advances in arrhythmia and electrophysiology implantable sensors

Advances in Arrhythmia and Electrophysiology

Implantable Sensors for Heart FailureFaisal M. Merchant, MD; G. William Dec, MD; Jagmeet P. Singh, MD, PhD

Heart failure (HF) affects more than 5 million Americans,and acute decompensated heart failure (ADHF) has

emerged as the leading cause of hospitalization among peopleover the age of 65 years.1 Importantly, HF is a leading publichealth concern, and hospitalization expenses related to man-agement of ADHF impose a substantial financial burden onthe health care system. Additionally, readmission rates afterhospitalization for ADHF may be as high as 50% at 6months,2 and insights from the ADHERE registry suggest thata majority of patients admitted with ADHF have a history ofheart failure.1 These data demonstrate that the majority ofpatients admitted with ADHF are known to the medicalsystem and to medical providers, thereby creating an oppor-tunity for upstream strategies that may be capable of detect-ing early HF destabilization and implementing therapies torestabilize the patient and avert hospitalization. The antici-pated changes in the health care system—with a focus onbundled payments for disease management—will necessitaterobust disease management programs to optimize therapy andminimize recurrent admissions once patients have been diag-nosed with HF. Perhaps even more importantly, avertingrepeat episodes of ADHF is likely to have a stabilizing effecton the progression of HF and may improve long-termmorbidity and mortality.

The transition from chronic HF to ADHF involves perturba-tions in multiple intersecting processes including neurohormonalcircuits, inflammatory mediators, cardiorenal interactions, andmyocardial performance. Concurrently, derangements in comor-bid illnesses such as coronary disease, atrial and ventriculararrhythmias, and hypertension also contribute to the patho-physiology of ADHF.3 Ultimately, these multiple pathwayslead to an elevation in ventricular filling pressures and signsof vascular congestion, which, in concert with symptomsfrom impaired cardiac output, lead to the clinical constella-tion of ADHF. This pathophysiologic paradigm highlightsmultiple opportunities for detecting early changes in theprocesses that lead to ADHF with the goal of upstreaminterventions to restabilize the patient and prevent hospital-ization. For example, subtle changes in ventricular fillingpressures or markers to detect early vascular congestion mayserve as important targets for heart failure monitoring. Evenmore proximal perturbations in neurohormonal pathways ormyocardial performance may be amenable to monitoring with

advanced sensor systems and may ultimately prove to be evenmore effective in aborting the ADHF cascade.

The concept of outpatient monitoring for early detectionand treatment of ADHF is not new. However, the question ofwhich parameters to monitor and what specific detectionstrategies should be used to prevent hospitalization has notbeen adequately addressed. Symptoms such as orthopnea andphysical examination signs such as pulmonary rales, periph-eral edema, and elevated jugular venous pressure reflectincreased ventricular filling pressures and vascular conges-tion and are often used for the diagnosis of ADHF. However,these findings have relatively poor sensitivity for detectingacute decompensation, particularly among patients withchronic HF syndromes.4 Additionally, these findings areoften relatively late-stage manifestations reflecting substan-tially elevated ventricular filling pressures. As such, depen-dence on symptoms and physical examination findings alonehas proven ineffective in averting ADHF hospitalizations.Serial monitoring of body weight has some utility in outpa-tient HF monitoring with increases in body weight detectableas early as 30 days before ADHF hospitalization.5 Serialassessment of body weight has been incorporated with otherclinical and physiological parameters (ie, symptoms, activitylogs, blood pressure, heart rate, and oxygen saturation) intomultidisciplinary remote patient monitoring programs to fa-cilitate early detection of ADHF and prevent recurrent hos-pitalizations. Most of these remote monitoring systems in-clude either patient-directed interventions in response to achange in a monitored parameter or direct contact (often viatelephone) with a health care provider to provide instructionson which interventions to undertake (ie, change in medicationor dietary regimen). However, despite the increasing use ofremote monitoring systems and the enhanced use of technol-ogy to facilitate remote monitoring and decision support,results have been variable. A meta-analysis of randomizedclinical trials of remote monitoring for HF patients suggestedreductions of �17% in total mortality, 7% in all-causehospitalization, and 29% for ADHF hospitalization.6 Al-though these results suggest a modest but significant im-provement in patients randomly assigned to remote monitor-ing, given the burgeoning HF epidemic, far more robuststrategies such as advanced sensor technology will be neces-

