WO1982003787A1

WO1982003787A1 - Pacemaker

Info

Publication number: WO1982003787A1
Application number: PCT/DE1982/000106
Authority: WO
Inventors: Mess & Therapieg Biotronik
Original assignee: Nettelhorst Herwig; Rexhausen Hermann; Roetter Juergen; Rupprecht Werner; Schaldach Max
Priority date: 1981-05-04
Filing date: 1982-05-04
Publication date: 1982-11-11
Also published as: EP0077802A1

Abstract

The pacemaker comprises a circuit (10) for formatting pulses, which is connected to the input of the input amplifier (5) so as to obtain, by the shortening of received stimulation pulses, a more rapid release of the input amplifier overloaded by those stimulation pulses in order to allow the amplifier to amplify the intracardiac signals.

Description

Pacemaker

Description

The invention relates to a pacemaker in accordance with the preamble of patent claim 1.

In known pacemakers there is the disadvantage that the input amplifier for intracardiac signals on one

OMH The stimulation pulse is not permeable for some time to signals picked up by the heart, so that, for example, separate electrodes with a separate input circuit are required to check the success of the stimulation.

The invention has for its object to reduce the disturbing overdrive of the input amplifier after sending out a stimulation pulse so that the signals that are important for the control of the pacemaker can be picked up without any problems.

This object is achieved by the pacemaker with the features of patent claim 1.

In the case of the invention, it was assumed that the input amplifier had to be caused to settle as shortly as possible after the stimulation pulse had been emitted, so that intracardiac signals can be further amplified and evaluated.

The input signals to be expected have such an amplitude spectrum that further processing would best be carried out with a DC voltage amplifier. However, since the stimulation pulse, which has a large DC voltage component, is not to be processed further, a lower cut-off frequency of 15 Hz is regarded as a good compromise for the combination of input amplifier and the preceding pulse. This combination is preferably an active high-pass filter with a 3dB cut-off frequency of 15 Hz and a gain in

^ j E Pass range of approximately 700 times. The settling time should be set such that the activation with a stimulation pulse extended to several hundred milliseconds no longer causes the settling amplitude after approx. 30 seconds to trigger the trigger circuit provided for evaluating intracardial signals in the corresponding detection device.

In-depth considerations have led to the result, since the requirement on the input amplifier to be able to process intracardiac signals after 30 msec is not or only very difficult to meet due to the addition of linear filters. Such a filter should have high-pass character. Usual filters according to Butterworth or Tschebyscheff are ruled out because of their non-linear phase curve. This also applies to the Bessel filter, which has only minor phase distortions in the pass band; However, the filter required here as a pulse shaper is not intended to transmit pulses in the passband, but the intercardiac pulse to be processed has a large part of its energy within the frequency spectrum in the cutoff or transition range of the required filter, namely in the range from 0 to 40 Hz. It is therefore important for a rapid settling of a filter to be connected upstream of the input amplifier as a pulse shaper that a phase response that is as linear as possible is approximated in this frequency range, but at least between 10 and 30 Hz. This condition is met by an orthogonal filter with a negative group delay.

The optimal setting of the filter parameters depends on the signal transmission conditions in the heart of the Patients, ie on the electrode fixation and the type of electrodes used. In order to enable the signal transmission ratios to be adapted to long-term or short-term fluctuations, an additional device is provided according to a preferred development of the invention, which adjusts the filter adjustment to the changing conditions. The adjustment criterion is the period of time that elapses before the signal at the output of the EKG amplifier has decayed after an artificial stimulation pulse. In this way it is achieved that the decay behavior of the input amplifier is optimally short when artificial stimulation pulses appear, so that signals from the heart can be picked up and evaluated again after a very short time.

