Implantable medical device employing electrode integrity monitoring

Abstract

The present invention relates to an implantable medical device (1) for stimulating a human or animal heart, comprising a processor (32), a memory unit (33), a stimulation unit (34) configured to stimulate a human or animal heart, and a detection unit (31) configured to detect an electric signal of the same heart. The memory unit (33) comprises a computer-readable program that causes the processor (32) to perform the following steps when executed on the processor (32): a) detecting a cardiac electric signal during a first time period, the detected cardiac electric signal comprising a plurality of events (11); b) evaluating a morphologic characteristic (15, 16) of at least some events (11) of the plurality of events (11) regarding fulfillment of a predetermined criterion; c) incrementing a counter if the predetermined criterion is fulfilled, and d) triggering a storing of at least a part of the detected cardiac electric signal in the memory unit (33) if the counter reaches a counter threshold to allow a later evaluation of the stored part of the cardiac electric signal.

Description

[0001] IMPLANTABLE MEDICAL DEVICE EMPLOYING ELECTRODE INTEGRITY MONITORING

[0002] The present invention relates to an implantable medical device according to the preamble of claim 1 and to a method for operating an implantable medical device according to the preamble of claim 15.

[0003] Prior art describes a plurality of possibilities to detect electrode integrity problems (electrode malfunctions) of an implantable medical device for cardiac stimulation.

[0004] According to an impedance-based approach, exceeding an absolute threshold or falling below an absolute threshold during painless monitoring of the shock and stimulation activity of an implantable medical device is used to detect an electrode integrity problem. Furthermore, a quasi-continuous impedance monitoring can be performed. Then, conspicuous, strong impedance changes on a sample-to- sample basis at a comparatively high sampling rate can provide information on an electrode integrity problem.

[0005] According to a timing-based approach, a short interval counter can be applied. An increased number of unphy si ologi cal short intervals (having a duration of approximately 110 ms to 140 ms) triggers further evaluation of the detected signal or directly indicates a suspected electrode integrity problem. Other timing patterns in the detected electric signal can also be used for triggering further evaluation of the detected signal.

[0006] While impedance-based approaches provide a strong indicator that an electrode integrity problem exists, they allow a continuous monitoring of the electrode integrity only under significantly increased power consumption. Exclusively time-based approaches, on the other hand, require that the electrode fault can be detected as oversensing in a bipolar derivation, typically between the right ventricular tip electrode pole and the right ventricular ring electrode pole of the used electrode.

[0007] It is an object of the present invention to provide a continuous energy-saving monitoring of cardiac electric signals to identify indicators for electrode integrity problems and to allow a later thorough analysis of the respective signals.

[0008] This object is achieved with an implantable medical device for stimulating a human or animal heart according to claim 1. Such an implantable medical device comprises a processor, a memory unit, a stimulation unit, and a detection unit. The stimulation unit serves for stimulating a human or animal heart. The detection unit serves for detecting an electric signal of the same heart, i.e., a cardiac electric signal.

[0009] According to an aspect of the present invention, the memory unit comprises a computer- readable program that causes the processor to perform the steps explained in the following when executed on the processor.

[0010] In a first step, a cardiac electric signal is detected during a first time period. The detected cardiac electric signal comprises a plurality of events. An appropriate example of such a cardiac electric signal is an intracardiac electrogram (IEGM).

[0011] In a further method step, a morphologic characteristic (characteristic feature) of at least some events of the plurality of events is evaluated. This evaluation serves to determine whether a predetermined criterion is fulfilled by one or more events of the plurality of events.

[0012] According to an embodiment, if the predetermined criterion is fulfilled, a counter (e.g. aligned with a daily analysis period) is incremented. According to an embodiment, the trend of stored daily counters can be analyzed and used as indicator for further activities, for instance by transmitting the data to an external server where the data is analyzed. The data analysis can be used for remote monitoring of the patient. Finally, storing (saving) at least a part of the detected cardiac electric signal in the memory unit is triggered if the counter reaches a counter threshold. Triggering such storing allows a later evaluation of the stored part of the cardiac electric signal, i.e., a deeper analysis of the cardiac electric signal, e.g., outside the implantable medical device with a computer having more computing capacity and being less limited in terms of power consumption than the implantable medical device is.

[0013] According to an embodiment of the invention, the counter threshold is in a range from 1 to 100, in particular 5 to 90, in particular 10 to 80, in particular 20 to 70, in particular 30 to 60, in particular 40 to 50.

[0014] For instance, the counter is incremented once in a predetermined time, as for instance once per 12 hours, 24 hours, or 48 hours, or 72 hours, if the predetermined criterion is fulfilled within said predetermined time.

[0015] This implantable medical device represents a highly performant system for early detection of electrode integrity problems (electrode failures) based on an evaluation of the detected cardiac electric signal, e.g., a detected IEGM. Since the system is open for a plurality of different predetermined criteria that are used for triggering the storing of at least a part of the detected cardiac electric signal, it enables an independent adjustment of the individual episodes triggers, i.e., the triggers of storing at least a part of the detected cardiac electric signal. Furthermore, the implantable medical device does not require a specific design of its system components. They can be chosen and operated such that the overall sensitivity and the overall specificity can be adjusted to the respective needs of the intended kind of application of the implantable medical device. Thus, irrespective of enabling a detailed analysis of suspicious areas of the detected cardiac electric signal to detect electrode integrity problems, the implantable medical device enables a 24 / 7 monitoring of a patient's heart (i.e., it can be operated with a high overall sensitivity). In addition, it does not require a complex sorting-out of false-positive trigger events for storing the detected cardiac electric signal that takes place when applying more demanding approaches or a higher calculation complexity based on a predetermined dataset (i.e., the implantable medical device can be operated with a high specificity). For the present application, the term electrode is used as synonym for an electrode lead or stimulation lead. It is an electrical device, comprising an elongated lead body which is electrically insulated, having a plurality of electrode poles which are located at the distal end of the lead body or elsewhere along the lead body. The electrode poles are used for sensing cardiac signals and / or stimulating cardiac tissue. The electrode has a connector at the proximal end for connecting the electrode with an implantable medical device.

