心脏植入式电子设备的寿命:一种新型建模工具
Cardiac implantable electronic devices' longevity: A novel modelling tool

原始链接: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0333195

## 心脏植入式电子设备 (CIED) 寿命:一种新的建模工具 本研究介绍了一种新的框架——功耗指数 (PCI),用于评估心脏植入式电子设备 (CIED) 的寿命,解决了患者、医疗保健提供者和支付方面临的一个关键挑战。目前,由于报告不一致和编程设置不同,比较不同设备之间的寿命具有困难。 PCI(计算方式为时间 x 电流/电池容量)通过分析功耗和电池容量,实现了标准化的寿命评估。研究人员分析了主要 CIED 制造商的用户手册,对各种设备和设置的电流消耗进行了建模。该模型通过瑞典设备注册处的真实数据进行了验证,证明了其在预测前代设备寿命方面的强大准确性。 结果表明,背景电流占功耗的 50% 以上,凸显了其重要性。PCI 模型成功预测了当前一代设备的寿命,揭示了不同制造商之间的差异以及特定编程选项(如远程监测和起搏算法)的影响。该工具有望改善临床医生个性化的设备选择,并为医疗保健系统的采购决策提供信息,最终优化患者护理和成本效益。

黑客新闻 新 | 过去 | 评论 | 提问 | 展示 | 招聘 | 提交 登录心脏植入式电子设备的寿命:一种新型建模工具 (plos.org)5 分,PaulHoule 发表于 3 小时前 | 隐藏 | 过去 | 收藏 | 1 条评论 et2o 发表于 3 分钟前 [–] 心脏病科住院医师。起搏次数越多,电池寿命越短?不明白这里的新颖之处是什么。回复 指南 | 常见问题 | 列表 | API | 安全 | 法律 | 申请 YC | 联系方式 搜索:
相关文章

原文

Abstract

Introduction

Longevity of Cardiac Implantable Electronic Devices (CIEDs) is an important issue for patients, who wish to avoid further surgery, and purchasers, who wish to optimize cost-effectiveness, and is therefore a relevant consideration for clinicians. It is appreciated that there are common discrepancies between declared (future) longevities of generators and their subsequent survival curve once implanted [1,2]. Despite calls for more transparency and industry-wide standardized reporting of longevity [35], comparisons of longevity between devices and manufacturers in different settings remains challenging. Given that 30–40% of all CIED procedures are generator replacements, there exists the risk of conflicts of interest for both manufacturers, and, in fee-for-service healthcare environments providers, that limit enthusiasm for transparency [6].

Although implanters and their patients appreciate the concept of battery capacity as a prime criteria for device longevity [7] they also recognize that energy drain plays a role [8,9]. However, the potential lifetime of the device is also determined by how energy is stored, and how efficiently it is delivered [10], along with the patient’s characteristics, all of which, add to the frustrating situation of complex and non-standardized user-manual declared longevity, with different programming as baseline across companies, making personalization of generator prescription impossible even in the presence of similar battery capacity. If one could reliably describe consumption and index this to battery capacity and pacing requirements, there remains the possibility of a reliable comparison of devices.

Based upon previous work defining the power consumption of CIEDs [11], which included a calculation for the inverse of longevity, we have developed the Power Consumption Index (PCI) (as defined by PCI = t x I/C (where t is a constant equal to 1 hour)) that aims to describe the intrinsic power consumption of the overall system (the pacing system coupled with the battery) during a normalized period (1 hour). The reciprocal of the PCI therefore allows a derivation of longevity (Fig 1 and S1 File).

thumbnail

Fig 1. PCI and nominal longevity.

Note: The Power Consumption Index for two illustrative devices. On the Y axis, the PCI index (τ x I/C with τ  = 1 h) is split according to each current contribution (τ x I background/C, τ x I PACING/C, etc.). On the Y’ axis, the green lines represent the corresponding longevity in years (a nonlinear “1/x scale”; the inverse of the PCI index τ x I/C multiplied by 106 (I is in µA) and divided by 365x24 gives the corresponding longevity (in years).

https://doi.org/10.1371/journal.pone.0333195.g001

The Power Consumption Index for two illustrative devices. On the Y axis, the PCI index (t x I/C with t = 1 h) is split according to each current contribution (t x I background/C, t x I PACING/C, etc..). On the Y’ axis, the green lines represent the corresponding longevity in years (a nonlinear “1/x scale”; the inverse of the PCI index t x I/C multiplied by 10^6 (I is in mA) and divided by 365x24 gives the corresponding longevity (in years).

