Marked Discordance Between Dynamic and Passive Diastolic Pressure-Volume Relations in Idiopathic Hypertrophic Cardiomyopathy
Background Dynamic diastolic pressure-volume curves measured during filling (PVRfill) in patients with idiopathic hypertrophic cardiomyopathy (HCM) are often considerably shallower than would be anticipated if one assumed high chamber stiffness. We hypothesized that these curves deviate markedly from the passive end-diastolic pressure-volume relation (EDPVR) and explored the mechanisms for such a discordance.
Methods and Results We used invasive pressure-volume analysis and conductance catheter methodology to study 42 patients. Nine had HCM, and the remaining patients comprised three comparison groups: 11 with normal left ventricular (LV) function, 13 with LV hypertrophy secondary to chronic hypertension (LVH-HTN), and 9 with idiopathic dilated cardiomyopathy (DCM). EDPVRs were recorded during balloon catheter obstruction of inferior vena cava inflow. In normal subjects, LVH-HTN patients, and DCM patients, PVRfill curves deviated only slightly from the passive EDPVR. In striking contrast, HCM patients displayed a flat PVRfill that was very different from the steep EDPVR. On reduction of preload, PVRfill relations in HCM shifted downward in parallel, with a net pressure decline at the same chamber volume of −10±4 mm Hg. This staircaselike shift was much less in the other patient groups (−2±2 mm Hg; P<.001). The unusual behavior in HCM could not be attributed directly to increased viscosity, enhanced pericardial constraint, or preload dependence of isovolumic relaxation. Regional heterogeneity of relaxation may play a role; however, we speculate that the major mechanism relates to the unique fiber and chamber architecture seen with HCM and possibly to enhanced ventricular interaction.
Conclusions Elevated LV filling pressures in HCM are not due simply to a stiff cavity but also reflect a major influence of offset pressures that vary with chamber loading. The large disparity between flat pressure-volume relations during filling and steep end-diastolic relations appears unique to HCM. This indicates that caution should be used in the interpretation of stiffness results derived from steady-state data and suggests that therapies that alter cavity geometry and/or reduce interaction may markedly influence LV diastolic pressures in HCM.
Idiopathic HCM is an inherited disorder whose pathological hallmarks are a profound thickening of the myocardium (due to myocyte enlargement and interstitial fibrosis) and an abnormal orientation of the muscle fibers.1 The disease often preferentially involves the interventricular septum, although concentric involvement can be observed. The combination of altered intrinsic myocardial wall properties and distorted chamber geometry is thought to increase chamber stiffness, with the result that LV end-diastolic pressures are raised and filling is limited.2 3 4 5 6 7 In addition, abnormalities in calcium handling and delayed pressure relaxation can contribute to diastolic dysfunction.8 9 10 11
The net effect of these abnormalities is an alteration of the diastolic PV relation. Precise characterization of this net change, however, remains remarkably limited in the literature. Prior studies6 10 11 12 13 14 uniformly presented data measured during filling of resting cardiac cycles, and surprisingly, the majority revealed fairly flat diastolic PV relations, which suggests a higher-than-anticipated compliance. In addition to chamber volume, however, diastolic pressures are influenced by extrinsic loads from the pericardium or right heart,15 16 17 abnormal relaxation,10 11 14 18 and viscoelastic behavior19 20 21 and potentially by the distorted cavity geometry itself.22 The passive PV relations measured at end diastole could be considerably stiffer, but such relations have not been reported previously in intact patients with HCM.
The purpose of the present study was to compare the dynamic diastolic PV curve measured during chamber filling with the passive EDPVR in patients with HCM. We used the conductance catheter method to measure both continuous PV data at rest and EDPVRs derived from multiple cycles at varying preloads.23 24 25 Data from HCM patients were compared with those from patients with acquired concentric hypertrophy due to chronic hypertension (to test for specificity of the behavior), patients with DCM (to test for the influence of chronic chamber enlargement), and normal control subjects. The results reveal a novel and unique abnormality of diastolic PV relations in HCM that suggests an exaggerated influence of external load and/or chamber geometric distortion on the dependence between diastolic pressure and volume during normal filling in these patients.
