The etiology of the fall in left ventricular stroke volume (LVSV) and arterial pressure with a negative intrathoracic pressure (NITP) during inspiration is controversial. An increase in LV afterload produced by NITP has been proposed as one explanation but is difficult to evaluate if preload is also altered. To test the hypothesis that a systolic event alone, i.e., a change in LV afterload or contractility, can reduce LVSV during inspiration independent of changes in LV preload, a rapid transient NITP confined to systole was produced by electrocardiogram-triggered phrenic nerve stimulation in eight anesthetized dogs. Intrathoracic descending aortic diameters were measured by sonomicrometry to transduce qualitative changes in aortic transmural pressure. With the airway completely obstructed systolic NITP resulted in a decrease in LVSV (-8.1%, P less than 0.001) but an increase in the systolic anteroposterior (0.54 mm, P less than 0.01) and right-to-left (0.45 mm, P less than 0.01) aortic diameters compared with preceding beat. Similar significant changes were observed with the airway unobstructed. These observations are consistent with an increased afterload imposed on the LV reducing LVSV and egress of blood out of the thorax. Prolonging NITP to include both systole and diastole, a profound fall in LVSV is observed, consistent with the independent influences of systolic and diastolic events combining to diminish LVSV.(ABSTRACT TRUNCATED AT 250 WORDS)

The etiology of the fall in left ventricular stroke volume (LVSV) with negative intrathoracic pressure (NITP) during inspiration has been ascribed to a reduction in LV preload. This study evaluated the effects of NITP with and without airway obstruction confined to early (ED), mid- (MD), or late diastole (LD) on the subsequent LVSV, anteroposterior (AP), and right-to-left (RL) aortic diameters (DAO) (series I, n = 6) as well as on phasic arterial blood flow out of the thorax (series II, n = 6) in anesthetized dogs. Transient NITP was obtained by electrocardiogram-triggered phrenic nerve stimulation. In series I, NITP applied for 60% of diastole with the airway obstructed caused decreases of LVSV during ED [-7.7 +/- 3.2% (SE) NS], MD (-11.7 +/- 3.9%, P less than 0.05), and LD (-14.6 +/- 1.5%, P less than 0.01) associated with significant increases of left ventricular end-diastolic pressures relative to both atmospheric and esophageal pressures during MD and LD. NITP increased DAO(AP) and DAO(RL), resulting in increases in diastolic aortic cross-sectional area by an average of 6.1-8.3% (P less than 0.01). Similar changes were seen with the airway unobstructed during NITP. In series II, NITP caused diminished diastolic antegrade carotid artery and/or descending aortic flow run off in all dogs. Transient retrograde arterial flows with NITP were observed in more than half of the animals consistent with increases in aortic diameters. We conclude that a decrease of intrathoracic pressure confined to diastole can 1) diminish the ensuing LVSV, presumptively reducing preload by ventricular interdependence; 2) distend the intrathoracic aorta; 3) diminish antegrade flow out of the thorax independent of effects on cardiac performance; and 4) cause transient retrograde carotid and aortic blood flow. The intrathoracic aorta and, presumably, the arterial intrathoracic vascular compartment can be viewed as an elastic container driven by changes in intrathoracic pressure.

Recent reports have suggested that positive end-expiratory pressure (PEEP) depresses left ventricular (LV) function or shifts LV pressure-volume (PV) relationships due to neural, humoral, or mechanical events. These studies have usually utilized pressures measured during expiration. To study the mechanical effects of PEEP in expiration and inspiration, the circulation was arrested in 12 open-chest dogs, and the coronaries were perfused with a cold cardioplegic agent. Balloons were placed in the ventricles to measure ventricular pressures. Shifts in cardiopulmonary blood volume were prevented by venting the atria to atmosphere. Having ablated neural reflexes and humoral changes, we varied ventricular volumes, chest wall compliance, tidal volume, and PEEP. We found that isovolumic ventricular pressures (relative to atmosphere) increase with PEEP (P less than 0.001), and heart-lung interaction with PEEP is significantly greater in inspiration (P less than 0.001). The effect of PEEP is modified by heart volume (P less than 0.01) and respiratory system compliance (P less than 0.01). We conclude that a mechanical compressive force can be applied to the heart by the lungs as they expand, and this may explain the previous reports of diminished LV function or LV diastolic compliance with PEEP and, in part, explain the decreased cardiac output associated with PEEP. The marked increase in mechanical compressive forces applied to the heart during inspiration with PEEP may have far greater hemodynamic consequences than events during expiration.

To clarify the effects of respiration on left ventricular (LV) dimensions and shortening, we studied chronically instrumented dogs with endocardial sonomicrometer crystals in the anterior-posterior (AP), septal to lateral (SL), and long axes (LA) following pericardiectomy. Ten anesthetized dogs were examined during spontaneous unobstructed respiration, partial inspiratory obstruction (PIO), and Mueller maneuvers (MM). During unobstructed inspiration, end-diastolic dimensions (EDD) demonstrated a significant increase in the AP and a similar decrease in the SL axis (i.e., noncongruent shape changes). During PIO only the SL EDD diminished significantly, while no significant changes occurred in any EDD during MM. Individual dogs also demonstrated noncongruent shape changes at end systole during inspiration. However, the end-systolic dimensions for the entire group demonstrated a significant increase in one dimension during each inspiratory mode with no significant changes in the other two axes suggesting an increased ventricular volume. Regional shortening declined only in the SL axis during both unobstructed respiration and PIO. Spontaneous sighs with large tidal volumes, yet smaller changes in pleural pressure than during the MM, were associated with marked noncongruent shape changes in both diastole and systole. We conclude that 1) estimates of LV volumes during respiration based on only one or two axes and assuming regional congruent shape changes may be misleading; and 2) lung volume changes can affect LV geometry independently of changes in pleural pressure.