Cardiac pacing for optimal intra-left ventricular resynchronization

[0001] This disclosure relates to implantable medical devices and more particularly to implantable cardiac pacemakers used to treat ventricular dysynchrony.

[0002] Heart failure affects approximately 5 million people in the United States. Many moderate to severe heart failure patients may also have a condition in which the two lower chambers of the heart (known as the left and right ventricles) are not beating together as they do normally. In medical terms, this condition is called “ventricular dysynchrony.” Ventricular dysynchrony disturbs the synchronous beating of the heart, and as a result the heart does not pump blood adequately to meet the needs of the body. More specifically, ventricular dysynchrony typically results from intraventricular conduction delays (IVCD) that disturb the synchronous beating of the ventricles. Typically, the IVCD has a left bundle branch block (LBBB) morphology.

[0003] One previous method for optimizing cardiac pacing for intra-left ventricular resynchronization involves modifying pacing to reduce QRS duration. Even though QRS duration can be reduced, reduction in QRS duration does not guarantee optimal hemodynamic cardiac operation. Kass et al., Improved left ventricular mechanics from acute VDD pacing in patients with dilated cardiomyopathy and ventricular conduction delay, Circulation 1999; 99:1567–73.

[0004] Another previous method for optimizing cardiac pacing for intra-left ventricular resynchronization involves modifying pacing to improve cardiac ventricle filling. Typically cardiac ventricle filling is measured by echocardiography. Although cardiac filling is one aspect of hemodynamics, tests have shown that optimal hemodynamic cardiac operation by pacing does not result from better cardiac filling. Auricchio et al., Cardiac resynchronization therapy restores optimal atrioventricular mechanical timing in heart failure patients with ventricular conduction delay, J. Am Coll Cardiol 2002; 39:1163–9.

[0005] One therapy to treat left ventricle dysynchrony is Cardiac Resynchronization Therapy (CRT). CRT is a new, proven treatment for selected patients with heart failure-induced conduction disturbances and ventricular dysynchrony. When used in combination with stable, optimal medical therapy, CRT is designed to reduce symptoms by restoring the mechanical sequence of ventricular activation. Cardiac resynchronization therapy (CRT) provides atrial-synchronized, biventricular pacing using standard pacing technology combined with a special third lead which is implanted via the coronary sinus and positioned in a cardiac vein to sense and pace the left ventricle. Following a sensed atrial contraction, both ventricles are stimulated to contract more synchronous. The resulting resynchronization of ventricular contraction reduces mitral regurgitation and optimizes left ventricular filling, thereby improving cardiac function.

[0006] Cardiac pacing for optimal intra-left ventricular resynchronization is performed as follows. Early paced inter-ventricular asynchrony is determined during ventricular pacing. Baseline inter-ventricular asynchrony is determined without pacing. Average inter-ventricular asynchrony is calculated by averaging the early paced inter-ventricular asynchrony and the baseline inter-ventricular asynchrony. Atrio-ventricular delay and ventricular-ventricular delay are adjusted during ventricular pacing to yield the average inter-ventricular asynchrony for optimal intra-left ventricular resynchronization. The elements above can be configured in software contained in an implantable medical device or embodied as a computer software product that includes a medium readable by a processor.

[0021] FIGS. 1 and 2 show an environmental view of an Implantable Cardiac Device (ICD) for optimal intra-left ventricular resynchronization. The ICD can be any ICD capable pacing both the right ventricle and the left ventricle known as bi-ventricular pacing. Implantable Cardiac Devices suitable for bi-ventricular pacing include certain pacemakers, cardioverters, and defibrillators configured for bi-ventricular pacing. For example, the ICD and be an InSync® III Model 8040 pacemaker or an InSync® Marquis cardioverter/defibrillator using two standard right heart electrical leads (in the right atrium and right ventricle) and one left-heart electrical lead such as an Attain™ Model 2187, 2188 or 4193 also available from Medtronic, Inc. in Minneapolis, Minn. USA. The left heart electrical lead is typically implanted transvenously via the coronary sinus and positioned in a cardiac vein to pace the left ventricle.