Hence, the objectives of this study were 1) to create and test a “universal” model, based on the concept of PCI, that offers the opportunity to model longevity for any CIED; 2) to validate this by comparing the modeled survival curves of previous generation CIEDs with real world data and; 3) apply this to predict product survival curves for new generations of CIEDs.

Methods

Firstly, we collected battery capacity and estimated current drain for a variety of devices from manufacturers’ user manuals to calculate PCI values and thereby longevity. The user manuals are available from the five major CIED manufacturers: Abbott (Abt, formerly St Jude Medical, Sylmar, CA, USA), Boston Scientific (BSc, St Paul, MN, USA), Biotronik (Btk, Berlin, Germany), Microport CRM (Mcp, formerly LivaNova and Sorin, Clamart, France), and Medtronic (Mdt, Minneapolis, MN, USA). We also used the only web-based longevity calculator available from BsC [12]. Detailed results are provided in S1 File.

Current drain modelling

Industry-reported CIEDs’ longevities in user manuals depend upon the programming (including the activation of specific algorithms such as rhythm storage, remote monitoring, and sensors). Current is not provided in manuals and therefore modeling was required to calculate the PCI for each CIED. Because nominal settings differ from one manufacturer to another, an additional step was necessary to specifically identify and estimate the current for background device activity and pacing (Ibackground, Ipacing) and optional settings (Iremote IIEGM Ialgo). For Ibackground, and Ipacing the evaluation was done by a regression analysis. For Iremote/IEGM/algo the current was estimated via the difference of longevity between activated and deactivated settings for each option.

Two categories of optional settings were considered. The first considered algorithms directly influencing the Ipacing drain such as automatic threshold management or reduction of right ventricular pacing (RVP) percentage (Ialgo) and the second explored optional settings such as remote monitoring and IEGM storage (Iremote and IIEGM). Unlike Ibackground or basic pacing, each factor needed to be analyzed individually (S1 File).

In addition, for the 20 previous generation single chamber and dual chamber pacemakers under investigation, 674 settings were considered, for 11 new generation devices 243 settings were explored and for 8 previous and 5 current CRT-P devices, respectively 294 and 177 settings. Finally for 3 leadless devices 156 settings were considered.

Validation of PCI model, using Monte-Carlo simulations

In order to validate the survival curves the model was applied not only according to nominal parameters but a variety of settings in order to reflect real-world patient characteristics and programming. A pool of fictitious patient sets (100,000 patients) was created via a Monte-Carlo simulation (programmed in Python™) with age, indication (sinus node dysfunction [SND], intermittent atrio-ventricular block [AVB], complete AVB) and programmed parameters based on available literature [1320]. The parameters used for the Monte-Carlo model are described in S2 File. Right ventricular pacing avoidance [RVPa] algorithms were assumed to be applied for SND patients eligible for DDD pacing. Remote monitoring was not standard for previous generation devices. We hypothesized a 50% adoption rate of remote for new generation current devices. The impact of settings (such as capture management, additional IEGM storage, RVPa for intermittent AVB for pacemaker, aCRT™ and MPP™ for CRT) on energy consumptions were studied.

For each device, longevity was calculated per patient using via the PCI energy consumption formula. Missing information which could not be derived from manuals was hypothesized, while assuming similarities among same generation devices. When longevity exceeded patient life expectancy, data were censored (since end of service uncommonly occurs simultaneously with death, and residual battery life of the device is rarely collected at death). The distribution of PCI and corresponding longevity across the pool of fictitious patients allowed the drafting of product survival curves for each cardiac implant. The PCI model was then validated for previous generation’s devices by comparing these modeled product survival curves and real-life data. For real-world survival curves, we used the Swedish registry [21] which was started in 1989 on the initiative of the Swedish Society of Cardiology. All the implanting clinics in Sweden report to the registry that compiles quarterly and annual reports of pacemaker use in Sweden. Every year there are about 5000 pacemaker procedures in Sweden. The real life product survival curves were extracted from these reports

Results

Current drain modeling

Background current (Ibackground) and pacing current (Ipacing).

