Abstract
The effects of continuous positive airway pressure (CPAP) on left (LV) and right ventricular (RV) volumes in patients with congestive heart failure (CHF) have not been studied. We hypothesized that CPAP would cause greater reductions in cardiac volumes in CHF patients with idiopathic dilated cardiomyopathy (IDC) than in those with ischemic cardiomyopathy (IsC), because their ventricles are more compliant. The effects of a 30-min CPAP application at 10 cm H2O on RV and LV end-diastolic (EDV) and end-systolic volumes (ESV), determined by radionuclide angiography, were therefore tested in 22 patients with CHF due to IsC (n = 13) or IDC (n = 9). CPAP-induced reductions in LVEDV, LVESV, RVEDV, and RVESV were significantly greater (p < 0.05) in the IDC than in the IsC group. Whereas in the IsC group CPAP caused no significant changes in LV or RV volumes, in the IDC group it induced significant reductions in RVEDV (527 ± 77 ml to 354 ± 50 ml, p = 0.03) and RVESV (400 ± 78 ml to 272 ± 54 ml, p = 0.04) that were greater than any reductions in LVEDV and LVESV. We conclude that CPAP causes greater short-term reductions in RV and LV volumes in CHF patients with IDC than in those with IsC, and that among patients with IDC, CPAP causes greater reductions in RV than in LV volumes. Mehta S, Liu PP, Fitzgerald FS, Allidina YK, Bradley TD. Effects of continuous positive airway pressure on cardiac volumes in patients with ischemic and dilated cardiomyopathy.
Long-term nightly application of nasal continuous positive airway pressure (CPAP) to patients with congestive heart failure (CHF) and central sleep apnea leads to improvements in cardiovascular mechanical and neurohumoral function. These improvements include increases in left ventricular (LV) ejection fraction, and reductions in functional mitral regurgitation, plasma atrial natriuretic peptide, and norepinephrine concentrations (1, 2). The mechanisms responsible for these beneficial long-term effects of CPAP, however, have yet to be fully elucidated.
One probable mechanism is a reduction in LV afterload resulting from the combined effects of increased intrathoracic pressure, and abolition of apnea-related surges in blood pressure (3, 4). Another potential mechanism is a reduction in cardiac preload. This possibility arises from the observations that CPAP reduces functional mitral regurgitation and atrial natriuretic peptide concentrations, both of which suggest reductions in cardiac volumes. Any such long-term reductions in cardiac volume would indicate beneficial reverse remodeling of the failing heart, and would portend an improved prognosis (5).
In experimental animal models with either normal cardiac function or pacing-induced CHF, acute application of CPAP or positive end-expiratory pressure (PEEP) causes reductions in LV end-diastolic volume (LVEDV), primarily by impeding venous return to the heart (6-8). However, the effects of CPAP on right ventricular end-diastolic volumes (RVEDV) have not been reported. Moreover, there are no published studies in which the effects of CPAP on cardiac volumes in patients with CHF have been documented.
In our previous long-term trials of CPAP for the treatment of CHF, subjects suffered from either ischemic or idiopathic dilated cardiomyopathy (IsC and IDC, respectively). Compared with patients with IsC, those with IDC have less gross and microscopic myocardial fibrosis (9, 10) and larger, possibly more compliant, LV (1, 9, 11, 12). This suggests greater potential for acute reductions in cardiac volumes in response to CPAP in patients with IDC. In addition, the right ventricle (RV) is a thinner walled and consequently more compliant structure than the LV in both IsC and IDC (10, 13-15). We therefore hypothesized that short-term application of CPAP would cause greater reductions in ventricular volumes in CHF patients with IDC than those with IsC, and that any reductions in RV volumes in response to CPAP would be greater than reductions in LV volumes.
We studied 22 consecutive patients with symptomatic CHF who were recruited from the Heart Function clinic of the Toronto Hospital. Inclusion criteria were: (1) a history of CHF caused by IsC or IDC of at least 6 mo duration documented by at least one clinical episode of heart failure characterized by cardiomegaly and pulmonary edema on a chest radiograph, dyspnea on exertion or at rest, and lower extremity edema; (2) reduced LV systolic function as evidenced by an LV ejection fraction ⩽ 45% at rest, determined by radionuclide angiography (RNA); and (3) stable clinical status as evidenced by an absence of acute exacerbations of dyspnea or medication change for at least 1 mo before entry. The diagnosis of IsC was based on the presence of flow-limiting stenosis (> 75%) involving one or more coronary arteries documented on cardiac angiography, or documented myocardial infarction, or a history of coronary artery bypass surgery. The diagnosis of IDC was based on an LVEDV > 80 ml/m2 determined by RNA (16), and either the absence of flow-limiting coronary artery stenosis on coronary angiography, or an absence of ischemic changes on myocardial perfusion imaging. Exclusion criteria included CHF secondary to primary valvular heart disease; unstable angina; and myocardial infarction or coronary bypass surgery within 3 mo prior to the study. The study was approved by the Human Subjects Review Committee of the University of Toronto and all patients provided written informed consent before the study.