[0016] It is not important for the presently claimed and described subject matter how the stored cardiac electric signal is finally evaluated or analyzed. Rather, the presently described and claimed solution merely aims in providing sensible and reliable causes of triggering the storing of at least a part of the detected cardiac electric signal in case of a suspected electrode integrity problem. The stored signal can afterwards be analyzed to find out whether there is indeed an electrode integrity problem. For instance, the analysis is carried out by an external server. According to an embodiment, the evaluation of the signals can be performed using various analysis tools and classifications of events, without the need of interacting with each independent implantable medical device.

[0017] In an embodiment, the implantable medical device is an implantable pulse generator (IPG), an implantable cardioverter-defibrillator (ICD), or a device for cardiac resynchronization therapy (CRT).

[0018] In an embodiment, the morphologic characteristic is an area of a signal section representing one of the events. Such an area can also referred to as area under the curve. An interesting finding from the applicant’s research efforts is that such an area is a particularly reliable measure for detecting an electrode integrity problem. It is thus particularly appropriate for solving the present object. In particular, when it comes to a lead failure due to an intermittent metal to metal contact (e.g. between two conductors of a lead, of multiple leads, of a lead conductor and the active implant's housing), the signal could be saturated for a while, followed by a decaying progression, which results in a large area under the curve representing the signal. In an embodiment, the peak-to-peak distance of the greatest peak of the signal section representing one of the events is used as morphologic characteristic. Typically, the R-wave of the cardiac electric signal represents the greatest peak. Such a peak-to-peak distance can be even easier determined than an area under the curve and also turned out to be a particular appropriate measure for detecting an electrode integrity problem, in particular in case of a peak-to-peak distance that is lower than expected.

[0019] According to an embodiment, the morphologic characteristic is an occurrence of a peak-to- peak distance below a lower threshold. The lower threshold is for instance determined according to a percentage of the measurement range for detecting the cardiac electric signal. According to an embodiment, the lower threshold lies in a range of 1-30% of the full measurement range for detecting the cardiac electric signal, preferably 2-5%. Alternatively, the lower threshold is determined according to a percentage value of peak-to-peak distances of previous cardiac electric signals, in particular a mean value calculated from previous cardiac electric signals. For instance, the mean value is calculated from 5-30, in particular 10, 20, or 30 previous cardiac electric signals. The threshold value is, according to embodiments, 1-30% of the mean value, preferably 2-10%.

[0020] In an embodiment, the predetermined criterion is fulfilled if not more than a certain percentage of cardiac electric signals among a plurality of successive cardiac electric signals show a peak-to-peak distance below a lower threshold. For instance, if not more than 1-30% of the cardiac electric signals within a plurality of cardiac electric signals comprise a peak- to-peak distance below a threshold value, preferably 2-8 percent, preferably 5%. For instance, the predetermined criterion is fulfilled if 5% of 20 cardiac signals have a peak-to- peak distance below the threshold value, that is 1 cardiac signal out of 20.

[0021] According to an embodiment, the respective cardiac electric signals are discarded from an analysis if a specific number of peak-to-peak distances fall below a minimum threshold. For instance, if 50-95% of a number of cardiac signals (e.g. 10-50) have a peak-to-peak distance below a minimum threshold, it is assumed that the measurement level for the cardiac electric signals is not sufficiently high for a reliable measurement outcome. In an embodiment , the cardiac electric signals are discarded from an analysis if a mean value of peak-to-peak distances fall below a minimum threshold.

[0022] In an embodiment, the morphologic characteristic is a spike difference vector comprising a peak-to-peak distance between two adjacent peaks of a signal section representing one of the events. This spike difference vector is - unlike the peak-to-peak distance between the highest and the lowest peak of the signal section - a measure that relies on a peak-to-peak distance between arbitrary adjacent peaks of the signal section. It can well be combined with other spike difference vectors of the same signal section to present a full image of the whole signal section rather than of its highest and lowest values. Therefore, the spike difference vector covers a broader physiologic range that might be important for an early detection of an electrode integrity problem.

[0023] In an embodiment, the morphologic characteristic is a normalized spike difference vector. This normalized spike difference vector is calculated by dividing the spike difference vector by the peak-to-peak distance between the highest and the lowest peak of the signal section. Thus, all individual peak-to-peak distances between adjacent peaks are normalized to the highest peak-to-peak difference of the signal section.

[0024] In an embodiment, the morphologic characteristic is a difference between the individual peak-to-peak distances between the highest and lowest peak of the respective signal section of a predetermined number of subsequent events (i.e., of subsequent signal sections). If the difference between the individual peak-to-peak distances is high, this indicates that individual signal sections or events represent a “correct” physiologic signal, whereas others represent a distorted signal that is due to an impaired electrode function.