Received August 26, 2010; accepted October 14, 2010.From the Cardiology Division (F.M.M.), Emory University School of Medicine, Atlanta, Ga; and the Cardiology Division (G.W.D., J.P.S.),

Massachusetts General Hospital Heart Center, Harvard Medical School, Boston, Mass.Correspondence to Jagmeet P. Singh, MD, Cardiac Arrhythmia Service, GRB 109, Massachusetts General Hospital Heart Center, 55 Fruit St, Boston,

MA 02114. E-mail [email protected](Circ Arrhythm Electrophysiol. 2010;3:657-667.)© 2010 American Heart Association, Inc.

Circ Arrhythm Electrophysiol is available at http://circep.ahajournals.org DOI: 10.1161/CIRCEP.110.959502

657

by guest on January 30, 2018http://circep.ahajournals.org/

Dow

nloaded from

http://circep.ahajournals.org/

sary to stem the tide of morbidity and mortality associatedwith episodes of ADHF.

Implantable SensorsThe rapidly expanding role of cardiac implantable electronicdevices (CIEDs) in HF patients presents an opportunity tobroaden the paradigm of outpatient HF monitoring. Tradi-tional roles for CIEDs in HF have focused on electrophysi-ological applications including pacing, antitachycardia ther-apies, and cardiac resynchronization. However, emergingtechnologies designed to couple various biological sensors toCIEDs have opened the door to novel HF monitoring strate-gies that can take broader advantage of the implantabledevices. In the simplest terms, in a closed system a “sensor”is an element that is attached to a device that detects a changeand signals to an “effector” to initiate a response. In theparadigm of HF disease management, the sensor could take anumber of different forms including the patient, a health careprovider, or the device itself. The use of implantable sensorshas several advantages over other monitoring paradigms. Forexample, self-monitoring systems in which the patient acts asthe “sensor” to detect changes in HF status are limited byseveral factors including subjectivity, variable patient ability,and often poor sensitivity for detecting subtle changes duringthe early destabilization process. Other strategies includefrequent in-clinic monitoring in which the health care pro-vider can act as the “sensor” during face-to-face contact.However, this strategy is logistically challenging, highlydependent on patient compliance with frequent clinic visits,and again limited by the poor sensitivity of symptoms andphysical examination findings for detecting early destabiliza-tion. In contrast, implantable sensor strategies have severaltheoretical advantages including continuous monitoring, ob-jectively measured metrics without the bias of subjectiveassessment, and the capability of providing a patient-specificclinical profile that can be analyzed serially overtime withrelative ease.

The concept of device-based monitoring and interventionrepresents a major opportunity for improving outcomes inchronic HF disease management. This article begins with adiscussion of implantable sensor strategies that are already invarious stages of clinical testing and then expands to adiscussion of future horizons in sensors for HF monitoring(Table).

Electrophysiologic SensorsBecause most currently available CIEDs are used for electro-physiological applications, it makes sense that much of thework on sensor strategies has also focused on monitoringelectrophysiological parameters. In the simplest terms, pres-ently available implantable pacemakers and defibrillators arecapable of sensing atrial and ventricular arrhythmias andunder certain circumstances, initiating therapy (either pacingor defibrillator shocks) in a closed loop system in which bothsensor and effector functions are contained within the device.Although detecting and treating rhythm disturbances mayhave a stabilizing effect in HF patients, monitoring otherelectrophysiological parameters may play a more importantrole in predicting and preventing episodes of ADHF. For