Cardiac function was assessed using gated RNA. Data analysis was performed by a technician blind to the patients' cardiac diagnosis. Patients were studied while awake and in the supine position. Gated R-wave synchronous equilibrium angiography using the in vivo red blood cell labeling technique was performed with two separate intravenous injections, one with stannous pyrophosphate, followed by a second with technetium-99m pertechnetate. Cardiac imaging was performed with a gamma camera (General Electric - Elscint 215 M or APEX 409) with a low-energy, all-purpose collimator in the left anterior oblique view that provided optimal ventricular separation. In patients with CHF who have RV dilatation, chamber separation is enhanced, because dilatation is associated with septal flattening and shift of the RV anteriorly. Gated images were collected with a computer in a 64-by-64 matrix at a rate of at least 16 frames/cycle; the total acquisition time was 5 min per view. A simultaneous blood sample (5 ml) for absolute blood volume determination was also obtained in a preweighed syringe, and counted 10 cm from the surface of the gamma camera with the exact time noted. Camera angulation and depth of the LV center of mass were determined using the method of Links (17). For data analysis, LV time–activity curves were constructed through a variable region of interest generated by a semiautomated edge-detection program for each frame of the composite cardiac cycle. End-diastolic and end-systolic frames were determined from the time–activity curve thus generated as the frames with maximal count at or immediately after the R wave and frame with the minimal count, respectively. RV volumes were determined using manual edge detection on end-diastolic and end-systolic frames. All of the volumes were determined in duplicate by two experienced technologists in a blinded fashion, and the results averaged. Ejection fraction was routinely determined for the left and right ventricles as the difference between the end-systolic (ES) and the end-diastolic counts (ED) divided by ED counts, all corrected for background (B) counts as follows: EF = [(ED-B) − (ES-B)]/(ED-B). To determine LVEDV, the ratio of the count rate from the LV and the concentration of radiotracer in the peripheral blood sample was calculated as: LVEDV = (Count rate from LV in end-diastolic frame/eud)/(Count rate/ml from blood sample), where u is the average linear attenuation coefficient and d is the depth of the center of the LV in the body (17). The linear attenuation coefficient was assumed to be equal to that of water (u = 0.15 cm−1). The count rate from the LV in the end-diastolic frame was calculated as: Total LV counts in end-diastolic frame/Time per frame × number of cycles acquired. The calculation of LV end-systolic, RV end-diastolic, and RV end-systolic volumes (LVESV, RVEDV, and RVESV, respectively) was analogous to that described for LVEDV. Determination of LV and RV volumes by this method is highly reproducible and has been validated against angiographically determined cardiac volumes (17, 18). Quality assurance studies performed in our Nuclear Cardiology Laboratory have established the standard error of the estimates for RV and LV ejection fraction to be less than 2% using semiautomated edge detection, and the RV and LV volume variation is less than 5% of the end-diastolic volume.
On the day of the study, RNA was performed to determine baseline RV and LV volumes. CPAP (BiPAP; Respironics Inc., Murrysville, PA) was then applied at a pressure of 10 cm H2O using a comfortably fitting nasal mask for 30 min. We used 10 cm H2O of CPAP because this is a pressure shown to be effective in improving LV function over time (2). Patients breathed through their noses with their mouths tightly closed. RNA images were then collected at the end of the 30-min CPAP application to determine its effects on cardiac volumes.
Data are presented as mean ± SEM. Two-tailed unpaired t tests were used to compare data between the IsC and IDC groups. Two-tailed paired t tests were used to compare data within groups. Relationships among variables were examined by least-squares linear regression analyses. A p value < 0.05 was considered statistically significant.
Baseline characteristics of the patients are shown in Table 1. Twenty-one were male and one female, 13 had IsC, and nine had IDC. Patients in the IDC group were significantly younger than those in the IsC group. Otherwise the two groups were similar, including their medication use.
