Device for monitoring the operation of a probe of an implantable active heart device

Description

Domaine technique

[0001] The present invention relates to a device for monitoring the operation of a probe of an active implantable cardiac device, in particular an implantable automatic defibrillator or a defibrillator for cardiac resynchronization. Arrière-plan technique

[0002] Leads are the critical part of an active implantable device (ICD) system. Indeed, patients with ICDs, particularly implantable cardioverter defibrillators or cardiac resynchronization therapy defibrillators, are exposed to a significant risk of complications (up to 30%), the majority of which are related to inappropriate shocks. These shocks are often due to damage to the defibrillation lead, weakening the detection of the cardiac signal and thus taking into account noise for arrhythmias.

[0003] These failures can come from abrasion following friction between two probes (causing a loss of insulation), from a break in the conductors or from a displacement resulting in poor contact between the end of the probe and the heart wall.

[0004] EP-A-1 857 143 discloses an active implantable medical device for cardiac stimulation, resynchronization, cardioversion and / or defibrillation comprising means for detecting lead fracture. This involves detecting the presence or absence of a cardiac contraction, based on the principle that each real cardiac cycle corresponds to a single cardiac contraction. The endocardial acceleration is analyzed, advantageously by detecting the presence or absence of a PEA I peak, to confirm the presence of mechanical activity of the heart upon detection of depolarization: such a detection which would not be followed by mechanical activity of the heart may have been generated by a disturbance produced by a lead fracture, it is therefore suspicious and must be diagnosed as such.

[0005] EP-A-3 081 257 discloses an active implantable medical device for cardiac stimulation comprising means for detecting a remodeling or reverse remodeling phenomenon in a patient. The device operates by comparative morphological analysis of depolarization signals collected in spontaneous rhythm on respective distinct channels, with two temporal components combined into a single 2D parametric characteristic of the VGM vectogram. Historical analysis means evaluate the variation over time of a descriptor parameter of the current VGM compared to a previous stored reference VGM. This variation is compared to predetermined thresholds to diagnose the occurrence of remodeling or reverse remodeling in the patient, and / or detect a lead breakage or the occurrence of ischemia.

[0006] US-A-2014 / 350620 discloses an implantable medical device capable of detecting cardiac signals and delivering cardiac electrical stimulation therapies, and capable of detecting a short circuit event. A signal is detected by a detection module coupled to electrodes. A controller detects a short circuit event in response to a slope of the detected signal exceeding a short circuit threshold. Description de l'invention

[0007] The present invention aims to improve the prediction of a faulty probe.

[0008] The object of the present invention is achieved by means of a device for monitoring the operation of a probe of an implantable active cardiac device according to claim 1, in particular an implantable automatic defibrillator or a defibrillator for cardiac resynchronization, comprising a parameter determination device for determining values of a plurality of parameters characterizing the probe. The monitoring device comprises a processing unit configured to determine representative values of at least one parameter of the plurality of parameters characterizing the probe based on at least two different time scales. The processing unit is further configured to compare a value called an analysis value of at least one parameter of the plurality of parameters characterizing the probe with the representative values of said parameter.

[0009] By comparing the analysis value with representative values according to two different time scales of at least one parameter characterizing the probe, the present device is configured to detect a failure of the probe.

[0010] Since the parameters to which the analysis value is compared are characteristic of the probe, a discrepancy between the representative values and the analysis value indicates a probe problem, and not a cardiac anomaly, for example. Thus, the present device makes it possible to improve the prediction of a faulty probe.

[0011] Comparison considering two different time scales allows to further refine the reliability of the detection of a probe failure.

[0012] The present invention, relating to a device for monitoring the operation of a probe of an active implantable cardiac device, can be further improved by means of the following embodiments.

[0013] According to one embodiment, a first representative value may be an average of a first predefined number of representative values determined before the analysis value that is compared by means of the processing unit.

[0014] Said first representative value is therefore determined so as to represent a trend of a parameter characterizing the probe which is prior to the analysis value. Said first representative value can thus constitute a comparative value.

[0015] According to one embodiment, a second representative value may be an average of a second predefined number of representative values determined before the analysis value which is compared by means of the processing unit, said second predefined number being greater than the first predefined number.

[0016] Thus, two representative values can be determined based on two different time scales.

[0017] Comparing the analysis value considering two different time scales allows to further refine the reliability of the detection of a probe failure. Indeed, the determination of a probe failure can depend on the time scale considered in relation to the analysis value.

[0018] According to one embodiment, a third representative value may be a rolling average based on an average of a third predetermined number of representative values determined before the analysis value that is compared by means of the processing unit, said average of a third predetermined number of values corresponding to one parameter of the plurality of parameters. The rolling average is a type of statistical average particularly suitable for analyzing time series, in particular by removing transient fluctuations so as to highlight longer-term trends.

[0019] In addition, the device is thus able to determine three representative values based on three different time scales.

[0020] According to one embodiment, the processing unit can be configured when determining the representative values such that a value among the values of the plurality of parameters characterizing the probe which exceeds a predefined limit value is not taken into account.

[0021] Thus, it is possible to discard values that would not be considered usable or as being within a range of viable values. Such values considered abnormal would then be advantageously discarded from the determination of representative values in order to improve their reliability.

[0022] According to one embodiment, the processing unit can be configured to compare the analysis value of at least one parameter of the plurality of parameters characterizing the probe with the representative values of said at least one parameter, the most recent value relative to the analysis value which is taken into account for the determination of the representative values being included in a first predetermined time interval.

