Pacemaker

<i>Technical field</i>

The present invention relates to a pacemaker comprising a pulse generating means for delivering stimulation pulses to a patient's heart, a controlling means for controlling the delivery of stimulation pulses from said pulse generating means, and a sensor provided to measure a parameter related to cardiac output of the patient.

<i>Background</i>

In multisite stimulation the PV-interval between the occurrence of an intrinsic P-wave and the application of a cardiac stimulation pulse in the ventricle can be different from the AV-interval between consecutive cardiac stimulation pulses to the atrium and the ventricle, and the PA-interval between the occurrence of an intrinsic P-wave in the right atrium and the application of a cardiac stimulation pulse in the left atrium can be different from the AA-interval between consecutive cardiac stimulation pulses to the right and left atria. By the expression AV-delay is meant in the following the interval between the occurrence of a paced or an intrinsic P-wave and the application of a cardiac stimulation pulse to the ventricle of the heart and by AA-delay the interval between the occurrence of a paced or an intrinsic P-wave in the right atrium and the application of a cardiac stimulation pulse to the left atrium of the heart.

In cardiac therapy there is a primary aim to stimulate the heart such that the natural way of a heart's functioning is re-established as far as possible. The heart is then working with a minimum waste of energy.

It is well known that cardiac output of a human being is depending on the AV-delay, see e.g. Swedish patent application no. 001534-7 .

In this application experimental data obtained from an animal study illustrate the dependence of several cardiac performance parameters on the AV-delay (or PV-delay). A changed AV-delay has an instant effect on the stroke volume SV but can also be observed on the mixed or central venous pressure P_vO₂ after a circulatory delay. A maximum P_vO₂ response was in the animal studies observed 50-80 s after the AV change. Thereafter it declined due to autonomic compensation. Thus, P_vO₂ mirrors the left heart's performance and reflects the cardiac output and its transient response can be used to optimize the AV-delay according to the algorithm described therein. Simultaneously the parameters oxygen saturation SO₂ and carbon dioxide CO₂ can be used for this purpose. The carbon dioxide concentration CO₂ will, however, decrease in the same degree as the oxygen concentration increases.

It is also known from e.g. US 4 928 688 to pace both ventricles of a heart to produce simultaneous contraction of both ventricles, thereby assuring hemodynamic efficiency. In this US patent only simultaneous pacing of both ventricles is proposed. In US 5 540 727 optimisation of the pacing by selective stimulation in both the atria and both ventricles with certain interatrial AA-, VV-, and AV-delays is described.

The purpose of the present invention is to improve the possibilities of determining the dependence of different cardiac performance parameters on at least one of the time-delays VV-delay and AA-delay, such that significant information can be extracted also from such parameters as the oxygen pressure (pO2) and oxygen saturation (SO2).

<i>Disclosure of the invention</i>

This purpose is obtained by a pacemaker of the kind set forth in the introductory portion of the description and having the characterising features of claim 1.

Thus, by using an analogous technique to the one described in the above-mentioned Swedish patent application no. 001534-7 significant information can also be extracted from such parameters as oxygen pressure pO2 or oxygen saturation SO2 for determining the dependency of cardiac output on the VV-delay and /or AA-delay. This is an important advantage since especially the pO2 sensor has appeared to be especially suitable to use for this kind of measurements. The technique according to the invention can of course also be applied on other measured cardiac performance parameters in order to get more distinct results, in particular if the measurement signals are affected by disturbances in one way or the other.

According to an advantageous embodiment of the pacemaker according to the invention said calculation means is adapted to form a first difference between average values obtained during the window positioned inside said time of operation with said first VV- or AA-delay value and obtained during the time window positioned inside said time of operation with said second VV- or AA-delay value, as well as a second difference between average values obtained during the time window positioned inside said time of operation with said second W- or AA-delay value and the time window positioned inside said time of operation after the return of said W- or AA-delay back to said first W- or AA-delay value for using said first and second differences in determining which VV- or AA-delay results in a higher cardiac output. In this way the possibility of extracting reliable information from the measurement signals is still more improved.

According to still another advantageous embodiment of the pacemaker according to the invention in which said altering means is adapted to repeatedly alter the VV- or AA-delay a plurality of times between said predetermined first and second W- or AA-delay values, the calculation means is adapted to form a mean value of a resulting plurality of said first differences and a mean value of a plurality of said second differences for use said mean values in determining which W-and/or AA-delay values result in a higher cardiac output. The possibility of the extracting information from the measurement signals is then further improved.