For conventional pacemakers, the difference between industry-reported longevities and those derived by regression was 0.1 years ± 4% for previous generation devices and −0.1 years ± 0.7% for new generation devices. For devices with a variety of configurations (Mdt and BSc via its longevity calculator website), the regression coefficient (R2) exceeded 90%, across all settings applied.

The Ibackground derived by regression analysis (Table 3) matched those reported by manufacturers for most devices with few exceptions. There was, however, a significant change between previous and new generation devices. Previous generation devices relied on a Ibackground exceeding 9–10 µA (except for Identity™ for which the Ibackground ranged between 5.72 µA and 6.19 µA, the Evia™-T for which the Ibackground ranged between 6 µA and 6.66 µA and Symphony™ with an Ibackground at 6µA). For new generation CIEDs, the Ibackground did not exceed 7 to 7.5 µA, except for BSc devices, which reached 9.7 µA to 10.4 µA. Similar results were observed for CRT-P. For leadless pacemakers, the Ibackground was much lower than for conventional pacemakers (ranging from 0.8 µA to 0.94 µA for SR and 1.75 µA for DR devices, including 0.81 µA to 1µA for communication between the atrial and the ventricular capsules).

The analysis conducted in S1 File (Pacing current) shows that for the following settings (60 bpm, pulse width 0.4 ms, lead impedance 500 ohms, with 100% pacing), the Ipacing was relatively consistent across all categories of contemporary pacemakers with average current drains of 1.98 ± 0.12µA at 2.5V pacing output and 4.37 ± 0.24 µA, at 3.5V.

Current from optional settings (Ialgo/remote/sensorIEGM).

Reduction of ventricular pacing algorithms (RVPa) or Adaptive CRT (aCRTTM) directly modulates the percentage of RV pacing. Depending on the amount of pacing avoided, Ipacing decreases from 1.98–4.37 µA, respectively at 2,5V and 3,5V (100% pacing), down to 0,89−1,97µA (45% of pacing) and 0,59−1,31µA (30% of pacing). User manuals report that these savings are achieved without energy cost. On the other hand, multipoint point pacing for CRT (MPPTM) increases Ipacing on the left ventricular channel up to 3,96 and 8,74µA respectively at both 2,5V and 3,5V.

Threshold algorithms whose sole objective is to guarantee capture, increase pacing outputs and pacing current (user manuals do not specify impact on longevity). Other automatic threshold algorithms aim at optimizing outputs and do this either daily (Capture ManagementTM from Mdt and Capture ControlTM from Btk) or on beat-to-beat basis (Auto captureTM from Abt, Automatic captureTM from Bsc). User manuals do not report an energy cost for the daily algorithms but an energy cost of 1 µA can be derived for beat-to-beat algorithms. The analysis conducted for BsC device (see S1 File: Algorithms influencing pacing current) shows that the beat-to-beat algorithm saves energy (current drain) only if the percentage of pacing is high (>60%) or outputs exceed 2,5V when the algorithm is deactivated.

Rate adaptive pacing usually relies on a G-sensor (accelerometer) to adapt pacing rate according to effort. Adaption of pacing rate can be optionally enhanced with the combination of a minute ventilation (MV) sensor. User manuals describe an estimation of the energy consumption by the MV sensor of (0.69 µA- 0.77 µA). The impact of rate adaptive technology on pacing is unknown and somewhat unpredictable.

For most suppliers, IEGM storage is embedded as a standard function and the energy cost related to EGM is already included in the current background. Only Mdt reports a specific impact on longevity (see S1 File: IEGM). While 6-month storage has minimal effects (0,11–0,34 µA for EnpulseTM, 0,04–0,11 µA for AzureTM), the optional additional use of pre-arrhythmia EGM storage, increases current drain and reduces projected service life by approximately by 34% or 4 months per year for Enpulse (equivalent to 5,67−6,16 µA for EnpulseTM 1,3–1,7µA for AzureTM).