[0023] It is thus possible to determine a first time interval which allows only values to be taken into consideration for the determination of representative values from an event which is not considered too old compared to the analysis value.

[0024] As a result, the reliability of the device, and therefore the prediction of a faulty probe, is further improved.

[0025] According to one embodiment, the processing unit can be configured to compare the analysis value of at least one parameter of the plurality of parameters characterizing the probe with the representative values of said at least one parameter, the oldest value relative to the analysis value which is taken into account for the determination of the representative values being included in a second predetermined time interval.

[0026] It is thus possible to determine a second time interval which allows only values to be taken into consideration for the determination of representative values by going back in time to an event which is not considered too old compared to the analysis value.

[0027] As a result, the reliability of the device, and therefore the prediction of a faulty probe, is further improved.

[0028] According to one embodiment, a parameter may be one of a detection signal amplitude, a lead continuity, a daily detection percentage, a number of non-sustained ventricular fibrillations, a number of untreated ventricular fibrillations, a number of treated ventricular fibrillations, a number of isolated extrasystoles, a number of total extrasystoles, a lead impedance and a pacing threshold. Thus, the present device is configured to determine and take into account only parameters characterizing a lead.

[0029] According to one embodiment, the plurality of parameters characterizing the probe may comprise at least two different parameters, in particular at least three different parameters.

[0030] Taking into account two different parameters, preferably three, characterizing the probe makes it possible to further improve the reliability and sensitivity of the present probe monitoring device.

[0031] According to one embodiment, the device may further comprise an alert unit for issuing an alert when the analysis value increasingly or decreasingly exceeds a limit value of at least one representative value and / or a limit threshold of at least one parameter of the plurality of parameters.

[0032] The device is thus configured to issue an alert when a probe failure is determined by exceeding the analysis value. The term "threshold limit" of a parameter includes two aspects: both that of limit value (for example: the parameter is beyond a limit value) and that of limit variation (for example: the parameter varies by more than this limit variation).

[0033] According to one embodiment, each of the parameters of the plurality of parameters may respectively have a limit threshold, the limit thresholds being grouped into a first group of limit thresholds for which the alert unit is configured to issue an alert in the event of exceeding a limit threshold of a single parameter, or a second group of limit thresholds for which the alert unit is configured to issue an alert in the event of concomitant exceeding of the limit thresholds of at least two different parameters.

[0034] Thus, this device is able to differentiate the need to issue an alert or not depending on the parameters whose limit threshold is crossed. Thus, only alerts considered justified are issued. As indicated above, the term "limit threshold" of a parameter includes two aspects: both that of limit value (for example: the parameter is beyond a limit value) and that of limit variation (for example: the parameter varies by more than this limit variation).

[0035] According to one embodiment, a limit threshold of a parameter assigned to the second group can be transferred to the first group if the exceeding of said limit threshold occurs successively a predetermined number of times.

[0036] Thus, it is possible to adapt the sensitivity and specificity of the alerts during probe monitoring according to the identified limit threshold exceedances.

[0037] According to one embodiment, the limit thresholds relating to lead impedance, lead continuity and the number of total extrasystoles may be part of the first group, and the limit thresholds relating to the amplitude of a detection signal, the percentage of detection, the pacing threshold, the number of isolated extrasystoles, the number of treated ventricular fibrillations, the number of sustained but untreated ventricular fibrillations, and the number of non-sustained ventricular fibrillations may be part of the second group.

[0038] Thus, this device is suitable for discriminating between limit thresholds relating to parameters which are sufficient in themselves to justify the issue of an alert.

[0039] According to one embodiment, a weighting value may be assigned to each parameter of the second group and wherein the alert unit may be configured to trigger an alert when the sum of the weighting values of the at least two parameters exceeds a predetermined number.

[0040] The weighting of the parameters in relation to each other makes it possible to determine whether their respective simultaneous exceedance of the limit threshold is sufficient to trigger the issue of an alert.

[0041] According to one embodiment, the alert unit may comprise a memory unit configured to save a limit threshold exceedance for a determined duration and to delete it after the elapse of said determined duration.

[0042] Considering previous events that have or have been likely to trigger the issuance of an alert allows for further improvement in the prediction of a faulty probe.

[0043] According to one embodiment, one of the plurality of parameters may include a first limit threshold and a second limit threshold, the first limit threshold being part of the first group and the second limit threshold being part of the second group.

[0044] Each of the limit thresholds may correspond, for example, to limit thresholds relating to different time scales. A first limit threshold may thus relate to a discrete time value while the second limit threshold may relate to a variation of the parameter.

[0045] According to one embodiment, limit thresholds among the limit thresholds of the second group may be linked to each other and others may not be linked to each other, such that the alert unit may be configured to trigger an alert in the presence of at least two limit threshold exceedances among thresholds of the second group which are not linked to each other.

[0046] Parameter threshold limits may be linked together if they reflect the same problem (e.g., detection proportion and detection amplitude). Exceeding two linked threshold limits is therefore not considered sufficient to issue an alert.

[0047] It may therefore be necessary to have the presence of at least two threshold exceedances among thresholds of the second group which are not linked to each other (such as for example the number of ventricular fibrillation episodes per day and the stimulation threshold) to issue an alert. Description des figures

[0048] The invention and its advantages will be explained in more detail in the following by means of preferred embodiments and with particular reference to the following accompanying figures, in which: There Figure 1 represents an operation monitoring device according to the present invention. The Figure 2 represents the analysis of the variation of a parameter on a second time scale called "medium term". The Figure 3 represents the analysis of the variation of a parameter on a third time scale called "long term". The Figure 4a represents a first part of a flowchart relating to the analysis of variations in values of a parameter on three different time scales according to the present invention and to the lifting of warnings. The Figure 4b represents a second portion of the organizational chart illustrated in Figure 4a . There Figure 5 represents a flowchart relating to an alert triggering based on the warnings raised according to the present invention. The Figure 6 represents a weighting table of the sufficiency of warnings between them.