The total study group consisted of 42 patients divided into four clinical subgroups. None of the patients had coronary artery or valvular heart disease. Nine patients had HCM; of these, 8 presented with NYHA class III or IV heart failure symptoms and 1 had exertional chest pain. Eight of the HCM patients had resting intracavitary pressure gradients. Chronic medications, such as β-adrenergic and calcium channel blockers, were discontinued 1 to 2 days before study. A comparison control group consisted of 11 patients with normal LV function who had been referred for evaluation of atypical chest pain. These patients all had normal ECGs, left ventriculograms, and echocardiograms. A second comparison group comprised 13 patients with LVH-HTN. Virtually all of these patients were referred for evaluation of exertional dyspnea. Finally, 9 patients had idiopathic DCM and NYHA class III symptoms.
All patients provided informed consent, and the protocol was approved by the Joint Committee on Clinical Investigation of the Johns Hopkins Medical Institutions. There were no complications associated with the study protocol.
Detailed methods for PV catheterization study have been reported.23 24 25 Briefly, after routine right- and left-heart catheterization, a multielectrode conductance catheter (Webster Labs) was positioned inside the LV with the tip of the catheter at the apex. In all patients with HCM and four with LVH-HTN, the conductance catheter had two distal side holes but no pigtail tip so that ventricular ectopy could be minimized after placement within the obliterating distal apex. Apical pressures were recorded through the side holes by use of a fluid-filled manometer. The frequency response of this system has been reported previously26 and is flat from 0 to 8 Hz. LV pressures used for PV analysis were measured with a second pigtail catheter with a 2F micromanometer-tipped catheter (Millar Instruments) advanced through its length, which was placed in the proximal LV chamber (well below the aortic valve but above the region of cavity obliteration). In all other patients, the conductance catheter terminated with a pigtail, and the micromanometer was advanced inside the lumen. The conductance catheter was connected to a stimulator /microprocessor (VCU, Cardiac Pacemakers Inc, or Sigma-5, CardioDynamics) that applied an alternating current to base and apex electrodes. Voltages measured at intervening electrodes were inversely proportional to segmental volumes, and their signals were added to generate total LV volume. We calibrated the steady-state volume signal by matching its cardiac output to that determined by thermodilution and its ejection fraction to that derived by left ventriculography.
To vary right- and left-heart load, a large balloon-tipped catheter (SP9516, Cordis) was placed within the right atrium and inflated with 10 to 25 mL CO2 while being drawn toward the IVC. After maximal inflow obstruction was achieved (generally ≤10 seconds), the balloon was deflated. Hemodynamic data were digitized at 200 Hz for analysis.
Baseline hemodynamics were analyzed from signal-averaged sequential cardiac cycles (≈6 beats) under resting conditions. End-systolic and -diastolic volumes were derived from averaged volumes during isovolumic relaxation and contraction, respectively. End-diastolic pressure was the value at the lower right-hand corner of the PV loop. End-systolic pressure was the pressure at maximal instantaneous elastance [P/(V−Vo)], where Vo is the volume-axis intercept of the ESPVR. The ESPVR was derived from the collection of end-systolic PV points from multiple cardiac cycles during transient preload reduction. We determined isovolumic relaxation from the negative inverse slope of the pressure-dP/dt plot using data between dP/dtmin and the onset of filling.27 In HCM patients, relaxation of proximal chamber pressures was well fit by a single time constant (τ). However, apical pressures required assessment in two phases: the initial 40 ms after dP/dtmin (Τ1), and a later phase that led up to the onset of filling (Τ2).13
Two diastolic PV relations were defined and evaluated: (1) the dynamic relation derived from resting (steady-state) data measured during diastolic filling (PVRfill), and (2) the passive relation that consisted of end-diastolic PV points obtained from multiple, variably volume-loaded cardiac cycles. Both PVRfill and EDPVR curves were fit to a monoexponential model to derive chamber stiffness (βEDPVR) given by where P and V are LV diastolic pressure and volume, respectively, and α is a scaling factor. Viscous properties were examined by refitting PVRfill data to a viscoelastic model19 21 in the form . Parameters PO, α, βss, βEDPVR, and γ were obtained by nonlinear regression analysis by use of a Marquardt optimization algorithm.
Parallel downward displacements of the PVRfill curves with minimal change in LV volume occur abruptly after IVC inflow obstruction, which reflects the sudden decrease of right-heart loading.28 Further displacement ensues as left-heart volumes decline, which also is not directly attributable to the passive dependence of LV pressure on volume. We quantified this displacement in the following manner (Fig 1⇓): Two PV loops were identified after the onset of IVC obstruction, one of which was at the baseline preload and the second at near maximal preload reduction. Both loops were measured at end expiration to remove effects of varying intrathoracic pressures. The data were superimposed, and an overlapping volume was selected such that the LV was in the middle third of filling for both beats. The LV pressures at this same volume were determined for each beat by linear interpolation, and the difference between them (ΔPdia) served as an index of the parallel displacement of the PVRfill associated with LV preload reduction.