For remote monitoring, assuming 2−4 transmissions per year, the current consumption is around 1,14−1,75 µA for RF solutions (BsC, Btk) while it is 0,09−0,59 µA for Bluetooth connectivity (Mdt, McP). Btk provides a unique solution as its devices transmit data daily (alerts are managed via its website) with a fixed energy cost close to 1,75 µA.

Estimation of nominal longevities, using the PCI model

After deriving current drain, the PCIs were computed and corresponding longevities were modelled, at nominal settings, across all devices. Standard settings, PCI value per device and corresponding longevity, evolution between previous and new generation devices, as well as a sensitivity analysis are reported in S2 File. Fig 2 compares each device per category (SR, DR, CRT-P) according to the PCI chart presented previously. Across all devices, the PCI values range from 26.9 (corresponding to a longevity of 4.2 years) (Enpulse) and 7.6 (with an estimated longevity of 15.1 years) for the single chamber Aveir ventricular device if the programmed output is set to 1.5V).

thumbnail

Fig 2. PCI and longevity with nominal settings for pacemaker devices.

The Power Consumption Index and longevity of previous and current devices describing the contribution of each setting. For conventional pacemakers, the settings considered were: basic rate: 60bpm, pacing threshold: 2.5V, pulse duration: 0.4ms and impedance: 500 ohms for both A&V. For VVI pacemakers, ventricular pacing: 90%. For dual chamber pacemakers, atrial pacing: 70% for SND, 30% for AVB (51% on average) and ventricular pacing assumptions accounted for the difference between AAI/DDD mode and other RVP algorithms (29% vs 47%). Options such as sensor, IEGM storage and remote monitoring (2 transmissions/year) are reported. For leadless VVI and DDD pacemakers, the settings used were: basic rate: 60bpm, pulse duration: 0.25ms for Micra™ and 0.4 ms for Aveir™, impedance: ~ 600 ohms for ventricle and ~300 ohms for the atrium. Pacing outputs were not reported in studies and two options were considered: 1.5 V or 2.5V reflecting the level of confidence of practitioners in adapting output (thresholds observed were typically low: 1.25 V at implant and 0.75 V weeks after). Pacing percentages were the same as for conventional pacemakers. Hysteresis mode was applied for DDD. For CRT-P, BIV was the standard pacing mode (60bpm, 50% A, 100% BIV, 500 ohms) with alternative options such as aCRTTM or MPPTM pacing. Ext. IEGM: extended IEGM, FUP: interrogation of cardiac implant via inductive during a face to face follow-up (one per year).

https://doi.org/10.1371/journal.pone.0333195.g002

The Power Consumption Index and longevity of previous and current devices describing the contribution of each setting.

For conventional pacemakers, the settings considered were: basic rate: 60bpm, pacing threshold: 2.5V, pulse duration: 0.4ms and impedance: 500 ohms for both A&V. For VVI pacemakers, ventricular pacing: 90%. For dual chamber pacemakers, atrial pacing: 70% for SND, 30% for AVB (51% on average) and ventricular pacing assumptions accounted for the difference between AAI/DDD mode and other RVP algorithms (29% vs 47%). Options such as sensor, IEGM storage and remote monitoring (2 transmissions/year) are reported. For leadless VVI and DDD pacemakers, the settings used were: basic rate: 60bpm, pulse duration: 0.25ms for Micra™ and 0.4 ms for Aveir™, impedance: ~ 600 ohms for ventricle and ~300 ohms for the atrium. Pacing outputs were not reported in studies and two options were considered: 1.5 V or 2.5V reflecting the level of confidence of practitioners in adapting output (thresholds observed were typically low: 1.25 V at implant and 0.75 V weeks after). Pacing percentages were the same as for conventional pacemakers. Hysteresis mode was applied for DDD. For CRT-P, BIV was the standard pacing mode (60bpm, 50% A, 100% BIV, 500 ohms) with alternative options such as aCRTTM or MPPTM pacing. Ext. IEGM: extended IEGM, FUP: interrogation of cardiac implant via inductive during a face to face follow-up (one per year).