According to another advantageous embodiment of the pacemaker according to the invention said altering means is adapted, after determination of which one of said first and second VV- or AA-delay values is indicating a higher cardiac output, to alter the W- or AA-delay between this better W- or AA-delay value of said first and second VV- or AA-delay values indicating a higher cardiac output and a third W- or AA-delay value and said determining means is adapted to then determine which one of said better W- or AA-delay value and said third AV-delay value results in an indication of a higher cardiac output, said altering means being adapted to repeat this procedure till W- and/or AA-delay values are determined which result in an indication of a highest cardiac output. Thus the pacemaker according to the invention will automatically find out the optimum VV-and/or AA-delay values and will then toggle around these optimum values. If conditions should change such other VV- and/or AA-delay values are needed to obtain a maximum cardiac output, the pacemaker will automatically find this new optimum delay values.

According to still another advantageous embodiment of the pacemaker according to the invention the altering means is adapted to alter, in addition to at least one of the time-delays W-delay and AA-delay, the AV-delay, and said determining means is adapted to determine from said detected changes in said parameter related to cardiac output when altering said VV-and/or AA-delays and said AV-delay which combination of VV-and/or AA-delay values and AV-delay values results in an indication of highest cardiac output. Thus the VV-delay or the AA-delay together with the AV-delay can be adjusted to give an optimum cardiac output, or all three time-delays VV-, AA- and AV-delays can be adjusted to give an optimum cardiac output.

According to yet another advantageous embodiment of the pacemaker according to the invention said altering means in adapted to alter said time-delay values after times of operation according to a pseudo-stochastic sequence. Since several biological variations and external disturbances have cyclic characters, which can interfere with the measurements, it is an advantage not to use a cyclic variation of the time-delays but pseudo-stochastic sequences of delay values. On patients having Left Bundle Branch Block LBBB only the VV-delay is optimized by a pseudo-sequence. On patients with both AV-block and LBBB it is possible to both optimize the AV-delay and the VV-delay by successive iterations.

According to another advantageous embodiment of the pacemaker according to the invention said time windows are positioned immediately before and immediately after an alteration of the respective time-delay values. In this way changes in the measured parameter just related to the change of the time-delay in question are determined and errors originating from variations in oxygen consumption's of the patient and sensor drift are minimised.

According to still other advantageous embodiments of the pacemaker according to the invention said sensor measuring a parameter related to cardiac output is one of the following kind of sensors, a pO2-sensor, a SO2-sensor, a mean blood pressure sensor, a coronary artery flow sensor or a sinus rate sensor, and said blood pressure sensor is preferably adapted to measure the blood pressure in the vena cava or the right atrium. Artery blood pressure, mean blood pressure, systolic and diastolic pressure all correlate to cardiac output.

<i>Brief description of the drawings</i>

To explain the invention in greater detail embodiments of the invention will now be described with reference to the drawings, on which <dl id="dl0001" compact="compact"><dt>Figures 1 and 2</dt><dd>illustrates generally a 3-chamber and a 4-chamber heart stimulator respectively,</dd><dt>Figure 3</dt><dd>shows qualitatively the time delay and the average pO2 as a function of time,</dd><dt>Figure 4</dt><dd>illustrates the result obtained by analysis of the data in figure 3 by the pacemaker according to the invention,</dd><dt>Figure 5</dt><dd>illustrates the differential technique implemented in the pacemaker according to the invention to obtain an optimum AV-delay,</dd><dt>Figure 6</dt><dd>illustrates different types of sampling measured parameter values,</dd><dt>Figure 7</dt><dd>shows a pseudo-stochastic AV-delay sequence,</dd><dt>Figure 8</dt><dd>shows the AV-delay sequence when an optimum time-delay is reached,</dd><dt>Figure 9</dt><dd>is a diagram illustrating the spread and medium value of pO2 differences obtained from the data shown in figure 3,</dd><dt>Figure 10</dt><dd>is a block diagram of a closed loop regulation system of the pacemaker according to the invention.</dd></dl>

<i>Description of preferred embodiments</i>

Figures 1 and 2 illustrate principally two examples of multisite stimulation. In the 3-chamber stimulation system shown in figure 1 electrodes 20, 22, 24 are positioned in the right atrium RA, the right ventricle RV and the left ventricle LV respectively. LA denotes the left atrium.