example, increases in mean heart rate have been demon-strated before episodes of ADHF and generally return back tobaseline after treatment during hospitalization.7 Such heartrate trends are easily sensed by an implantable device, and,when coupled to either a patient-based or a health careprovider–based effector to initiate therapy, have the potentialto avert hospitalization. Similarly, cardiac autonomic tone asmeasured by short-term heart rate variability (HRV) is apowerful predictor of sudden cardiac death and total mortalityin patients with chronic HF.8,9 HRV can be measured fromCIEDs with atrial leads by determining the standard deviationof 5-minute median atrial-atrial intervals (SDAAM) or con-secutive ventricular (N-N) intervals (SDANN) over a 24-hourperiod. Periods of atrial pacing or high atrial rate episodes,including atrial fibrillation, are excluded from HRV analysis.Reductions in HRV have been shown to predict cardiovas-cular hospitalization with changes in HRV detectable as earlyas 3 weeks before hospitalization.10 The use of an algorithmto detect subtle changes in HRV and predict hospitalizationdemonstrated a sensitivity of 70%, with a false-positivedetection rate of 2.4 events per patient-year. These valueswere derived by applying a single threshold for change in

Table. Sensor Modalities for Heart Failure Monitoring

Sensor Examples

Currently available sensors

Heart rate derivatives Mean heart rate, nocturnal heart rate

Heart rate variability (SDAAM, SDANN)

HRV footprint

Accelerometers Physical activity level

Impedance monitors RV-Can (OptiVol)

LV-RV, LV-can impedance

Minute ventilation

Hemodynamic Right ventricle pressure (Chronicle IHM)

RV dP/dtmax (ePAD)

Left atrial pressure (HeartPOD)

Pulmonary artery pressure (Champion)

Cardiac output Doppler

RV O2 saturation monitor

Heart sounds Peak endocardial acceleration

Emerging modalities

Chemicals PO2, PCO2, pH

Electrolytes, glucose

Biomarkers Natriuretic peptides (BNP, NT-proBNP, ANP)

Inflammatory markers (TNF-a, IL-6, hsCRP)

Troponin

Metabolomic/signalingcascades

Apoptosis/caspase signaling

Glycolysis

Microtubule assembly pathways

SDAAM indicates standard deviation of 5-minute median atrial-atrial inter-vals; SDANN, standard deviation of 5-minute median ventricular-ventricularintervals; LV, left ventricle; RV, right ventricle; HRV, heart rate variability; RV,right ventricle; LV, left ventricle; ePAD, estimated pulmonary artery diastolicpressure; BNP, B-type natriuretic peptide; NT-proBNP, N-terminal proBNP;ANP, atrial natriuretic peptide; TNF-a, tumor necrosis factor-a; IL-6,interleukin-6; and hsCRP, high-sensitivity C-reactive protein.

658 Circ Arrhythm Electrophysiol December 2010


Dow

nloaded from


HRV to all patients. Extension of these findings and thepotential for customized detection thresholds based on indi-vidual patient trends holds promise for further improving thesensitivity and specificity of these algorithms. Additionally,cardiac resynchronization therapy (CRT) improves HRVamong clinical responders, with a shift in autonomic controlaway from sympathetic dominance,11–13 suggesting that HRVmonitoring may also play an important role in predictingresponse to therapy. Future work will be necessary to deter-mine whether specific interventions on the part of either thepatient, health care provider, or device can be used toimplement therapy in response to a sensed change in HRVand avert an episode of ADHF.

Beyond prediction of short-term risk of ADHF, measure-ment of HRV from CRT devices is also a significant predictorof long-term mortality such that patients demonstrating thelowest baseline HRV have significantly worse long-termoutcomes.10,14 Recent work has suggested that device-derivedHRV along with heart rate and physical activity data can beused to risk-stratify patients with HF and CIEDs15 (Figure 1).In this manner, implantable sensors may also play an impor-tant role not only in predicting short-term risk of destabiliza-tion but also in refining long-term risk stratification.

