CRT programming during exercise

Generalities

In a resynchronized patient, the main objectives of the programming during exercise are: 1) to maintain a permanent and effective biventricular capture for high heart rates, 2) to ensure a good contribution of atrial systole to cardiac output, and 3) to allow for appropriate heart rate acceleration which is the fundamental adaptive mechanism of the cardiac output during exercise, in particular in heart failure patients.

Checking for the maintenance of a permanent biventricular capture during exercise must be part of the standard CRT patient assessment. The recording of episodes of ventricular sensing occurring at high sinus rate in the device memory suggests the loss of LV capture at exercise. There are various causes that can lead to a loss of biventricular pacing during exercise: 1) atrial undersensing, 2) frequent PVCs, 3) atrial or ventricular arrhythmias, 4) a shortening of the intrinsic PR interval below the programmed AV delay, 5) a maximum tracking rate programmed too low for the patient’s capacity. In patients with permanent atrial fibrillation and no His bundle ablation, the presence of an effective capture at rest does not guarantee the presence of a LV capture during exercise, when a competition between the improvement of the intrinsic atrioventricular conduction and the acceleration of the pacing rate dictated by the rate response algorithm may occur.

Different kinds of stress test can be performed in a resynchronized patient. The ideal is to have a simultaneous recording of the surface electrocardiogram and intracardiac electrograms (with the programmer) to detect in real time any dysfunction and correct it appropriately. The real-time analysis of the tracings during exercise is now facilitated by the development of wireless technology and the possibility of remote telemetric interrogation of the device.

• Traditional stress tests: they allow for evaluating the overall behavior of the device and the sensor during exercise. It seems that adaptation of heart rate during exercise is closer to that observed in real life with a treadmill than with a cycle ergometer. Indeed, the bicycle mobilizes essentially the muscles of the lower limbs, far from the implantation site of the device. Therefore, the activation of the sensor (accelerometer) is delayed. At equal workload, the heart rate acceleration is greater on a treadmill as compared with the cycle ergometer.
• Standardized tests in daily life: a large number of resynchronized patients are relatively old and have a limited capacity for everyday life efforts (walking, toilet, cooking, household...). In these patients, it is essential to ensure the good function of the device during everyday life stresses. The stress test type used to evaluate device function must be adapted to these specificities. Different types of exercise under electrocardiographic monitoring can be performed: walking, climbing stairs, steps…

Rate-Responsive Pacing

Some patients present with chronotropic incompetence characterized by a heart rate that does not adapt to their level of physical activity. This inability to increase heart rate during exercise may be associated with various symptoms like a shortness of breath, fatigue or a reduced exercise capacity… In resynchronized patients with chronotropic incompetence (defined by the inability to achieve 85% of the theoretical maximum heart rate), programming a rate responsive pacing aims to ensure an increase of cardiac output according to the metabolic need imposed by ongoing effort. Programming a rate adaptive pacing must be systematic in the presence of severe chronotropic incompetence (inability to achieve 70% of the theoretical maximum heart rate).

The functioning of the different sensors has been detailed in the manuscript about standard pacemakers.

Maximal tracking rate

The operating function of the device when the heart rate exceeds the maximal tracking rate depends on the quality of the atrio-ventricular conduction.

In a resynchronized patient with a complete AV block and a preserved chronotropic function, when the sinus rate accelerates and exceeds the maximal tracking rate, ventricular stimulation at the end of the programmed AV Delay would be associated with a heart rate above the maximum value programmed, which is impossible. The ventricular rate can no longer follow the atrial rate in a 1/1 ratio. To overcome this limitation, the device extends the AV delay and a Wenckebach phenomenon occurs. As the sinus rate increases beyond the maximal tracking rate, the ventricular pacing rate remains at the maximal tracking value and the sensed AV delay is prolonged at each cycle of stimulation. After several pacing cycles, an atrial sensed event falls inside the PVARP and is not coupled with a ventricular pacing, resulting in a missing ventricle. The next P-wave falls outside the refractory period and initiates a programmed AV delay. This pattern repeats itself as long as the sinus rate remains higher than the maximum maximal tracking rate programmed. Missing beats occurs less frequently when the sinus rate is only slightly higher than the maximal tracking rate and more frequently when the sinus rate exceeds largely the maximal tracking rate. When the sinus rate falls below the maximum frequency, the AV synchronicity (1:1 ratio) is restored. The Wenckebach behavior can be characterized by the rate at which the first missing beat will occur and by the ratio of detected atrial beats and paced ventricular beats (for example 8:7, 7:6, 6:5 or 3:2). If the increase in the heart rate reaches the 2:1 block point, an important drop of heart rate occurs with a Atrium / Ventricle ratio of 2:1.

In a resynchronized patient with a good chronotropic function and a preserved AV conduction, when the sinus rate accelerates and exceeds the maximal tracking rate, the programmed AV delay lengthens progressively favoring the appearance of a fusion with the spontaneously conducted ventricles. At higher rates, the ventricular sensing inhibits biventricular pacing. Four elements characterize this operation:

• the intervals between 2 VS are shorter than those corresponding to the maximal tracking rate;
• for a given cycle, the PR interval (AS-VS) is shorter than the theoretical AS-VP;
• there is no P-wave undersensing or P-wave falling inside the PVARP as in a typical Wenckebach period in a patient with atrioventricular block;
• biventricular pacing restarts only when the atrial rate falls below the maximal tracking rate. 