[0049] The invention will now be described in more detail using advantageous embodiments in an exemplary manner and with reference to the figures. The described embodiments are merely possible configurations and it should be borne in mind that individual features as described above may be provided independently of one another or may be omitted altogether when implementing the present invention.

[0050] There Figure 1 illustrates an active implantable cardiac device 1 and a processing unit 2 forming a device for monitoring the operation of a probe 4 according to the present invention.

[0051] The active implantable cardiac device 1 may be an implantable cardioverter defibrillator or a defibrillator suitable for cardiac resynchronization.

[0052] The implantable cardiac device 1 comprises a housing 3. The housing 3 comprises in particular electronic circuits and a battery, for example of the lithium / iodine type. The housing 3 also comprises a connector part 5 into which an implantable probe 7 can be inserted and then screwed.

[0053] Although the Figure 1 represents an example of an active implantable cardiac device comprising an implantable probe 7, it should be borne in mind that in a variant (not shown) several implantable probes can be connected to the connector part 5 of the housing 3. The device 4 for monitoring the operation of a probe is configured for an active implantable cardiac device comprising several probes. The device 4 for monitoring the operation of a probe is thus configured to determine a failure resulting from abrasion following friction between two probes, causing at least partial loss of their insulation.

[0054] The implantable probe 7 comprises several electrodes 8a, 8b, 8c - the number of electrodes illustrated in the Figure 1 not being limiting - which constitute means of detection and stimulation and / or defibrillation of the implantable cardiac device 1.

[0055] The implantable probe 7 may be a defibrillation probe.

[0056] The implantable probe 7 is configured to measure values of a plurality of parameters characterizing it, such as impedance values for example.

[0057] According to the present invention, the plurality of parameters characterizing the implantable probe 7 may at least comprise: the amplitude of the detection signal, the continuity of the probe, a daily percentage of detection, a number of non-sustained ventricular fibrillations, a number of untreated ventricular fibrillations, a number of treated ventricular fibrillations, a number of isolated extrasystoles, a number of total extrasystoles, an impedance of the probe and a stimulation threshold.

[0058] An isolated extrasystole is defined as a cardiac cycle with a single extrasystole.

[0059] Furthermore, in an embodiment in which the implantable lead 7 is a defibrillation lead, the following parameters may also be considered: the continuity of the defibrillation lead, the number of treated ventricular fibrillations, the number of sustained but untreated ventricular fibrillations and the number of non-sustained ventricular fibrillations.

[0060] Treated ventricular fibrillation is defined as ventricular fibrillation that persists as such and has been treated with electric shock.

[0061] Untreated ventricular fibrillation is defined as ventricular fibrillation that persists as such but has not been treated with electrical shock.

[0062] Non-sustained ventricular fibrillation is defined as ventricular fibrillation that does not persist and has not been treated with electrical shock.

[0063] Note that although not all of the aforementioned parameters characterizing the probe are available for all types of probes, and as will be explained further in the following, this does not influence the multifactorial analysis implemented by the probe operation monitoring device 4 of the present invention.

[0064] The implantable cardiac device 1 thus provides a parameter determination device for determining values of a plurality of parameters characterizing the implantable probe 7.

[0065] The line breaks 9 indicate that the length of the implantable probe 7 is not fully represented at the Figure 1 no worries about drawing scale.

[0066] The implantable probe 7 is connected to the connector part 5 of the housing 3 by means of a pin 11. Partial screwing or insufficient insertion of the pin 11 into the connector part 5 of the implantable cardiac device 1 may cause connection problems.

[0067] As will be explained in more detail below, the device 4 for monitoring the operation of a probe according to the present invention is configured to detect this type of failure.

[0068] To do this, the device 4 for monitoring the operation of a probe according to the present invention further comprises a processing unit 2.

[0069] The processing unit 2 may be implemented in the implantable cardiac device 1 or in an external device, such as a computer.

[0070] The implantable cardiac device 1 and the processing unit 2 are configured to communicate with each other, for example by telemetry 6.

[0071] The processing unit 2 is configured to determine representative values of at least one parameter of the plurality of parameters characterizing the implantable probe 7 based on at least two different time scales, in particular three time scales. The analysis of variations in the values of a parameter on different time scales is further described with reference to Figures 2 et 3 .

[0072] The processing unit 2 is further configured to compare a value called the analysis value of at least one parameter of the plurality of parameters characterizing the implantable probe 7 with the values representative of said parameter.

[0073] According to the present invention, the analysis of each of the parameters characterizing the implantable probe 7 can be carried out according to several factors such as a maximum or minimum threshold, an absolute ascending or descending variation called "short term" (for example over one day), an absolute or relative ascending or descending variation called "medium term" (for example over one week) and a relative ascending or descending variation called "long term" (for example over one month). The study of the variations of the parameters is described with reference to Figures 2, 3 And 4 .

[0074] Secondly, a combination of all these analyses relating to the parameters characterizing the implantable probe is carried out in order to raise an alert, which is described with reference to Figures 5 And 6 .