[0022] FIG. 3 shows a block diagram of an implantable cardiac device 20 for optimal intra-left ventricular resynchronization. The cardiac pacemaker 20 comprises a housing 26, a controller 28, memory 30, pacing electronics 32, sensing electronics 34, a first electrical lead 36, a second electrical lead 38, and software 40. The housing 26 has a power supply 42 carried in the housing 26 and a feedthrough 44. The controller 28 is carried in the housing 26 and coupled to the power supply 42. Memory 30 is coupled to the controller 28. The pacing electronics 32 are coupled to the controller 28 and the feedthrough 44. The sensing electronics 34 are coupled to the controller 28 and the feedthrough 44.

[0023] The first electrical lead 36 is coupled to the feedthrough 44 and configured for positioning in the right ventricle 22. The first electrical lead 36 has at least a first electrode for pacing the right ventricle 22, and the first electrical lead 36 can also have additional electrodes for functions such as pacing the right atrium. The first electrode can also serve as a sensing electrode to provide sensing signals of right ventricle 22 timing to the sensing electronics 34.

[0024] The second electrical lead 38 is coupled to the feedthrough 44 and configured for positioning in the left ventricle 24. The second electrical lead 38 has at least a second electrode for pacing the left ventricle 24, and the second electrical lead 38 can also have additional electrodes for functions such as pacing the left atrium. The second electrode can also serve as a sensing electrode to provide sensing signals of left ventrical 24 timing to the sensing electronics 34.

[0025] In addition to sensors placed on the first electrical lead 36 and second electrical lead 38, a variety of other electrical or mechanical sensors can be used to sense asynchronies such as a vectorcardiogram, bodysurface mapping, echocardiography, heartsounds, tissue Doppler Imaging, and the like.

[0026] Software 40 is stored in memory 30 that contains a first sequence of instructions 46, a second sequence of instructions 48, a third sequence of instructions 50, and a fourth sequence of instructions 52. The software 40 can be configured in the implantable medical device 20 or embodied as a computer software product that includes a medium readable by a processor. The first sequence of instructions 46 when executed by the controller 28, causes the controller 28 to receive early paced inter-ventricular asynchrony data during ventricular pacing through the first electrical lead 36 and sensing electronics 34. The first sequence of instructions 46 serves as a means for determining paced inter-ventricular asynchrony during ventricular pacing. The second sequence of instruction 48 when executed by the controller 28, causes the controller 28 to receive baseline inter-ventricular asynchrony without pacing data through the first electrical lead 36, the second electrical lead 38, and sensing electronics 34. The second sequence of instructions 48 serves as a means for determining baseline inter-ventricular asynchrony without pacing. The third sequence of instruction 50 when executed by the controller 28, causes the controller 28 to calculate average inter-ventricular asynchrony by averaging the early paced inter-ventricular asynchrony data and the baseline inter-ventricular asynchrony data. The third sequence of instructions 50 serves as a means for calculating average inter-ventricular asynchrony by averaging the paced inter-ventricular asynchrony and the baseline inter-ventricular asynchrony. The forth sequence of instructions 52 when executed by the controller 28, causes the controller 28 to adjust atrio-ventricular delay and ventricular-ventricular delay during ventricular pacing to yield the average inter-ventricular asynchrony for optimal intra-left ventricular resynchronization. The forth sequence of instructions 52 serves as a means for adjusting atrio-ventricular delay and ventricle-ventricle delay during ventricular pacing to yield the average inter-ventricular asynchrony for optimal intra-left ventricular resynchronization.