[0025] In an embodiment, the morphologic characteristic is the percentage of differences exceeding a predetermined threshold, wherein the differences are calculated between the peak-to-peak distances of adjacent pairs of a predetermined number of subsequent events. In this context, the peak-to-peak distances are again calculated between the highest and the lowest peak of the signal section. If the percentage of differences exceeding the predetermined threshold is high, this indicates an undesired and non-physiologic variation between adjacent pairs of events that is due to an electrode malfunction. In an embodiment, the predetermined threshold that is exceeded if the difference between the peak-to-peak distances lies in a range of from 30 % to 100 % difference, in particular 40 % to 90 % difference, in particular 50 % to 80 % difference, in particular 60 % to 70 % difference.

[0026] In an embodiment, the morphologic characteristic is an absolute or normalized difference between the maximum positive slope and the maximum negative slope of the curve representing the signal section.

[0027] In an embodiment, the morphologic characteristic is the absolute or normalized area (i.e., area under the curve) of the signal section between the points of maximum positive slope and maximum negative slope of the signal section.

[0028] In an embodiment, the morphologic characteristic is the skew of the distribution of the amplitude values of the peaks within the signal section.

[0029] In an embodiment, the morphologic characteristic is the absolute normalized mean deviation of a peak amplitude from the mean peak amplitude of the respective signal section.

[0030] In an embodiment, the morphologic characteristic is the highest absolute or normalized value of an extremum of the peaks of the signal section.

[0031] In an embodiment, the morphologic characteristic is the kurtosis of the distribution of the amplitude values of the peaks of the signal section.

[0032] In an embodiment, the morphologic characteristic is the mean signal amplitude normalized by the maximum absolute peak value of the signal section.

[0033] In an embodiment, the morphologic characteristic is the absolute or normalized maximum positive slope of the curve representing the signal section. In an embodiment, the morphologic characteristic is the mean of the absolute slope normalized by the maximum absolute peak value of the signal section.

[0034] In an embodiment, the morphologic characteristic is the length of an interval between the points of maximum positive slope and maximum negative slope of the signal section.

[0035] In an embodiment, the morphologic characteristic is the absolute or normalized area of the negative signal portions of the signal section.

[0036] In an embodiment, the morphologic characteristic is the absolute or normalized area of the positive signal portions of the signal section.

[0037] In an embodiment, the morphologic characteristic is the absolute or normalized maximum negative slope of the curve describing the signal section.

[0038] In an embodiment, the morphologic characteristic is the number of amplitude values of the peaks of the signal section that are higher than 0.5 to 1.0 times, in particular 0.6 to 0.9 time, in particular 0.7 to 0.8 times a maximum amplitude value of the highest peak of the signal section.

[0039] In an embodiment, the morphologic characteristic is the length of the Q-R interval.

[0040] In an embodiment, the morphologic characteristic is the absolute or normalized sum of the absolute values of the differences between adjacent signal sections, in particular differences of the maximum peak amplitudes between adjacent signal sections.

[0041] In an embodiment, the morphologic characteristic is the absolute or normalized slope between the Q point and the R point of the signal section.

[0042] In an embodiment, the morphologic characteristic is the absolute or normalized slope between the Q point and the S point of the signal section. In an embodiment, the morphologic characteristic is the absolute or normalized slope between the R point and the S point of the signal section.

[0043] In an embodiment, the morphologic characteristic is the length of the Q-S interval of the signal section.

[0044] In an embodiment, the morphologic characteristic is the length of the R-S interval of the signal section.

[0045] In an embodiment, the morphologic characteristic is the number of zero crossings of the curve describing the signal section.

[0046] The maximum peak amplitude (also referred to as maximum signal value or maximum peak value) is particularly appropriate for normalizing any of the before-mentioned absolute values to obtain a normalized value. In an embodiment, the maximum absolute peak value is used for calculating a normalized value from an absolute value of the parameter that is used as morphologic characteristic according to any of the precedingly explained embodiments.

[0047] All of the precedingly explained embodiments of the morphologic characteristic are likewise very well suited to be used for detecting an electrode integrity problem of the implantable medical device.

[0048] In an embodiment, the predetermined criterion comprises exceeding a predetermined upper threshold or falling below a predetermined lower threshold. In this context, the threshold can be set according to the respective needs. It can be a fixed threshold or a variable threshold that is adjusted according to the course of the cardiac electric signal over a predetermined time period. Then, it is possible to adapt the threshold to the specific needs of an individual patient and thus to provide a particularly physiologic criterion, the fulfilment of which is decisive for triggering the storing of at least a part of the detected cardiac electric signal. In an embodiment, the computer-readable program causes the processor to evaluate whether the areas of a plurality of signal sections exceed an upper threshold. In this context, each signal section represents one of the events. In addition, the predetermined criterion is fulfilled if at least one area exceeds the upper threshold. Thus, this embodiment relies on the morphologic characteristic of a conspicuous area under the curve for triggering the storing of at least a part of the detected cardiac electric signal. It turned out that the area under the curve is a particularly appropriate measure for determining electrode integrity problems at an early stage and is thus very appropriate to solve the present object.

[0049] In an embodiment, the computer-readable program causes the processor to consider the predetermined criterion as being fulfilled only if x areas of y subsequent areas exceed the upper threshold. In this context, x is 1 to 4, in particular 2 to 3. In addition, y is 2 to 40, in particular 3 to 35, in particular 4 to 30, in particular 5 to 25, in particular 6 to 20, in particular 7 to 15, in particular 8 to 12, in particular 9 to 10. If the implantable medical device does not rely on a single evaluated area only, the probability that there is indeed an electrode integrity problem is significantly increased as in case of relying on a single area only. However, even when only a single area exceeds the predetermined threshold, this is a strong indicator that there is an electrode integrity problem that needs to be further examined and that might require at the very end an exchange of the electrode or of the implantable medical device with connected electrode.