[0075] Note that the processing unit 2 is configured when determining the representative values in such a way that a value among the values of the plurality of parameters characterizing the probe that exceeds a predefined limit value is not taken into account. A value exceeding such a predefined limit value is referred to as a “non-usable point,” that is, a point whose value is outside a range of viable values or whose value is not available. On the contrary, the value exceeding such a predefined limit value is referred to as an “abnormal point” when it is a point whose value exceeds a maximum or minimum threshold. The other values, which do not exceed a predefined limit value, are considered “normal” and therefore usable for the analysis on the variation.

[0076] The analysis of the variation on a first time scale called "short term" is done by analyzing the variations between a maximum point and a minimum point of the same day. For example, processing unit 2 takes into account four impedance measurements during a day. In a variant, processing unit 2 also takes into consideration other parameters such as the stimulation threshold or the detection amplitude. The measurement unit can take more or less than four measurements of a parameter over a day.

[0077] The processing unit 2 further comprises an alert unit (not shown in the Figure 1 ). The alert unit is configured to issue an alert when the analysis value increasingly or decreasingly exceeds a limit value of at least one representative value and / or a limit threshold of at least one parameter of the plurality of parameters. The term "limit threshold" of a parameter includes two aspects: both that of limit value (for example: the parameter is beyond a limit value) and that of limit variation (for example: the parameter varies by more than this limit variation).

[0078] The processing unit 2 also includes a memory unit (not shown in the Figure 1 ) by means of which data can be saved.

[0079] There Figure 2 represents the analysis on the variation according to a second time scale called “medium term”.

[0080] The second time scale according to the present invention is different from the first time scale in that it relates to the analysis of variation over more than one day, in particular over a week.

[0081] The analyses of variations (relative or absolute) on the second time scale called "medium term" are carried out between an analysis point Pa and a baseline Lm called the medium term baseline.

[0082] The analysis point Pa corresponds to a point representative of the daily average.

[0083] The baseline Lm corresponds to a number n of last points each representative of the daily average. This baseline Lm only includes points qualified as "normal", that is to say points A1, A2, A3 of the Figure 2 are excluded from this Lm baseline because they each have a value exceeding a maximum threshold (represented by a dotted horizontal line on the Figure 2 ). Points exceeding a minimum threshold may also be excluded from the Lm baseline.

[0084] The Lm baseline being a “medium-term” Lm baseline, i.e. relating to one week, the medium-term Lm baseline corresponds to the last seven points Pm1 to Pm7 qualified as normal of the Lm baseline. As explained above, each of the seven points Pm1 to Pm7 corresponds to a daily average.

[0085] On the example of the Figure 2 , Pm1 represents the most recent point of the medium-term Lm baseline relative to the analysis point Pa. Pm7 represents the oldest point of the medium-term Lm baseline relative to the analysis point Pa.

[0086] In order not to compare the analysis point Pa with events considered too old, a first time interval Δm1 can be determined between the analysis point Pa and the most recent point Pm1 of the medium-term baseline Lm. This first time interval Δm1 can correspond to a duration of seven days. Otherwise, the variation analysis can be suspended.

[0087] In addition, a second time interval Δm2 can be determined between the most recent point Pm1 of the medium-term Lm baseline and a point Pm of the Lm baseline which could correspond to the oldest point of the medium-term Lm baseline. This second time interval Δm2 can correspond to a duration of fourteen days. Otherwise, the variation analysis can be suspended.

[0088] There Figure 3 represents the analysis of the variation of values on a third time scale called “long term”.

[0089] The processing unit 2 of the device 4 for monitoring the operation of a probe according to the present invention considers a signal S obtained by means of the device for determining parameters of said monitoring device 4. The signal S can be a raw signal or a signal processed by known filtering means.

[0090] Processing unit 2 is configured to operate a sliding average, in particular over seven points, on the signal S. The Cm curve of the Figure 3 represents the averaged curve thus obtained. Averaging makes it possible to avoid alterations due to so-called “short-term” variations, i.e. variations over a day.

[0091] The Cm curve of the Figure 3 corresponds to the curve taken into consideration for the analysis of the variation on the third time scale called long term.

[0092] Since averaging creates a shift, the Cm curve can be recentered over three days in order to realign the Cm curve.

[0093] The Cm curve illustrated in the example of the Figure 3 is a descending curve. In a variant the Cm curve can be an ascending curve.

[0094] As for the analysis of the variation on the second time scale described with reference to the Figure 2 , points considered abnormal were excluded before the averaging operation. Points considered normal correspond to a weekly average.

[0095] As illustrated in the figure 3 , the predetermined time interval ΔI1 can include the last seven points PI1 to PI7 considered normal, the first point PI1 corresponding to the analysis point Pa, and the point PI7 corresponding to the oldest point relative to the analysis point Pa.

[0096] As illustrated in the Figure 3 , in order not to compare the analysis point Pa with events considered too old, the first time interval ΔI1 is determined between the analysis point Pa and the most recent point PI1 of the long-term LI baseline. This first time interval ΔI1 can correspond to a duration of seven days. Otherwise, the variation analysis can be suspended.

[0097] Thus, by limiting the predetermined time interval ΔI1 to the seven most recent points in relation to the analysis point Pa, there cannot be more than seven points between the analysis point Pa and the last point PI7, which makes it possible not to compare a weekly maximum or minimum in relation to events considered too old.

[0098] Points PI1 to PI7 of the predetermined time interval ΔI1 are compared to a so-called long-term baseline LI.