| IsC (n = 13 ) | IDC (n = 9) | |||
|---|---|---|---|---|
| Age, yr | 69.4 ± 2.5 | 57.3 ± 4.7† | ||
| Sex, male/female | 13/0 | 8/1 | ||
| BMI, kg/m2 | 29.6 ± 1.6 | 26.4 ± 1.9 | ||
| NYHA classification, n | ||||
| I | 1 | 1 | ||
| II | 2 | 1 | ||
| III | 9 | 5 | ||
| IV | 1 | 2 | ||
| Heart rate, beats/min | 64.0 ± 3.6 | 68.8 ± 3.8 | ||
| Cardiac rhythm, n | ||||
| Sinus | 9 | 8 | ||
| Atrial fibrillation | 3 | 1 | ||
| Medication use, n | ||||
| Diuretics | 12 | 9 | ||
| Digoxin | 7 | 7 | ||
| Beta blockers | 2 | 1 | ||
| Vasodilators | ||||
| ACE inhibitors | 11 | 7 | ||
| Nitrates | 7 | 2 | ||
| Ca antagonists | 3 | 0 | ||
| Hydralazine | 2 | 2 | ||
| Antiarrhythmic agents | 5 | 2 | ||
| Warfarin | 6 | 5 | ||
| Aspirin | 4 | 0 |
Baseline LV and RV volumes and ejection fraction data are presented in Table 2. No significant differences were found between the IsC and IDC groups in baseline LV ejection fraction (EF), RV ejection fraction (EF), LVESV, RVESV, or RVEDV. However, the IDC group had significantly higher baseline LVEDV, almost twice that of the IsC group (p = 0.03), even when the IsC and IDC groups were age matched by excluding data from the two oldest patients in the IsC group (age: 67 ± 2 versus 57 ± 2 yr, respectively, p = 0.06; LVEDV: 307 ± 40 versus 521 ± 100 ml, respectively, p = 0.047). Severe LV and RV systolic functional impairment was observed in both groups, as indicated by mean LVEF and RVEF less than 30%. Neither group experienced significant changes in LVEF or RVEF while on CPAP. In the IDC group, changes in LVEF and RVEF with the application of CPAP were −0.2 ± 1.5% and −0.8 ± 1.6%, respectively. In the IsC group, the changes in LVEF and RVEF were −1.4 ± 0.8%, and 3.5 ± 1.9%, respectively. Although the response to CPAP varied within the IsC group, such that cardiac volumes decreased in some patients and increased in others, there were no significant changes in the mean LV or RV volumes in response to CPAP. In contrast, in the IDC group CPAP caused significant reductions in mean RVEDV (527 ± 77 to 354 ± 50 ml, p = 0.03) and RVESV (400 ± 78 to 272 ± 54 ml, p = 0.04). In addition, CPAP-induced changes in the IDC group in mean LVEDV, LVESV, RVEDV, and RVESV were all significantly greater than the mean changes in the IsC group (Figure 1). In the IDC group, the mean percent changes in LVEDV, LVESV, RVEDV, and RVESV during CPAP were −8%, −6%, −25%, and −24%, respectively. In the IsC group the corresponding mean percent changes were +9%, +13%, +0.8%, and −2%, respectively.
| IsC (n = 13 ) | IDC (n = 9) | |||
|---|---|---|---|---|
| RVEDV, ml | 490 ± 54 | 527 ± 77 | ||
| RVESV, ml | 370 ± 44 | 400 ± 78 | ||
| RVEF, % | 24 ± 3 | 27 ± 4 | ||
| LVEDV, ml | 309 ± 34 | 521 ± 100† | ||
| LVESV, ml | 236 ± 29 | 430 ± 96 | ||
| LVEF, % | 25 ± 2 | 22 ± 5 |
Fig. 1. Change in cardiac volumes from baseline with nasal CPAP application in patients with IsC and IDC. *p < 0.05 versus change in volume in the IsC group. #p < 0.05 versus baseline volume in the IDC group.
[More] [Minimize]Within the IDC group, the change in both RV and LV volumes with CPAP application correlated significantly with baseline volumes (Figure 2). The larger the baseline volume, the greater the reduction in volume during CPAP application. In contrast, no such relationship was observed in the IsC group. In both the IsC and IDC groups, the change in LVESV and RVESV while on CPAP correlated significantly with the change in LVEDV and RVEDV, respectively (Figures 3 and 4). There were no significant correlations between the changes in RVEDV and LVEDV, or changes in RVESV and LVESV.
Fig. 2. Relationships between baseline cardiac volumes and change (Δ) in cardiac volumes with nasal CPAP application in patients with IDC. The data are normally distributed.
[More] [Minimize]
Fig. 3. Relationships between changes (Δ) in end-diastolic volumes and end-systolic volumes with nasal CPAP application in patients with IsC.
[More] [Minimize]
Fig. 4. Relationships between changes (Δ) in end-diastolic volumes and end-systolic volumes with nasal CPAP application in patients with IDC.
[More] [Minimize]Heart rate did not change significantly in either group during CPAP application. Exclusion of patients with atrial fibrillation from data analysis and matching the IsC and IDC groups more closely for age did not affect the statistical significances of volume changes or correlations among variables.