In a further advantageous embodiment of the invention, means are provided in order to use the signal obtained with the pulse generator shown to monitor the success of the pacemaker, that is to say to influence the behavior by signals which were recorded immediately after the delivery of a stimulation pulse. This arrangement can be used particularly advantageously in connection with an automatic amplitude control of the output pulses of the pacemaker, the period of time after which intra-cardiac signals recorded following a stimulation pulse can be processed again. The invention can thus also be used in particular with pacemakers which have a Carry out double-chamber stimulation, since here too a complete separation of the signals to be recorded is required. The time between the stimulation pulse and the usable intracardiac signal becomes additional shortened by the provision of a circuit for discharging the output capacitor of the stimulation circuit, as is known per se.

An advantageous exemplary embodiment of the invention is described in more detail below, including preferred further developments. Show it:

FIG. 1 shows a circuit diagram of an exemplary embodiment of the pacemaker according to the invention in a block diagram,

FIG. 2 shows the principle according to which such a filter is preferably constructed,

3 shows, as an exemplary embodiment, a circuit diagram for the pulse shaper constructed according to the principle of FIG. 2, which is designated by 10 in FIG. 1, and

Figure 4 shows an embodiment of an additional module for automatic adaptation of the filter properties to changing signal transmission conditions in the heart with means for varying the stimulation energy depending on the success.

The signal-inhibited demand pacemaker shown as a block diagram in FIG. 1 initially contains the essential components of conventional known pacemakers, such as an astable multivibrator which forms a timer 1 and which triggers the output stage 2 and therefore retrieves a stimulation pulse and a detection unit 3 which when registering an R wave, reset timer 1

O PI sets, so that the action cycle (850 msec) is initiated anew.

The stimulation pulses emitted by output stage 2 reach the heart via a pacemaker electrode (connection 4). Conversely, this Schrittmacherelek¬ trode the intracardiac electrogram recorded and an input amplifier (ECG amplifier) 5 Detek¬ tion unit 3 supplied for resetting (De and-functional) ^"of the timer respectively at the beginning of the Stimula¬ tion cycle, wherein the reset is carried out via the reed switch M located in the switch position shown. An upstream refractory time unit 6, for example likewise in the form of an astable multivibrator, prevents the reset of the timer 1 within the first 400 msec after stimulation and detection, for example 3 downstream interference filter 7 identifies electrical and magnetic alternating fields and switches (via the other switch position of the reed switch M) the timer 1 to a fixed-frequency mode of operation before there can be interference with the detection unit 3 and pulse suppression of the reed scha The detector bypasses lters M with the help of a magnet, resulting in a firm stimulation.

Finally, an energy source provides the power supply for the device shown in FIG. 1.

The stimulation pulses required for stimulating the heart have a voltage amplitude of approximately 5 V with a

_ OMP Pulse duration from about 0.1 to 1 ms. The electrical signals caused by the functioning of the heart (intracardiac signals), which are taken from the pacemaker electrode 4 and then serve to control the pacemaker, on the other hand only have amplitudes in the millivolt range, so that they are amplified by the input gain 5 before further processing need to be raised. However, since the output stage of the timer 1 serving as a pulse generator is directly connected to the input of the input amplifier 5 in known cardiac pacemakers, the input amplifier is heavily overdriven during the transmission of a stimulation pulse. In addition, there is the fact that the electrolyte surrounding the pacemaker electrode 4 causes the effects of a stirulation pulse to subside after several hundred milliseconds due to the polarization voltage that is developing.

The input 11 of the block diagram according to FIG. 2, which corresponds in principle to the pulse shaper 10 of FIG. 1, receives its input voltage U _e superimposed both from the output stage 2 and from the input terminal 4 in FIG. 1. The output voltage U _a at Output 12 of the pulse shaper 10 according to FIG. 2 reaches the detection device 3 in FIG. 1.

The pulse shaper 10 according to FIG. 2 contains all-passes AP] _ and AP2, which are connected in series with one another and each cause a phase rotation by 90 ° at a frequency of 15 Hz, so that the voltage O- versus U _e by 90 ° and the voltage U2 is rotated 180 ° in phase with respect to U _e . Such an all-pass has a transmission function

, ÜΪU Uι / U _e = Ü ₂ / Ü! = (ls) / (l + s),

where s can be equal to jω in the damping-free ideal case (t being the angular frequency).