[0099] In the example shown in Figure 3 , the so-called long-term LI baseline corresponds to the last twenty-eight points considered normal on the Cm curve preceding point PI7. The so-called long-term LI baseline in the example of the Figure 3 is thus included between point PI7 and point PI35.

[0100] In a variant of the present invention, the so-called long-term baseline LI could comprise more or less than twenty-eight points, at least more points than the so-called medium-term baseline Lm.

[0101] In a variant, the so-called long-term LI baseline cannot include more than fifty-six points in order to avoid events considered too old being taken into consideration.

[0102] The variation analysis on the third time scale LI, called "long-term", is carried out between the maximum, minimum or average of points considered normal over a predetermined time interval ΔI1 and the long-term baseline LI or the maximum, minimum or average of points considered normal over a predetermined time interval ΔI3. The points of the Cm curve included in the predetermined time interval ΔI1 are compared to points of the Cm curve included in a predetermined time interval ΔI2 or ΔI3.

[0103] The predetermined time interval ΔI2 includes the points between PI7 and PIn, PIn corresponding to the oldest point relative to Pa. For illustration, ΔI2 on the Figure 3 represents an interval comprising fifty-six points between PI7 and PIn.

[0104] The predetermined time interval ΔI3 includes the last seven points considered normal on the Cm curve starting from point PI35 which is the oldest point on the long-term baseline. In the example of the Figure 3 , the predetermined time interval ΔI3 thus includes points PI28 to PI35.

[0105] When the Cm curve is descending as in the example of the Figure 3 , the minimum of points PI1 to PI7 of the predetermined time interval ΔI1 can be compared to the maximum of points PI28 to PI35 of the predetermined time interval ΔI3.

[0106] Alternatively, the average of points PI1 to PI7 of the predetermined time interval ΔI1 may be compared to the average of points PI28 to PI35 of the predetermined time interval ΔI3.

[0107] THE Figures 4a And 4billustrate a flowchart 100 representative of the analysis of variations and threshold values on three different time scales according to the present invention. The flowchart 100 is represented in two figures 4a , 4b solely for the sake of clarity of the drawings. The figure 4b illustrates the sequence of steps shown in the figure 4a . Thus, step 114 of fig 4a is followed by step 116 illustrated in figure 4b .

[0108] The flowchart 100 comprises steps implemented by the processing unit 2 of the device 4 for monitoring the operation of a probe of an active implantable cardiac device 1 as described above. This thus involves the analysis of variations in values of a parameter characterizing the implantable probe 7. As a result, the elements with the same numerical references already used for the description of the Figures 1 à 3 will not be described again in detail, and reference is made to their descriptions above.

[0109] At a first step 102 of the variation analysis, the analysis point Pa is taken into account by the processing unit 2.

[0110] In a step 104, it is determined whether the value of the analysis point Pa corresponds to a usable value. If the value of the analysis point Pa exceeds a predefined limit value, it is qualified as a “non-usable point”, that is to say at a point whose value is outside a range of viable values or whose value is not available. In this case, the analysis is suspended in a step 105. A next point will then be considered for the analysis point Pa in a step 130.

[0111] If the value of the analysis point Pa is considered usable, the analysis continues.

[0112] In a step 106, it is determined whether the value of the analysis point Pa corresponds to a value that can be considered a “normal value”. For this purpose, the value of the analysis point is compared to a predefined limit, maximum or minimum threshold. If the value of the analysis point Pa exceeds the predefined maximum or minimum threshold, the value of the analysis point Pa is considered abnormal. In this case, a warning indicating a threshold exceedance is raised in a step 107 and a next point will then be considered for the analysis point Pa.

[0113] If the value of the analysis point Pa is considered normal, the analysis continues at step 130.

[0114] In a step 108, it is determined whether the variation on the first time scale exceeds a predetermined limit threshold, in this case a limit variation. The first time scale may relate to a day. The variation on the first time scale then corresponds to the variation between a maximum point and a minimum point of the same day.

[0115] If the predetermined threshold for the first time scale is actually exceeded, a warning indicating an overshoot relating to a variation is raised at a step 109.

[0116] Whether or not the predetermined threshold for the first time scale has been exceeded in step 108, a baseline on a second time scale, which is different from the first time scale, is determined in a step 110. The second time scale may relate to a duration of one week.

[0117] At a step 112, it is determined whether the number of days between the analysis point Pa and the most recent point Pm1 of the baseline Lm of the second scale called “medium term” (see Figure 2 ) is within the predefined time interval Δm1. Preferably, Δm1 is equal to seven days.

[0118] If this is not the case, the analysis is suspended at a step 113. A next point will then be considered for the analysis point Pa at a step 130.

[0119] Otherwise, the analysis continues.

[0120] At a step 114, it is determined whether the number of days between the most recent point Pm1 and the oldest point Pm of the baseline Lm of the second scale called “medium term” (see Figure 2 ) is included in the predefined time interval Δm2. Preferably, Δm2 is equal to fourteen days.

[0121] If this is not the case, the analysis is suspended at a step 115. A next point will then be considered for the analysis point Pa at a step 130.

[0122] Otherwise, the analysis continues.

[0123] In a step 116, it is determined whether the variation on the second time scale exceeds a predetermined limit threshold, in this case a limit variation.

[0124] If the predetermined threshold for the second time scale is actually exceeded, a warning indicating an overshoot relating to a variation is raised at a step 117.

[0125] Whether or not the predetermined threshold for the second time scale has been exceeded in step 116, a baseline on a third time scale, which is different from the first time scale and the second time scale, is determined in a step 118. The third time scale may relate to one month and is considered "long term".

