Silent Ischemia as a Central Problem: Regional Brain Activation Compared in Silent and Painful Myocardial Ischemia

Soon after the publication of Heberden's classic description of angina pectoris [1], the discrepancy between the extent of anatomical coronary artery disease and patients' symptoms was noted [2]. It has since been recognized that myocardial ischemia and infarction can occur without pain; infarction is sometimes discovered by electrocardiography, and myocardial ischemia might be assumed retrospectively after the postmortem demonstration of substantial coronary artery disease [3-5]. In particular, reversible silent myocardial ischemia has been investigated by using ambulatory electrocardiographic monitoring [6-8] and is a common finding. As many as 70% of episodes of myocardial ischemia in patients with coronary artery disease may be asymptomatic, and the incidence of painless events for acute myocardial infarction is estimated to be about 30% [4-8]. Silent ischemia often coexists with painful ischemia in the same patient; this precludes any simple explanation of silent ischemia based on the particular characteristics of an individual patient. Silent myocardial ischemia is clinically important because it is associated with poor prognosis after an event such as an episode of unstable angina [5-9] or myocardial infarction [10, 11]. Most dramatically, it has been assumed to exist in patients in whom coronary artery disease presents as sudden cardiac death [12]. Silent ischemia has also been found during exercise in survivors of cardiac arrest and in patients with life-threatening arrhythmias [13]. The pathophysiologic basis of silent ischemia has not been established. From the observation [7] that patients with stable angina can have many more painless than painful episodes of ST-segment depression, it was hypothesized that silent myocardial ischemia represented less severe ischemia [5, 7, 8, 14, 15]. However, in a recent study [16] that assessed the significance of chest pain in patients with coronary artery disease who had a high a priori likelihood of inducible ischemia, investigators found that the differences in objective measurements of ischemia (made using ambulatory electrocardiographic monitoring and thallium-201 single-photon emission computed tomography exercise testing) between patients with angina and patients with silent ischemia were insignificant. The higher incidence of silent ischemia in patients with diabetes [17] implicated peripheral neuropathy in the process; differences in autonomic nerve function have also been described in nondiabetic patients with silent myocardial ischemia [18]. Conversely, silent ischemia can be shown in many nondiabetic persons who have no evidence of neuropathy. The discovery of the brain's endogenous opiate system [19] led to the idea that a higher pain threshold due to enhanced central opiate activity could explain the silence of silent myocardial ischemia. However, results of the opiate studies to date have been equivocal [14, 20-25], often having been based on measurements of plasma -endorphin levels. In addition, psychological and personality factors may play a role in the perception of angina [26]. The anatomical pathways for the transmission of adequate peripheral painful stimuli have been established largely on the basis of invasive experiments in animals [27-29]. However, beyond the thalamic level, the central connections mediating visceral pain perception and the affective response to it are unclear. To investigate this further, we adopted an interdisciplinary approach to the study of the cerebral regulation of cardiac function. Positron emission tomography is a powerful technique for assessing regional brain function. The actual variable quantified by positron emission tomography is regional cerebral blood flow. (In most circumstances, regional cerebral blood flow is a highly reliable index of cerebral glucose consumption, which increases regionally when a given cerebral territory is activated [30, 31]. Glucose consumption is, in turn, coupled with Na/K-dependent adenosine triphosphatase and therefore with neuronal firing rates [32]). Measurements of regional cerebral blood flow have largely been confined to neurologic studies and used to investigate, for example, responses of the brain to various motor tasks or auditory or visual stimuli [31, 33-35]. However, after applying this approach to the investigation of the cardiovascular system, we recently reported a pilot study in which we used positron emission tomography with 15O-labeled water to define the functional central nervous pathways activated by dobutamine-induced angina pectoris [36]. One important finding was that thalamic activation persisted after the symptoms and signs of myocardial ischemia had ceased. This prompted us to propose that gating [37] of painful signals may occur at the thalamic level. In the present study, we used the same methods to investigate patients with silent myocardial ischemia. Our aim was to assess whether any