[0027] FIG. 4 shows a flowchart of a method for cardiac pacing for optimal intra-left ventricular resynchronization. The implantable cardiac device 20 such as a pacemaker operates by performing a method for optimal intra-left ventricular resynchronization that comprises the elements of determining early paced inter-ventricular asynchrony 54, determining baseline inter-ventricular asynchrony 56, calculating average inter-ventricular asynchrony 58, and adjusting delay during ventricular pacing 60. The implantable cardiac device 20 can perform intra-left ventricular resynchronization optimization automatically or through the control of a clinician. The early paced inter-ventricular asynchrony is determined 54 during ventricular pacing. The early paced inter-ventricular asynchrony is determined 54 by applying the short atrio-ventricular delay and then measuring the activation delay between the left-ventricle 22 and the right-ventricle 24. The short atrio-ventricular delay is less than 50 ms. The baseline inter-ventricular asynchrony is determined 56 without pacing. The baseline inter-ventricular asynchrony is determined 56 by measuring the activation delay between the right-ventricle 22 and the left-ventricle 24. The average inter-ventricular asynchrony is calculated 58 by averaging the early paced inter-ventricular asynchrony 54 and the baseline inter-ventricular asynchrony 56. The average inter-ventricular asynchrony 58 optimizes cardiac function by minimizing intra-left ventricular asynchrony.

[0028] FIG. 5 shows a flowchart of a method for adjusting atrio-ventricular delay and ventricular-ventricular delay. The atrio-ventricular delay and ventricular-ventricular delay are adjusted during ventricular pacing to yield the average inter-ventricular asynchrony for optimal intra-left ventricular resynchronization. The atrio-ventricular delay and ventricular-ventricular delay are adjusted by adjusting atrio-ventricular delay 62, measuring the inter-ventricular asynchrony 64, comparing the inter-ventricular asynchrony 66, adjusting atrio-ventricular delay 68, and setting the atrio-ventricular delay 70. The atrio-ventricular delay 62 is adjusted by an atrio-ventricular delay increment. The ventricular-ventricular delay can be increased or decreased by a ventricular-ventricular delay increment. The inter-ventricular asynchrony is measured 64 using a sensor such as a left-ventricular pacing lead and a right-ventricular pacing lead. The inter-ventricular asynchrony is compared 66 to the average inter-ventricular asynchrony. The inter-ventricular asynchrony is compared 66 to the average inter-ventricular asynchrony by minimizing the absolute difference between the inter-ventricular asynchrony and the average inter-ventricular asynchrony. The atrio-ventricular delay is adjusted 68 by an atrio-ventricular delay increment and increasing ventricular-ventricular delay by a ventricular-ventricular delay increment. The atrio-ventricular delay increment and the ventricular-ventricular delay increment are typically less than about 30 ms. The atrio-ventricular delay and ventricular-ventricular delay is set once the average inter-ventricular asynchrony is reached.

[0029] FIG. 6 shows a chart relating atrio-ventricular delay, intra-ventricular asynchrony, and inter-ventricular asynchrony during left ventricular pacing. The upper panel shows intrinsic (also known as endogenous) left ventricle (LV) pacing induced impulse propagation during left bundle branch block (LBBB,+LV pacing. The middle and lower panel show the relation between paced AV-delay and intra-ventricular asynchrony and inter-ventricular asynchrony during LBBB+LV pacing.

[0030] Table 1 data was used to develop the relationships shown above in FIG. 6. Table 1 below shows hemodynamic and asynchrony parameters before and after creation of LBBB and during LBBB+LV pacing for all experiments. Table 1 shows LBBB significantly decreased LVdP/dt|_max and SW while intraVA and interVA significantly increased. During LBBB+optimal LV pacing LVdP/dt|_max and SW restored when pacing with an AV-delay equal to the baseline PQ-time. At the optimum hemodynamic response, intraVA was restored to pre-LBBB values while a significant residual interVA remained present.

