategies have been utilized to minimize the effects of respiratory motion during abdominal/thoracic SBRT, such as abdominal compression, amplitude- and phased-based gating, and breath-hold techniques, among others.1,3-5

Voluntary breath-hold techniques are attractive for motion management during abdominal SBRT. With this strategy, the beam is intermittently enabled only when the patient is holding their breath, which is coordinated via instructions from the radiation therapists. As the tumor and target tissue are effectively stationary during beam-on, there is no need for an internal target volume margin, which minimizes the volume of irradiated normal tissue to achieve adequate tumor target coverage. This reduction in motion translates to improved on-board cone-beam CT (CBCT) image quality, allowing more accurate patient alignment prior to SBRT treatment.6-8

Both inspiratory breath-hold (IBH)5,9,10 and expiratory breath hold (EBH)11-13 techniques have been successfully utilized for SBRT treatments. While EBH is more reproducible and minimizes target motion compared with IBH,13-19 it is generally more challenging for patients to perform an EBH of sufficient duration compared with IBH.²⁰ Physiological studies of breath holding have shown that supplemental oxygen and mild hyperventilation can significantly improve breath-hold durations,²⁰ with each technique adding incrementally to the improvement in breath-hold duration. Several pilot studies have demonstrated the clinical effectiveness of supplemental oxygen and mild hyperventilation for deep-inspiratory breath hold (DIBH) treatment in patients with breast cancer,21-23 but the effectiveness of this technique for patients undergoing EBH is unknown.

In this article, we report the experience of the first 20 patients treated with abdominal SBRT using a supplemented EBH technique (EBH_supp) with supplemental oxygen and mild hyperventilation. We evaluated data on individual patient EBH durations and treatment times, and we compared this data with a cohort of similar patients treated with EBHs without supplementation (EBH_RA, room air, no mild hyperventilation). We hypothesized that the EBH_supp technique would prolong EBHs and reduce overall treatment time compared with EBH_RA.

Materials and Methods

Treatment Procedure Details

Prior to August 2020, patients were treated with standard EBH without oxygen supplementation or coaching/prompting of their respiratory rate (RR) prior to EBH (designated EBH_RA, where RA signifies room air). After August 2020, as part of a quality improvement initiative in our department and after a successful proof-of-concept study in healthy volunteers (see Supplementary Figure 1, available in the online version of this article at www.appliedradiationoncology.com), patients were offered treatment with a supplemented oxygen, mild hyperventilation EBH technique (EBH_supp). During both CT simulation and treatment, patients received 50% supplemental oxygen (50% fraction inspired oxygen or F_iO₂) via Venturi mask with an appropriate adapter and were instructed to synchronize their breathing at an RR of 18 breaths/min (ie, mild hyperventilation) with audio cues from an online metronome that was beamed into the simulation or treatment room (see Figure 1 and Supplementary Figure 2, available in the online version of this article at www.appliedradiationoncology.com, an online metronome available at https://www.imusic-school.com/en/tools/online-metronome, set at 36 beats/min with stress on the first beat to give a different audio cue for inhalation vs exhalation). The oxygen content of 50% F_iO₂ and RR of 18 breaths/min were chosen based on the known safety of these parameters in humans and the fact that similar parameters have been utilized in cancer patients performing inspiratory breath holds during radiation treatments.21,22 While breathing to the beat of the metronome was encouraged by radiation therapists, there were no measures to forcibly maintain the patient at a strict RR of 18 breaths/min prior to EBH. Initiation of supplemental oxygen and mild hyperventilation in patients undergoing EBH_supp was initiated just prior to setup imaging and stopped right after treatment beam-off (with no prolonged time for oxygenation prior to treatment start).

Figure 1.

