INTRODUCTION
According to the American Heart Association, nearly 5 million Americans are living with heart failure, and approximately 550,000 new cases are diagnosed each year. Diastolic heart failure (DHF) contributes to approximately 40% to 50% of congestive heart failure (CHF) patients admitted to hospitals. DHF is characterized by clinical signs and symptoms of heart failure, normal ejection fraction (EF), and evidence of abnormal left ventricular (LV) relaxation, filling, diastolic distensibility, or diastolic stiffness.
Zile et al. studied 63 patients with DHF with cardiac catheterization and echocardiography. Almost all had one or more abnormal indices of diastolic function. A recent population study showed that by comparison with systolic heart failure (SHF), DHF is more likely distinguished by older age, female sex, and a history of hypertension, as well as atrial fibrillation. Although the adjusted 1-year mortality rate was higher for systolic than diastolic dysfunction, the heart failure readmission rate was similar at 30 days (4.5% vs. 4.7% p = 0.66) and 1 year (13.5% vs. 16.1% p = 0.09).
More than 40% of heart failure patients have preserved systolic function, suggesting that diastolic dysfunction may be responsible for the clinical manifestation of heart failure in these patients. Persson et al. studied 312 patients (mean age, 66; EF, 50%; female, 34%) over a median 569 days. Patients with normal diastolic function and mild diastolic dysfunction (abnormal relaxation) had a good prognosis, whereas moderate and severe diastolic dysfunctions (restrictive physiology) were associated with increased cardiovascular morbidity and mortality. Moderate and severe diastolic dysfunctions were the only predictors for cardiovascular death or heart failure hospitalization ( p < 0.003). Apart from prognosis, the ventricular myocardium of DHF differs from that of SHF. DHF is associated with myocyte hypertrophy, higher myofibrillar density, elevated myocyte relative thickness, and increased myofilamentary Ca 2+ sensitivity.
Since the beginning of pacing therapy in 1958, tremendous improvement in technology has produced more sophisticated, efficient, and compact pacemakers. The pacing sites, pacing modes, and baseline heart function influence diastolic heart function, and various authors have reported the effects of pacing on one or more diastolic function parameters. This chapter will discuss and explore this evolving understanding of the effects of pacing therapy on diastolic function and DHF.
PATHOPHYSIOLOGY
Basics of Pacing and Cardiac Resynchronization Therapy
Pacemakers generate an electrical charge through electrodes that produce action potentials in myocardial cells. These action potentials, if above threshold, stimulate surrounding myocardium (capture), and a wave of electrical discharge (depolarization) moves away from the electrode to energize cardiac chambers. Electromechanical coupling follows in the direction of depolarization. A unichamber lead is usually in the right ventricular apex (RVA), from which a wave of depolarization travels posteriorly and toward the cardiac base. Although it affects ventricular contraction, this can be at the cost of ventricular dyssynergy.
Dual-chamber pacemakers have an additional lead in the right atrium. Depending on the mode, the atrial lead will sense intrinsic atrial depolarization and be inhibited as well as activate the ventricular lead at a preset time interval (atrioventricular [AV] delay) if no intrinsic ventricular depolarization is sensed. This allows for AV synchrony and improved diastolic filling of the left ventricle. AV synchrony is achieved by single-chamber atrial rate-responsive (AAIR) or dual-chamber rate-responsive (DDDR) programming.
Cardiac resynchronization therapy (CRT) requires placement of two ventricular leads: one positioned at the RVA (or RV outflow tract [RVOT]) and the other in the posterolateral ventricular wall (via the coronary sinus). CRT has been advocated for patients with New York Heart Association (NYHA) stage III/IV heart failure, left ventricular (LV) dysfunction (EF ≥35%) refractory to optimal drug therapy, and prolonged QRS interval (>120 ms). CRT has resulted in improvement in quality of life, 6-minute hall walk, EF, and reversal of LV remodeling and mortality.
The principal mechanism underlying CRT is that LV dysfunction is often associated with QRS prolongation. Such conduction delay can cause dyssynchronous contraction of the ventricle, resulting in inefficient function. Delay of different myocardial segments of the ventricular wall can be seen on standard two-dimensional echocardiographic imaging, but more sophisticated Doppler modalities have proven useful and predictive of a favorable response to CRT when measured prior to a planned implant. More specifically, tissue velocity imaging has proven useful, as longitudinal velocity timing can be sampled and compared between opposing basal segments of the left ventricle, resulting in a measure of intraventricular dyssynchrony. This provides a more precise parameter for identifying dyssynchronous contraction than QRS morphology or duration.