[0031]

TABLE 1
Hemodynamic and Asynchrony Parameters
	pre-LBBB	post-LBBB	LBBB + LVP
LVdP/dt|max[mmHg/s]	1627 ± 644	1345 ± 413*	1929 ± 815†
SW [ml]	3480 ± 1062	1972 ± 635*	3224 ± 950†
interVA [ms]	−6 ± 9	−28 ± 9*	−18 ± 10*†
intraVA [ms]	5 ± 2	18 ± 3*	4 ± 2†
PQ-time|LBBB-AV-delay	—	—	1 ± 6
[ms]
*,† vs. pre and post LBBB resp. (P < 0.05)

[0032] In FIG. 6, the data graphed is from one animal experiment with the solid lines showing the following conceptual relationship. I: LBBB+LV pacing at short AV-delays; the ventricles are completely activated by pacing induced activation resulting in a fixed degree of intraVA and interVA (X). II: LBBB (or LBBB+LV pacing with long AV-delays); endogenous activation entirely determines ventricular activation resulting in a fixed degree of intraVA and interVA (Y). III: LBBB+LV pacing at intermediate AV-delays (ranging from A and B); fusion between pacing induced and endogenous activation. Minimal intraVA occurs when pacing induced and endogenous activation meet halfway. This results in a residual degree of interventricular asynchrony equal to (X+Y)/2 which occurs at an AV-delay equal to (A+B)/2.

[0033] Based on the results from animal experiments on asynchrony during pacing, a model for the behavior of asynchrony as a function of the timing of pacing was developed. With this model, the degree of residual inter-ventricular asynchrony at the optimal cardiac pump function can be predicted. This allows for optimization of pacing therapy based on measurements of inter- instead of intra-ventricular asynchrony. An advantage of this approach is that measures for inter-ventricular asynchrony are relatively simple compared to the complex techniques required for accurate intra-ventricular asynchrony measurements. Inter-ventricular asynchrony can be assessed by non-invasive measures such as the timing differences in opening of aortic and pulmonary valves by echocardiography, Tissue Doppler, heart sound, and the like. Unfortunately, MRI technology is not suitable for most pacemaker patients.

[0034] FIG. 7 shows bulls-eye plots of left-ventricular (LV) endocardial activation times during left bundle branch block (LBBB)+LV lateral wall pacing with short, intermediate and long AV-delays and during baseline in one animal. The inner radius of the bulls-eye plots represents the LV apex and the outer radius the LV base. Location of the Septum, Anterior, Lateral and Posterior wall is indicated by S, A, L and P respectively in the baseline LBBB plot. The arrows in the plots represent the Activation Delay Vectors (ADV), which is the directional sum of activation times. The amplitude of the ADV is a measure for the degree of interVA and the angle of the ADV represents the main direction of conduction. FIG. 8 shows a graph of inter-ventricular asynchrony (interVA) as a function of the atrio-ventricular delay (AV-delay) during left bundle branch block (LBBB)+left ventricle (LV) pacing.

[0035] FIG. 9 shows a flowchart of optimization of interventricular asynchrony for maximal hemodynamic effect during left-ventricle (LV) pacing. The degree of interventricular asynchrony is measured 72 during LV pacing with a short AV-delay and measured 74 during the unpaced situation resulting in interVAs of X and Y, respectively. Then, starting with pacing with the longest possible AV-delays, the AV-delays is gradually decreased while measuring interventricular asynchrony (Z) 76. Once Z equals (X+Y)/2 78 pacing is optimal. The time of 10 ms is an arbitrary value. To account for noise/errors in interventricular asynchrony during the decision calculation (X, Y and Z), this may be replaced by minimization of the absolute difference between Z and (X+Y)/2. If the minimum is reached, LV pacing is optimal.