The input voltage U _e and the output voltages of the all-passes are weighted in evaluation elements B ₀ , Bj, B ₂ with passive coefficients K _Q , KJ_, K ₂ , and added to the input voltage U _a in an adder A according to the sign. For a high-pass filter, the coefficients are selected, for example, as follows:

^κ o = ~ ^κ l = 0.5; K ₂ = 0.

The following applies accordingly to a bandpass filter:

Ki as. 0; K _Q = -K ₂ = 0.5.

The filter chosen as a pulse shaper according to FIG. 2 has the advantage of low component expenditure with a large variation range of the possible filter setting. This filter is also predestined for an adaptive setting of the filter coefficients. If the coefficients are chosen so that K _{0 has} the opposite polarity to K] _, the filter has high-pass characteristics.

Instead of the embodiment shown in FIG. 2, a corresponding digital filter can also be used, which, although it requires more effort, is easier to integrate. The detailed embodiment shown in FIG. 3 has approximately one gain in the pass band and serves to "shorten" the stimulation pulse, while the actual increase in the amplitude value takes place in the input amplifier 5 to be connected (in FIG. 1). Since can be inserted between the output 12 in FIG. 3 and the input amplifier 5 in FIG. 1, a limiter circuit which keeps voltage peaks with a voltage greater than 1.2 V away from the input amplifier 5, so that extreme overloads are avoided .

The exemplary embodiment shown in FIG. 3 has a pulse shaping part between the input terminal 11 and the output 12 and an upstream bandpass consisting of resistors R1 and R2 and capacitors C1, C2 with a downstream isolating amplifier 13. The bandpass has an upper 3dB Limit frequency of approximately 40 Hz and a lower 3dB limit frequency of approximately 15 Hz. By using this bandpass, even better settling times could be achieved compared to the sole use of the actual pulse shaping part between terminals 11 and 12. In addition, a standard-compliant input resistance and a limitation to higher frequencies were achieved.

3 is formed by resistors R3 to R 5 and a capacitor C3 in conjunction with an operational amplifier 14. A second all-pass connected downstream is formed by resistors R6 to R8 and one Capacitor C4 in connection with an operational amplifier 15. Correspondingly - lo ¬

2, the voltages U _e , U_, U _{2 are} also fed via evaluation elements to a summing circuit which contains, inter alia, an operational amplifier 16 and a resistor R9 as a negative feedback path. The first evaluation element (corresponding to the coefficient K ₀ ) consists of a voltage divider R11-R12, which is also part of the summing circuit containing the operational amplifier 16 and is connected to the non-inverting input of the operational amplifier 16. The second and third evaluation _elements (corresponding to K _{x and} κ _{2) are formed by} voltage dividers P1-R13-R14 or P2-R15-R12. The first evaluation element is connected to the inverting and the second to the non-inverting input of the operational amplifier 16. The resistors (R) have resistance values of 100 kilohms and the capacitors (C) have capacitance values of 0.1 microfarads, unless other values are given in FIG. 3.

The setting of the coefficients is carried out in such a way that the above-mentioned conditions for operation with conventional electrodes can be met, the above-mentioned values on the one hand representing a favorable mean value, but on the other hand an adaptation to special electrodes can take place accordingly.

In the input amplifier 5 shown in Fig. 1 an FET switch is inserted, which is controlled by the output signal of the output stage 2 such that the signal path into the main amplifier is blocked during the settling time of the pulse shaper 10 - for example for one

C -__ 1 Time up to and including 30 msec. From the transmission of the stimulation pulse from the output stage 2. Such an FET switch is shown in FIG. 3 as the output of the pulse shaper. It shorts the output signal of the filter via a resistor R 16 with a high resistance. The FET switch is controlled from the output of output stage 2. The corresponding connecting line is shown in FIG. 1. After this "blanking time" has ended, the amplifier is switched on again.