Hemodynamic SensorsFundamental to the pathophysiology of ADHF is an elevationof ventricular filling pressures and therefore, much of thefocus of chronic HF disease management has centered onachieving the lowest ventricular filling pressures possiblewithout compromising cardiac output or further activatingneurohormonal cascades. Serial monitoring of intracardiacand pulmonary pressures may be useful in optimizing medi-cal and device-based therapy. However, to date, obtainingsuch hemodynamic data has required repeated invasive rightheart catheterization (RHC), which is not feasible as part ofan outpatient disease management program, particularlygiven the rapidly increasing numbers of patients with chronicHF. Other options for assessing ventricular filling pressuresinclude noninvasive surrogates such as natriuretic peptide

testing or Doppler echocardiography indices. However, cor-relation between these noninvasive markers and invasiveassessment of pulmonary capillary wedge pressure has beenmodest.16 Serial natriuretic peptide testing has shown promisefor improving outcomes in chronic HF management,17,18 butresults have not been consistent19,20 and readmission ratesremain high. With this background, the opportunity forcontinuous monitoring of intracardiac pressures via implant-able sensors is an attractive option. From an implantabledevice perspective, coupling a pressure transducer to the rightventricle (RV) lead of a pacemaker or defibrillator providesthe simplest form of continuous intracardiac pressure moni-toring. Early work demonstrated the feasibility and safety oflong-term RV pressure monitoring and also documentedexcellent correlation between pressure measurements ob-tained from the implantable sensor and serial invasive RHCmeasurements out to 1 year.21 Additionally, resting pressuremeasurements obtained from the RV sensor between mid-night and 4 AM correlate most closely with resting supinehemodynamic data at the time of RHC,22 allowing directextrapolation of the vast clinical experience with chronic HFmanagement, based on invasive hemodynamic data. Perhapsmost importantly, a validated technique has been developedfor deriving estimated pulmonary artery diastolic (ePAD)pressure from the first derivative of the RV pressure curve(RV dP/dtmax), thus allowing a closer surrogate of left atrialfilling pressures to be extrapolated from an RV pressuresensor.23

A commercially available implantable pressure sensor(Chronicle IHM, Medtronic Inc, Minneapolis, Minn) hasbeen tested across a range of clinical settings. Early datasuggested that the use of a continuous RV pressure sensormay reduce ADHF hospitalizations.24 A subsequent random-ized trial of 274 patients with New York Heart Association(NYHA) class III or IV heart failure demonstrated a nonsig-nificant 21% reduction in HF events among patients ran-domly assigned to continuous hemodynamic monitoring.25 Asecondary efficacy analysis from this study did demonstrate asignificant 36% relative risk reduction in the time to first HFhospitalization among patients implanted with the ChronicleIHM (P�0.03) (Figure 2). Failure to meet the primaryefficacy end point may have been related to a lower-than-expected event rate in the control cohort and suggests thatfuture studies, potentially with increased statistical power,will be necessary to more clearly define the utility ofimplantable hemodynamic monitoring in this patient popula-tion. Additionally, beyond attempting to reduce HF hospital-izations, data from studies on continuous RV and ePADpressure monitoring have also yielded important insights intothe pathophysiology of HF with reduced versus preservedejection fraction (EF), demonstrating important differences inbaseline ventricular filling pressures and ventricular distensi-bility between the groups.26 Patients in both EF groupsdemonstrated significant increases in ePAD before episodes ofADHF, although the number of days of advanced notice wassignificantly less (19.3�17.3 versus 29.1�22.3, P�0.05) andthe rate of rise in filling pressures was significantly greater inpatients with preserved EF compared with those with low EF.These findings may have important implications for tailoring

Figure 1. Prognostic utility of a risk score based on 4 simplesensor-derived parameters (SDANN, HRV footprint, HR, andphysical activity) to risk-stratify HF patients for long-term mortal-ity. From Singh et al.15

Merchant et al Implantable Sensors for Heart Failure 659


Dow

nloaded from


chronic HF management strategies for patients in different EFsubgroups.