Comparison of Catheter to Ventriculographic Volumes in HCM Hearts
Prior studies29 reported that the volume catheter signal reliably tracks relative changes in chamber volume in humans with normal and abnormal wall motion. This correlation depends on a stable catheter position, and in HCM patients with distorted chamber geometry and distal cavity obliteration, this was assisted by use of a straight-tipped catheter that was more easily anchored at the apex. To further test the reliability of the relative volume changes measured by catheter in these patients, contrast ventriculogram–derived LV volumes were compared with conductance volumes. Fig 2⇓ displays an example from a patient with a resting intracavitary gradient of 120 mm Hg, an ejection fraction of 82%, and marked cavity obliteration. The two volume curves displayed good agreement, with an overall correlation of 0.98. Similar results were obtained in the other HCM patients.
Data from the four patient groups were analyzed by ANOVA to test for differences between group means. If the ANOVA revealed a significant influence of patient group, post hoc comparisons were made between the individual groups by use of Fisher's least significant difference multiple comparisons test. Changes in hemodynamic parameters induced by preload reduction were compared by use of Student's paired t test. All data are presented as mean±SD, with significance reported for values of P<.05.
Baseline Characteristics and Hemodynamics
Baseline clinical characteristics for each patient group are provided in Table 1⇓. Both LVH-HTN and HCM patients had increased wall thickness, with the greatest hypertrophy documented in the HCM group. The mean intracavitary pressure gradient from the eight HCM patients who displayed gradients at rest was 79±40 mm Hg. Two LVH-HTN patients also had provocable (postextrasystolic) gradients of ≥50 mm Hg, but none had gradients at rest. No other patient had resting or provocable intracavitary pressure gradients.
Table 2⇓ compares baseline hemodynamic data for the four patient groups. Each of the three disease groups had similarly elevated LV end-diastolic pressures compared with control subjects. RAP was also increased in each disease group but was highest in HCM patients. Ejection fraction and heart rate were altered only in DCM patients. Ees was significantly higher in HCM compared with control subjects. Ees was also elevated in LVH-HTN patients, but this reached borderline significance (P=.06) due to greater interpatient variability. Ees was reduced in DCM subjects.
The relaxation time constant (τ) derived from the proximal midchamber pressures in HCM subjects was not significantly different from values measured in control subjects or LVH-HTN subjects. Relaxation of distal cavity pressures was biphasic, with an initially slow decay (Τ1=92.1±65.5 ms) followed by a more rapid rate (Τ2=51.7±25.9 ms). The latter was more synchronous and numerically similar to the relaxation of proximal chamber pressures, whereas the initial slower rate corresponded to the decay in the systolic intracavitary pressure gradient. Examples of raw pressure and dP/dt tracings and pressure-dP/dt plots from which these time constants were derived are displayed in Fig 3⇓. Thus, by the onset of filling, relaxation rates in HCM were faster and comparable in proximal and distal cavity locations.
Diastolic PV Relation Analysis
Fig 4A⇓ shows PV data recorded during transient obstruction of IVC blood inflow for a representative patient in each group. The diastolic portion is replotted on an expanded scale in Fig 4B⇓. These graphs demonstrate the major new finding of the present study. For normal, LVH-HTN, and DCM patients, the diastolic PV data from multiple cardiac cycles aligned along a single relation. PVRfill and the EDPVR, therefore, were very similar. However, this differed strikingly from HCM patients, in whom individual PVRfill curves were remarkably flat and then shifted downward in parallel as preload volume declined progressively. This unusual behavior was observed in nearly every HCM subject. Thus, early and late diastolic pressures in HCM patients were quite similar, and their elevation was less dependent on a given chamber volume than on overall chamber loading (both right ventricular loading and LV loading). It is particularly noteworthy that this behavior was not observed in patients with secondary LVH due to hypertension.