[0050] In an embodiment, the computer-readable program causes the processor to consider the predetermined criterion as being fulfilled only if additionally a relative difference between a) a mean or median of the area of all y subsequent events on the one hand and b) a mean or median of all x areas exceeding the upper threshold on the other hand lies in a range of from 10 % to 200 %, in particular from 15 % to 190 %, in particular from 20 % to 180 %, in particular from 30 % to 170 %, in particular from 40 % to 160 %, in particular from 50 % to 150 %, in particular from 60 % to 140 %, in particular from 70 % to 130 %, in particular from 80 % to 120 %, in particular from 90 % to 110 %, in particular from 100 % to 105 %. This additional requirement decreases the sensitivity of the detection of an electrode integrity problem but increases the specificity. An inadvertent exceeding of the upper threshold by an evaluated area (which is, however, not due to an electrode integrity problem) can thus be easier disregarded, wherein relevant exceeding events of the upper threshold will still be detected.

[0051] In an embodiment, the computer-readable program causes the processor to normalize the areas of the plurality of signal sections. This is done by dividing each of the areas by a peak- to-peak distance, in particular by the peak-to-peak distance between the highest and the lowest peak, of the signal section for which the respective area has been determined. Such normalization results in a normalized area that can be better compared to other normalized areas than non-normalized areas can be compared.

[0052] In an embodiment, the computer-readable program causes the processor to evaluate whether the peak-to-peak differences of the plurality of signal sections fall below a lower threshold. As explained above, each signal section represents one of the events of the cardiac electric signal. In this embodiment, the predetermined criterion is fulfilled if 1 % to 100 %, in particular 1% to 50%, in particular 1% to 30%, in particular 5% to 20% percent. If not more than such number of peak-to-peak differences fall below the lower threshold, this is typically an indication of an electrode integrity problem. According to an embodiment, if a certain higher number of peak-to-peak differences fall below a lower threshold, a low signal quality can be the reason. In such cases, the respective cardiac signal shall be discarded from evaluation. For instance, if 30% 100%, in particular 30% to 90%, in particular 40% to 80%, in particular 45% to 75%, in particular 50% to 60% of the peak-to-peak differences of the plurality of signal sections fall below a lower threshold, the respective cardiac signals shall be discarded from further evaluation.

[0053] In an embodiment, the plurality of signal sections comprises the latest n sections, wherein n lies in a range from 2 to 40, in particular from 3 to 35, in particular from 4 to 30, in particular from 5 to 25, in particular from 6 to 20, in particular from 7 to 15, in particular from 8 to 12, in particular from 9 to 10.

[0054] In an embodiment, the computer-readable program causes the processor to analyze a plurality of events of the detected cardiac electric signal, wherein the analysis window is defined for each event to be analyzed. In this context, the analysis window covers ± 200 ms, in particular ± 150 ms, in particular ± 100 ms, in particular ± 50 ms around an event sensing signal. This event sensing signal indicates that the event has detected by the detection unit.

[0055] In an embodiment, the analysis window is asymmetrically arranged around the event sensing signal. In an embodiment, the smaller part of the analysis window lies before the event sensing signal, wherein a bigger part of the analysis window lies after the event sensing signal. In an embodiment, the analysis window covers a timeframe from -150 ms to ±200 ms around the event sensing signal, in particular a timeframe from -100 ms to ±150 ms around the event sensing signal. Then, all peaks of the signal section (i.e., of the event within the electric cardiac signal) lie within the analysis window and can thus be properly evaluated.

[0056] In an embodiment, the detection unit comprises an electrode having a first electrode pole, in particular a distal tip electrode pole, a second electrode pole, in particular a ring electrode pole, that is located proximally of the first electrode pole, and a third electrode pole, in particular a coil electrode pole, that is located proximally of the second electrode pole. By such an electrode arrangement, a first sensing vector is defined from the second electrode pole to the first electrode pole. In addition, a second sensing vector is defined from the third electrode pole to a housing of the implantable medical device serving as counter electrode. The second sensing vector is in some cases called the “far field” signal. It is then particularly easy possible to derive electric signals along the first sensing vector and / or along the second sensing vector. In doing so, cardiac electric signals derived along the first sensing vector are near field cardiac electric signals, wherein the signals sensed along the second sensing vector are far field cardiac electric signals. These electric signals have different electrical properties. A combination of those signals is particularly appropriate to properly evaluate any impairment of electrode function and thus particularly electrode integrity.

[0057] In an embodiment, the event sensing signal is generated based on cardiac electric signals sensed along the first sensing vector. In addition, the event of the cardiac electric signal is analyzed based on cardiac electric signals sensed along the second sensing vector. Thus, the signals sensed along the first sensing vector are used as trigger for defining an analysis window in which the cardiac electric signals sensed along the second sensing vector are analyzed. This splitting of triggering signals on the one hand and signals to be analyzed on the other hand allows a particularly reliable detection of electrode integrity problems. In an embodiment, the cardiac electric signal is analyzed with a sampling frequency lying in a range of from 60 Hz to 550 Hz, in particular from 64 Hz to 512 Hz, in particular from 100 Hz to 300 Hz, in particular from 128 Hz to 256 Hz.

