Impaired myocardial perfusion in atrial fibrillation cause or effect?

This editorial refers to ‘Myocardial perfusion imaging does not adequately assess the risk of coronary artery disease in patients with atrial fibrillation’ by M.D. Smit et al. , on page 643. Atrial fibrillation (AF) is the most common sustained arrhythmia, particularly in the elderly. It is associated with increased morbidity and mortality and frequently requires pharmacological (antiarrhythmic drugs) or interventional (catheter ablation) therapy. 1 Patients with AF frequently present with angina-like chest pain and transient ischaemic-type ST-segment depression, irrespective of ischaemic, non-ischaemic, or idiopathic origin of AF. In addition, elevated markers of myocardial damage have been reported after cardiac decompensation due to tachyarrhythmias. However, these findings were not always associated with significant coronary artery disease. Atrial fibrillation is more likely to develop on the basis of structural heart disease such as coronary artery disease, dilated cardiomyopathy (DCM), valvular heart disease, and others. In patients with pre-existing left ventricular dysfunction, AF may further diminish cardiac performance and aggravate heart failure, resulting in more frequent hospitalizations and impaired prognosis. On the other hand, restoration and maintenance of sinus rhythm may improve myocardial perfusion and perfusion reserve resulting in amelioration of left ventricular function. Patients with AF and concomitant coronary artery disease are at increased risk of cardiovascular impairment and complications. Therefore, the early identification and appropriate treatment of ischaemic heart disease underlying the arrhythmia is of major clinical and prognostic importance. However, non-invasive diagnosis of coronary artery disease is hampered in patients with AF. Bicycle or treadmill exercise tests provide imperfect information on myocardial ischaemia in patients with AF due to inappropriate tachyarrhythmia and non-specific ST-segment depression. Imaging modalities such as echocardiography and magnetic resonance imaging with pharmacological stress (dobutamine or adenosine) are limited due to impaired image quality for the assessment of ischaemia-induced regional left ventricular dysfunction during AF with or without tachyarrhythmia. Similar restrictions with regard to the correct identification of significant coronary artery disease apply to radionuclide single-photon emission computed tomography (SPECT) perfusion imaging, which has, however, proved to provide additional prognostic information in patients with AF. 2 Smit et al . 3 reported a study investigating the value of myocardial perfusion SPECT imaging in the coronary artery disease risk assessment. In this study, 129 patients with a history of AF were compared with gender and age-matched control subjects. Primary endpoint was a positive SPECT result with significant signs of myocardial ischaemia. In summary, a positive SPECT result was similar in patients with AF and controls. However, in AF patients, a positive SPECT outcome was less often correlated with significant coronary artery disease in those patients who subsequently underwent coronary angiography. This finding raises the problem of false-positive SPECT results in the setting of AF. Furthermore, a negative SPECT outcome did not exclude significant coronary artery disease in either group (AF vs. control). Although the SPECT results were not validated by comparison with other exercise or stress imaging modalities or invasive coronary angiography, the study results confirm previous reports indicating myocardial ischaemia and/or impairment of myocardial microcirculation in the absence of significant coronary artery disease in patients with AF. Early experimental studies in dogs demonstrated that artificially induced AF resulted in an impaired vasodilator response. Early clinical studies in patients with mitral valve stenosis and persistent AF indicated a reduction of coronary blood flow. Recent pathophysiological studies by our group using quantitative positron emission tomography and radioactively labelled water (H 215 O-PET) in male patients with persistent AF and/or DCM 4 , 5 confirmed abnormal myocardial blood flow (MBF), coronary flow reserve (CFR) and coronary vascular resistance (CVR) in patients with persistent AF and/or left ventricular dysfunction when compared with male control subjects in sinus rhythm without structural heart disease or history of AF. In patients with AF and exclusion of coronary artery disease, resting MBF was reduced by ∼20%, hyperaemic MBF (adenosine infusion) by ∼40%, and CFR by ∼22%, whereas CVR was increased during hyperaemia by ∼62% in AF patients when compared with the control group ( Table 1 ). 