Results for ΔPdia, the maximal parallel downward EDPVR shift measured at an identical middiastolic volume before and after preload reduction (cf Fig 1⇑), are displayed in Fig 5⇓. HCM patients exhibited more than a fivefold greater decline compared with each of the other three groups (eg, −9.7±3.8 versus −2.4±2.1 mm Hg in LVH-HTN). ΔPdia for the three comparison groups was similar.
The exaggerated parallel downward shift of the PVRfill data in HCM subjects introduced a major ambiguity regarding assessment of chamber stiffness. Typically, PVRfill data measured from a single condition at rest are fit to a monoexponential to derive stiffness. However, stiffness could also be estimated from the passive EDPVR. The theoretical advantage of this latter approach is that it avoids earlier diastolic data that can be contaminated by ongoing relaxation and filling-dependent phenomena (eg, viscosity). Fig 6⇓ demonstrates both approaches from a control subject and an HCM patient. Although some disparity between these fits exists in all patients, it was particularly striking in HCM patients. Table 3⇓ compares chamber stiffness coefficients derived from PVRfill (βss) and EDPVR data (βEDPVR). βEDPVR was smaller than βss in each group except HCM patients, in whom it was much larger. Thus, the flat PVRfill yielded a much lower estimate of chamber stiffness (higher compliance) than that derived from the passive EDPVR (P<.004).
Role of Preload-Induced Change in Hemodynamic Parameters
The enhanced ΔPdia in HCM patients could have resulted from an abnormal hemodynamic response to preload reduction, particularly in relaxation, RAP, or intracavitary pressure gradients. Table 3⇑ provides mean percent changes induced by preload reduction for major hemodynamic parameters. Relaxation time was minimally (<10%) and insignificantly altered in each patient group (Table 3⇑). Relaxation times derived from the apical pressures in HCM patients were also minimally changed by preload reduction (Τ1=111.7±16.2 ms, Τ2=44.5±18.6 ms; both P>.30 versus baseline). The percent decline in RAP was similar in the three disease groups. In particular, both HCM and DCM patients had nearly the same drop in RAP (−7 versus −6.6 mm Hg) despite very different EDPVR behavior. Similarly, responses in end-diastolic pressure, cardiac output, stroke volume, heart rate, end-systolic pressure, end-diastolic volume, and end-systolic volume were indistinguishable between HCM patients and one or more of these comparison groups. Finally, mean intracavitary systolic pressure gradient also was altered minimally by preload reduction (87±40 versus 79±40 mm Hg at baseline; P=.34).
Role of Viscoelasticity
Another potential contributor to enhanced ΔPdia in HCM patients was greater viscosity. More rapid filling at higher preloads could raise diastolic pressures if viscosity were also enhanced. Reduction of preload would lessen this effect and potentially increase ΔPdia. To test this, PVRfill data were fit to a four-term viscoelastic model (Table 3⇑). The results showed few clear differences among the groups. The one coefficient that was markedly elevated in the HCM patients was the pressure offset, Po (P<.05 versus control patients and P=.08 versus LVH-HTN). However, the steady-state chamber elastic modulus (βss) and viscosity parameter (γ) were not increased in HCM patients.
Influence of LV Pressure Recording Site
The pressure data used for EDPVR analysis in HCM subjects were obtained in the proximal-half chamber rather than the distal cavity to avoid potential effects of direct muscle compression on the transducer. However, this differed from the distal site used in the other patients, which raised a question as to whether the recording site itself contributed to the unusual behavior. Although systolic pressures in proximal and distal sites often differed in the HCM patients, this was not true of pressures during diastolic filling (see Fig 3A⇑). The mean difference between proximal and distal pressures throughout the diastolic filling period averaged only 1.8±2.1 mm Hg. When the analysis was restricted to the middle third of diastole, during which ΔPdia generally was measured, this pressure difference was even smaller (0.1±0.5 mm Hg). Thus, similar diastolic results would likely have been obtained from the use of distal pressures as well.
The present study presents a novel finding concerning LV diastolic PV behavior in patients with HCM. In contrast to the other patient groups examined, HCM patients displayed a striking discordance between the dynamic diastolic PV relation measured during filling of any single cardiac cycle and the passive (end-diastolic) chamber properties. The former was remarkably flat throughout filling and yielded a lower estimate of chamber stiffness. When right- and left-heart filling were reduced, there was an exaggerated, parallel, downward displacement of these curves. The result was a much greater decline in LV diastolic pressure for a given change in preload volume than that predicted from the steady-state PVRfill and a correspondingly stiffer EDPVR. These findings highlight a marked influence of factors other than passive chamber stiffness on the resting diastolic behavior in HCM and raise concerns regarding the traditional analysis of chamber stiffness based on single steady-state beats.