The main purpose of the present study was to assess and compare the effects of short-term application of 10 cm H2O of CPAP on ventricular volumes in patients with IsC and IDC. The second purpose was to compare the effects of CPAP on LV and RV volumes. We found that the effects of CPAP on cardiac volumes differed between patients with IsC and those with IDC, and between RV and LV volumes in patients with IDC. Four main observations were made. First, LVEDV is larger in patients with IDC than in IsC. Second, the IDC group experienced nonsignificant reductions in LV volumes, and significant reductions in RV volumes in response to CPAP. In these patients, reductions in RV volumes exceeded reductions in LV volumes. Third, changes in LVEDV, LVESV, RVEDV, and RVESV induced by CPAP were significantly greater in patients with IDC than in those with IsC. Finally, IDC patients with the largest cardiac volumes experienced the greatest CPAP-induced decreases in cardiac volumes.
Our observation that baseline LVEDV is larger in patients with IDC than IsC is consistent with previous reports (9, 11). Although baseline LVESV was also larger by approximately 200 ml in the IDC group than in the IsC group, the difference was not statistically significant. In patients referred for heart transplantation, Warner-Stevenson and Perloff (11) reported that LVEDV was significantly greater in patients with IDC than in those with IsC. Bulkley and coworkers (9) also found, in an autopsy study, that heart weight and cardiac chamber size were greater in patients with IDC than in those with IsC. In contrast, Iskandrian and coworkers (19) observed no difference in LVEDV measured by RNA between patients with IsC or IDC. However, their data were derived from a retrospective study in which the presence or absence of myocardial infarctions was not documented in either group, so that the diagnoses of IsC and IDC were not firmly established. In our study, on the other hand, objective evidence of coronary artery disease either by coronary angiography or documented myocardial infarction allowed more reliable classification of patients as having IsC or IDC.
In the IDC group, baseline RV and LV volumes were similar. This finding is in keeping with the primary pathology of this disease, which is a diffuse cardiomyopathy involving both ventricles (19). In contrast, RV volumes were greater than LV volumes in the IsC group, just as they are in subjects with normal heart function (20, 21). Dilatation of the RV secondary to pulmonary hypertension is the most likely explanation for this phenomenon. The LV is less compliant in IsC than in IDC, from either ischemia or prior infarction, resulting in more diastolic dysfunction, higher pulmonary artery pressures and, as a consequence, passive dilatation of the more compliant RV (14, 15). In view of the severe LV dysfunction in the patients with IsC (Table 2), many probably suffered from pulmonary hypertension. However, this possibility can be neither confirmed nor excluded because pulmonary artery pressures were not directly measured.
The effects of 10 cm H2O of CPAP on cardiac volumes differed between the IsC and IDC groups as well as between the right and left ventricles. We found that CPAP induced reductions in LVEDV, LVESV, RVEDV, and RVESV in the IDC group, and the absolute changes from baseline values were significantly greater than in the IsC group. Indeed, within the IsC group, CPAP had no significant effect on any of these volumes. Within the IDC group, CPAP caused significant reductions in RVEDV and RVESV; however, it had no significant effect on LVEDV or LVESV. The reductions in RV volumes were almost certainly not caused by overlap of the LV, because in patients such as ours, RV dilatation enhanced ventricular separation by causing flattening of the interventricular septum and an anterior shift of the RV. Moreover, lung inflation by CPAP would tend to artifactually increase the calculated RV volume, rather than decrease it, owing to reductions in background counts and attenuation. There are only a few studies in which the effects of CPAP or PEEP on cardiac volumes have been examined, and these were performed exclusively in animals and humans with normal cardiac function. Furthermore, most of these studies focused on the LV. As described subsequently, those few that evaluated RV dimensions reported inconsistent results. Although we have discussed articles evaluating the effects of CPAP and PEEP on cardiac volumes collectively, the effects of these two interventions on cardiac volumes may not be identical. Because inspiratory intrathoracic pressure becomes negative during CPAP, but remains positive during mechanical ventilation with PEEP, venous return may be better preserved during CPAP.
Jardin and colleagues (22) used echocardiography to investigate the effects of 15 cm H2O of CPAP on LV and RV volumes in a group of healthy volunteers. CPAP produced a marked decrease in LV, and an increase in RV dimensions, suggesting an increase in RV afterload. In a subsequent study (23) the same group showed that 20 cm H2O of PEEP did not alter RVEDV when applied to patients with reduced lung compliance. In contrast, Huemer and colleagues (24), also using echocardiography, observed significant reductions in both RV and LV end-diastolic dimensions in healthy volunteers subjected to stepwise increases in PEEP up to 12.5 cm H2O. However, all of these results must be interpreted with caution because echocardiography is a relatively poor technique for determining RV and LV volumes. Furthermore, LV and RV dimensions would likely be influenced artifactually by positive airway pressure–induced lung inflation that alters the configuration of the heart and its orientation toward the echocardiographic probe. Pinsky and colleagues did not observe any significant changes in RV volumes, measured by a modified pulmonary artery catheter, with the application of 15 cm H2O of PEEP in 12 intubated patients after thoracotomy for cardiac surgery (25). In another study in which LVEDV and RVEDV were measured directly by a conductance catheter in anesthetized pigs, positive airway pressure did not affect LVEDV but caused a significant reduction in RVEDV (26), just as we observed in our IDC group. The inconsistent findings of these previous reports are probably accounted for by differences in and limitations of the techniques used to measure cardiac volumes, including potential artifacts caused by lung inflation by positive airway pressure, as well as differences in species and experimental preparations. It is nevertheless apparent that the effect of positive airway pressure on ventricular volumes is complex and is probably affected by the type of experimental preparation and underlying cardiorespiratory function.