[0126] At a step 120, an average of the last seven points of the so-called long-term LI baseline (i.e. an average of the seven points from the oldest point of the so-called long-term LI baseline - see Figure 3 ) is determined. Alternatively, a maximum point or a minimum point of the last seven points of the so-called long-term LI baseline may be determined in step 120.

[0127] At a step 122, it is determined whether the number of days between the analysis point Pa corresponding to the most recent point PI1 and the oldest point PIn considered for the analysis on the third time scale (see Figure 3 ) is included in the predefined time interval ΔI1. Preferably, ΔI1 is equal to seven days.

[0128] If this is not the case, the analysis is suspended at a step 123. A next point will then be considered for the analysis point Pa at a step 130.

[0129] Otherwise, the analysis continues.

[0130] At a step 124, it is determined whether the number of days between the most recent point PI1 and the oldest point PI of the baseline LI of the third scale called "long term" (see

[0131] Figure 3 ) is included in the predefined time interval ΔI2. Preferably, ΔI2 is equal to fifty-six days.

[0132] If this is not the case, the analysis is suspended at a step 125. A following point will then be considered as an analysis point Pa at a step 130.

[0133] Otherwise, the analysis continues.

[0134] In a step 126, it is determined whether the variation on the third time scale exceeds a predetermined limit threshold, in this case a limit variation.

[0135] If the predetermined threshold for the third time scale is actually exceeded, a warning indicating an overshoot relating to a variation is raised at a step 127.

[0136] Otherwise, no warning is raised at step 128.

[0137] In both cases, whether the predetermined threshold has been exceeded or not, the analysis continues by considering a next point for the analysis point Pa at a step 130.

[0138] The variation analysis illustrated by flowchart 100 thus includes successive variation analyses on different time scales, from the shortest time scale to the longest time scale.

[0139] There Figure 5 represents a flowchart 300 relating to an alert triggering based on the warnings raised in steps 107, 109, 117 and 127 of the flowchart 100.

[0140] The flowchart 300 comprises steps implemented by the processing unit 2 of the device 4 for monitoring the operation of a probe of an active implantable cardiac device 1 as described above. Therefore, the elements with the same numerical references already used for the description of the Figures 1 à 4 will not be described again in detail, and reference is made to their descriptions above.

[0141] Flowchart 300 illustrates how the various warnings previously raised during steps 107, 109, 117 and 127 of flowchart 100 are combined in order to optimize the sensitivity and specificity of the alerts sent to the physician. In other words, in order to trigger only justified alerts.

[0142] According to the present invention, a warning is different from an alert. The alert is immediately communicated to the physician to indicate a potential probe failure, for example by a visual or audible message. A warning is not necessarily communicated to the physician. However, as will be explained below, the concomitance of warnings may lead to the triggering of an alert.

[0143] Thus, while an analysis of values above or below a limit threshold can immediately raise an alert (for example in the case of impedance or continuity), an analysis of variation of this same parameter will have to go through a warning stage.

[0144] Two types of warning can be taken into account by the processing unit 2 of the monitoring device 4 of the present invention: variation warnings (at a step 301) and threshold warnings (at a step 302).

[0145] As described with reference to step 107 of the figure 4a , a threshold warning corresponds to the crossing of a limit threshold by the value of the analysis point Pa.

[0146] As described with reference to steps 109 of the figure 4a and at steps 117 and 127 of the figure 4b , a variation warning, corresponds to the crossing of a limit threshold by the variation according to a time scale of the value of a parameter characterizing the implantable probe 7.

[0147] At least one limit threshold is determined for each parameter characterizing the implantable probe 7.

[0148] Several threshold limits can be determined for the same parameter. Thus, a parameter can have a first threshold limit, the exceeding of which generates a warning, and a second threshold limit, the exceeding of which generates an alert.

[0149] According to the present invention, the limit thresholds of the parameters characterizing the probe can be classified into two groups.

[0150] The first group includes the limit thresholds for which the alert unit of the monitoring device 4 is configured to issue an alert in the event of exceeding a limit threshold of a single parameter. For example, the limit thresholds relating to lead impedance, lead continuity and the number of total extrasystoles are part of the first group.

[0151] The second group groups together the limit thresholds for which the alert unit of the monitoring device 4 is configured to issue an alert in the event of concomitant exceeding of the limit thresholds of at least two different parameters. For example, the limit thresholds relating to the amplitude of a detection signal, the detection percentage, the pacing threshold, the number of isolated extrasystoles, the number of treated ventricular fibrillations, the number of sustained but untreated ventricular fibrillations, and the number of non-sustained ventricular fibrillations are part of the second group.

[0152] Note that a limit threshold of a parameter assigned to the second group can be transferred to the first group if the exceeding of said limit threshold occurs successively a predetermined number of times.

[0153] It should also be noted that the threshold limit itself for raising an alert can vary. This is the case, for example, for the impedance of a left ventricular lead: a warning can be raised for a unipolar vector on a lower threshold limit than for a bipolar vector.

[0154] As illustrated in the Figure 5 , if at a step 302 of the flowchart 300 it is detected that a threshold warning has been raised (at step 107 of the flowchart 100), it is determined at a step 304 whether the raised threshold warning relates to a value of a parameter classified in the first group.