[0036] FIG. 10 shows a flowchart of an alternative approach to maximize hemodynamic effect of left-ventricular (LV) pacing by optimization of the AV-delay via interventricular asynchrony measurements. During pacing with the shortest possible AV-delay, interVA is measured (X) 80. Next, the AV-delay is stepwise increased while measuring interVA (Z) 82. Increasing the AV-delay continues until Z no longer equals X (AV-delay=A) 84. Similarly, during pacing with the longest possible AV-delay interVA is measured (Y) 86. Next, the AV-delay is stepwise decreased while measuring interVA (Z) 88. Decreasing the AV-delay continues until Z no longer equals Y (AV-delay=B) 90. The optimal AV-delay for LV-pacing equals (A+B)/2 92. The time of 10 ms is an arbitrary value. To account for noise/errors in interventricular asynchrony during the decision calculation (X, Y and Z), the decision calculation can be replaced by the comparison (Z−X)<5 ms and (Z−Y)<5 ms (5 ms value arbitrary).

[0037] FIG. 11 shows a flowchart of optimization of interventricular asynchrony for maximal hemodynamic effect during bi-ventricular (Bi-V) pacing. The degree of interventricular asynchrony is measured during LV pacing with a short AV-delay 94 and during the unpaced situation 96 resulting in interVAs of X and Y, respectively. Interventricular asynchrony is measured (Z) 98 with BiV pacing having the longest possible AV-delays (e.g. 200 ms) and the longest possible VV-timing (e.g. +50 ms). VV-timing is stepwise decreased until the shortest possible VV-timing is achieved (e.g. −50 ms) while interVA is measured (Z). Once the degree of interventricular asynchrony (Z) equals (X+Y)/2 pacing is optimal 100. If this does not occur for all tested VV-timing the process is repeated but now during BiV pacing with an AV-delay that is decreased by 10 ms 102. In this approach all combinations of AV-delay and VV-timing are tested. However, it is likely that the condition Z=(X+Y)/2 is fulfilled at more than one combination of VV-timing and AV-delay. Therefore, instead of starting with the longest possible AV-delay, an arbitrary choice may be made for the AV-delay to begin with. The time of 10 ms is an arbitrary value. To account for noise/errors in interventricular asynchrony (X, Y and Z) this may be replaced by minimization of the absolute difference between Z and (X+Y)/2. If the minimum is reached, LV pacing is optimal.

[0038] Table 2 below shows a list of symbols that will be used in equations to show theoretical background to cardiac pacing for optimal intra-left ventricular resynchronization.

[0039]

TABLE 2
List of Symbols
	Symbol	Definition
	interVA	interventricular asynchrony [ms] {variable}
	interVA|opt	predicted degree of interventricular
		asynchrony [ms] at optimal LV function
		{constant}
	interVA|BL	interventricular asynchrony during baseline
		[ms] (no pacing) {constant}
	interVA|LVP	interventricular asynchrony during LV pacing
		[ms] {variable}
	interVA|RVP	interventricular asynchrony during RV pacing
		[ms] {variable}
	interVA|BiVP	interventricular asynchrony during BiV pacing
		[ms] {variable}
	VV	VV-timing [ms] {variable}
	AVD	AV-delay [ms] {variable}
	AVD|LV	AV-delay of the LV [ms] {variable}
	AVD|RV	AV-delay of the RV [ms] {variable}
	AVD0	shortest possible AV-delay [ms] {constant}
	LVF	LV pump function (e.g. LVdP/dt|max, LV
		Stroke Work or Aortic Pulse Pressure)
		{variable}

[0040] Prediction of optimal inter-VA during LV and BiV pacing. Based on the conceptual relation between AV-delay and ventricular asynchrony the following equation can be derived to predict the degree of interVA at optimal hemodynamic response during LV pacing;
interVA|_opt=½·[interVA(AVD0)|_LVP+interVA|_BL]  Equation 1
with AVD0, the shortest possible AV-delay, interVA(AVD0)|_LVP the degree of interVA during LV pacing at AVD0 and interVA|_BL, the degree of interVA during baseline (unpaced LBBB). The same optimal degree of interVA is also valid during BiV pacing since, animal experiments showed that at similar degrees of interVA, LV and BiV pacing produced an equal hemodynamic response;
LVF(interVA)|_BiVP=LVF (interV)|_LVP  Equation 2
with LVF, LV function during LV pacing (LVP) and BiV pacing (BiVP). Consequently, the same interVA|_opt is valid for BiV pacing.