On average, the PCI for conventional pacemakers is lower for current generations compared with previous generations leading to an increase in longevity for both SR and DR devices (10.8 years vs. 15.4 years for SR and 11.2 years vs. 14.3 years for DR). Unlike standard pacemakers, the average PCI for CRT-P increased (from 12.5 to 14.1) with the introduction of remote monitoring leading to a reduction of longevity (from 8.3 years to 7.8 years). For leadless devices, the PCI reached 14.2 (corresponding to a longevity of 8.8 years) for dual chamber and 11.7 (10.6 years) for single chamber devices demonstrating the consequence of energy cost of transmission between capsules in the two-chamber system.

The split of PCI per current highlights a strong impact of the Ibackground for all categories of pacemaker: more than 50% of PCI is due to Ibackground. The reduction of total PCI for SR/DR between previous and new generations of conventional CIEDs resulted primarily from the reduction of the Ibackground. For conventional pacemakers, Ipacing accounted for only 20% of the total PCI for SR/DR pacemakers and for 30% of the PCI for CRT-P.

Among the contemporary devices, Accolade™ SR/DR had the highest PCI, and thus the lowest estimated longevity (PCI of 13.5 and longevity of 8.5 years for SR, PCI of 14.2 and longevity of 8 years for DR). This is because this device had the highest Ibackground (10µA) as compared with other devices in the same category. On the other hand, Accolade™ DR EL, even with activated remote monitoring, had the lowest PCI (8.9), and thus the longest longevity (12.8 years), thanks to the high battery capacity at 1.6Ah.

For the other devices, differences were primarily due to optional settings such as extended IEGM, sensor or remote monitoring. In the past, IEGM storage negatively impacted Mdt device longevity. This has been significantly improved upon for standard pacemakers. Moreover, remote monitoring power consumption is three times higher for RF solutions than Bluetooth solutions (PCI for remote monitoring 1.2 vs 0.4). For example, the estimated longevity for Edora™ 8 reached 9.7 years for SR and 9.1 years for DR. Azure™ benefits from a low remote monitoring power consumption from Buetooth and reached 12.7 years for SR and 11.7 years for DR (extended IEGM turned “Off”). Alizea™ SR and DR benefit from additional battery capacity especially if remote monitoring is switched off, such that nominal longevity reached 12.9 years, similar to that of the Accolade™ DR EL device. The impact of activating the G-sensor was not different between devices.

For leadless SR and DR devices, assumptions included a basic rate of 60bpm, a pulse duration of 0,25 ms for Micra™ (Mdt) and 0,4 ms for Aveir™ (Abt), an impedance of ~600 ohms for ventricular pacing and ~300 ohms for atrial pacing [2225]. Pacing outputs were not reported in studies and two options were considered: 1.5V or 2.5V, reflecting the level of confidence of practitioners in adapting output (thresholds observed were typically low: 1.25 V at implant and 0.75 V weeks after). Pacing percentages were the same as the one for conventional pacemakers. Hysteresis mode was applied for DDD.

Unlike conventional pacemakers, PCI related to pacing in leadless pacemakers accounted for more of the total PCI (40% on average, higher with 2.5V outputs). This is the consequence of a lower battery capacity and thus, a greater proportion of total available energy is required for pacing. Consequently, PCI and longevity significantly changed depending on pacing output assumptions (Aveir™ SR: 15.1 years if at 1.5V vs. 10.6 years if at 2.5V; Micra™ SR:10.2 years vs. 6.6 years, respectively; Aveir™ ventricular device (DDD mode): 12.5 years vs. 10.6 years; Aveir™ atrial capsule (DDD mode) 7.3 vs. 5.4 years, respectively).