Diastolic Dysfunction
Diastolic dysfunction is associated with functional abnormalities of diastolic relaxation, filling, or distensibility, while DHF is associated with signs and symptoms of heart failure, normal EF, and abnormalities of one or more parameters of diastolic function.
It is important to understand that physiologically, diastole is considered to begin from the period of reduced ventricular ejection, encompassing isovolumic relaxation and filling phases of ventricles. It begins when calcium ions are taken up into the sarcoplasmic reticulum so that myocyte relaxation dominates over contraction, and LV pressure starts to fall. Performance of the ventricle as a pump is highly dependent on diastolic filling of the ventricle (preload). Frank-Starling’s law of the heart dictates that stroke volume is related to end diastolic volume. The greater the initial LV volume, the greater the peak pressure reached and the faster the rate of relaxation (lusitropic effect).
There are two major disease processes of diastole that affect filling or distensibility: abnormal relaxation and increase in myocardial stiffness. Abnormal relaxation is usually affected early in the disease process and can be measured by cardiac catheterization or echocardiographic techniques. An increase in myocardial stiffness represents advanced diastolic dysfunction and is usually associated with an increase in LV end diastolic pressure. LV stiffness refers to a change in diastolic LV pressure relative to diastolic LV volume (dp/dv) and equals the slope of the diastolic pressure-volume relation (P-VR). It is also inversely proportional to deceleration time (DT). Thus, a rapid DT indicates elevated LV early diastolic chamber stiffness.
Echocardiography has emerged as a simple and reliable non-invasive method of evaluating systolic and diastolic function. Detailed description of the role of echocardiography in the classification of diastolic dysfunction is discussed in Chapters 6 and Chapter 10 , Chapter 11 , Chapter 12 .
CLINICAL RELEVANCE
Diastolic dysfunction is a feature representing abnormal diastole; however, DHF as an entity needs only one or more features of diastolic dysfunction. The currently accepted definition of DHF requires signs and symptoms of heart failure and normal EF (≥50%). Historically, EF has been regarded as a vital component of normal systolic function, but other markers (e.g., ventricular systolic synchrony, LV end systolic volume [LVESV], dp/dt) are also important. Interestingly, DHF has been shown to have many features typically regarded as markers of systolic dysfunction.
Recent studies have dispelled the commonly held belief that patients with DHF have only abnormalities of relaxation. Synchronicity studies, especially based on tissue Doppler imaging (TDI), have shown that systolic asynchrony is a relatively common finding (33%-39%) in DHF. Yu et al. reported the prevalence of isolated diastolic asynchrony in 22%, isolated systolic asynchrony in 25%, and coexisting diastolic and systolic asynchrony in 14% of patients with DHF. The corresponding prevalence in the SHF group was 17%, 31%, and 26%, respectively ( Figs. 29-1 and 29-2 ).
Biventricular (biV) pacing produces improvement in LV function and symptomatic status by reducing systolic dyssynchrony in patients with SHF. Patients with CRT have consistently shown an improvement in many echocardiographic systolic as well as diastolic parameters. Though studies are brimming with reports of improvement in systolic parameters (e.g., EF, LVESV, time to peak systolic myocardial velocities) in CRT patients, a similar demonstration for diastolic parameters has been less consistent. A recent wave of articles has explored the subtle but significant role of diastolic function in heart failure. This chapter attempts to uncover current evidence for the role and interaction of pacing, particularly CRT, in influencing diastolic dysfunction and DHF.
Overall Effects of Pacing on Cardiac Function
Pacing has multiple effects on ventricular function depending on pacing site, pacing configuration, and disease status. The effect may be positive, as seen with atrial pacing in patients with sinus node dysfunction, where it results in an increase in heart rate and cardiac output. However, the effect may be negative, as seen with right ventricular (RV) apical pacing in patients with bradycardic indications and normal baseline LV function, which is thereby impaired. A detailed overview of the effects of pacing on diastolic cardiac function (as it pertains to diastolic physiology) is given in Tables 29-1 and 29-2 .