The amplifier can also perform the task of eliminating the effects of the polarization voltage on the pacemaker electrode. It is then designed to act like a first-order high-pass filter. However, this causes a phase distortion that results in a large increase in group delay.

However, since the phase distortion of the overall system should be as small as possible, in this case the pulse shaper 10 upstream of the main amplifier must compensate for this group delay at the lower frequency limit by a negative group delay. The generation of such a negative group delay is - like the measurement of the filters used - known per se from the dissertations J. Rotter "Adjustable Group Delay Filters, Realization and Approximation", 1978, University of Kaiserslautern, and W. Rupprecht: "Linear networks with negative group term", 1961, University of Karlsruhe.

The output stage reproduced in block circuit form in FIG. 1 is provided with a circuit for shortening the Recovery time equipped after a stimulation pulse, as shown in the European patent application EP-Al-0 000 989. In connection with such a circuit, the minimum time between stimulation and intracardial signal recording is additionally shortened considerably.

FIG. 4 shows an exemplary embodiment of an arrangement which is suitable for changing individual parameters of the pulse shaper stage as a function of the decay time. Furthermore, additional means are provided which change the output amplitude of the pacemaker in response to cardiac actions which have been stimulated by a previous pulse of the pacemaker. Depending on the success of the stimulation, the output amplitude of the stimulation pulses is increased or decreased. A step-up takes place if no stimulation success has been achieved, while if the stimulation was successful, the amplitude is reduced at intervals by a fixed amount. The reduction is initially carried out periodically on a trial basis after a certain number of stimulation pulses. If there is no successful stimulation after a reduction in the stimulation amplitude, the stimulation is carried out again in the immediately following stimulation in order to ensure the necessary stimulation security for the patient.

The means for changing the output amplitude as a function of success - as shown in FIG. 4 - are independent of the means for varying the pulse shaping parameters. The representation of the two functions was in Embodiment realized on the basis of a single assembly, different components being able to be used with regard to both functions, so that simplifications resulted.

The output signal of stage 2 in FIG. 1 is fed to a frequency divider stage 17 which, by generating a control pulse sequence which is lower than the stimulation frequency, ensures that, depending on the stimulation pulses emitted, the adaptation of the pulse former stage or the reduction in the stimulation amplitude is reduced not repeated with every pulse, but only at regular intervals. In the absence of stimulation pulses in demand mode, no adjustment is necessary. A counter 18, which is driven by a clock generator (not shown) with higher-frequency timing pulses, is reset to an initial state by a first output signal of the frequency divider 17 and activated for counting. In the present example, this counting process is initiated every sixth pulse emitted by the pacemaker.

The output signal of the amplifier 5 downstream of the pulse shaper 10 (FIG. 1) reaches a Schmitt trigger or a similar component which gives a signal as long as an output signal is emitted at the output of the amplifier 5, the amount of which differs sufficiently from zero or for as long an input signal has not decayed to a negligible "value. The output signal of the trigger stage 19 arrives at a pulse shaper 20, which the trailing edge during the

O converting the sounding signal of the present pulse into a short-term pulse, which is supplied as an input signal in subsequent processing stages and activates it accordingly, so that the point in time in question can serve as a criterion for the decay behavior of the output signal of the pulse former.

The further outputs x and y are further frequency-divided compared to the signal fed to the counter 18 and therefore change in even slower cycles. In the example shown, the outputs x and y may additionally be divided by a factor of 16 (output x) and 32 (output y) compared to the aforementioned output, so that four are connected by means of AND gates 21 to 24 by correspondingly linking the signals at their outputs further signals are available which appear alternately in succession in a time cycle for a quarter of the total period. The signal present at the AND gate 21 defines a first measurement cycle, during which it is determined whether a count value stored in a memory 25 still corresponds to the time that the input circuit needs to reach the idle state after an artificial stimulation pulse. For this purpose, the counter 25 is increased to its maximum value by means of a corresponding input signal by means of a further pulse shaper 26, which emits a signal on the rising edge of the output signal of the AND gate 21.