[0058] In an embodiment, the word width used for each sampling value lies in a range of from 5 bits to 10 bits, in particular from 6 bits to 9 bits, in particular from 7 bits to 8 bits. A particular appropriate combination is a sampling frequency of 128 Hz and a word width of 8 bit. Then, 32 sample values can be obtained within 250 ms, wherein each sample value can adopt a value in the range from -127 to +128 (or, alternatively, from -128 to +127), i.e., covering 256 values including 0.

[0059] In an embodiment, the computer-readable program causes the processor, upon being triggered to do so, to store the latest 5 to 60 seconds, in particular the latest 10 to 50 seconds, in particular the latest 15 to 45 seconds, in particular the latest 20 to 40 seconds, in particular the latest 25 to 30 seconds of the cardiac electric signal before that event that caused the fulfilment of the predetermined criterion. Typically, such range of the detected cardiac electric signal is fully sufficient to evaluate by a subsequent analysis whether there is indeed an electrode integrity problem as assumed by the internal evaluation of the events of the cardiac electric signal.

[0060] In an embodiment, the computer-readable program causes the processor to store the next 1 to 10 seconds, in particular the next 2 to 9 seconds, in particular the next 3 to 8 seconds, in particular the next 4 to 7 seconds, in particular the next 5 to 6 seconds of the detected cardiac electric signal after that event that caused the fulfilment of the predetermined criterion. Thus, in this embodiment, the implantable medical device does not only store the history of the event that was considered to be indicative for an electrode integrity problem, but also the post-history. This allows an even better subsequent evaluation of the cardiac electric signal and a distinction between an unusual course of the signal due to a physiologic reason and an unusual course of the signal due to the suspected electrode integrity problem. In an embodiment, the implantable medical device does not only store the detected cardiac electric signal in the memory unit upon receiving the trigger signal upon fulfillment of the predetermined criterion, but also additional information such as marker time points and marker types. This additional information is stored, in an embodiment, in a data-reduced or compressed manner to save storing space and to reduce the power being necessary for storing, reading and / or transferring such additional data.

[0061] In an embodiment, the memory unit comprises a ring buffer for short-term caching the detected cardiac electric signal. In addition, the memory unit comprises a long-term memory, wherein the storing of at least a part of the detected cardiac electric signal comprises writing the part of the detected cardiac signal to be stored from the ring buffer into the long-term memory. This guarantees that the stored part of the detected cardiac electric signal is available at a later time for evaluation remote from the implantable medical device. The long-term memory can be generally a volatile or a non-volatile memory. To reduce the risk of data loss, it is typically a non-volatile memory.

[0062] In an embodiment, the implantable medical device comprises a data communication unit that enables a data transfer of data stored within the memory unit to an external device or remote system. Such external device is, in an embodiment, a telemedicine system such as a home monitoring system. The data transferred from the implantable medical device to this remote system can then be evaluated on the remote system. To allow a better visualization for the medical staff engaged with the evaluation of the data, the events in the cardiac electric signals having suspicious features can be displayed in a highlighted manner.

[0063] In an embodiment, a data transfer takes place immediately after having stored the part of the cardiac electric signal into the memory unit. In an embodiment, the stored cardiac electric signal is transferred to the remote system upon the next opportunity to do so, i.e., upon confirmed contact with a transfer unit forming part of a data transfer network and / or being operatively coupled with the remote monitoring system.

[0064] In an embodiment, the data communication unit serves for transferring data to the remote system in a wireless manner. All standard data transmission protocols or specifications are appropriate for such a wireless data communication. Examples of standard data transmission protocols or specifications are the Medical Device Radiocommunications Service (MICS), the Bluetooth Low Energy (BLE) protocol, the Zigbee specification, the long range wide area network (LoRaWAN) protocol, the wireless personal area network (WPAN) specification, the low-power wide-area network (LPWAN) specification, the wireless local area network (WLAN) specification, the Global System for Mobile Communications (GSM) specification, the Long-Term Evolution (LTE) standard, and the fifth-generation technology standard for broadband cellular networks (5G).

[0065] In an aspect, the present invention relates to a method for operating an implantable medical device for stimulating a human or animal heart, in particular for operating an implantable medical device according to the preceding explanations. This method is characterized by the steps explained in the following.

[0066] In a first step, a cardiac electric signal is detected during a first time period. The detected cardiac electric signal comprises a plurality of events. An appropriate example of such a cardiac electric signal is an intracardiac electrogram (IEGM).

[0067] In a further method step, a morphologic characteristic of at least some events of the plurality of events is evaluated. This evaluation serves to determine whether a predetermined criterion is fulfilled by one or more events of the plurality of events.

[0068] Finally, storing at least a part of the detected cardiac electric signal in the memory unit is triggered if the predetermined criterion is fulfilled. Triggering such storing allows a later evaluation of the stored part of the cardiac electric signal, i.e., a deeper analysis of the cardiac electric signal, e.g., outside the implantable medical device with a computer having more computing capacity and being less limited in terms of power consumption than the implantable medical device is.

[0069] All embodiments of the implantable medical device can be combined in any desired way and can be transferred either individually or in any arbitrary combination to the described method. Likewise, all embodiments of the described method can be combined in any desired way and can be transferred either individually or in any arbitrary combination to the implantable medical device.

