Chronotropic response to vasodilator-stress in patients submitted to myocardial perfusion imaging: impact on the accuracy in detecting coronary stenosis.

Abstract

A lower heart rate response (HRR) during vasodilator MPI has been shown to have a relevant adverse prognostic impact. We sought to evaluate the interaction among individual HRR to vasodilator stress and myocardial perfusion imaging (MPI) accuracy in patients with suspected ischemic heart disease (IHD). One hundred and sixty-five consecutive patients were submitted to vasodilator-stress MPI on a cardiac camera equipped with cadmium-zinc-thelluride detectors and coronary angiography. A coronary stenosis >70 % was considered significant. In every patient, the summed difference score (SDS) was computed from MPI images. Patients were categorized according to the tertiles of the distribution of individual HRR during dipyridamole: "Group 1" (HRR < 8 bpm; lowest tertile); "Group 2" (8 ≤ HRR ≤ 12 bpm; middle tertile); "Group 3" (HRR >12 bpm; highest tertile). Significant coronary artery disease (CAD) was present in 102 (62 %) patients. In the overall population, MPI showed a significant accuracy (AUC: 0.81, 95 % CI 0.74-0.86; p < 0.001) in unmasking the presence of significant coronary stenosis. Interestingly, in patients with a blunted HRR during dipyridamole ("Group 1") MPI showed a significantly lower sensitivity (68 %) in detecting CAD than in those with a higher HRR ("Group 3") (91 %, p = 0.007), despite a preserved specificity (76 % vs 77 %, P=NS). Similarly, the correlation among CAD extent and post-stress LV functional stunning was limited to "Group 3" patients, while it disappeared in those with blunted HRR. In patients with suspected IHD, MPI sensitivity is strongly influenced by the magnitude of patient heart rate increase to the pharmacologic stressor, suggesting an interaction among blunted HRR and lower accuracy in unmasking CAD.