The greater reductions in cardiac volumes in the IDC group compared with the IsC group are likely related to differences between the two disease processes. IDC is characterized by diffuse myocardial disease with far less focal scarring and greater myocardial compliance than in patients with IsC (9, 10). In fact, Bortone and coworkers (12) found that myocardial compliance can be normal in patients with IDC despite severely depressed systolic function. It follows that CPAP would have a greater impact on cardiac volumes in the relatively compliant ventricles of patients with IDC than in those with IsC. However, because we did not make direct myocardial compliance measurements, we cannot confirm this hypothesis. Prior studies evaluating the effects of positive airway pressure in CHF have not distinguished between patients with IsC and IDC. However, our data indicate that it is essential to do so because of differences in cardiac pathophysiology and responses to CPAP between these two diseases.
Within the IDC group only RV volumes decreased significantly in response to CPAP, and those reductions were significantly greater than reductions in LV volumes. These findings are in agreement with those of Shi and coworkers (26) who observed in pigs that positive pressure ventilation primarily reduced RVEDV significantly but did not affect LVEDV. In our study, CPAP probably caused greater reductions in RV volumes for two reasons. First, CPAP increases intrathoracic pressure, which reduces the pressure gradient for venous flow and directly compresses the inferior vena cava (27), both of which reduce venous return to the right heart. Second, the RV free wall is significantly thinner than that of the LV in normal hearts (3 to 5 mm compared with 8 to 12 mm, respectively) (13), as well as in explanted IDC hearts (4 mm versus 12 mm) and IsC hearts (6 mm versus 10 mm) (10). It follows, therefore, that the RV is much more compliant than the LV and would be more sensitive to any preload or afterload reducing effects of CPAP (14, 15). Therefore, positive intrathoracic pressure should cause a relatively greater reduction in RV than in LV volumes.
Another interesting observation from our study was that IDC patients with the largest RV and LV volumes experienced the greatest CPAP-induced reductions in these volumes. This relationship held true for both end-diastolic and end-systolic volumes. The most likely explanation for the reduction in end-diastolic volumes is that the more dilated hearts are more compliant, and therefore, when exposed to a given positive intrathoracic pressure, would decrease in volume more than the smaller, less compliant ventricles. On the other hand, reductions in end-systolic volumes are more likely owing to reductions in RV and LV afterload. We have previously shown that CPAP causes significant reductions in LV transmural pressure (an important determinant of afterload) by increasing intrathoracic pressure (3). If contractility is constant, a decline in LV systolic transmural pressure should cause a decrease in end-systolic ventricular volume. With respect to the RV, CPAP should also reduce transmural pressure by increasing intrathoracic pressure. In addition, Lenique and coworkers (28) have shown that CPAP causes reductions in pulmonary artery pressure in patients with CHF, which would also reduce RV afterload. Such a reduction in RV afterload in our patients with IDC could have contributed to the decrease in RVESV while on CPAP. Another possible explanation for greater reductions in end-systolic volumes in the more dilated hearts is the greater reduction in end-diastolic volumes, which by reducing the radius of curvature, would reduce wall stress (i.e., afterload) via Laplace's relationship (29).
Despite reductions in biventricular volumes in the IDC group, there was no measurable change in ejection fraction. The most likely explanation for this finding is that CPAP application is associated with reductions in the degree of tricuspid and mitral regurgitation (1), which many of these patients have. This would tend to decrease ejection fraction because of a reduced “apparent” stroke volume. The lack of change in RVEF as RVEDV changed in patients with IDC in our study is consistent with the findings of a previous report. Dhainaut and coworkers (30) showed in septic patients that RVEF remained constant as RVEDV changed. This was presumably because changes in RVESV were proportional to changes in RVEDV. Thus, in the current study, the striking correlation between the changes in EDV and ESV during CPAP application in both the IsC and IDC groups is consistent with other published reports (25, 30). Clearly preload and afterload are coupled, such that a CPAP-induced reduction in RVEDV will reduce afterload, via Laplace's relationship, by reducing the radius of curvature of the RV. As a result, RVESV will decrease. Similarly, any CPAP-induced reduction in RV afterload will be associated with a fall in the RV force of contraction, which will reduce RVEDV via Starling's law.