[0070] Further details of aspects of the present invention will be explained in the following making reference to exemplary embodiments and accompanying Figures. In the Figures:

[0071] Figure 1 A schematically shows a system comprising an implantable medical device;

[0072] Figure IB schematically shows different components of the implantable medical device of Figure 1A;

[0073] Figure 2A shows a first electric derivation obtained along a first sensing vector illustrated in Figure 1A;

[0074] Figure 2B schematically shows a first electric derivation obtained along a second sensing vector illustrated in Figure 1 A;

[0075] Figure 3A shows a second electric derivation obtained along a first sensing vector illustrated in Figure 1 A;

[0076] Figure 3B schematically shows a second electric derivation obtained along a second sensing vector illustrated in Figure 1 A;

[0077] Figure 3 C shows a graphic representation of the values of areas of the individual events of Figure 3B;

[0078] Figure 4A shows again the second electric derivation obtained along a first sensing vector illustrated in Figure 1 A;

[0079] Figure 4B schematically shows again the second electric derivation obtained along a second sensing vector illustrated in Figure 1 A; Figure 4C shows a graphic representation of the values of areas of the individual events of Figure 4B;

[0080] Figure 5 A shows a third electric derivation obtained along a first sensing vector illustrated in Figure 1A;

[0081] Figure 5B schematically shows a third electric derivation obtained along a second sensing vector illustrated in Figure 1 A;

[0082] Figure 5C shows the peak-to-peak distances of the individual events of the cardiac electric signal of Figure 5B; and

[0083] Figure 6 is a schematic illustration of the stored parts of the detected cardiac electric signal.

[0084] Figure 1A shows a system comprising a device for cardiac pacing, including a cardiac defibrillation function (implantable cardioverter defibrillator, ICD device) 1 as example of an implantable medical device for stimulating the human or animal heart. The system further comprises a home monitoring system 2 serving as remote system. It is possible for the implantable medical device 1 to establish a wireless data communication with the home monitoring system 2.

[0085] The implantable medical device 1 comprises a housing 3 with a header 4 and an electrode 5 connected to the header 4. The electrode 5 comprises a tip electrode pole 6, a ring electrode pole 7 that is proximally arranged from the tip electrode pole 6, and a coil electrode pole 8 that is proximally arranged from the ring electrode pole 7. Due to the different electrode poles 6, 7, 8 of the electrode 5, a first sensing vector 9 and a second sensing vector 10 are defined. The first sensing vector 9 extends from the ring electrode pole 7 to the tip electrode pole 6. The second sensing vector 10 extends from the coil electrode pole 8 to the housing 3. While electric cardiac signals sensed between the tip electrode pole 6 and the ring electrode pole 7, i.e., along the first sensing vector 9, are directly recorded within a heart chamber, typically the right ventricle, the cardiac electric signals derived along the second sensing vector 10 are far field signals.

[0086] Figure IB schematically illustrates individual components of the implantable medical device 1 that are comprised within the housing 3. Thus, the housing 3 houses a detection unit 31 (also referred to as sensing unit) that comprises an analog-to-digital converter, a bandpass filter, and an offset compensation. The detection unit 31 is operatively connected with a processor 32 that has access to a memory unit 33. The memory unit 33 serves for storing instructions for the processor 32 as well as data detected by the detection unit 31. The housing 3 further comprises an evaluation unit 34 that can also be part of the processor 32 and that serves for extracting features from the detected cardiac electric signal. The housing 3 further comprises a stimulation unit 35 that serves for stimulating the heart from which the detection unit 31 detects electric signals. The electrode 5 along with its electrode poles 6, 7, 8 (confer Figure 1A) forms part of the detection unit 31 and the stimulation unit 35. Additionally, the housing 3 comprises a data communication unit 36 that serves for data transfer to the home monitoring system 2 (confer Figure 1 A).

[0087] Figure 2A a shows an electric derivation obtained along the first sensing vector 9 illustrated in Figure 1A. The sensing unit 31 (confer Figure IB) has detected four different events 11 and has assigned an event sensing signal 12 to each of these events 11. In this context, the individual event sensing signals 12 are placed at the beginning of each of the events 11. In addition, an analysis window 13 is defined around each of the event sensing signals 12. Each of the analysis windows 13 extends from -100 ms to +150 ms around the respective event sensing signal 12.

[0088] Figure 2B shows an electric derivation of the same cardiac activity that caused the cardiac electric signal of Figure 2A. However, the signal represented in Figure 2B is obtained along the second sensing vector 10 (confer Figure 1A), i.e., it represents a far field signal. Here, each event 11 is represented by a signal section 14. Thus, four sections 14 each having a plurality of peaks form part of the electric derivation illustrated in Figure 2B. Typically, each signal section 14 comprises a P wave 141, a Q wave 142, an R wave 143, an S wave 144, and a T wave 145. Each of these waves can also be denoted as peak. In the depiction of Figure 2B, only the individual peaks of one of the four signal sections 14 carries the corresponding numeral references for better visualization.

[0089] The hatched area 15 of the first signal section 14 is the area under the curve representing the derived cardiac electric signal. In the embodiment shown in Figure 2B, the cardiac electric signal is sampled with a sampling rate of 128 Hz. Since the analysis window 13 has a width of 250 ms (100 ms before the event sensing signal 12 plus 150 ms after the event sensing signal 12), each signal section 14 is described by 32 sample values. By applying a word width of 8 bit, each of these 32 sample values can adopt a value between -127 and +128. Consequently, a total area of 32*128 = 4096 is possible. In the exemplary embodiment shown in Figure 2B, the storing of the detected cardiac electric signal shall be triggered if the area exceeds a threshold of 2500 (corresponding to approximately 60 % of the theoretic maximum of the area 15 of each individual signal section 14).