Conceptually, continuous measurement of RV and extrap-olated pulmonary pressures from an RV transducer representsthe most straightforward form of implantable sensor-basedhemodynamic monitoring. However, most patients admittedfor ADHF have pulmonary congestion caused by elevated leftatrial pressure.27 Therefore, continuous monitoring of leftatrial pressure (LAP) may provide a more robust target forimplantable sensor-based strategies. The HeartPOD (St JudeMedical Inc, Minneapolis, Minn) is a permanently implant-able LAP sensor inserted during transseptal cardiac catheter-ization (Figure 3). The LAP sensor is coupled via a subcuta-neous antenna to a hand-held patient advisor module, which

contains a range of patient alerts with reminders to takemedications or obtain additional LAP recordings. Addition-ally, the patient advisor module is able to generate a custom-ized patient prescription for several different parametersincluding medication dosage, activity level, sodium and fluidintake, and physician contact. The prescription is a physician-directed, patient self-management program that is customizedon the basis of LAP measurements. In an early stagesafety/efficacy observational study, 40 patients with NHYAclass III or IV HF were implanted with the LAP sensor andfollowed for a median of 25 months.28 All patients underwentsuccessful device placement without any major device-related complications. LAP-guided therapy resulted in im-provements in several important efficacy end points includingreductions in mean daily LAP, improvements in functionalcapacity, and higher doses of neurohormonal antagonists withreductions in daily diuretic requirements. Additionally, ele-vations in LAP were clearly identifiable during the monthbefore episodes of ADHF, suggesting that continuous LAPmonitoring may be effective in reducing HF hospitalizations.Although not specifically designed to assess reductions inhospitalizations or other clinical end points, landmark analy-sis from this study also demonstrated a significant improve-ment in event-free survival (death or ADHF hospitalization)during the period of LAP monitoring compared with theobservation period of standard HF therapy (event-free sur-vival: 95% versus 77%, P�0.012) (Figure 4). Further studieswill be necessary to validate the impact of LAP sensors onclinical outcomes.

Another emerging technology for implantable hemody-namic sensors is direct pulmonary pressure measurement viaan implantable pulmonary artery (PA) pressure transducer(Champion, CardioMEMS, Atlanta, Ga). In a pivotal trialrecently reported in preliminary form, 550 patients with classIII HF were randomly assigned to PA pressure monitoring orstandard care.29 The safety profile of the PA sensor was

Figure 2. Significant improvement in time to first HF hospitaliza-tion associated with chronic right ventricle pressure monitoring(Chronicle IHM) compared with standard HF therapy. Reprintedwith permission from J Am Coll Cardiol.25

Figure 3. A, Chest radiograph of a patient implanted with an LAP monitoring system via the right subclavian vein and a biventricularpacemaker/defibrillator via the left subclavian vein. The LAP sensor module (SM) is implanted percutaneously across the atrial septum.Coil antenna (CA), biventricular pacemaker/implantable cardioverter-defibrillator (CRT-D), right atrial lead (RA), right ventricular lead(RV), and left ventricular lead (LV) are shown. B, Intracardiac echocardiography image of the LAP sensor module positioned across theatrial septum with the sensor diaphragm secured to the left atrial (LA) side of the septum.



Dow

nloaded from


excellent, without any major device-related adverse events,and all devices functioning appropriately during a median15-month follow-up. PAP-guided management demonstrateda significant 30% relative risk reduction in HF hospitalizationat 6 months (the primary efficacy end point) (31% versus44%, P�0.001). This trial is the first to demonstrate asignificant and clinically important reduction in ADHF epi-sodes with an implantable sensor–based strategy and servesas an important proof of concept for future sensor modalities.

