Utility of Dobutamine Stress CMR Imaging for Preoperative Right Ventricular Functional Assessment in LVAD Candidates

Background: Right ventricular failure (RVF) is a major cause of morbidity and mortality following durable LVAD implantation. Contemporary preoperative imaging has poor predictive value in assessing for RV contractile reserve. Cardiovascular magnetic resonance (CMR) provides accurate RV assessment, but its role under pharmacologic stress is not well established. We hypothesize that dobutamine stress cardiovascular magnetic resonance (DS-CMR) will be safe in a high-risk end stage heart failure (HF) population. We also hypothesize we will see a variance in contractile response to inotropy and thus could serve as an effective variable to quantify RV reserve.

Methods: Patients with end stage systolic HF with concomitant resting RV dysfunction undergoing LVAD evaluation at a quaternary care medical center underwent resting and DS-CMR imaging prior to decision on LVAD candidacy. Exclusion criteria included vasopressor support or ≥2 anti-arrhythmic medications. Dobutamine was infused in incremental doses up to 20 µg/kg/min. RV function, LV function, and RV size parameters were measured using SSFP or GRE sequences in patients with ICDs. RV fractional area change (FAC) was quantified at incremental doses of dobutamine. RV contractile reserve was defined as the percent change in FAC from baseline to peak dose. Safety measures- including incidence of sustained arrhythmia, symptomatic hypotension, or hypoxemia- and magnitude of percent change based on resting RV function were analyzed. Secondary outcomes of RVF and mortality were collected.

Results: A total of 18 consecutive patients underwent DS-CMR as part of LVAD evaluation (median age 63.5 years; median LVEF 16.5%; ischemic cardiomyopathy 33.3%, Table 1). All achieved peak dobutamine infusion of 20 µg/kg/min. No sustained arrhythmias, symptomatic hypotension, or hypoxemia events occurred. Median increase in FAC from baseline to peak dose was 12%, with variance in degree of RV contractile response (Graph 1). There was no significant change in FAC in 28% of the patients, defined as less than 50% increase in FAC from rest to peak dobutamine infusion. Partial augmentation in RV function was observed in 22% of the patients, defined as >50% increase in FAC but without normalization of RV function (Figure 1a). There was normalization in RV function in 50% of the patients (Figure 1b).

Conclusion: In an end stage HF patient population, DS-CMR is a safe and effective method of quantifying RV contractile reserve. This technique revealed heterogeneity in RV response to inotropy, which may have prognostic implications. To our knowledge, this is the first study describing DS-CMR as part of LVAD perioperative assessment, and providing a proof of concept, thereby setting the stage for implementing DS-CMR protocols in larger LVAD multi-center trials. The degree of RV reserve, measured by FAC with low dose dobutamine, needs further evaluation to assess risk of RVF and mortality after LVAD implantation, and potentially after other high risk cardiothoracic surgeries.

Table 1

	Mean ± SD	Median
Age	59.5 ± 13.43	63.5 (53-68.5)
BMI (kg/sq m)	30.83 ± 6.32	30.26 (28.88-34.7)
NT ProBNP (pg/mL)	7875.35 ± 9480.53	4286 (2236-7949)
LVEF (echo, %)	18.22 ± 6.78	17.5 (13.25-23.75)
TAPSE (CMR, %)	13.93 ± 3.22	14 (12.25-15.75)
RVEDV (CMR, mL)	224.5 ± 91.29	215 (160.5-295.5)
LVEF (CMR, %)	19.33 ± 7.47	16.5 (15-23)
RVEF (CMR, %)	25.89 ± 10.16	26.5 (18.25-32.25)
FAC at rest (CMR, %)	21.67  7.86	22.5 (16.5-28.75)

Figure 1
Graph 1. Quadratic relationship of RV FAC variance demonstrating the distribution and variability of FAC values across the cohort, highlighting the greatest FAC in patients with moderate baseline RV dysfunction.