1356705 - Novel exercise CMR device for under $50: a comprehensive validation

Exercise cardiovascular magnetic resonance (CMR) techniques have gained increased attention in clinical practice for a wide range of purposes. However, broader adoption of exercise CMR is limited, at least in part, by the expense of CMR-compatible ergometers. We have developed a cheap and simple MR-compatible exercise apparatus for use in the CMR environment. The aim of this pilot validation study was to compare the exercise parameters obtained using this apparatus against the gold-standard of an MR-conditional bicycle ergometer.



Methods:

We carried out CMR-cardiopulmonary exercise testing (CMR-CPET) using both the novel and conventional exercise equipment in 5 healthy volunteers. The two exercise tests were performed within one month of one another. The novel apparatus consists of two half-pipes fixed to a wooden base that fits in a recess in the scanner table (Figure 1). Participants wore knee-length socks with a 0.5kg weight placed in the front of each sock. Studies using the novel exercise apparatus were performed by a ramped protocol, with the participant moving their legs back and forth in time with a metronome of gradually increasing speed until exhaustion. Studies performed with conventional exercise equipment used a supine ergometer following a ramped protocol of increasing resistance until exhaustion.

Aortic flow was measured at rest and at peak exercise using real-time phase-contrast MR (PCMR). PCMR was performed using a uniform density golden-angle spiral sequence, with a compressive sensing (CS) reconstruction. Breath-by-breath gas exchange analysis was performed using a commercially available CPET system with a modified umbilicus passed through the waveguide from the gas analyzer to the facemask. Values for peak heart rate (HR), peak oxygen consumption (VO2) and peak cardiac output (CO) were obtained at rest and peak-exercise for all subjects with each technique.



Results:

Five healthy volunteers successfully underwent CMR-CPET by both techniques. All participants reached anaerobic threshold (RER >1) at both tests. Excellent correlation coefficients were observed for peak HR (HR-bike 143±21 beats per minute (bpm) versus HR-pipe 144±25 bpm, r=0.94) and CO (CO-bike 11.4±1.1 L/min versus CO-pipe 11.9±2.1 L/min, r=0.92, Figure 2). There was good corelation for peak VO2 (VO2-bike 24.8 ±6.2 mL/min/kg versus VO2-pipe 17.6±4.1 mL/min/kg, r=0.59) using the two techniques. However, the peak VO2 using the novel exercise equipment was statistically significant lower than when using the bike (-7.2±5.1 mL/min/kg, p=0.033).



Conclusion:

We have created a novel, robust and low-cost piece of exercise equipment that can achieve similar peak heart rates and cardiac output readings as conventional MR-conditional bicycle ergometers. Given its simplicity, portability and cheap cost to build, this may allow increased dissemination of CMR exercise in clinical and research CMR practice.