[0041] FIG. 12 shows a graph of the relative change in LVdP/dt|max in percent during left ventricle (LV) and bi-ventricle (BiV) pacing in a canine heart with left bundle branch block (LBBB). The graph shows typical examples of the hemodynamic response as a function of interVA in one animal experiment and in a patient during LV and BiV pacing. This graph illustrates that the hemodynamic response during BiV pacing equals the hemodynamic response during LV pacing at equal interVA. During BiV pacing however, the degree of interVA that could be achieved was smaller than during LV pacing.

[0042] Effect of AV-delay and VV-timing on interVA during BiV pacing. During BiV pacing the degree of interVA at each AV-delay, was equal to the average of the degree of interVA during LV pacing alone and the degree of interVA during RV pacing alone at the same AV-delay. Thus;
interVA(AVD)|_BiVP=½·[interVA(AVD)|_LVP+interVA(AVD)|_RVP]  Equation 4
with AVD, the AV-delay.

[0043] FIG. 13 shows a graph of inter-ventricular asynchrony (interVA) during bi-ventricular (BiV) pacing versus the average interVA during left ventricle (LV) and right ventricle (RV) pacing for all atrio-ventricular delays (AV-delays). This graph shows a typical example of interVA during BiV pacing versus the average degree of interVA during LV and RV pacing. Linear regression resulted in a near unity slope and only a small intercept of 3 ms.

[0044] To extend the application to different VV-timing which can be applied during BiV pacing it should be noted that;
VV=AVD|_RV−AVD|_LV  Equation 4
with AVD|_RV and AVD|_LV the AV-delays of the RV and LV respectively. By definition VV>0 if the LV is activated prior to the RV. Based on Equation 3 and Equation 4 we postulate;
interVA(VV)|_BiVP=½·[interVA(AVD|_LV)|_LVP+interVA(AVD|_RV)|_RVP]  Equation 5
Because pacemakers don't differentiate between LV and RV AV-delays it should be noted that
if VV>0 AVD|_LV=AVD
AVD|_RV=AVD+VV
if VV<0 AVD|_LV=AVD−VV
AVD|_RV=AVD  Equation 6
In this case Equation 5 changes to
if VV>0 interVA(VV,AVD)|_BiVP=½·[interVA(AVD)|_LVP+inter VA(AVD+VV)|_RVP]
if VV<0 interVA(VV,AVD)|_BiVP=½·[interVA(AVD−VV)|_LVP+interVA(AVD)|_RVP]Equation 7

[0045] FIG. 14 shows three-dimensional graph of the relationship between interVA during BiV pacing as a function of AV-delay (x-axis) and VV-interval (y-axis). The colors in the 3 D-plane represent the level of interVA during BiV pacing, which is also displayed on the z-axis. In this example characteristic parameters of the heart were set to user defined values (interVA(AVD0)|_LVP, interVA(AVD0)|_RVP and interVA|_BL). The transparent 2 D-plane represents a constant interVA value at which BiV pacing is optimal. Cross section of this plane with the colored 3 D-plane represents the combination of AV-delays and VV-interval at which BiV pacing is optimal. FIG. 14 was constructed based on the above equations and knowledge of a few characteristic asynchrony parameters of the heart (interVA(AVD0)|_LVP, interVA(AVD0)|_RVP and interVA|_BL) which were user defined in this example. The transparent plane in this figure represents the degree of interVA|_opt at optimal hemodynamic response. The intercept with this transparent plane represents all combinations of the AV-delay and VV-timing leading to an optimal hemodynamic response.

[0046] Thus, embodiments of the cardiac pacing for optimal intra-left ventricular resynchronization are disclosed. One skilled in the art will appreciate that the present invention can be practiced with embodiments other than those disclosed. The disclosed embodiments are presented for purposes of illustration and not limitation, and the present invention is limited only by the claims that follow.