Now that the time-determining counter 18 has been started with the delivery of an artificial stimulation pulse, the output signal of the counter 18 is constantly compared by means of a comparator 27 with the content of the counter 25 .. The comparator 27 outputs an output signal as long as the counter reading of the counter 18 is less than the numerical value contained in the memory 25.

If this signal is not output when the pulse shaper 20 a pulse as a sign of completion of the delivery of an output signal ^• om amplifier 5 appears, it is by means of an AND gate 25 to which the output signal of the comparator 27 and also the signal of the AND gate 21 is supplied, a transfer stage 29 is activated, which overwrites the content of the counter 18 into the memory 25 at this time. The memory 25 thus contains the numerical value which corresponds to the current decay time of the arrangement and characterizes the signal transmission behavior in the heart and pulse shaping stage. While the output signal of the AND gate 21 appears, this process is repeated with each stimulation pulse.

It is thus ensured that the shortest decay time is determined in the case of several stimulation pulses, and if artefacts or interference signals occur, a longer signal duration will appear after a stimulation pulse, which will have no effect on the reduction of the count value in memory 25. Deleting the memory in the meantime ensures that the transmission behavior also follows such changes in the transmission behavior in the heart - for example with changes in the electrode effect - which result in an extension of the decay behavior. Subsequent to this measurement period, characterized by an output signal from the AND gate 22, there is a period with which an attempt is made to improve the signal decay behavior by changing a parameter of the pulse shaper stage 10 (in FIG. 1).

In order to clarify the function of the principle of adapting the parameters of the pulse shaper 10, the variant of the adjusting element P1 of the pulse shaper 10 shown in FIG. 3 must first be explained. Instead of a continuously variable resistor, a decade of resistance u is used here, in which switches are connected in parallel to the individual resistors and are controlled by a counting and decoding arrangement 28. The counting and decoding apparatus 28 includes a so-counting it, which are incremented in single counts or _decremented (inputs "+1" or ^π -l "in the figure). The indices" can p "and" t "are intended to indicate that the incrementing or incrementing takes place permanently or temporarily via the inputs concerned. In the case of the temporary increment or decrementing, the changed counter reading is only retained as long as the corresponding signal is present Realize practice by means of an additional circuit to a conventional counter which, when the signal denoted by "t" appears, causes the counter to increment or decrement and when the signal ends - on its trailing edge - generates a pulse which triggers the reverse counting direction The outputs of the counter correspond to its binary switching states, the resistances in the decade of the potentiometer Pl 'being so

are chosen so that their value ratios correspond to the binary values of the outputs of the counter which control the associated switches. The total resistance value of the potentiometer P1 'is thus dimensioned such that it increases or decreases in accordance with the increase or decrease in the counter reading.

If a signal is present at the output of the AND gate 22, this leads to a temporary reduction in the ^" state of the counter 28. By reducing the count value, the total resistance of the potentiometer Pl 'is reduced by one step and thus the behavior of the pulse shaper 5 is slightly The downstream counter 18 is reset and started by the corresponding output of the divider 17. The output signal of the pulse shaper 20 stops the counter after the signal of the input amplifier 5 has subsided (input "Stop") and with the same pulse after delay a delay element 30 activates the comparator 27, which emits an output signal if the content of the counter 18 is less than the content of the memory 25 corresponding to the smallest measured value duration determined. In this case, a signal reaches the transmission element 29, which changes the memory 25 according to the last status of the counter 18. Di The intermediate AND gate is activated via an OR gate 30. At the same time, the output signal of the comparator also reaches AND gates 31 and 32, which control the inputs "+ lt" and "-p" when the AND gates 22 or 23 a signal is present. In this case, the counter 28 would be reduced by one value via the AND gate 32 - as a sign of this. that an improved decay behavior of the pulse shaper 19 could be achieved with the assigned setting of the potentiometer Pl ¹ .