In the 4-chamber system shown in figure 2 electrodes 26, 28, 30 and 32 are positioned in the right atrium RA, the right ventricle RV, the left atrium LA and the left ventricle LV respectively.

In figure 1 biventricular 3-chamber stimulation is thus illustrated and in figure 2 biatrial and biventricular 4-chamber stimulation is shown. In the 3-chamber stimulation according to figure 1 two time delays are of interest, viz. the ordinary AV-delay, t1 in the figure, and the VV-time delay, t2 in the figure. In the 4-chamber system shown in figure 2 three time delays are of interest, viz. the ordinary AV-delay t1, the VV-delay t2, and the AA-delay t4.

The AV-delay t1 on the right side of the heart can be different from the AV-delay t3 on the left side of the heart. It should also be noted that t1 ≈ t2+t3, see Koglec.

With the technique described below each of the above mentioned time-delays or time intervals can be optimized, and by successive iterations the optimal timing can be obtained by the method described in the following.

It should also be noticed that the time settings t1 - t4 above are different for sensed and stimulated invents as pointed out above and therefore should be optimized separately.

Figure 3 shows qualitatively the time interval VV- or AA-delay as a function of time. In this diagram dots are shown representing the average value of the measured pO2 during a predetermined period of time immediately before and immediately after each change in the time-delay. The marked values are average values during a predetermined time immediately preceding the position of the dot.

Figure 4 shows the average value of the difference pO2-diff between each couple of average values from either sides of transitions between two time delay values for negative transitions, to the left in figure 4, and for positive transitions, to the right in the figure. From figure 4 it appears that a negative time transition increases the p02- difference whereas a positive time step decreases the pO2-difference. This significant difference between the two pO2-differences indicates that the shorter timedelay is more effective than the longer delay in this example.

In figure 9 the minimum and maximum values of the used pO2 differences are shown as well as the 25%-75% spread of these values and the median values.

Corresponding diagrams can be obtained for the SO₂ difference, the central venous pressure difference, the mean blood pressure difference, the carotid artery flow difference and the sinus rate difference.

As mentioned above it is possible to identify hemodynamic improvements of cardiac output by measuring changes in 02 after a step change of a time-delay. The oxygen contents can be measured electrochemically, by a pO2-sensor, or optically by a SO2 sensor.

Thus in the example described above the time delay, or the timing of the pulse generator of the pacemaker, is altered between two settings, e.g. a level A and a level B, one of which is the "better" one. By forming the difference between the average oxygen content, Δp (AB) during a time window of predetermined length immediately before a change of the time delay and during a time window of the same length immediately after the change from level A to level B, and comparing this result with the corresponding difference, Δp (BA), when the AV-delay is changed in the opposite direction from level B to level A, it is possible to find the best setting, A or B. The described procedure can be repeated and the results averaged for obtain a better resolution as described above.

If Δp(AB) &lt; Δp(BA) setting A of the time delay gives a higher O2 content indicating a better ventricular performance.

If, on the other hand, Δp(AB) &gt; Δp(BA) setting B of the time delay give a better cardiac performance.

In figure 5 n is defined as the first oxygen measurement p(n) after a step change of the time delay from level A to level B. The sample # m is defined as the first oxygen measurement p(m) after a change of time delay from level B to level A. The number of averaged O2 samples is denoted by i, in figure 5, i =4.

Figure 5 illustrates the differential technique according to the invention for selecting the best time delay value as explained above.

After # n + i - 1 measurements of O2 it is possible to calculate the O2 response of a change of the time delay from level A to level B according to the equation.<maths id="math0001" num=""><img file="EP1304136A2_D0001.tif" /></maths>

After # m + i -1 O2 measurements it is possible to calculate the O2 response of a change of the time delay from level B to level A according to equation<maths id="math0002" num=""><img file="EP1304136A2_D0002.tif" /></maths> By repeating this procedure several times u the risk of fault decisions is practically eliminated. The corresponding mean values ψ (A, B) are given by the following equations<maths id="math0003" num=""><img file="EP1304136A2_D0003.tif" /></maths><maths id="math0004" num=""><img file="EP1304136A2_D0004.tif" /></maths>

Thus if Ψ(AB) &lt; Ψ(BA) the time delay value B does not improve the heart performance compared to the situation with a time-delay value equal to level A.