[0090] Another criterion for triggering storing the cardiac electric signal is an evaluation of the peak-to-peak distance 16 that represents the distance between the highest amplitude of a peak of the signal section 14 and the lowest amplitude of a peak of the signal section 14. Typically, the R wave 143 has the highest amplitude, wherein the S wave 144 typically has the lowest amplitude. In such a case, the peak-to-peak distance 16 extends from the maximum value of the R wave 143 to the minimum value of the S wave 144.

[0091] Concrete exemplary embodiments of an evaluation of the area 15 or the peak-to-peak distance 16 for triggering the storing of the detected cardiac electric signal will be explained in connection with Figures 3 A to 5C. In these and in all other Figures, similar elements will be denoted with the same numeral reference.

[0092] Figure 3A corresponds to Figure 2A so that reference is made to the explanations given above with respect to Figure 2A.

[0093] The far field signals detected along the second sensing vector 10 (confer Figure 1A) are represented in Figure 3B. The areas 15 of the individual signal sections 14 are calculated as explained with respect to Figure 2B. The resulting area values are depicted in Figure 3C. The values of the areas 15 of the first, the second and the third signal section 14 lie below a threshold 17 which is set to 2500 (confer explanations with respect to Figure 2B). However, the area 15 of the fourth signal section 14 has a value exceeding the threshold 17. This is taken as trigger for storing the detected cardiac electric signal to allow a later evaluation. The unusual high area 15 of the fourth signal section 14 is indicative of an impaired electrode integrity.

[0094] Figures 4A to 4C illustrate a similar embodiment like Figures 3A to 3C. In fact, Figure 4A corresponds to Figure 3 A, and Figure 4B corresponds to Figure 3B. Therefore, reference is made to the explanations given above with respect Figure 3 A and Figure 3B.

[0095] Figure 4C illustrates that the predetermined threshold 17 is not the only criterion for triggering storing of the detected cardiac electric signal. Rather, an average value 18 is calculated from the four area values calculated in this exemplary embodiment (generally, any other number of areas for calculating the average value 18 can be used). Only if a predetermined distance 19 to the average value 18 is exceeded by the value of a specific area 15, storing of the detected cardiac electric signal is triggered. As can be seen from Figure 4C, the fourth signal section 14 has an area that exceeds both the threshold 17 and the distance 19 to the average 18. Consequently, also in this embodiment, storing of the detected cardiac electric signal is triggered to allow a later evaluation of the signal.

[0096] Figures 5 A to 5C illustrate a further embodiment, wherein Figure 5 A corresponds to Figures 2A, 3A and 4A so that reference is made to the explanations given above with respect to those Figures.

[0097] As can be seen from Figure 5B, this embodiment does not rely on a calculation of the area 15, but rather on the peak-to-peak distance 16. This peak-to-peak distance 16 is calculated for each of the signal section 14 detected in the cardiac electric signal derived along the second sensing vector 10 (confer Figure 1 A), as illustrated in Figure 5B.

[0098] Figure 5C illustrates the values of the calculated peak-to-peak distances for each of the four signal sections 14. Here, the peak-to-peak distances of the first, the second and the fourth signal section 14 lie above the threshold 17, whereas the peak-to-peak distance of the third signal section 14 falls below the threshold 17. This is taken as the trigger for storing the detected cardiac electric signal for later evaluation since such an unusual low peak-to-peak distance is indicative for an impaired electrode functionality or lacking electrode integrity. According to an embodiment, the trigger for storing the detected cardiac electric signal is only activated if not more than a certain percentage of cardiac electric signals among a plurality of successive cardiac electric signals show a peak-to-peak distance below threshold 17. For instance, if not more than 1-30% of the cardiac electric signals within a plurality of cardiac electric signals comprise a peak-to-peak distance below threshold 17, preferably 2- 8 percent, preferably 5%.

[0099] According to an embodiment, the cardiac electric signals are discarded from an analysis if a specific amount of peak-to-peak distances fall below a minimum threshold (not depicted). For instance, if 50-95% of a number of 10-50 cardiac signals have a peak-to-peak distance below a minimum threshold, it is assumed that the measurement level for the cardiac electric signals is not sufficiently high for a reliable measurement outcome. In an embodiment, the cardiac electric signals are discarded from an analysis if a mean value of peak-to-peak distances fall below a minimum threshold.

[0100] Figure 6 illustrates an embodiment of storing the detected cardiac electric signal. In this embodiment, both the history 20 and the post-history 21 around that event 11 that caused triggering of storing of the cardiac electric signal are stored. Typically, the history 20 comprises 5 to 60 seconds of the cardiac electric signal, wherein the post-history comprises 1 to 10 seconds of the cardiac electric signal. Such a part of the detected cardiac electric signal is typically sufficient to subsequently analyze the signal more thoroughly for an electrode integrity problem. This is typically done in the remote monitoring system 2 (confer Figure 1A) since this remote monitoring system 2 allows computationally more complex feature extractions and classifications from the stored cardiac electric signal than this is possible within the implantable medical device 1.

Claims

Claims1. Implantable medical device (1) for stimulating a human or animal heart, comprising a processor (32), a memory unit (33), a stimulation unit (34) configured to stimulate a human or animal heart, and a detection unit (31) configured to detect an electric signal of the same heart, characterized in that the memory unit (33) comprises a computer-readable program that causes the processor (32) to perform the following steps when executed on the processor (32): a) detecting a cardiac electric signal during a first time period, the detected cardiac electric signal comprising a plurality of events (11); b) evaluating a morphologic characteristic (15, 16) of at least some events (11) of the plurality of events (11) regarding fulfillment of a predetermined criterion; c) incrementing a counter if the predetermined criterion is fulfilled, and d) triggering a storing of at least a part of the detected cardiac electric signal in the memory unit (33) if the counter reaches a counter threshold to allow a later evaluation of the stored part of the cardiac electric signal.