As long as the signal appears at the output of the AND gate 22, it is repeatedly determined whether a further reduction in the decay behavior of the input circuit can be achieved after a stimulation pulse. If the outputs x and y change their state with the progress of the counter 17, so that. the AND gate 23 becomes conductive, a trial increase of the counter 28 follows and a corresponding check is carried out to determine whether an improvement can be achieved by a further change in the parameters of the transmission part 19. Changes in the transmission of the signals recorded in the heart can thus be taken into account and the variable coefficient of the pulse shaping circuit 19 can be adapted with sufficient speed. Artifacts and interference signals always lead exclusively to an increase in the decay time after a stimulation pulse, so that a change in the parameters is not caused thereby. A change in the transmission conditions in the sense that an extended decay time has to be taken into account is taken into account in that in an initial period (characterized by switching through the AND gate 21 a readjustment from the point of view that an optimization only takes place in a range of decay times can be carried out which are longer than the previously found values. A fourth measuring period (characterized by switching the AND gate 24) is periodically available in order to control the effectiveness of the stimulation pulses. By shortening the decay of the input stage 5, it is namely ^'now possible to determine the success of the artificial stimulation pulses from the following signal response of the heart directly.

A monoflop 34 is set with an artificial stimulation pulse, the pulse duration of which corresponds to the time during which a response signal appearing in response to a stimulation pulse is to be expected from the heart. The output signal of the monoflop 34 reaches an AND gate 35, at the other input of which the output signal of the pulse shaping stage 20 is present. If, therefore, a pulse appears after a stimulation pulse, which indicates the decay of the signal picked up by the heart, this pulse passes via the AND gate 35 to a flip-flop 36, which is then set. The following pulses now appearing at the output of the Schmitt trigger 19 thus arrive at a further AND gate 37, which is connected downstream of the Q output of the flip-flop 36, and cause a flip-flop 38 to be set, with its ent speaking input, the output of the AND gate 37 is connected. Setting this flip-flop corresponds to successful stimulation.

The trailing edge of the pulse emitted by the monoflop 34 is converted into a short-lasting pulse by means of a pulse shaper 39, which indicates the end of the measuring period for the signal recording from the heart to be carried out here and is delayed by a delay element 40 by a few milliseconds. When the flip-flop 38 is set, this pulse reaches the decrementing input (-) of a control circuit 42 via an AND gate 41, the outputs of which are connected to a control circuit 43 for the output amplitude of the stimulation pulses. The control circuit 43 may be either as shown from a connected resistor chain - but made in accordance with the potentiometer Pl ¹ -or- ■ from a controllable voltage multiplier in stages. ^'. With the

Decrementing reduces the amplitude of the stimulation pulses when the stimulation is successful, whereas it increases when there is no cardiac response to a stimulation. This occurs when the flip-flop 38 is not set in response to an output signal of the pulse shaping stage 39, the output signal of which in this case reaches the incrementing input of the counter 42 via an AND gate 44, because the AND gate 44 thus has its inverting input connected to the is connected to the Q output of the flip-flop 38.

While the amplitude of the stimulation pulses can be continuously increased by incrementing the counter 42 (the permissible input values being adjustable by selecting the type of pacemaker or by external programming means), in order to stimulate the electrode contact as quickly as possible ¬ To be able to provide pulses of greater intensity, the reduction in the amplitudes depends on the output signal of the AND gate 24, which during a fourth measurement period, for the two outputs x and y of the

^"ü H ^ Counter 17 lead no output signal, is activated via its two inverting inputs. For this purpose, the output signal of the AND gate 24 arrives at an additional input of the AND gate 41, at the output of which a signal only appears when the relevant measurement period is pending. In this way, the stimulation amplitude is only reduced periodically “on a trial basis” in order to avoid excessive energy consumption in the long term. If there is no stimulation success due to a stimulation with a reduced amplitude, the amplitude is increased again immediately after the "test period". In order to avoid fluctuation effects, the counter is temporarily decremented by more than one level (input -2t) during the "test period" during which the behavior with reduced stimulation amplitude is examined. It can thus be achieved that the stimulation amplitude always has a certain "amplitude reserve" and that no reduction is made which could result in unsafe stimulation conditions. The pulse shaping circuit according to the invention can thus be used for effectiveness control directly via the stimulation electrode.