For CRT-P, biventricular pacing is the standard pacing mode albeit there are additional options such as aCRTTM or MPPTM pacing. Biventricular pacing with the Visionist CRT-P device is associated with a PCI of 10.7 and a longevity of 10.6 years, whereas the Percepta™ device with aCRT activated achieves a PCI of 12.3 and longevity of 9.3 years. Not surprisingly Quadra Allure™ with MPP™ activated suffered a considerable increase in PCI and a corresponding reduction in longevity (PCI: 16.9; longevity of 6.8 years).

Sensitivity analysis of longevity related to fluctuations of currents (see S2 File) reveal a standard deviation close to 3–4% (ratio: sigma divided by nominal longevity) across all devices. For a small, clinically achievable sample size this led to a 95% CI for nominal longevity of 0,3–0,7 years, whereas a simulation of 100 revealed a 95%CI of 0,04–0,10 years and for a simulation of 40,000 the 95%CI was 0,003–0007 years.

Validation of PCI model, using Monte-Carlo simulations

Previous generation devices.

Modeled survival curves with standard assumptions fitted Swedish registry data for multiple models with few exceptions (S2 File). For conventional pacemakers, modeled survival curves departed from real-life data for following models: Enrythm™ DR, Evia™ DR-T and Identity™ DR 5370. For Enrythm™, the programming of IEGM storage was the only parameter explaining the difference. For Evia™ DR-T and for Identity™ DR Adx, the difference between modeled survival curves could be explained by the automatic threshold management mode. Overall, the model showed a good fit for CRT-P available in the Swedish registry (InSync III™, Invive™, Frontier II™, Anthem™). This comparison could not be pursued further since real-world programming of implants is not available on the Swedish registry website. Nevertheless, the model showed consistency between real-world and modelled survival curves for most CIEDs from previous generations (Fig 2).

Estimation of current devices’ longevity, using the PCI model

Conventional pacemakers.

Survival curves for current devices are shown in Fig 3 and reported in S3 File. The 95% confidence intervals are extremely low, due to the size (100,000) of the simulated population, allowing a comparison across devices (S3 File). The aggregated survival curves for conventional pacemakers showed wide differences between devices and manufacturers. Accolade™ SR and DR have the shortest estimated lifespan while the extended longevity DR version of Accolade™ offered the best longevity. The impact of programming options is reported in S2 File.

thumbnail

Fig 3. Modelled survival curves for current generation devices without (3a) and with (50% remote transmission adoption and 2 yearly transmission) (3b) activated.

(*) No daily check, 2 Radiofrequency transmissions/year. The figures show the modelled curves for current generation devices.

https://doi.org/10.1371/journal.pone.0333195.g003

The figures show the modelled curves for current generation devices.

Among possible settings, the beat-to-beat automatic threshold management algorithm (Auto captureTM, Automatic captureTM) tended to straighten product survival curve with energy savings on side of the inflexion point and energy cost on the other side, suggesting that optimal programming could extend median product longevity. Reduction of ventricular pacing for intermittent AVB via AAI/DDD mode (available for Btk, Mcp and Mdt devices) extended median longevity for the corresponding devices and reduced the difference between Accolade™ DR EL and Azure™ XT DR or Alizea™ DR. The number of remote transmissions had a marginal effect on longevity. Two devices (Edora™, Alizea™) showed a reduced longevity simply by activation of remote monitoring (by a one-off increase of current for Edora™ or by a reduction of battery capacity for Alizea™).

Discussion

The current study firstly describes a novel way to estimate generator longevity by combining current and battery capacity, and then validates this model across a variety of programming options and previous generation of devices by comparing the modeled data with observed longevity from a country-wide registry, and finally provides estimations of the longevities of currently implanted devices for which there are no reliable observed data.

Longevities from user manuals are difficult to use for implant decision making because manufacturers provide these with a variety of settings, pacing options and configurations. In addition, the lack of a common framework does not facilitate an understanding of the determinants of longevity and a comparison between devices. Calculations of longevity not only should use settings reflecting clinical practice but also split power consumption according to unavoidable current usage (background current) and optional algorithms to help practitioners in their implant decision and subsequent programming. We summarize here the key findings and a few recommendations.