| ATRIAL PACING | RIGHT VENTRICULAR PACING | BIVENTRICULAR PACING | LEFT VENTRICULAR PACING |
|---|---|---|---|
|
|
|
|
|
|
|
|
| DYSSNCHRONOUS PACING, RVA PACING | SYNCHRONOUS PACING, BIV PACING | |
|---|---|---|
| Cellular Level |
|
|
| Tissue Level |
|
|
| Organ Level |
|
|
| Body Level |
|
|
Pacing and Diastolic Function
Hay et al. studied short-term effects of RV pacing (RVA and RVOT), LV free wall pacing, and biV pacing in patients with heart failure (EF 14%-30%). Systolic function improved in all modes, but more so in the biV group. However, only the biV group showed improved diastolic function (isovolumic relaxation and diastolic filling times). The authors suggested that single-site LV pacing, in contrast to biV pacing, might induce some intraventricular dyssynchrony that could impact relaxation.
Myocardial relaxation is influenced by chamber load and homogeneity of activation. Auricchio et al. studied 27 CHF patients with conduction disorders. RV, LV, and biV pacing were compared. BiV and LV pacing increased LV (+)dp/dt and aortic pulse pressure more than did RV pacing. LV diastolic performances also changed in all pacing configurations, but the changes were small and inconsistent (LV end diastolic pressure [LVEDP] decreased and absolute value of LV (−)dp/dt also decreased, indicating slower relaxation). The important variables predicting outcome were pacing site, appropriate AV delay, and prolonged QRS width.
Simantirakis et al. studied 12 patients with AV node ablation (due to atrial fibrillation) but with normal LV systolic function. LV-based pacing (LV free wall or biV) was compared with RV apical pacing. LV-based pacing improved indices of LV systolic function more than did RV pacing. Indices of LV diastolic filling (EDP, EDV) were better during LV-based pacing, whereas LV diastolic function indices such as (−)dp/dtmax, Tau [τ], and passive diastolic chamber stiffness did not change significantly. This reflects a more complex influence of preactivation and atrial contraction on chamber load present during sinus rhythm.
Effect of Cardiac Resynchronization Therapy on Diastolic Function
CRT has been shown to improve NYHA functional class, quality of life indicators, and EF. It is also associated with improved LV geometry (reduction in LV end diastolic and systolic diameter as well as volume) and reduction in mitral regurgitation. Patients who respond to CRT are often those with positive structural and functional LV remodeling.
The mechanisms of benefit of CRT are several, including synchronization of systolic function as well as enhancement of diastolic function. The literature is replete with studies supporting the beneficial effects of CRT on systolic function but is less prolific with respect to diastolic function. Table 29-3 provides a list of effects of CRT on diastolic parameters.
| DIASTOLIC DYSFUNCTION PARAMETERS | MARKERS | INFLUENCE OF CRT | REFERENCES |
|---|---|---|---|
| Velocities | Mitral E velocity | Yes | , , |
| Mitral A velocity | No | ||
| Flow propagation velocity | No | ||
| Mean Em velocity (Global Em) | Yes | ||
| Mean Am velocity | No | ||
| Time | Deceleration time | Yes | , |
| Diastolic filling time | Yes | , , | |
| Isovolumic relaxation time | Yes | , | |
| Tau, time constant of relaxation | Yes | ||
| Negative dp/dt | Yes | ||
| Q-Em time diffierence (SD in Te) | Yes | , | |
| Te-difierence between 2 wall segments | No | ||
| Interventricular mechanincal delay | Yes | ||
| Ratios | Mitral E/A ratio | Yes | , |
| Mitral E/Em ratio | Yes | ||
| Mitral E/Vp ratio | Yes | , | |
| PVs/Pvd ratio | Yes | ||
| Deceleration slope | Yes | ||
| Myocardial performance index | Yes | , | |
| Volume | LVEDV | Yes | |
| LVESV | Yes | , , |
CRT does appear to have a beneficial effect on diastolic properties of the failing heart. CRT enhances diastolic filling patterns in patients with systolic dysfunction. In a recent study, 23 patients were evaluated at 1 week prior as well as 1 and 6 months after implantation. Significant clinical improvement was noted in all patients. Compared with baseline, the ratio of early to late peak velocities (E/A) decreased significantly (1.5 to 0.8) at 6 months. The pulmonary systolic flow to diastolic flow ratio (PVs/PVd) increased from 0.9 to 1.3 at 6 months, and the ratio of early peak velocity to LV flow propagation velocity (E/Vp) decreased from 2.7 to 2 at 1 month, then to 1.9 at 6 months. Patients who demonstrated improvement in EF of more than 25% were designated responders ( N = 17, 74%). In these patients, the E wave and pulmonary venous (PV) atrial reversal velocity decreased, E-wave deceleration time increased, and the E/Vp ratio improved significantly (all diastolic parameters), whereas in the nonresponder population, changes in LV diastolic parameters remained insignificant.