Implementation details of the supplemented expiratory breath hold technique (EBH_supp). Supplemental oxygen was delivered at 50% F_iO₂ via a Venturi mask and appropriate adapter (red arrow) connected to in-house oxygen flowmeters (yellow arrow) in the treatment delivery vault (A). Varian’s Real-Time Position Management (RPM) system was used to track patient respiratory motion with an infrared tracking camera (light blue arrow) and reflective marker (dark blue arrow) placed on the patient’s abdomen (B). Patients were encouraged to breathe at a respiratory rate of 18 breaths/min with the use of an online metronome. The metronome was set at 36 beats/min with stress on the first beat so there would be a distinct sound cue for both the start of inhalation and the start of exhalation (C). An example of breath-hold tracing during patient treatment via the RPM system is shown (D). Permission was given by a patient volunteer for the use of photographs in this article.

For both EBH_RA and EBH_supp, patients were screened at the time of CT simulation, per standard departmental protocol, to verify their ability to perform repeated EBHs of more than 20 seconds’ duration. This was assessed by radiation therapy staff with the patient on the CT simulation table and in the treatment position with full immobilization gear and tracking of abdominal excursion via Varian’s Real-Time Position Management system (RPM). Completion of 2 consecutive, 20+ second EBHs within a 5-mm-amplitude window (as tracked by RPM) was required for the patient to move forward with EBH CT simulation and treatment. Patients not able to complete EBH simulation were treated with phased-based respiratory gating and are not described in this article. To be eligible for EBH-based SBRT treatments at our institution, all patients (both for EBH_RA and EBH_supp) were required to have either implanted fiducials or radio-opaque transarterial chemoembolization (TACE) material within or directly adjacent to the target to allow for intrafraction kV x-ray real-time monitoring of motion in addition to RPM amplitude gating.

Treatment Delivery

On the days of treatment, the general treatment workflow for both EBH_RA and EBH_supp treatments was as follows (see Figure 2): initial alignment to bony structures with orthogonal EBH kV images, EBH CBCT for final target/fiducial alignment, and finally delivery of the treatment with 2-3 coplanar arcs. For EBH_supp patients, supplemental oxygen and metronome-cued breathing started right before setup imaging and stopped right after treatment beam-off, with no prolonged oxygenation prior to therapy.

Figure 2.

Treatment timeline example and definition of time points. The shaded gray area shows an example “Session Timeline” screenshot from Varian Eclipse treatment planning software from a patient treated with a supplemented expiratory breath-hold (EBH) technique. The individual EBHs performed by the patient during cone-beam CT (CBCT) and delivery of treatment arcs are shown in the amplitude tracings at the top of the figure (EBH#1 to EBH#6). Treatment time definitions are shown toward the bottom of the figure (and are described in the “Materials and Methods” section).

The RPM amplitude gating (5 mm) and intrafraction fiducial/TACE tracking with triggered kV images obtained every 10/20 degrees of gantry rotation were used to confirm breath-hold position during treatment (see Figure 1). Patients were instructed by therapy staff to breathe to the beat of the metronome in between EBH. A minimum of 6 breaths was recommended between each attempted EBH, but the timing of EBH attempts was left to the discretion of the radiation therapy staff. Additional images (eg, repeat kV or CBCT) were acquired as clinically necessary. All data regarding EBH duration and treatment time were automatically logged into the OIS.

Results

We identified a total of 46 patients meeting inclusion criteria who received 3-fraction SBRT with EBH treatment in our department after January 2018. Prior to the initiation of the EBH_supp technique in August 2020, 26 patients were treated with 28 treatment plans via standard EBH_RA (2 patients had 2 liver tumors that were treated with separate treatment plans), accounting for a total of 83 EBH_RA treatments (28 treatment plans × 3 fractions = 84 treatments minus 1 patient who received a liver transplant after the completion of only 2 fractions). At the time of this analysis, 20 patients were treated with the EBH_supp technique with 24 treatment plans (4 patients had 2 liver tumors who were treated with separate treatment plans) for a total of 72 EBH_supp treatments (24 treatment plans × 3 fractions = 72 treatments).