Pacing current

The present analysis did not reveal major differences between manufacturers within each category of device, but, by exploring the differences between conventional pacemakers and leadless pacemakers, demonstrated the relative importance of energy consumption and its impact on device longevity as long as lower pacing outputs (<1.5V) can be achieved without loss of capture. As pacing output reaches 2.5V, power consumption increases and longevity is significantly impacted particularly in those devices with lower battery capacity, emphasizing the need to target pacing output close to 1−1.5V. Particularly in leadless dual chamber devices, the atrial capsule which has a modest energy capacity has the potential to limit longevity if atrial pacing is above 80% due to the additional costs of inter-capsule communication. Presently therefore, a dual chamber approach to SND requires careful consideration, especially given the younger age of the affected population. On the other hand, AV Block, with reliable intrinsic atrial activity is likely to be a more useful application for these until further technological developments improve the energetic demand of this connection.

Limitations

Supporting information

S2 File. Power consumption index rational; Power consumption and nominal longevities; PCI and longevity model sensitivity analysis; Survival curves generated by the Monte-Carlo modelling; Settings and distribution used for Monte-Carlo simulations; Survival curves generated for previous generation devices, Impact of settings for previous generation standard pacemaker devices.

https://doi.org/10.1371/journal.pone.0333195.s002

(DOCX)