Atherton and colleagues recently assessed ventricular volumes by RNA before and during 5 min of lower body negative pressure of −30 mm Hg in patients with chronic CHF (31). In a minority of patients (nine of 21) they observed a paradoxical increase in LVEDV in association with the expected decrease in RVEDV, and suggested that this finding was a result of diastolic ventricular interaction. When this interaction occurs, acute volume unloading, such as with lower body negative pressure or positive intrathoracic pressure, results in a reduction in RVEDV, as expected, but LVEDV may paradoxically increase because of a shift of the interventricular septum toward the RV owing to pericardial constraint. In our study, there was no such tendency for LVEDV to increase in the face of a reduction in RVEDV. The most likely explanation for the lack of a paradoxical increase in LVEDV while on CPAP in our study is that whereas lower body suction causes pure preload reduction, in the failing heart CPAP-induced increases in intrathoracic pressure can cause reductions in both preload and afterload (3, 28). This additional effect of reduced LV afterload would counteract any tendency toward LV dilation in accordance with Starling's law. Thus, our findings do not favor ventricular interdependence as a mechanism by which CPAP influences LV volumes in IDC (32).
We used RNA to determine ventricular volumes off and on CPAP because it is a noninvasive technique that is highly reproducible and has been validated against angiographically determined volumes of both the left (17, 33, 34) and right ventricles (18, 35). This technique has several advantages over echocardiography for this purpose. First, it provides a more direct assessment of ventricular volumes than echocardiography. Second, optimal determination of cardiac volumes by echocardiography requires a breath hold that is usually accompanied by glottic closure. Under these conditions positive airway pressure would not be transmitted to the intrathoracic airways and heart. In contrast, cardiac volume assessment by RNA is performed during spontaneous breathing, so that the heart and other intrathoracic structures are exposed to positive airway pressure during CPAP application. Third, unlike echocardiography, determination of ventricular volumes by RNA is not subject to artifacts in the setting of increased lung inflation as occurs during CPAP application (36). Finally, the inability to obtain multiple adequate windows during echocardiography precludes accurate assessment of RV dimensions in adults (37, 38). A potential drawback of RNA for determination of RV volumes, however, is that the RV cavity overlies the LV in the anterior view and the right atrium in the left anterior oblique view, which could result in spuriously elevated RV counts. Nevertheless, where care has been taken to obtain maximal separation of the right and left ventricles in the best septal view (as in our study) others have demonstrated excellent correlations of RVEDV (r = 0.91) and RVESV (r = 0.93), determined by RNA, with those obtained by a cast-validated biplane cineventriculographic method (18).
Taken together, our data indicate that short-term application of CPAP does not reduce LV preload in patients with CHF, except in a subset of IDC patients with very large LVEDV. In contrast, we (3) and others (28) have demonstrated, in patients with CHF due to either IsC or IDC, that CPAP causes consistent reductions in LV afterload by increasing intrathoracic pressure. It further decreases LV afterload in those CHF patients with coexisting sleep apnea, by reducing sympathetic nervous system activity (2, 39) and systemic blood pressure (4). Accordingly, primary CPAP- induced reductions in LV afterload and sympathetic nervous system activity are more likely to explain long-term increases in LVEF, and reductions in mitral regurgitation and atrial natriuretic peptide levels (1) over time than are primary reductions in LV preload, especially in those patients who had IsC. Nevertheless, among IDC patients with very large LV volumes, it is possible that primary CPAP-induced reduction in LV volumes could contribute to these long-term effects. The significance of the CPAP-induced reductions in RV volumes in patients with IDC is uncertain, but suggests that short-term CPAP can reduce overall cardiac volume, which could contribute to reductions in atrial stretch and atrial natriuretic peptide secretion (40). In conclusion, short-term application of CPAP is more likely to reduce both RV and LV preload in CHF patients with IDC than in those with IsC. Whether this predicts more beneficial long-term effects of CPAP in patients with IDC than in those with IsC is uncertain, but merits further investigation.