Relaxation and ΔPdia
Prolonged isovolumic relaxation superimposed on early diastolic filling can elevate the initial portion of the PVRfill, particularly when a rapid heart rate compromises the diastolic period.18 Parallel downward displacement of resting PVRfill in HCM patients (ie, ΔPdia) has been reported after treatment with verapamil12 13 or nifedipine11 14 and ascribed to improved relaxation. If preload reduction had accelerated relaxation in HCM patients similarly, it could have contributed to the downward, staircaselike shift. However, when preload was lowered by IVC inflow obstruction, it did not significantly alter relaxation rates in any of the patient groups. In the HCM group, this applied to both proximal and distal cavity pressures. This finding is supported by isolated muscle data that have shown relaxation to be fairly insensitive to preload change.30 31
Baseline isovolumic relaxation for the proximal chamber pressure in HCM was not prolonged in the present study, whereas many previous investigations10 11 12 13 14 have reported delayed relaxation with this disease. This discrepancy most likely is related to the chamber location and timing at which the calculations were made. Reported data most often are derived from distal LV cavity pressures; this reveals a complex relaxation process that requires a biexponential fit.13 32 The initial, slower relaxation reflects the decline in high distal intracavitary pressures, whereas the second, more rapid rate occurs as proximal and distal cavity pressures equilibrate. Similar differences were documented in the present study. Because our investigation focused on the PVRfill during middiastole rather than initial rapid filling, the latter, faster rate arguably was more relevant. By that time, the effects of prior discrepancies in relaxation between proximal and distal chambers had largely dissipated. This is shown by example in Fig 3A⇑; it was further supported by minimal mean pressure differences between proximal and distal cavity sites during much of diastolic filling. This result, particularly in relation to mid to late diastole, is consistent with previous published data.33
It is possible that heterogeneity of relaxation, which might not have been reflected adequately in proximal or distal chamber relaxation times, contributed to the unusual PVRfill behavior in HCM patients. Recent studies have documented enhanced regional dyssynchrony of diastolic wall motion34 and estimated stress relaxation35 in HCM patients. A higher coefficient of variation in either parameter was found to correlate directly with prolonged chamber relaxation. Regional dyssynchrony indicates a stretch of some portions of the chamber while others are still relaxing, which could elevate the initial portion of the PVRfill and contribute to a staircaselike appearance on reduction of preload.
Viscoelasticity and ΔPdia
Increased viscous properties of the heart have been reported, particularly in enlarged chambers,20 21 and are thought to stem from structural proteins that are external as well as internal to the myocyte.36 However, viscoelastic model analysis did not demonstrate a significant difference in the viscous term in HCM compared with the other patient groups. Examination of the PV data also failed to reveal a pressure rise during early filling or late in diastole with atrial systole. This is consistent with a prior study by Hess et al37 that reported little difference in the viscosity term between normal control subjects and patients with concentric hypertrophy due to aortic stenosis.
Parallel downward shifts of PVRfill often accompany administration of vasodilators such as nitroprusside and are typically ascribed to decreased pericardial constraints.38 39 The IVC occlusion method can rapidly reduce right-heart load, which often results in a downward shift of the PVRfill before left-heart volumes are diminished.28 This is thought to be due to release of pericardial and right-heart loading effects on the resting PVRfill. ΔPdia determined in the present study represents an additional downward displacement as LV volumes subsequently declined. It is difficult to relate this shift to a release of pericardial constraint for several reasons. RAP did not decline disproportionately more on IVC occlusion in HCM patients. Large epicardial volumes due to massive LV hypertrophy conceivably could exacerbate a pericardial constraint and contribute to the parallel shift phenomenon. However, ΔPdia was small in DCM patients despite large volumes, and RAP decline after IVC obstruction was nearly identical in DCM and HCM groups, which makes this possibility less likely.