Different types of differentiation of the O2 signal can be used. In the example above and in situations "A" and "B" in figure 6 the time windows, in which the measurements are performed, i.e. the measurements samples are taken, do not overlap. In the type of differentiation illustrated at "C"and "D" in figure 6 the same measurement samples are sometimes used twice, since adjacent time windows, in which the measurements are carried out, overlap. The time windows in question are marked tw. Curve A illustrates a situation with an intermediate time interval between each couple of time windows in which measurements are performed, curve B illustrates a situation in which consecutive measurement time windows directly follow each other, curve C illustrates a situation with partially overlapping measurement time windows and curve D illustrates a situation in which the time windows are totally overlapping.

It is an advantage not to use a cyclic variation of the time delays, since many biological variations and external disturbances are cyclic which consequently can interfere with the measurements. It may therefore be an advantage to change the time delays according to a pseudo-stochastic sequence as illustrated in figure 7.

When a time delay value has been selected as the "better" one, this "better" value is used in a new comparison procedure in which it is compared with another time delay. In each step of this selection procedure the time delay value is favoured which results in the highest O2 value, and when the optimum time delay value is reached, the time delays will toggle around this optimum value, as illustrated in figure 8. In the shown example a time delay equal to B is supposed to give a better O2 value than a time delay equal to A. Further, a time delay equal to B is also supposed to give a better O2 value than a time delay equal to C.

From the description above it appears that, if the situation of the patient changes such that the optimum time delays are changed, the pacemaker according to the invention will automatically found out new optimum values and the pacemaker will operate with these new delays.

The differential technique described above is a simple way of eliminating errors originating from variations in oxygen consumption of the patient and sensor drift. It is also advantages to use this technique because the variations of the oxygen content due to time-delay changes are much smaller that changes in the oxygen content due to metabolic variations and also other factors.

Figure 10 shows a block diagram of the principal layout of the pacemaker according to the invention. The pacemaker comprises a sensor 2, preferably a pO2-sensor, for measuring a parameter related to cardiac output of the patient (=e.g. oxygen pressure). The measurement signals are processed in a suitable signal processing unit 4 and the average value during predetermined time windows immediately before and after a change in a time delay is calculated in the calculating means 6. In this calculating means 6 the difference between average values obtained in the respective time window on each side of a time delay transition are calculated, and finally a mean value is determined for this difference for "negative" transitions and "positive" transitions in the time delay, respectively, as described above, for determining which one of the time delay values results in a higher cardiac output. This result is supplied to a controlling means 8 including an altering means 10.

The timing of the pulse generator 12 is then controlled by the altering means 10 to change the time delay in question between this "better" time delay value and a new time-delay value.

The length of the time window in which the measurements are performed as well as the sampling frequency can also be varied from the controlling means 8.

An IEGM detector 14 is also connected to the controlling means 8, a.o. to synchronise the measurements to the cardiac cycle of the patient.

The optimum time delay at rest is normally longer than at exercise. Different activity levels therefore have to be distinguished. Considerable variations in activity level can, however, give rise to problems. The easiest way to avoid such problem is to inhibit the time delay selection function of the pacemaker when an activity sensor indicates a work of the patient exceeding a predetermined threshold level. An activity sensor 16 is therefore connected to the controlling means 8. This activity sensor can be the O2 sensor used and/or e.g. a movement sensor. If an increased activity persists over a longer time it is possible with the pacemaker according to the invention to search a new optimum time delay(s) at this defined level of workload. In such a state it is probably advantageous to have a shorter decision time for deciding the optimum time delay(s) than at rest condition.

The sensor used in the pacemaker according to the invention is preferably an electrochemically pO2-sensor of the kind described in WO 98/14772. The invention is, however, not limited to the use of such a sensor. A pO2-sensor can be implanted together with an implanted pacemaker, e.g. for measuring the oxygen concentration in the right atrium. However, variations of the oxygen content in e.g. the ventricle or arteria pulmonaris can be continuously measured in a corresponding way. Further, with a pO2-sensor the measuring pulses can be made so short that synchronisation of the measurement to the cardiac cycle is possible, and the pO2-sensor has proved to give reliable measurement results over time.