Patient demographic, clinicopathologic, and treatment parameters are shown in Table 1. Overall, there were no significant differences in any parameter between patients in the EBH_RA and EBH_supp groups. The mean patient age was approximately 62 years old in both groups (62.4 vs 63.2, P = .788), and most patients were male (76.9% vs 90.0%, P = .246) and carried a diagnosis of hepatocellular carcinoma (73% vs 70%, P = .883). In all cases, patients received EBH SBRT to the liver. Patient comorbidity burden was similar between the groups as judged by the Charlson comorbidity index (CCI, 7.31 vs 7.60, P = .590). Treatment parameters, including dose per fraction and MU delivered per treatment, were similar between the groups, with nearly all patients receiving treatment with 2 co-planar arc VMAT plans (2 patients in the EBH_supp group were treated with 3 co-planar arc VMAT plans).

A total of 1735 individual EBHs were extracted from the OIS (Supplementary Figure S3, available in the online version of this article at www.appliedradiationoncology.com) and showed a significant increase in EBH duration for patients treated with the EBH_supp technique (Figure 3, Table 2). Maximum EBH for patients was significantly increased for patients treated with EBH_supp, with a difference in maximum EBH of ~18 seconds (34.5 s vs 52.8 s, P < .001). Mean (18.2 s vs 25.1 s, P < .001) and median (18.7 s vs 24.9 s, P = .002) EBH were also significantly increased for patients treated with EBH_supp compared with EBH_RA. There was a corresponding decrease in the number of EBHs required to complete each treatment (12.7 vs 8.9, P < .001). Table 2 shows a percentile breakdown of patient EBH between the 2 groups with a statistically increased EBH observed with EBH_supp treatment for all individual patient EBH percentiles except for the lowest (10th) percentile (10th percentile: 6.4 s vs 7.8 s, P = .219).

Figure 3.

Duration and number of individual expiratory breath holds (EBHs) during stereotactic body radiation therapy treatments. The data illustrate the average maximum (Max EBH) (A), mean (Mean EBH) (B), median (Median EBH) (C), expiratory breath-hold duration of individual patients, as well as the average of the total number of expiratory breath holds required for individual patients to complete treatment for patients treated with standard (EBH_RA) or supplemented (EBH_supp) EBH technique (D). Error bars in the box and whisker plot represent the range (min to max), bars represent the interquartile range, and line represents the median. P values were calculated via a Student t-test.

Examination of treatment times between the 2 groups (Figure 4) showed a significant reduction in the time required for completion of treatment in patients treated with EBH_supp compared with EBH_RA, with a mean reduction of ~5 minutes (21.8 min vs 16.7 min, P = .025) and a median reduction of ~2 minutes (18.1 min vs 16.2 min). While the time required for completion of kV radiographs (“kV Time”) was not different between the 2 groups (P = .325), the time required for completion of CBCT (“CBCT Time,” P < .001) and each treatment arc (“Treatment Arc Time,” P < .001) was significantly reduced in patients treated with the EBH_supp technique.

Figure 4.

Times for completion of individual treatment components and for total treatment. Pie charts showing the mean times required for each component of treatment and the total treatment time for individual patients treated with standard expiratory breath hold (EBH) (EBH_RA) (A) or supplemented EBH (EBH_supp) techniques (B). Box and whisker plots of time to completion of orthogonal kVs (kV Time) (C), time to completion of cone-beam CT (CBCT Time) (D), time to completion of treatment arcs (Treatment Arc Time) (E), and total treatment time (F). Error bars in the box and whisker plot represent the range (min to max), bars represent the interquartile range, and line represents the median. P values were calculated via Mann-Whitney U test.

Discussion

Expiratory breath hold is an effective method for motion management in patients undergoing abdominal SBRT treatments but is underutilized due to patients’ difficulty in performing repeated EBH of sufficient duration.¹ In this study, we found that patients undergoing abdominal SBRT with supplemental oxygen and mild hyperventilation exhibited prolonged EBH durations compared with patients treated with nonsupplemented EBH. Patients in the EBH_supp group performed EBH of longer duration by all reported metrics (maximum EBH, mean EBH, and median EBH) and required less total EBH to complete treatments. These results agree with previously published experiences utilizing similar techniques in patients undergoing DIBH for breast cancer treatments20-22 as well as a recent randomized study of volunteers undergoing EBH that showed an improvement in median EBH duration from 24 seconds to 49 seconds with supplemental oxygen and mild hyperventilation.²⁴

