Echocardiographic AV-interval optimization in patients with reduced left ventricular function

Since introduction of the DDD pacemaker in the early 1980s, researchers have repeatedly attempted to optimize the atrioventricular (AV) interval, for the purpose of maximizing patient hemodynamic performance. Cannon waves may be induced by programming excessively short AV intervals, and diastolic mitral regurgitation may occur with excessively long programmed AV intervals. The AV interval is considered optimal (AV_opt) if it allows maximum cardiac output.

The duration of the optimal AV interval varies throughout a wide range among individuals, primarily the result of appreciable differences in interatrial conduction [1‐4].

An extensive variety of techniques has been employed to optimize AV delay, including acquisition and analysis of essential hemodynamic parameters by means of aortic-valve Doppler signals, impedance cardiography [9‐11], Swan-Ganz catheterization [12‐15], and especially the stroke volume [5‐8]. Leman et al. [16] have demonstrated that it is also possible to utilize measurement of left ventricular ejection fraction and stroke volume by myocardial thallium scintigraphy as a means of AV interval optimization. A further possibility involves detection of left atrial depolarization by an esophageal electrode recording [17, 18]. During recent years, the use of Doppler echocardiography in conjunction with the mitral valve inflow profile has been investigated as means of AV interval optimization: i.e., Ritter's method [19]. Previous investigations have evaluated Ritter's method in patients with normal left ventricular ejection fractions. During recent years, cardiac resynchronization therapy (CRT) has increasingly gained in significance for patients with chronic heart failure (CHF) [20]. In cases without ventricular desynchronicity, normal DDD pacemakers (or ICDs with DDD pacemaker function) will in future continue to be implanted in patients with reduced left ventricular ejection fraction. The goal of our study was accordingly to apply Ritter's method – until now validated only for patients with normal EF – for patients with reduced left ventricular ejection fraction (EF < 35%).

We studied 20 DDD pacemaker patients within the context of in-office follow-up. Table 1 shows the baseline characteristics and Table 2, the inclusion criteria. We classified patients into two groups, according to left ventricular ejection fraction results obtained by echocardiography. Group 1 consisted of 10 patients with normal left ventricular ejection fraction, or with moderately reduced EF (EF > 35%). Group 2 comprised 10 patients with appreciably reduced left ventricular ejection fraction (EF < 35%).

We performed ejection fraction analysis by RNV and Ritter's method to achieve AV optimization, for 5 AV intervals in the range of 100 to 250 ms. We performed all measurements within 15 minutes of AV interval programming for every patient. All patients were permanently stimulated in the right atrium and right ventricle (binodal disease). Patient heart rate remained constant during the measurement period at programmed pacemaker lower rate (60 – 70 beats/min).

A number of studies have documented the importance of AV synchronization for maximizing the left ventricular ejection fraction in pacemaker patients [21‐24]. Despite CRT, the implantation of a DDD pacemaker (or ICD with DDD-pacemaker function) is still justified for patients with reduced left ventricular function and a lack of ventricular asynchrony.

The goal of our study was accordingly to apply Ritter's method for patients with reduced left ventricular ejection fraction. In every subject of Groups 1 and 2, it was possible on the basis of the left ventricular ejection fraction to define the optimal AV delay by means of RNV. The method of AV delay optimization by RNV has been previously verified [16]. The cost and complexity of this method, however, have hindered its extensive clinical application. On the basis of minimal inter- and intraobserver variability this method is nevertheless very well suited as a reference method.

Our application of Ritter's method enabled definition of the optimal AV interval for all patients. We further determined that Ritter's method can be reliably employed even in cases of reduced left ventricular systolic function. The AV interval calculated by Ritter's method correlated well with data obtained by RNV: both for normal (with intra-class coefficient of 0.8965) as well as for reduced left ventricular EF (intra-class coefficient of 0.9228).

Since Ritter's initial publication in 1994, AV interval optimization on the basis of the mitral valve inflow profile has been reported in one additional study [19]. In 1997 Kindermann et al. compared results calculated from Ritter's formula with those obtained from impedance cardiography [10]. This study established a high degree of correlation between the results for the optimal AV interval determined by the two different methods. The mean deviation in optimal AV interval between the results from Ritter's formula and determination of stroke volume by impedance cardiography was ± 26 ms for the atrial-triggering mode, and ± 30 ms for the AV sequential mode. Kindermann et al. criticized the fact that it is possible to apply Ritter's method only for patients with ventricular stimulation.

In comparison to time-consuming and expensive RNV, and AV-interval optimization by Swan-Ganz catheterization (with the associated risks of an invasive procedure), Ritter's method offers the following advantages: it is non-invasive and can be quickly performed (approx. 5 min.). It does not require long years of echo experience, and it is cost-effective. Even with patients not readily amenable to sonographic detection, the mitral valve inflow profile is almost always qualitatively satisfactory enough to allow application of Ritter's method. The only noteworthy disadvantage of this method is the necessity for continuous ventricular stimulation: which means that it can be used only for patients with a complete AV block. Patients with only intermittent high-grade AV blocks are accordingly not suited for Ritter's method.

In our patients, the mean optimal AV interval in Group 1 (EF > 35%) was 190 ± 28.5 ms. In comparison, the optimal AV interval among the patients with chronic heart failure in Group 2 was 180 ± 35 ms. Data in the literature are not consistent on the duration of the optimal AV delay. Kindermann [10] considers AV_opt = 88 ms ± 35 ms with atrial triggering, and AV_opt = 143 ms ± 41 ms for the AV sequential mode. Knorre [18] has determined AV_opt = 100.5 ± 27.8 ms for atrial triggering, and AV_opt = 169 ± 24.5 ms for the AV sequential mode. Haskel [5] has established the best AV interval to be 150 ms. Janosik [6] considers AV_opt = 144 ± 48 ms with atrial triggering, and AV_opt = 176 ± 44 ms for the AV sequential mode. Ishikawa [15] has determined AV_opt = 161 ± 26 ms.

Our results on the length of the optimal AV delay lie within the range found in the literature. The variance in data observed in some cases emphasizes the highly individual nature of the optimal AV delay: indeed, it results from the interatrial conduction period specific to each patient, and the potential delays induced by pacing versus intrinsic depolarization and conduction in a given patient [1‐4].

As a result, the mean optimal AV intervals determined by us cannot be applied to other patient cohorts with the same basic disease. Individual programming of the AV interval is therefore necessary.

In summary, our findings confirm that Ritter's method can be reliably applied for patients with normal and with reduced left ventricular pump function. The only prerequisite is a continuous ventricular stimulation.