Clinical stress markers vs. quantitative myocardial blood flow for assessing adequate hyperemia during stress CMR
M Becker, E Cani, T Schroeder, G Moutzoukis, M K Lorenz, R Bekeredjian, N Schmid, H Mahrholdt, A SeitzAbstract
Introduction
Accurate detection of myocardial ischemia with vasodilator stress cardiovascular magnetic resonance imaging (S-CMR) relies on adequate hyperemia, typically induced pharmacologically with adenosine or regadenoson. Current practice guidelines recommend monitoring clinical markers (heart rate (HR) increase >10 bpm or systolic blood pressure (SBP) drop >10 mmHg) (1) as surrogate for adequate hyperemia. However, clinical response alone may not reliably reflect true myocardial hyperemia, and previous studies have shown that HR and SBP changes are rather poor surrogate markers for vasodilator-induced perfusion augmentation. Recently, a small study suggested quantitative myocardial blood flow (MBF) assessment (stress MBF ≥1.43 ml/g/min in ≥1 AHA segment) as a novel and superior marker for adequate hyperemia (2).
Purpose
This study aimed to assess clinical and MBF response in a large real-world population of consecutive patients undergoing vasodilator S-CMR.
Methods
This was a retrospective secondary analysis among 2,247 all-comer patients (age 65 ± 15, 64% male) who underwent S-CMR using either standard-dose adenosine infusion (140 µg/kg/min; n=1,577) or fixed-dose regadenoson bolus injection (400 µg; n=670). HR and SBP were recorded at rest and during vasodilator stress, and quantitative MBF was measured across all myocardial segments using automated inline pixel-wise perfusion mapping. Adequate hyperemia was defined as stress MBF ≥1.43 ml/g/min in ≥1 AHA segment as previously proposed (2).
Results
In the overall study population, 11% and 16% of patients undergoing adenosine and regadenoson stress CMR, respectively, were non-responders according to clinical stress criteria. Based on MBF stress criteria, overall non-responder rates were 5% for adenosine and 7% for regadenoson studies, respectively. MBF non-respondance was more frequent in clinical non-reponders as compared to clinical responders (adenosine: 9.8% vs. 4.6%, p=0.003; regadenoson: 12.5% vs. 6.5%, p=0.033). Conversely, however, MBF data confirmed adequate stress response in 88% and 90% of adenosine and regadenoson studies despite evidence of non-respondance according to clinical criteria (Figure 1). Accordingly, clinical stress criteria demonstrated high sensitivity but very low specificity for predicting adequate hyperaemia according to MBF (adenosine: sensitivity 90%, specificity 21%; regadenoson: sensitivity 85%, specificity 26%).
Conclusions
Clinical stress criteria do not reliably confirm adequate hyperemia during vasodilator S-CMR. While they demonstrate high sensitivity, their very low specificity indicates limited ability to identify true non-responders. Notably, the vast majority (88–90%) of clinical non-responders were adequately stressed according to quantitative MBF. These findings support the routine integration of quantitative perfusion analysis into clinical S-CMR protocols to verify adequate hyperemia.