Right-heart–left-heart interaction is also directly mediated by the septum and shared epicardial fibers. Although much attention has been paid to the role of the abnormal septum for generation of outflow tract obstruction and intracavitary gradients in HCM, the septum may also contribute to altered right-heart–left-heart interdependence. Both the abnormal geometry of the septum (saddle shape40 ) and altered fiber structure may be responsible for enhancement of ventricular interdependence. If the transmission of right-heart–left-heart interaction was delayed because of abnormal chamber and muscle architecture caused by HCM, this could lead to a staircaselike parallel shift of PVRfill. Such delay was not evident in the time required to reduce LV flow after IVC inflow obstruction (series interaction), which was 11.5±4.5 beats for HCM and 9.3±5.4 beats for normal subjects (P=NS). In animal models of pressure-overload LV hypertrophy, direct ventricular interaction appears to be reduced.41 This may apply to LVH-HTN subjects, but the present results suggest that hypertrophy associated with HCM behaves differently.
Flat Diastolic PV Relation with HCM and ΔPdia
A fourth mechanism that could explain the increased ΔPdia in HCM relates to the very flat PVRfill observed in individual beats. This result seems at odds with the widely accepted notion that HCM hearts are exceedingly stiff, yet it is remarkably consistent with the majority of published data of such relations.10 11 12 13 14 32 At least one prior study6 demonstrated a clear, steep diastolic PV relation in HCM, but in that instance, chamber volumes also were profoundly reduced. Importantly, the flat virtually assured an increased If one assumes that each reduction in preload moved the end-diastolic PV point down along its passive relation (EDPVR, as shown in Fig 6⇑), then prior dynamic filling pressures would be similar, which would generate the staircase appearance.
The mechanism for the flat EDPVR observed in HCM patients remains unclear. The most common explanation is that it is due to slow relaxation.18 Although delayed relaxation can elevate early diastolic pressures, at resting heart rates of 70 to 80 bpm and with relaxation times of 50 to 100 ms (typically reported values), this is not to be anticipated beyond early-middiastolic filling. Furthermore, patients with DCM also typically have very abnormal relaxation kinetics42 43 44 (see also Table 1⇑), yet we did not observe a similar flat PVRfill or enhanced ΔPdia in such patients. As noted above, regional heterogeneity of relaxation,34 35 perhaps from fiber disarray and geometric distortion, can result in dyssynchronous wall stretch and contribute to early filling pressure elevation. However, reported stress relaxation time constants are also fairly short (≈60 ms)35 and thus are unlikely to explain behavior observed at midfilling (ie, time of ΔPdia calculation). The present study also suggests that enhanced viscoelasticity is not responsible for the flat PVRfill.
An alternative explanation for the PVRfill “flatness” may relate to marked LV shape changes that occur during early diastole.3 22 Gibson and Brown22 reported that wall stress is comparably low in normal subjects and LVH patients during diastole until more than 90% of end-diastolic circumference is reached, at which point the wall stress in LV hypertrophy rises significantly higher than in normal subjects. If the distal chamber is virtually emptied by end systole, then unfolding of the chamber (much like flower petals) in early diastole could accommodate substantial volumes by pure shape change without increasing the endocardial surface area and thus without stretching the myocardium.3 22 This may account for the fact that there was very little change in LV pressure during early filling in HCM hearts, yielding the shallow PVRfill.
There are several limitations to the present study. Data were obtained in a group of HCM patients who primarily presented later in the course of their disorder, ie, with class III to IV heart failure symptoms. The extent to which they accurately represent a broader group of patients who often suffer from cardiac arrhythmias and chest discomfort is unclear. However, this patient group represented virtually every patient referred to the Johns Hopkins Cardiac Catheterization Laboratory for evaluation of HCM over the past 4 years. The only patients not studied were missed due to unavailability of the investigators; thus, this was not a select group. Second, the role of ventricular interaction was supported primarily by the exclusion of other mechanisms, because it was impossible to directly alter the interaction by pericardiectomy, myectomy, or other procedures in a clinical study of this nature. Finally, the conductance catheter volume method, like all volumetric methods, is subject to error. However, we performed catheter calibration in each individual subject using external standards (thermodilution output and ejection fraction by ventriculography); thus, absolute volumes were no better or worse than those determined from these measures. Furthermore, relative changes in catheter-derived volume were in close agreement with ventriculographic volumes in HCM patients, which supports the validity of the signal.
The diastolic PV behavior of patients with HCM displays a novel abnormality that is not observed in patients with LVH-HTN or in normal or DCM patients. It is characterized by relatively flat diastolic PV relations that shift up and down in parallel with directionally similar changes in right-heart loads. At the very least, these results complicate the interpretation of diastolic PV data in HCM, as well as conclusions regarding the influence of therapies based on analysis on single cardiac cycles. More importantly, they raise the specter that better control of right-heart loading and maneuvers to reduce right-heart–left-heart interaction hold the promise of substantial benefit to patients with this disorder.