[0155] If this is the case, this condition is sufficient for an alert to be issued by the alert unit of the monitoring device 4 at a step 306. As explained previously with reference to the limit thresholds of the first group, an analysis of values exceeding a limit threshold can in fact make it possible to immediately raise an alert (for example in the case of an impedance or continuity). Thus, a very high or very low value of a parameter (for example a probe impedance greater than 2000 ohms) is as such characteristic of a probe problem (in favor of a fracture). This factor can therefore be sufficient in itself to trigger an alert indicating a potential probe fault.

[0156] Otherwise, it is checked in a step 308 whether the raised threshold warning relates to a parameter comprising a second limit threshold, the exceeding of which is likely to trigger an alert. Indeed, as explained above, certain parameters, for total extrasystoles for example, may have a first limit threshold, the exceeding of which generates a warning, and a second limit threshold, the exceeding of which generates an alert. In this case, it is checked in a step 310 whether the value of the analysis point crosses the second limit threshold. If so, an alert is triggered in step 306.

[0157] Cases not covered above by warnings that may alone generate an alert are described below.

[0158] As illustrated by flowchart 300 in step 301, a warning of variations in a parameter is not sufficient in itself to raise an alert.

[0159] This parameter therefore needs to have at least one second concomitant parameter for an alert to be triggered.

[0160] It is thus determined at a step 312 whether a warning relating to a second parameter has been detected concomitantly with the warning of step 301. This second parameter may also not be sufficient in the case where it reflects the same problem (for example the detection proportion and the detection amplitude). It is then said that the first and second parameters are “linked”.

[0161] Therefore, in a step 314 it is determined whether the first parameter and the second parameter are linked to each other.

[0162] If they are not linked to each other, then the concomitance of a warning relating to a first parameter with a warning relating to a second parameter, not linked to the first parameter, causes an alert to be triggered at step 306.

[0163] It is therefore necessary to have at least one second unrelated parameter, such as the number of ventricular fibrillation episodes per day or the pacing threshold, to raise an alert.

[0164] The pairs of parameters characterizing the probe which are not sufficient together to trigger an alert because they are "linked" are illustrated by means of the gray boxes in the Figure 6 , which will be further described below.

[0165] Three pairs of so-called "linked" parameters are thus defined. The first pair corresponds to the detection / day and the amplitude of the wave. The second pair corresponds to a sustained episode and an untreated episode. Finally, the third pair corresponds to isolated extrasystoles and total extrasystoles.

[0166] If it is determined in step 314 of flowchart 300 that the two parameters are linked, or even that in step 312 a warning relating to a second parameter had not been detected concomitantly, it is determined in a step 316 whether the warning relating to the first parameter (that of step 301) is triggered every day.

[0167] For this purpose, warnings relating to the first parameter are saved in a memory unit of processing unit 2.

[0168] It should be noted that the analysis of the various parameters is done simultaneously. When one of the parameters raises a warning, it remains active for a predetermined period, for example seven days.

[0169] If the warning is lifted for several days in a row, then it will remain active for seven days after it is lifted.

[0170] If multiple warnings from different parameters are active at the same time (i.e. concurrently), this may activate an alert.

[0171] This system of concurrent alerts makes it possible to detect a failure that can occur in different ways, at different times.

[0172] In step 318, it is determined whether the warning was triggered more than seven days ago. If so, said warning is deactivated (turned off) in step 320. Otherwise, the analysis continues in step 322, taking into consideration the next point.

[0173] There Figure 6 represents a weighting table of the sufficiency of warnings between them.

[0174] In order to calculate the sufficiency of the warnings between them, in particular at step 314 of the flowchart 300, a pair weighting system is put in place.

[0175] All warnings related to probe parameters appearing in a day are listed in alphabetical order.

[0176] The striped boxes of the Figure 6 represent a single warning, not two warnings raised from the same parameter, such as variation on two different scales of the same parameter.

[0177] The "weights" assigned to each pair in the table are added together. If the result of said addition is greater than and different from 3, the alert is raised at step 306 of flowchart 300 (see Figure 5 ).

[0178] You must then place yourself on the line of the first parameter, then add the weight of each of the pairs formed to the own weight of the first parameter (indicated in the hatched boxes). Examples are provided below.

[0179] In a first example, the first parameter corresponds to the impedance and the second parameter corresponds to the stimulation threshold. For the first example, you must then place yourself on the impedance line, and add the impedance's own weight (i.e. 2, see the hatched box) and the weight of the pair formed with the stimulation threshold (i.e. 4). The result of the addition, which is 6, is greater than 3: an alert is therefore raised.

[0180] In a second example, the first parameter corresponds to the signal amplitude and the second parameter corresponds to the daily proportion of signal detected, i.e. the daily proportion of signal in spontaneous rhythm. For the second example, it is then necessary to place oneself on the line of the amplitude of the wave, and add the own weight of the amplitude of the wave (i.e. 2, see the hatched box) and the weight of the binomial formed with the detection / day (i.e. 1). The result of the addition being equal to 3, the alert is not raised.

[0181] In one embodiment of the invention, the device for monitoring the operation of a probe takes into consideration at least two different parameters characterizing the probe.

[0182] In another embodiment of the invention, the device for monitoring the operation of a probe takes into consideration at least three different parameters characterizing the probe. Thus, a third example is described below in which three different parameters are taken into account.

[0183] In the third example, the first parameter corresponds to the wave amplitude, the second parameter corresponds to the detection / day and the third parameter corresponds to the continuity. For the third example, you must then place yourself on the wave amplitude line, and add the wave amplitude's own weight (i.e. 2, see the hatched box), the weight of the binomial formed with the detection / day (i.e. 1) and the weight of the binomial formed with the continuity (i.e. 4). The result of the addition being equal to 7, i.e. greater than 3, the alert is raised.