Claims

P a t e n t a n s r u c h e

1. pacemaker with electrode connection for receiving intracardial signals and an input amplifier (5) connected to the connection for the intracardial signals,

characterized ,

that the input amplifier (5) is followed by a pulse shaper (10) in order to shorten the period of time after which intracardiac signals recorded following a stimulation pulse can be processed again, whereby

the combination of pulse shaper (10) and input amplifier (5) dampens static and very low-frequency components of its input signal relatively much more strongly than signal components with frequencies around 25 Hz

and has an approximately linear frequency dependence of the phase profile in the transition region between these frequencies.

2. Pacemaker according to claim 1, characterized in that the pulse shaper (10) is constructed as an orthogonal filter and has a negative group delay. 3. Cardiac pacemaker according to one of the preceding claims, characterized in that the pulse shaper contains a chain circuit of all-passports (APi, AP ₂ ), of whose input and output signals, portions via evaluation elements (B ₀ , Bi, B ₂ ) and a common adder (A) are fed to the pulse shaper output (12).

4. pacemaker according to claim 3, d a d u r c h g e k e n n z e i c h n e t that the all-passes (APi, AP) each cause a phase rotation of the input signal of 90 ° at about 15 Hz.

5. Pacemaker according to one of the preceding claims, ie that the pulse shaper (FIG. 3) has a high-pass character with a lower 3dB cut-off frequency of approximately 15 Hz.

6. pacemaker according to claim 5, characterized in that the connected at the input of the first allpass evaluation member (Rll, R12 in Figure 3) outputs signals to the adder (A), which are phase-shifted compared to the signals from the user ¬ tion element (P1-R13-R14) are supplied, which is connected to the output of this Allpass.

7. pacemaker according to claim 1, d a d u r c h g e k e n n z e i c h n e t that the pulse shaper has an upstream bandpass filter (Cl, R2, C2).

- 8. Cardiac pacemaker according to one of the preceding claims, characterized in that a switch (FET) is provided in the signal path behind the pulse shaper for brief signal path interruption with the occurrence of a stimulation pulse.

9. A pacemaker according to one of the preceding claims, ie that a circuit for discharging the coupling capacitor of the stimulation output stage (2) is provided in a period following a stimulation.

11. Cardiac pacemaker according to one of the preceding claims, characterized in that means (FIG. 4) are provided which vary at least one parameter of the pulse shaper depending on the period after which intracardiac signals recorded following a stimulation pulse can be processed again , and initially maintain this change when a shortening of the time period is achieved compared to a previously determined value, this value being determined by an amplitude criterion for the signal emitted by the pulse shaper in response to a stimulation pulse.

12. A pacemaker according to claim 11, d a d u r c h g e k e n n z e i c h n e t that the amplitude of the output signal of the first all-pass can be influenced as a parameter.

13. Cardiac pacemaker according to one of the preceding claims, characterized in that switching means (FIG. 4) are provided which reduce the energy of subsequent stimulation pulses, if on a preceding stimulation pulse after the expiry of the time period, after the intracardiac recorded following a stimulation pulse signals are again verarbeit¬ bar, a signal appears at the output of the pulse shaper er¬ which originates from a heart action, or set up ^'the stimulation amplitude if such a signal does not appear.

14. A pacemaker according to claim 13, which also means that the pulse energy is reduced after each, a number of successful stimulations, but is increased after a single failure of a cardiac action after stimulation.