This research was supported by an American College of Cardiology/Merck Fellowship Award (Dr Pak), National Public Health Service grants HL-47511 and AG-12249 (Dr Kass), and an American Heart Association Established Investigator Award (Dr Kass). The authors gratefully thank Amit Nussbacher, MD, Siguemituzo Arieˆ, MD, Chun-Peng Liu, MD, and Chih-Tai Ting for allowing us to include some of their patient data for the comparison groups.
Selected Abbreviations and Acronyms
|EDPVR||=||end-diastolic pressure-volume relation|
|ESPVR||=||end-systolic pressure-volume relation|
|IVC||=||inferior vena cava|
|LV||=||left ventricle, left ventricular|
|LVH-HTN||=||left ventricular hypertrophy secondary to chronic hypertension|
|=||difference between left ventricular pressures at middiastolic volume|
|PVRfill||=||dynamic diastolic pressure-volume curves measured during filling|
|RAP||=||right atrial pressure|
|τ||=||relaxation time constant|
|Τ1||=||time constant for apical pressures assessed 40 ms after dP/dtmin|
|Τ2||=||time constant for apical pressures assessed at a later phase that led up to the onset of filling|
- Received August 28, 1995.
- Revision received December 7, 1995.
- Accepted January 2, 1996.
- Copyright © 1996 by American Heart Association
Stewart S, Mason DT, Braunwald E. Impaired rate of left ventricular filling in idiopathic hypertrophic subaortic stenosis and valvular aortic stenosis. Circulation. 1968;37:8-14.
Sanderson JE, Gibson DG, Brown DJ, Goodwin JF. Left ventricular filling in hypertrophic cardiomyopathy: an angiographic study. Br Heart J. 1977;39:661-670.
St John Sutton MG, Tajik AJ, Gibson DG, Brown DJ, Seward JB, Giuliani ER. Echocardiographic assessment of left ventricular filling and septal and posterior wall dynamics in idiopathic hypertrophic subaortic stenosis. Circulation. 1978;57:512-520.
Bonow RO, Rosing DR, Bacharach SL, Green MV, Kent KM, Lipson LC, Maron BJ, Leon MB, Epstein SE. Effects of verapamil on left ventricular systolic function and diastolic filling in patients with hypertrophic cardiomyopathy. Circulation. 1981;64:787-796.
Gwathmey JK, Warren SE, Briggs GM, Copelas L, Feldman MD, Phillips PJ, Callahan M Jr, Schoen FJ, Grossman W, Morgan JP. Diastolic dysfunction in cardiomyopathy. J Clin Invest. 1992;87:1023-1031.
Paulus WJ, Lorell BH, Craig WE, Wynne J, Murgo JP, Grossman W. Comparison of the effects of nitroprusside and nifedipine on diastolic properties in patients with hypertrophic cardiomyopathy: altered left ventricular loading or improved muscle inactivation? J Am Coll Cardiol. 1983;2:879-886.
Hess OM, Murakami T, Krayenbuehl HP. Does verapamil improve left ventricular relaxation in patients with myocardial hypertrophy? Circulation. 1986;74:530-543.
Bonow RO, Ostrow HG, Rosing DR, Cannon RO III, Lipson LC, Maron BJ, Kent KM, Bacharach SL, Green MV. Effects of verapamil on left ventricular systolic and diastolic function in patients with hypertrophic cardiomyopathy: pressure-volume analysis with a nonimaging scintillation probe. Circulation. 1983;68:1062-1073.
Lorell BH, Paulus WJ, Grossman W, Wynne J, Cohn PF. Modification of abnormal left ventricular diastolic properties by nifedipine in patients with hypertrophic cardiomyopathy. Circulation. 1982;65:499-507.
Glantz SA, Misbach GA, Moores WY, Mathey DG, Lekven J, Stowe DF, Parmley WW, Tyberg JV. The pericardium substantially affects the left ventricular diastolic pressure-volume relationship in the dog. Circ Res. 1978;42:433-441.
Slinker BK, Goto Y, LeWinter MM. Direct diastolic ventricular interaction gain measured with sudden hemodynamic transients. Am J Physiol. 1989;256:H567-H573.
Grossman W, Lorell BH, eds. Diastolic Relaxation of the Heart. Boston, Mass: Martinus-Nijhoff; 1988.