Supported by operating grant MT 11607 from the Medical Research Council of Canada.
| 1. | Tkacova R., Liu P. P., Naughton M. T., Bradley T.D.Effect of continuous positive airway pressure on mitral regurgitant fraction and atrial natriuretic peptide in patients with heart failure. J. Am. Coll. Cardiol.301997739745 |
| 2. | Naughton M. T., Liu P. P., Benard D. C., Golstein R. S., Bradley T. D.Treatment of congestive heart failure and Cheyne-Stokes respiration during sleep by continuous positive airway pressure. Am. J. Respir. Crit. Care Med.15119959297 |
| 3. | Naughton M. T., Rahman A., Hara K., Floras J., Bradley T. D.Effect of continuous positive airway pressure on intrathoracic and left ventricular transmural pressures in patients with congestive heart failure. Circulation91199517251731 |
| 4. | Tkacova R., Rankin F., Fitzgerald F. S., Floras J. S., Bradley T. D.Effects of continuous positive airway pressure on obstructive sleep apnea and left ventricular afterload in patients with heart failure. Circulation98199822692275 |
| 5. | Pfeffer M. A.Left ventricular remodeling after acute myocardial infarction. Ann. Rev. Med.461995455466 |
| 6. | Fewell J. E., Abendschein D. R., Carlson C. J., Murray J. F., Rapaport E.Continuous positive-pressure ventilation decreases right and left end-diastolic volumes in dog. Circ. Res.461980125132 |
| 7. | Genovese J., Huberfeld S., Tarasiuk A., Moskowitz M., Scharf S. M.Effects of CPAP on cardiac output in pigs with pacing-induced congestive heart failure.Am. J. Respir. Crit Care. Med.152199518471853 |
| 8. | Johnston W. E., Viton-Johansen J., Santamore W. P., Case L. D., Little W. C.Mechanism of reduced cardiac output during positive end-expiratory pressure in the dog. Am. Rev. Respir. Dis.140198912571264 |
| 9. | Bulkley B. H., Hutchins G. M., Bailey I., Strauss H. W., Pitt B.Thallium 201 imaging and gated cardiac blood pool scans in patients with ischemic and idiopathic congestive cardiomyopathy. Circulation551977753760 |
| 10. | Parodi O., DeMaria R., Oltrona L., Testa R., Sambuceti G., Roghi A., Merli M., Berlingheri L., Accinni R., Spinelli F., Pellegrini A., Baroldi G.Myocardial blood flow distribution in patients with ischemic heart disease or dilated cardiomyopathy undergoing heart transplantation. Circulation881993509522 |
| 11. | Warner-Stevenson L., Perloff J. K.The dilated cardiomyopathies: clinical aspects. Cardiol. Clin.61988187218 |
| 12. | Bortone A. S., Hess O. M., Chiddo A., Gaglione A., Locuratolo N., Caruso G., Rizzon P.Functional and structural abnormalities in patients with dilated cardiomyopathy.J. Am. Coll. Cardiol.141989613623 |
| 13. | Williams, P. L., L. H. Bannister, M. M. Berry, P. Collins, M. Dyson, J. E. Dussek, and M. W. J. Ferguson. 1995. Gray's Anatomy, 38th ed. Churchill Livingstone, New York. 1480,1482. |
| 14. | Konstam M. A., Cohen S. R., Salem D. N., Conlon T. P., Isner J. M., Das D., Zile M. R., Levine H. J., Kahn P. C.Comparison of left and right ventricular end-systolic pressure–volume relations in congestive heart failure. J. Am. Coll. Cardiol.5198513261334 |
| 15. | Konstam M. A., Weiland D. S., Conlon T. P., Martin T. T., Cohen S. R., Eichhorn E. J., Isner J. M., Zile M. R., Salem D. N.Hemodynamic correlates of left ventricular versus right ventricular radionuclide volumetric responses to vasodilator therapy in congestive heart failure secondary to ischemic or dilated cardiomyopathy. Am. J. Cardiol.59198711311137 |
| 16. | Manolio T. A., Baughman K. L., Rodeheffer R., Pearson T. A., Bristow J. D., Michels V. V., Abelmann W. H., Harlan W. R.Prevalence and etiology of idiopathic dilated cardiomyopathy (summary of a National Heart, Lung, and Blood Institute workshop). Am. J. Cardiol.69199214581466 |
| 17. | Links J. M., Becker L. C., Shindledecker G., Guzman P., Burow R. D., Nickoloff E. L., Alderson P. O., Wagner H. N.Measurement of absolute left ventricular volume from gated blood pool studies. Circulation6519828290 |
| 18. | Dell'Italia L. J., Starling M. R., Walsh R. A., Badke F. R., Lasher J. C., Blumhardt R.Validation of attenuation-corrected equilibrium radionuclide angiographic determinations of right ventricular volume: comparison with cast-validated biplane cineventriculography. Circulation721985317326 |
| 19. | Iskandrian A. S., Helfeld H., Lemlek J., Lee J., Iskandrian B., Heo J.Differentiation between primary dilated cardiomyopathy and ischemic cardiomyopathy based on right ventricular performance. Am. Heart J.1231992768773 |