In the Multicenter Insync Randomized Clinical Evaluation (MIRACLE) study, 323 patients were randomized into two groups: control ( n = 151) and CRT ( n = 172). Echocardiographic parameters were measured at baseline, then 3 and 6 months after intervention. Significant reduction in LVEDV and LVESV occurred at 3 months in the CRT group compared with the control group. The reduction in LVEDV and LVESV continued between 3 and 6 months. There were significant increases in EF of 2.6% at 3 months and 3.6% at 6 months in the CRT group. There was also a significant increase in deceleration slope and deceleration time of the E wave (during rapid filling) at 3 and 6 months in the CRT group. The maximum E-wave velocity and myocardial performance index decreased significantly at follow-up, consistent with improved ventricular function. There was no significant change in peak A-wave velocity, E/A wave velocity ratio, and isovolumic relaxation time (IVRT) in the CRT or the control group.
Of note, optimization of the AV delay and synchronous biV pacing resulted in prolongation of the duration of LV filling (diastolic filling time), separation of the rapid filling phase from atrial systolic contraction, concomitant shortening of interventricular mechanical delay (IVMD), and simultaneous ventricular depolarization that coordinated contraction and relaxation. There was also a significant improvement in the diastolic deceleration slope and deceleration time of the rapid ventricular filling wave velocity (E wave), leading to better diastolic filling.
In CRT patients, there was a significant improvement in the myocardial performance index (Tei index) at 3 months that continued between 6 and 12 months compared with the control group. The Tei index is a parameter representing both systolic and diastolic performance and is the summation of isovolumic contraction time and IVRT divided by the ejection time. CRT failed, however, to show significant remodeling in patients with SHF and restrictive LV diastolic filling (characterized by a short deceleration time and peak E-wave velocity >1 m/sec).
Reduction in ventricular systolic dyssynchrony is one of the major mechanisms by which CRT provides benefit. Various methods of measuring synchrony are being investigated. TDI has provided an easy-to-use tool to quantify dyssynchrony both in systole and in diastole. Echocardiographic studies have concentrated on the basal and middle segments of the heart, ignoring the apex, because of limitations of TDI when applied to the apical segments. Dyssynchronous contraction and relaxation have been measured as standard deviations of the time to peak segmental myocardial velocity of 12 myocardial segments, the maximum time difference between opposing myocardial segments, or the average of peak myocardial velocities of myocardial segments.
Waggoner et al. studied diastolic function in patients with severe SHF after CRT. Diastolic LV filling indices (mitral E/A, E/Em, E/filling pressure [FP] ratios), as well as diastolic synchrony (Q-Em velocity difference) changed favorably at 4 months, especially in the non-ischemic cardiomyopathy group. The cohort with ischemic cardiomyopathy had significant improvement in only Em global velocity (a measure of diastolic asynchrony) at 4 months. Diastolic filling indices were also predictive of long-term event rates (heart failure hospitalization). In another study, by Yu et al., in which patients were divided between responders (45%) and nonresponders (55%) based on LV reverse remodeling (reduction of LVESV ≥15%) after CRT, responders had significant reduction in E/Em ratio (a marker of LV FP; 30 to 22, p < 0.05). Responders also had significant prolongation in LV filling time (381 to 441 msec, p < 0.05). Patients with baseline abnormal relaxation patterns had greater reduction of LV volumes than those exhibiting a pseudonormal pattern (29% vs. 11%, p < 0.05). Another interesting finding was that nonresponders differed significantly from responders with respect to baseline E/A ratio, IVRT, mean Em, and mean Am ( Table 29-4 ). However, in this uncontrolled trial, systolic (but not diastolic) dyssynchrony was a predictor of favorable reduction in LVESV. As yet, there is no indisputable evidence that CRT improves either systolic or diastolic function in DHF. It is hoped that recognition of diastolic dyssynchrony and subsequent intervention may improve outcome in patients with heart failure.
