Ventricular mechanoreflex and chemoreflex alterations in chronic heart failure

Cardiac and arterial baroreflex control of the circulation is abnormal in both human and experimental heart failure. Ventricular vagal afferents mediate mechanical and chemical reflexes, which result in bradycardia and hypotension. The aim of the present study was to evaluate the changes that occur in ventricular mechanoreflexes and chemoreflexes in a conscious canine model of chronic heart failure. Dogs were instrumented for the measurement of left ventricular pressure, left atrial pressure, arterial pressure, and heart rate. Vascular occluders were placed on the ascending thoracic aorta, on the descending thoracic aorta, and on the thoracic inferior vena cava. A chronic left circumflex coronary artery catheter was also implanted. Finally, a pacing lead was secured to the left ventricular free wall. After recovery from surgery (10 to 14 days), the dogs were subjected to complete arterial baroreceptor denervation. The responses to vascular occlusions and intracoronary administration of prostacyclin (PGI₂) were carried out before and after heart failure was induced by chronic cardiac pacing at 250 beats per minute. PGI₂ was used as a chemical stimulus for ventricular afferents; ascending aortic occlusion was used as a mechanical stimulus. Before chronic pacing, ascending aortic occlusion resulted in a decrease in heart rate of 36.1±12.3 beats per minute (mean±SD, P<.001). After heart failure was induced, the heart rate response to ascending aortic occlusion was almost completely abolished. The slope of the linear relation between pulse interval and left ventricular end-diastolic pressure was reduced by 90.5% from a control value of 11.3±6.9 ms/mm Hg after heart failure had been induced. In three dogs, the bradycardia had returned after cessation of the pacing. On the other hand, there was no change in the response to intracoronary PGI₂ after heart failure had been induced. The primary effect was a leftward shift in the PGI₂ and heart rate dose-response curve in the heart failure state. At the midrange dose, which averaged 9.5±6.7 ng/kg, heart rate was reduced by 9.6±10.4 beats per minute before pacing versus 36.0±8.9 beats per minute after heart failure (P<.05). Because there were directionally different changes in heart rate evoked by mechanoreflexes and chemoreflexes in heart failure, it is suggested that the change in these reflexes is not mediated by abnormal efferent vagal function but rather by alterations in different types of ventricular vagal afferents.