| 20. | Benchimol A., Desser K. B., Hastreiter A. R.Right ventricular volume in congenital heart disease. Am. J. Cardiol.3619756775 |
| 21. | Pietras R. J., Kondos G. T., Kaplan D., Lam W.Comparative angiographic right and left ventricular volumes. Am. Heart J.1091985321326 |
| 22. | Jardin, F., J.-C. Farcot, P. Gueret, J.-F. Prost, Y. Ozier, and J.-P. Bourdarias. 1984. Echocardiographic evaluation of ventricles during continuous positive airway pressure breathing. J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. 56:619–627. |
| 23. | Jardin F., Brun-Ney D., Hardy A., Aegerter P., Beauchet A., Bourdarias J.-P.Combined thermodilution and two-dimensional echocardiographic evaluation of right ventricular function during respiratory support with PEEP. Chest991991162168 |
| 24. | Huemer G., Kolev N., Kurz A., Zimpfer M.Influence of positive end-expiratory pressure on right and left ventricular performance assessed by doppler two-dimensional echocardiography. Chest10619946773 |
| 25. | Pinsky M. R., Desmet J., Vincent J. L.Effect of positive end-expiratory pressure on right ventricular function in humans. Am. Rev. Respir. Dis.1461992681687 |
| 26. | Shi W., Lutz J., Ondulick B. W., Pinsky M. R.Positive-pressure ventilation selectively reduces right ventricular end-diastolic volume (abstract). FASEB J.91995A329 |
| 27. | Fessler H. E., Brower R. G., Shapiro E. P., Permutt S.Effects of positive end-expiratory pressure and body position on pressure in the thoracic veins. Am. Rev. Respir. Dis.148199316571664 |
| 28. | Lenique F., Habis M., Lofaso F., Dubois-Rande J.-L., Harf A., Brochard L.Ventilatory and hemodynamic effects of continuous positive airway pressure in left heart failure. Am. J. Respir. Crit. Care Med.1551997500505 |
| 29. | Hurst, J. W. 1986. The Heart, 6th ed. McGraw-Hill, New York. 328–329. |
| 30. | Dhainaut J.-F., Pinsky M. R., Nouria S., Slomka F., Brunet F.Right ventricular function in human sepsis: a thermodilution study. Chest112199710431049 |
| 31. | Atherton J. J., Moore T. D., Lele S. S., Thomson H. L., Galbraith A. J., Belenkie I., Tyberg J. V., Frenneaux M. P.Diastolic ventricular interaction in chronic heart failure. Lancet349199717201724 |
| 32. | Bradley T. D., Holloway R. M., McLaughlin P. R., Ross B. L., Walters J., Liu P. P.Cardiac output response to continuous positive airway pressure in congestive heart failure. Am. Rev. Respir. Dis.1451992377382 |
| 33. | Dehmer G. J., Lewis S. E., Hillis L. D., Twieg D., Falkoff M., Parkey R. W., Willerson J. T.Nongeometric determination of left ventricular volumes from equilibrium blood pool scans. Am. J. Cardiol.451980293300 |
| 34. | Massie B. M., Kramer B. L., Gertz E. W., Henderson S. G.Radionuclide measurements of left ventricular volume: comparison of geometric and count-based methods. Circulation651982725730 |
| 35. | Dehmer G. J., Firth B. G., Hillis L. D., Nicod P., Willerson J. T., Lewis S. E.Nongeometric determination of right ventricular volumes from equilibrium blood pool scans. Am. J. Cardiol.4919827884 |
| 36. | Butler G. C., Naughton M. T., Rahman M. A., Bradley T. D., Floras J. S.Continuous positive airway pressure increases heart rate variability in congestive heart failure. J. Am. Coll. Cardiol.251995672679 |
| 37. | Lewis J. F., Webber J. D., Sutton L. L., Chesoni S., Curry C. L.Discordance in degree of right and left ventricular dilation in patients with dilated cardiomyopathy: recognition and clinical implications. J. Am. Coll. Cardiol.211993649654 |
| 38. | Sun J. P., James K. B., Yang X. S., Solankhi N., Shah M. S., Arheart K. L., Thomas J. D., Stewart W. J.Comparison of mortality rates and progression of left ventricular dysfunction in patients with idiopathic dilated cardiomyopathy and dilated versus nondilated right ventricular cavities. Am. J. Cardiol.80199715831587 |
| 39. | Somers V. K., Dyken M. E., Clary M. P., Abboud F. M.Sympathetic neural mechanisms in obstructive sleep apnea. J. Clin. Invest.96199518971904 |
| 40. | Wilkins M. A., Su X. L., Palayew M. D., Yamashiro Y., Rolli P., McKenzie J. K., Kryger M. H.The effects of posture change and continuous positive airway pressure on cardiac natriuretic peptides in congestive heart failure. Chest1071995909915 |
P. P. Liu is supported by an Endowed Research Chair of the Heart and Stroke Foundation of Ontario.