Hess OM, Grimm J, Krayenbuehl HP. Diastolic simple elastic and viscoelastic properties of the left ventricle in man. Circulation. 1979;59:1178-1187.
Pouleur H, Karliner JS, LeWinter MM, Covell JW. Diastolic viscous properties of the intact canine left ventricle. Circ Res. 1979;45:410-419.
Gilbert JC, Glantz SA. Determinants of left ventricular filling and of the diastolic pressure-volume relationship. Circ Res. 1989;64:827-852.
Gibson DG, Brown DG. Relation between diastolic left ventricular wall stress and strain in man. Br Heart J. 1974;36:1066-1077.
Kass DA, Midei M, Brinker J, Maughan WL. Influence of coronary occlusion during PTCA on end-systolic and end-diastolic pressure-volume relations in humans. Circulation. 1990;81:447-460.
Kass DA. Clinical evaluation of left heart function by conductance catheter technique. Eur Heart J. 1992;13(suppl E):57-64.
Sharir T, Marmor A, Ting CT, Chen JW, Liu CP, Chang MS, Yin FCP, Kass DA. Validation of a method for noninvasive measurement of central arterial pressure. Hypertension. 1993;21:74-82.
Raff GL, Glantz SA. Volume loading slows left ventricular isovolumic relaxation rate. Circ Res. 1981;48:813-824.
Odake M, Takeuchi M, Takaoka H, Hata K, Hayashi Y, Yokoyama M. Determination of left ventricular volume using a conductance catheter in the diseased human heart. Eur Heart J. 1992;13(suppl E):22-27.
Zile MR, Conrad CH, Gaasch WH, Robinson KG, Bing OH. Preload does not affect relaxation rate in normal, hypoxic, or hypertrophic myocardium. Am J Physiol. 1990;258:H191-H197.
Brutsaert DL, Sys SU. Relaxation and diastole of the heart. Physiol Rev. 1989; 69:1228-1315.
TenCate FJ, Serruys PW, Mey S, Roelandt J. Effects of short-term administration of verapamil on left ventricular relaxation and filling dynamics measured by a combined hemodynamic-ultrasonic technique in patients with hypertrophic cardiomyopathy. Circulation. 1983;68:1274-1279.
Murgo JP, Alter BR, Dorethy JF, Altobelli SA, McGranahan GM Jr. Dynamics of left ventricular ejection in obstructive and nonobstructive hypertrophic cardiomyopathy. J Clin Invest. 1980;66:1369-1382.
Betocchi S, Hess OM, Losi MA, Nonogi H, Krayenbuehl HP. Regional left ventricular mechanics in hypertrophic cardiomyopathy. Circulation. 1993;88:2206-2214.
Hayashida W, Kumada T, Kohno F, Noda M, Ishikawa N, Kojima J, Himura Y, Kawai C. Left ventricular regional relaxation and its nonuniformity in hypertrophic nonobstructive cardiomyopathy. Circulation. 1991;84:1496-1504.
Hess OM, Schneider J, Koch R, Bamert C, Grimm J, Krayenbuehl HP. Diastolic function and myocardial structure in patients with myocardial hypertrophy. Circulation. 1981;63:360-371.
Brodie BR, Grossman W, Mann T, McLaurin L. Effects of sodium nitroprusside on left ventricular diastolic pressure-volume relations. J Clin Invest. 1977;59:59-68.
Hutchins GM, Bulkley BH. Catenoid shape of the interventricular septum: possible cause of idiopathic hypertrophic subaortic stenosis. Circulation. 1978;58:392-397.
Slinker BK, Chagas ACP, Glantz SA. Chronic pressure overload hypertrophy decreases direct ventricular interaction. Am J Physiol. 1987;253(Heart Circ Physiol 22):H347-H357.
Parker JD, Landzberg JS, Bittl JA, Mirsky I, Colucci WS. Effects of β-adrenergic stimulation with dobutamine on isovolumic relaxation in the normal and failing human left ventricle. Circulation. 1991;84:1040-1048.
Hasenfuss G, Reinecke H, Studer R, Meyer M, Pieske B, Holtz J, Holunbarsch C, Posival H, Just H, Drexler H. Relation between myocardial function and expression of sarcoplasmic reticulum Ca2+-ATPase in failing and nonfailing human myocardium. Circ Res. 1994;75:434-442.