[0184] The present invention thus allows the consideration of multiple parameters (electrical and rhythmic) characteristic of an implantable probe on different time scales in order to improve the prediction of a failure of the implantable probe.

Claims

1. A device (4) for monitoring operation of a probe (7) of an implantable active cardiac device (1), in particular an implantable automatic defibrillator or a defibrillator for cardiac resynchronization, comprising: a parameter determining device for determining values of a plurality of parameters characterizing the probe (7); a processing unit (2) configured to determine representative values of at least one of the plurality of parameters characterizing the probe (7) based on at least two different time scales, wherein the processing unit (2) is further configured to compare a value as an analysis value of at least one of the plurality of parameters characterizing the probe (7) with the representative values of said parameter.

2. The device for monitoring operation of a probe of an implantable active cardiac device according to claim 1, wherein a first representative value is an average of a first predefined number of representative values determined prior to the analysis value that is compared by means of the processing unit.

3. The device for monitoring operation of a probe of an implantable active cardiac device according to claim 2, wherein a second representative value is an average of a second predefined number of representative values determined prior to the analysis value that is compared by means of the processing unit, said second predefined number being greater than the first predefined number.

4. The device for monitoring operation of a probe of an implantable active cardiac device according to at least one of claims 1 to 3, wherein a third representative value is a rolling average based on an average of a third predetermined number of representative values determined prior to the analysis value compared by means of the processing unit, said average of a third predetermined number of representative values corresponding to one of the plurality of parameters.

5. The device for monitoring operation of a probe of an implantable active cardiac device according to at least one of claims 1 to 4, wherein the processing unit is configured during the determination of the representative values in such a way that one of the values of the plurality of parameters characterizing the probe that exceeds a predefined limit value is not taken into account.

6. The device for monitoring operation of a probe of an implantable active cardiac device according to at least one of claims 1 to 5, wherein the processing unit is configured to compare the analysis value of at least one of the plurality of parameters characterizing the probe with the representative values of the at least one parameter, the most recent value relative to the analysis value that is taken into account for the determination of the representative values being included within a first predetermined time interval.

7. The device for monitoring operation of a probe of an implantable active cardiac device according to at least one of claims 1 to 6, wherein the processing unit is configured to compare the analysis value of at least one of the plurality of parameters characterizing the probe with the representative values of the at least one parameter, the oldest value relative to the analysis value that is taken into account for the determination of the representative values being included within a second predetermined time interval.

8. The device for monitoring operation of a probe of an implantable active cardiac device according to at least one of claims 1 to 7, wherein a parameter is one of an amplitude of the detection signal, a continuity of the probe, a daily detection percentage, a number of non-sustained ventricular fibrillations, a number of untreated ventricular fibrillations, a number of treated ventricular fibrillations, a number of isolated extrasystoles, a number of total extrasystoles, an impedance of the probe, and a pacing threshold, and / or wherein the plurality of parameters characterizing the probe comprises at least two different parameters, in particular at least three different parameters.

9. The device for monitoring operation of a probe of an implantable active cardiac device according to at least one of claims 1 to 8, further comprising an alert unit for issuing an alert when the analysis value exceeds, in an increasing or decreasing manner, a limit value of at least one representative value and / or a limit threshold of at least one of the plurality of parameters.

10. The device for monitoring operation of a probe of an implantable active cardiac device according to claim 9, wherein each parameter of the plurality of parameters respectively has a limit threshold, the limit thresholds being grouped into: a first group of limit thresholds for which the alert unit is configured to issue an alert in case of exceeding a limit threshold of a single parameter; or a second group of limit thresholds for which the alert unit is configured to issue an alert in case of simultaneously exceeding the limit thresholds of at least two different parameters.

11. The device for monitoring operation of a probe of an implantable active cardiac device according to claim 10, wherein a limit threshold of a parameter assigned to the second group is transferred to the first group if exceedance of said limit threshold occurs successively a predetermined number of times, and / or wherein: the limit thresholds relating to the impedance of the probe, the continuity of the probe, and the number of total extrasystoles are part of the first group; and the limit thresholds relating to the amplitude of a detection signal, the detection percentage, the pacing threshold, the number of isolated extrasystoles, the number of treated ventricular fibrillations, the number of sustained but untreated ventricular fibrillations, and the number of non-sustained ventricular fibrillations are part of the second group.

12. The device for monitoring operation of a probe of an implantable active cardiac device according to claim 10 or 11, wherein a weighting value is assigned to each parameter of the second group, and wherein the alert unit is configured to trigger an alert when a sum of the weighting values of the at least two parameters exceeds a predetermined number.

13. The device for monitoring operation of a probe of an implantable active cardiac device according to at least one of claims 10 to 12, wherein the alert unit comprises a memory unit configured to store a limit threshold exceedance for a determined duration and to delete it after the expiration of said determined duration.

14. The device for monitoring operation of a probe of an implantable active cardiac device according to at least one of claims 10 to 13, wherein a parameter of the plurality of parameters comprises a first limit threshold and a second limit threshold, the first limit threshold being part of the first group and the second limit threshold being part of the second group.

15. The device for monitoring operation of a probe of an implantable active cardiac device according to at least one of claims 10 to 14, wherein limit thresholds among the limit thresholds of the second group are linked to each other and others are not linked to each other, in such a way that the alert unit is configured to trigger an alert in the presence of at least two exceedances of threshold limits among thresholds of the second group that are not linked to each other.