The reduction in treatment times observed in the EBH_supp group compared with the EBH_RA group is likely related to improve EBH using the supplemented technique. This is supported by significant improvements in time to complete tasks that required prolonged EBH (eg, CBCT, treatment arcs) and a lack of significant improvement in tasks that did not require prolonged EBH (eg, kV acquisition/alignment). There were no significant differences in patient clinical/pathological characteristics between the groups (eg, age, diagnosis, and comorbidity index) nor in treatment-related parameters that might be expected to influence treatment time (eg, dose per fraction, MU delivered, and number of treatment arcs). During the study period, there were no other changes in departmental protocols or treatment-planning techniques as an alternate explanation for the reduction in treatment time observed. While the EBH_supp technique reduced treatment times in our study, there was an initial time investment (~10-15 min) at the time of CT simulation for added patient training for breathing with a metronome and the use of a Venturi mask.

The reported EBH_supp technique was safe in our study population. We had no patients who had any issues or symptoms related to the breath-hold component of their treatments (eg, no lightheadedness, syncope, tingling, tetany, or other concerns), which is in agreement with other studies that have shown the safety of breath holds (both inspiratory and expiratory) with mild hyperventilation and supplemental oxygen.20-25 The mild hyperventilation used in this study was chosen based on the RRs previously utilized and found to be safe in breast cancer patients undergoing DIBH treatment with prolonged hyperventilation.22,23 We specifically avoided more rapid hyperventilation given the theoretical increased risk of tetany that can occur with more aggressive hyperventilation and associated hypocapnia.²⁶ Similarly, the level of supplemental oxygen of 50% F_iO₂ was selected based on safety as supplemental oxygen levels above 60% are associated with an increased risk of absorptive atelectasis.25,27-30

One safety issue that should be mentioned for all breath-hold treatments (not just supplemented ones such as the EBH_supp method) is the well-established increase in blood pressure during prolonged breath holds.25,31,32 However, the risk and severity of blood pressure rise during breath holds do not appear to be worsened by supplemented techniques, including a similar technique with mild hyperventilation and supplemental oxygen.²⁵ Further, a recent randomized study of EBHs in volunteers did not find a significant change in blood pressure during EBH, possibly due to the relatively modest prolongation of EBHs with supplemented techniques (< 1 min) compared with DIBH (up to 5 min prolonged breath holds reported).²⁴ In general, patient cardiopulmonary comorbidities should be considered by the treating radiation oncologist prior to proceeding with breath-hold treatment with discussion of the technique with the patient’s other involved physicians (eg, cardiologist and pulmonologist) for those with significant cardiopulmonary comorbidities.

There are several limitations to this study. First, this was not a randomized trial and. therefore, there is a possibility of biases (eg, selection bias) that could have partially influenced the results between the EBH_RA and EBH_supp groups, though patient groups were well balanced overall with respect to both clinical parameters and treatment parameters. Second, all the patients described here received liver SBRT and, therefore, the results may not be generalizable to all patients receiving abdominal SBRT, though we have additionally treated several patients with primary pancreatic cancer with pancreas SBRT with the EBH_supp method with similar experience to those treated with liver SBRT. Lastly, this was a “real-world” study of EBH_supp implementation in a busy radiation oncology clinic, and we did not attempt to capture nor control all aspects of respiratory physiology that govern breath-hold capacity. While we encouraged patients to breathe at a rate that would normally correspond to mild hyperventilation, we did not forcibly control patient RR or tidal volumes and did not measure partial pressures of carbon dioxide; therefore, whether hyperventilation/hypocapnia was achieved for each patient is unknown. Future mechanistic studies of patients undergoing repeated, supplemented EBH with real-time measurement of these parameters will be helpful in further optimizing supplementation/hyperventilation protocols and ensuring uniformity of technique among individual patients.

Conclusions

Patients receiving supplemental oxygen and mild hyperventilation exhibited prolonged EBH time and reduced overall treatment time during abdominal SBRT. This intervention is simple, inexpensive, safe, and may improve individual patient breath-hold times, reduce treatment time, and increase the number of patients eligible for EBH-based abdominal SBRT.