4 However, no regional differences in left ventricular MBF were demonstrated in these studies. Myocardial perfusion, flow reserve, and coronary vascular resistance in atrial fibrillation and dilative cardiomyopathy 4 , 5 ADO, adenosine infusion; AF, atrial fibrillation; CFR, coronary flow reserve; CVR, coronary vascular resistance; DCM, dilative cardiomyopathy; MBF, myocardial blood flow; SR, sinus rhythm. °Not significant (ns); * P < 0.05; ** P < 0.01; *** P < 0.001 vs. controls. Myocardial perfusion, flow reserve, and coronary vascular resistance in atrial fibrillation and dilative cardiomyopathy 4 , 5 ADO, adenosine infusion; AF, atrial fibrillation; CFR, coronary flow reserve; CVR, coronary vascular resistance; DCM, dilative cardiomyopathy; MBF, myocardial blood flow; SR, sinus rhythm. °Not significant (ns); * P < 0.05; ** P < 0.01; *** P < 0.001 vs. controls. In patients with AF and additional DCM and left ventricular dysfunction, the impairment of MBF, CFR, and CVR was even more pronounced, particularly under hyperaemic conditions ( Table 1 ). 5 These findings imply an independent negative effect of DCM on hyperaemic MBF. Additional investigations using H 215 O-PET in patients before and after electrical cardioversion of AF demonstrated partial reversibility of impaired MBF and CVR after a mean of 4.1 months of sustained sinus rhythm. 4 These data indicate that in patients with AF with or without DCM, compared with normal control subjects, diminished myocardial perfusion and elevated coronary vascular resistance during hyperaemia are most likely the consequence rather than the cause of AF. Experimental and clinical pathophysiological investigations provide convincing evidence that AF reduces myocardial perfusion, most likely by constriction of the microvasculature, possibly triggered by increased sympathetic activity and/or elevated circulating vasoactive peptides. 4 , 5 Atrial fibrillation in the absence or presence of structural heart disease (i.e. DCM or coronary artery disease) is a common clinical problem that may severely affect cardiac function and the patients’ quality of life. Several studies demonstrated a relationship between AF and cardiac events or mortality independent of the results of SPECT imaging. 2 , 6 However, patients with both AF and myocardial ischaemia are considered to have the worst prognosis. Therefore, a reliable non-invasive test for the detection of concomitant coronary artery disease is highly desirable to optimize treatment strategies and to improve the prognosis in AF patients. The clinical dilemma, however, is that none of the tests available provides sufficient accuracy to predict or exclude significant artery coronary disease in the setting of AF because they lack both sensitivity and specificity. This is due to non-specific ST-segment depression mimicking myocardial ischaemia, diminished image quality and interpretation during the arrhythmia, and the impairment of myocardial perfusion and perfusion reserve caused by AF itself. These myocardial perfusion abnormalities secondary to the presence of AF together with the technical limitations of imaging during AF (i.e. unreliable triggering) may explain and contribute to the great number of false-positive SPECT results reported by Smit et al . Therefore, these results support the previous reports 2 indicating that stress SPECT imaging is not useful as a screening or routine test in patients with AF and without clinical symptoms suggestive of coronary disease, because false-positive results may trigger unnecessary invasive procedures. On the other hand, in patients with AF and signs or complaints suggestive of myocardial ischaemia, referral for coronary angiography should not be restricted because of normal findings in myocardial perfusion imaging since false-negative SPECT results were not infrequently reported. As a clinical consequence, in ambiguous or equivocal cases with symptoms or findings suspicious for coronary artery disease, coronary angiography remains the diagnostic gold standard until new non-invasive techniques may overcome the current limitations of myocardial perfusion imaging in patients with AF. Conflict of interest: none declared.