References

  1. 1. Schaer BA, Koller MT, Sticherling C, Altmann D, Joerg L, Osswald S. Longevity of implantable cardioverter-defibrillators, influencing factors, and comparison to industry-projected longevity. Heart Rhythm. 2009;6(12):1737–43. pmid:19959121
  2. 2. Alam MB, Munir MB, Rattan R, Adelstein E, Jain S, Saba S. Battery longevity from cardiac resynchronization therapy defibrillators: differences between manufacturers and discrepancies with published product performance reports. Europace. 2017;19(3):421–4. pmid:27009183
  3. 3. Munawar DA, Mahajan R, Linz D, Wong GR, Khokhar KB, Thiyagarajah A, et al. Predicted longevity of contemporary cardiac implantable electronic devices: A call for industry-wide “standardized” reporting. Heart Rhythm. 2018;15(12):1756–63. pmid:30063990
  4. 4. Boriani G, Merino J, Wright DJ, Gadler F, Schaer B, Landolina M. Battery longevity of implantable cardioverter-defibrillators and cardiac resynchronization therapy defibrillators: technical, clinical and economic aspects. An expert review paper from EHRA. Europace. 2018;20(12):1882–97. pmid:29757390
  5. 5. Censi F, Calcagnini G, Mattei E, Ricci RP, Zoni Berisso M, Landolina M, et al. Estimate and reporting of longevity for cardiac implantable electronic devices: a proposal for standardized criteria. Expert Rev Med Devices. 2021;18(12):1203–8. pmid:34854774
  6. 6. Dean J, Sulke N. Pacemaker battery scandal. BMJ. 2016;352:i228. pmid:26848134
  7. 7. Ellis CR, Dickerman DI, Orton JM, Hassan S, Good ED, Okabe T, et al. Ampere Hour as a Predictor of Cardiac Resynchronization Defibrillator Pulse Generator Battery Longevity: A Multicenter Study. Pacing Clin Electrophysiol. 2016;39(7):658–68. pmid:26875541
  8. 8. Lau EW. Technologies for Prolonging Cardiac Implantable Electronic Device Longevity. Pacing Clin Electrophysiol. 2017;40(1):75–96. pmid:27943326
  9. 9. Paton MF, Landolina M, Billuart JR, Field D, Sibley J, Witte K. Projected longevities of cardiac implantable defibrillators: a retrospective analysis over the period 2007-17 and the impact of technological factors in determining longevity. Europace 2020;22:149-155. doi: 10.1093/europace/euz222. Erratum in: Europace. 2020 Jan 1;22(1):155.
  10. 10. Evans JM, Cleves A, Morgan H, Millar L, Carolan-Rees G. ENDURALIFE-Powered Cardiac Resynchronisation Therapy Defibrillator Devices for Treating Heart Failure: A NICE Medical Technology Guidance. Appl Health Econ Health Policy. 2018;16(2):177–86. pmid:29086228
  11. 11. Lau EW. Longevity decoded: Insights from power consumption analyses into device construction and their clinical implications. Pacing Clin Electrophysiol. 2019;42(4):407–22. pmid:30802980
  12. 12. Boston Scientific Longevity calculator, https://www.bostonscientific.com/en-EU/medical-specialties/electrophysiology/device-longevity/longevity-calculator.disclaimer.html
  13. 13. Udo EO, van Hemel NM, Zuithoff NPA, Barrett MJ, Ruiter JH, Doevendans PA, et al. Incidence and predictors of pacemaker reprogramming: potential consequences for remote follow-up. Europace. 2013;15(7):978–83. pmid:23419656
  14. 14. Auricchio A, Ellenbogen KA. Reducing Ventricular Pacing Frequency in Patients With Atrioventricular Block: Is It Time to Change the Current Pacing Paradigm? Circ Arrhythm Electrophysiol. 2016;9(9):e004404. pmid:27637555
  15. 15. Stockburger M, Defaye P, Boveda S, Stancak B, Lazarus A, Sipötz J, et al. Safety and efficiency of ventricular pacing prevention with an AAI-DDD changeover mode in patients with sinus node disease or atrioventricular block: impact on battery longevity-a sub-study of the ANSWER trial. Europace. 2016;18(5):739–46. pmid:26612882
  16. 16. Biffi M, Melissano D, Rossi P, Kaliska G, Havliĉek A, Pelargonio G, et al. The OPTI-MIND study: a prospective, observational study of pacemaker patients according to pacing modality and primary indications. Europace. 2014;16(5):689–97. pmid:24500787
  17. 17. Benezet-Mazuecos J, Iglesias JA, Rubio JM, Cortés M, de la Cruz E, de la Vieja JJ, et al. Limitations of the AutoCapture™ Pacing System in patients with cardiac stimulation devices. Europace. 2014;16(10):1469–75. pmid:24789975
  18. 18. Pierantozzi A, Landolina M, Agricola T, Lunati M, Pisanò E, Lonardi G, et al. Automatic adjustment of stimulation output in resynchronization therapy: impact and effectiveness in clinical practice. Europace. 2011;13(9):1311–8. pmid:21515591
  19. 19. Zoppo F, Gagno G. Left ventricle automatic pacing threshold management in CRT systems: A comprehensive review. J Cardiovasc Electrophysiol. 2020;31(9):2489–98. pmid:32573048
  20. 20. Martin DO, Lemke B, Birnie D, Krum H, Lee KL-F, Aonuma K, et al. Investigation of a novel algorithm for synchronized left-ventricular pacing and ambulatory optimization of cardiac resynchronization therapy: results of the adaptive CRT trial. Heart Rhythm. 2012;9(11):1807–14. pmid:22796472
  21. 21. Swedish ICD, Pacemaker Registry, https://www.pacemakerregistret.se/icdpmr/start.do
  22. 22. Reynolds D, Duray GZ, Omar R, Soejima K, Neuzil P, Zhang S, et al. A Leadless Intracardiac Transcatheter Pacing System. N Engl J Med. 2016;374(6):533–41. pmid:26551877
  23. 23. Piccini JP, Stromberg K, Jackson KP, Laager V, Duray GZ, El-Chami M, et al. Long-term outcomes in leadless Micra transcatheter pacemakers with elevated thresholds at implantation: Results from the Micra Transcatheter Pacing System Global Clinical Trial. Heart Rhythm. 2017;14(5):685–91. pmid:28111349
  24. 24. Knops RE, Reddy VY, Ip JE, Doshi R, Exner DV, Defaye P, et al. A Dual-Chamber Leadless Pacemaker. N Engl J Med. 2023;388(25):2360–70. pmid:37212442
  25. 25. Shantha G, Brock J, Singleton MJ, Schmitt AJ, Kozak P, Bodziock G, et al. A comparative study of the two leadless pacemakers in clinical practice. J Cardiovasc Electrophysiol. 2023;34(9):1896–903. pmid:37522245
联系我们 contact @ memedata.com