Therefore, it seems logical to program a high maximal tracking rate to avoid a Wenckebach behavior or the loss of biventricular pacing. In order to limit the increase in heart rate during exercise, it does not seem logical to limit the maximal tracking rate but rather to optimize the medical therapy with rate control agents. Programming a maximal tracking rate too low in coronary patient probably does not have any protective value… In contrast it may rather favor the loss of biventricular capture or the occurrence of intermittent blocked P-waves that can promote disabling symptoms by increasing myocardial oxygen consumption and inducing a fall in cardiac output.

AV-delay optimization during exercise

It is possible to program different AV delay at rest and during exercise. This programming presents three goals: 1) hemodynamic optimization with the research of the AV delay at exercise allowing for the best exercise capacity 2) maintenance of a biventricular pacing at exertion (programmed AV delay shorter than the PR) 3) maintaining a regular atrioventricular synchronicity for high sinus rates (avoid to exceed the point of 2:1).

The requirements differ depending on the quality of the atrioventricular conduction. In a patient with a permanent complete atrioventricular block, there is no risk of resumption of the atrioventricular conduction at exercise and therefore no risk of loss of biventricular pacing. In contrast, the AV delay must be programmed short during exercise to enable a 1/1 AV relationship (and avoid a sudden drop in heart rate when the atrial rate exceeds the 2:1 block point) and to allow for an optimal hemodynamic response. In a patient with a preserved atrioventricular conduction, there is no risk of rate drop after the occurrence of the 2:1 block point. However, the improvement of the intrinsic atrio-ventricular conduction can promote the loss of biventricular capture.

Optimizing the AV delay in order to improve the patient hemodynamic is difficult at rest… It is even more difficult during exercise. There is currently no method of reference for this optimization and the results in the literature are contrasted and conflicting. In a patient with a healthy heart but conduction disorders, the positive effect of shortening the AV delay during exercise has clearly been demonstrated. Various studies have analyzed the dynamics of the optimal AV delay in resynchronized patients with sometimes different results leading to conflicting recommendations: based on patients and studies, the optimal AV delay could be shorter, identical but also sometimes longer during exercise than at rest. Considering the physiological characteristics of the atrioventricular node in healthy patients, these disparate results are unexpected. Some methodological limitations may have contributed to these conflicting results. However, patients with heart failure and left bundle branch block represents an heterogeneous group of patient, which could have lead to a variable impact of the stress test on the intra-atrial conduction, atrioventricular or intraventricular times. In patients with ischemic heart disease particularly, the impact of exercise on heart rate, load conditions and autonomic tone may have a variable effects from patient to patient on the conductive and contractile tissue.

It seems therefore difficult to establish firm and final recommendations for the hemodynamic optimization of the AV delay at exercise. From a practical point of view, it seems inconceivable to consider an individualized dynamic optimization based on the echocardiographic pattern or other invasive or non-invasive technique. An automatic optimization at rest and during exercise of the AV delay by the device itself may be the solution but requires the development of specific algorithms and their clinical validation with the demonstration of clinical benefit provided in terms of response to the CRT and / or effort capacity.

In practice, what are the possibilities with current devices? Two options are possible in most devices: maintaining a fixed AV (sensed or paced) delay both at rest and during exercise, or programming an adaptive AV delay which allows for a linear shortening of the AV delay with the increasing heart rate. Even if the demonstration of the hemodynamic benefit of dynamic AV delay remains controversial in resynchronized patients, programming a dynamic AV delay has certain advantages and is frequently proposed in clinical practice: 1) among non-dependent patients with a shortening of the PR interval at exercise, a dynamic AV delay will maintain biventricular pacing and avoid the PR interval to become shorter than the AV delay; 2) among pacemaker dependent patients, programming a dynamic AV delay and a PVARP Auto allows to postpone the occurrence of 2:1 block point beyond the maximum capacity of the patient.

Atrial sensitivity

Maintenance of a synchronous 1:1 atrioventricular relationship during exercise requires a perfect detection of the atrial activity. The exertion and the increased respiratory movements are often associated with an alteration of the atrial detection. The diagnosis relies on the absence of AR signals (2:1 block) and the absence of a marker despite the presence of an atrial signal. It is therefore necessary in these patients to increase the sensitivity and the margin as compared with the detection threshold measured at rest.

Premature ventricular contraction in patients with a long PR interval

At exercise, the acceleration of the sinus rate in combination with the occurrence of PVC in a patient with long PR interval may result in prolonged loss of biventricular pacing. Indeed, the P-wave following a PVC can fall inside the PVARP which is extended to 400 ms (marker AR); no AV delay will thus be triggered and explain the absence of biventricular pacing; the following spontaneous ventricle will also be considered as a PVC and the PVARP will be extended. The longer is the PR interval, the faster is the atrial rate, the more important are the chances to see this phenomenon maintained (succession of cycles AR-VS).