cortical or subcortical cerebral activation accompanied silent myocardial ischemia and, if so, whether different patterns of activation of central neural structures might distinguish painful from painless myocardial ischemia. Methods Study Sample No patients in either study group had diabetes or any systemic disease. The baseline characteristics of the patients are shown in Table 1 and Table 5. Table 1. Patients Characteristics and Hemodynamic Data* Table 5. continued from table 1 Patients with Silent Ischemia We studied nine right-handed men (mean age SD, 62 7 years) with significant coronary artery disease at angiography (at least one stenosis more than 50% of luminal diameter). Five patients had three-vessel disease, two patients had two-vessel disease, and two patients had had coronary bypass grafting but had recently developed occlusion of important grafts. Eight patients had been identified during outpatient investigation of stable exertional breathlessness; breathlessness had been found in two of these eight, who had previously had coronary artery bypass surgery for breathlessness, mild angina, and reduced effort tolerance. One patient had previously had an inferior myocardial infarction. The patients with silent ischemia were enrolled consecutively over a 14-month period. All had painless myocardial ischemia demonstrable by the development of ischemic electrocardiographic changes (> 0.1 mV downsloping or rectilinear ST-segment depression 80 milliseconds after the J point) and new regional wall motion abnormalities during dobutamine stress echocardiography in the complete absence of chest pain. None of the nine patients with silent ischemia showed a mixed clinical picture (angina on some occasions and painless ischemia on others). Resting ventricular function was normal in all patients except for the one who had sustained the infarction; inferolateral hypokinesia could be seen in that patient. Patients with Angina We also studied nine right-handed patients with stable angina pectoris caused by coronary artery disease (seven men and two women; mean age, 61 7 years) who were recruited from the outpatient clinic. The data on these patients have been previously presented, in part [36]. All nine patients had typical angina pectoris, developed ischemic electrocardiographic changes with exercise stress, and had clinically significant, predominantly single-vessel, coronary artery disease. One patient with angina had had an inferior myocardial infarction with persistent hypokinesis of this wall, but resting ventricular function was normal in the other eight. Positron Emission Tomography Scanning Protocol Using a technique described previously [34, 35], we used dynamic positron emission tomography with H2 15O (from inhaled 15O-labeled carbon dioxide [C15 O2]) to make six regional cerebral blood flow measurements for each patient. For each measurement, a 30-second background frame was recorded; scanning was then continued for an additional 2 minutes, during which time C15O2 was administered at 500 mL/min and 6 MBq/mL activity. The scanner was an emission computed 931-08/12 positron tomograph (Computer Technology and Imaging-Siemens, Knoxville, Tennessee), whose characteristics have been described previously [38]. Myocardial ischemia was provoked by the increase in cardiac work caused by infusion of the 1-agonist dobutamine, a drug that does not cross the blood-brain barrier [39] and has a short half-life (2.4 minutes) [40]. Scan Sequence Before every scan, a 12-lead electrocardiogram and a brief echocardiogram (Challenge 7000, Esaote Biomedica, Florence, Italy) were recorded. Optimal echocardiographic views were chosen, on the basis of the previous dobutamine stress echocardiogram, as those that would best show the development of a new wall motion abnormality. The following sequence of scans was done, with approximately 12 minutes between scans, to allow for the complete decay of 15O radioactivity. All scans were done while the patient was lying on the scanner couch in a dimly lit room with eyes closed. 1. Baseline scan 1. 2. Placebo scan: done after 6 minutes of a saline infusion. Patients were unaware of the identity of the infused substance and were warned of the possibility of pain or unusual sensations in the chest due to the infusion, which was continued during the scan. The placebo scan was done to control for possible changes in regional cerebral blood flow caused by anticipation. 3. Baseline scan 2. 4. Low-dose dobutamine scan: done during intravenous infusion of low-dose dobutamine. The dose of dobutamine was 5 g/kg per minute for 3 minutes and then 10 g/kg per minute for 3 minutes. The latter dose was continued throughout the scan. Patients were blinded to the identity of the infused substance and were warned of the possibility of pain or abnormal sensations in the chest. This infusion was done to control for the effects of dobutamine on regional cerebral blood flow at doses insufficient to cause myocardial ischemia. 5. High-dose dobutamine scan: done during