2. Implantable medical device according to claim 1, characterized in that the morphologic characteristic (15, 16) is chosen from the group consisting of an area (15) of a signal section (14) representing one of the events (11); a peak-to-peak distance (16) between the highest peak (143) and the lowest peak (144) of a signal section (14) representing one of the events (11); a spike difference vector comprising a peak-to- peak distance between two adjacent peaks of a signal section (14) representing one of the events (11); a normalized spike difference vector that is calculated by dividing the spike difference vector by the peak-to-peak distance; a difference between the peak- to-peak distances between the highest peak (143) and the lowest peak (144) of a signal section (14) of a predetermined number of subsequent events (11); and a percentage of differences between the peak-to-peak distances between the highest peak (143) and the lowest peak (144) of a signal section (14) of adjacent pairs of a predetermined number of subsequent events (11), wherein the differences exceed a predetermined threshold (17).

3. Implantable medical device according to claim 1 or 2, characterized in that the computer-readable program causes the processor (32) to evaluate whether the areas (15) of a plurality of signal sections (14) exceed an upper threshold (17), wherein each signal section (14) represents one of the events (11), wherein the predetermined criterion is fulfilled if at least one area (15) exceeds the upper threshold (17).

4. Implantable medical device according to claim 3, characterized in that the computer- readable program causes the processor (32) to consider the predetermined criterion as being fulfilled only if x areas (15) of y subsequent plurality of signal sections (14) exceed the upper threshold (17), wherein x is 1 to 4, and y is 2 to 40.

5. Implantable medical device according to claim 4, characterized in that the computer- readable program causes the processor (32) to consider the predetermined criterion as being fulfilled only if additionally a relative difference (19) between i) a mean (18) or median of the areas (15) of all y subsequent events and ii) a mean or median of all x areas exceeding the upper threshold lies in a range from 10 % to 200 %.

6. Implantable medical device according to any of claims 3 to 5, characterized in that the computer-readable program causes the processor (32) to normalize the areas (15) of the plurality of signal sections (14) by dividing each of the areas (15) by a peak-to- peak distance of the signal section (14) for which the respective area (15) has been determined.

7. Implantable medical device according to claim 1 or 2, characterized in that the computer-readable program causes the processor (32) to evaluate whether the maximum peak-to-peak distance (16) of a plurality of signal sections (14) falls below a lower threshold (17), wherein each signal section (14) represents one of the events (11), wherein the predetermined criterion is fulfilled if 1 % to 30% of the peak-to- peak distances (16) fall below the lower threshold (17).

8. Implantable medical device according to claim 7, characterized in that the plurality of signal sections (14) comprises the latest 2 to 100 sections (14).

9. Implantable medical device according to claim 7 or 8, characterized in that the predetermined criterion is fulfilled if not more than 5% of the peak-to-peak distances (16) fall below the lower threshold (17).

10. Implantable medical device according to any of the claims 7 to 9, characterized in that the respective cardiac electric signals are discarded from an analysis if 50%-95% of 10 to 50 peak-to-peak distances fall below a minimum threshold.

11. Implantable medical device according to any of the preceding claims, characterized in that the computer-readable program causes the processor (32) to analyze a plurality of events (11) of the detected cardiac electric signal, wherein an analysis window (13) is defined for each event (11) to be analyzed, wherein the analysis window (13) covers ± 200 ms around an event sensing signal (12) indicating that the event (11) has been detected by the detection unit (31).

12. Implantable medical device according to any of the preceding claims, characterized in that the detection unit (31) comprises an electrode (5) having a first electrode pole (6), a second electrode pole (7) located proximally of the first electrode pole (6), and a third electrode pole (8) located proximally of the second electrode pole (7), wherein a first sensing vector (9) is defined from the second electrode pole (7) to the first electrode pole (6), and wherein a second sensing vector (10) is defined from the third electrode pole (8) to a housing (3) of the implantable medical device (1).

13. Implantable medical device according to claim 9 and 10, characterized in that the event sensing signal (12) is generated based on cardiac electric signals sensed along the first sensing vector (9), and in that the event (11) is analyzed based on cardiac electric signals sensed along the second sensing vector (10).

14. Implantable medical device according to any of the preceding claims, characterized in that the computer-readable program causes the processor (31) to store the latest 5 to 60 seconds of the cardiac electric signal prior to detecting that event (11) that caused the fulfillment of the predetermined criterion.

15. Method for operating an implantable medical device (1) for stimulating a human or animal heart, in particular an implantable medical device (1) according to any of the preceding claims, characterized by the following steps: a) detecting, with a detection unit (31) of the implantable medical device (1), a cardiac electric signal during a first time period, the detected cardiac electric signal comprising a plurality of events (11); b) evaluating a morphologic characteristic (15, 16) of at least some events (11) of the plurality events (11) regarding fulfillment of a predetermined criterion; c) incrementing a counter if the predetermined criterion is fulfilled, and d) triggering a storing of at least a part of the detected cardiac electric signal in a memory unit (33) of the implantable medical device (1) if the counter reaches a counter threshold to allow a later evaluation of the stored part of the cardiac electric signal.

Images