Impedance MonitoringAnother potential target for implantable sensors is thoracicimpedance (Z) monitoring. Impedance monitoring is based onthe premise that a drop in the electric impedance (ohms, �)across the thoracic cavity reflects an increase in tissue fluidcontent in the interpositioned pulmonary tissue and signals astate of volume overload and fluid retention.30 Early studiesused a noninvasive impedance cardiograph to measure tho-racic electric impedance using electrodes on the body surfaceand demonstrated significant associations between changes inimpedance and risk of near-term HF decompensation.31

Unfortunately, noninvasive impedance monitoring requiresrepeated clinic visits for serial monitoring and is also limitedby poor reproducibility caused by changes in electrodeposition. However, impedance monitoring can be easilycoupled to a CIED. By passing an electric current between thedevice can and the RV coil, Z can be measured in a safe andreproducible manner and yields results comparable to nonin-vasive impedance cardiography. The OptiVol Fluid StatusMonitoring system (Medtronic Inc, Minneapolis, Minn) usesa proprietary system that compares daily impedance measure-ments with a baseline reference value. The reference value isa slow-moving average of preceding impedance values. The

difference in impedance between the daily measurement andthe reference value is summed to generate an OptiVol fluidindex such that an increase in OptiVol fluid index corre-sponds to a cumulative decrease in impedance measurements,reflecting a gradual increase in pulmonary fluid content(Figure 5). Notably, a decrease in impedance is typically seenduring the first several weeks after device implantation and isattributed to pocket edema and local inflammation.32 Imped-ance recordings during this early period are not used forclinical decision-making or for determining the referenceimpedance value.

A strong inverse correlation has been defined betweenintrathoracic impedance and pulmonary capillary wedge pres-sure and between impedance and net fluid loss amongpatients hospitalized for ADHF.33 Additionally, this worksuggested that an OptiVol threshold of 60 �-days yielded thegreatest diagnostic accuracy in predicting impending ADHFhospitalization without substantially compromising specific-ity by increasing the “false-positive” detection rate (ie,crossing the threshold without subsequent hospitalization).Several studies have demonstrated significant reductions inimpedance preceding episodes of ADHF, with a sensitivityranging from 60% to 77% using the 60 �-day threshold forpredicting impending hospitalization with a relatively lowfalse detection rate ranging from 0.2 to 1.5 alert eventswithout hospitalization per patient-year.33–35 Additionally, atleast 1 study has suggested that a strategy of audible devicealerts based on impedance monitoring might reduce HFhospitalization.34 Although the published sensitivity rates ofimpedance monitoring for predicting ADHF suggest that thistechnology may be an important adjunct in HF management,the positive and negative predictive values have been rela-tively modest, compromised by both “false-positive” and“false-negative” events. Early studies have suggested thatepisodes classified as “false positives” are often due to verymild signs or symptoms of congestive HF and may in facttrigger changes in treatment but do not necessarily lead tohospitalization for ADHF.33,35 It is likely that many of theseevents are clinically relevant and therefore should not beclassified as “false positives.” Analysis of “false-negative”events (ie, hospitalization for ADHF without crossing theimpedance threshold) suggests that these events cluster inpatients who present more often with signs of peripheraledema or symptoms of low cardiac output and less frequentlywith pulmonary congestion.35 These findings highlight one ofthe major challenges in evaluating the efficacy of sensorstrategies and that is the lack of a true gold-standard for thediagnosis of ADHF that can be broadly applied to all patients.To date, most studies have relied on a combination of clinicalevaluation and adjunctive testing (natriuretic peptides, chestradiography, RHC…) for adjudicating whether a hospitaliza-tion is due to ADHF or not. However, all methods fordetecting ADHF, including clinical evaluation4 and natriuret-ic peptide testing,20 are imperfect and therefore pose a majorlimitation when serving as the gold standard for testing newdiagnostic modalities. New technologies such as impedancemonitoring or hemodynamic sensors that are designed to detectvery early changes may be inappropriately penalized for detect-ing subtle changes in volume status or ventricular loading

Figure 4. Landmark analysis from the HOMEOSTASIS studydemonstrating trend toward improved event-free survival (deathor ADHF hospitalization) during the period of left atrial pressuremonitoring compared with the observational period of standardHF care (event-free survival: 95% versus 77%, P�0.012).Reprinted with permission from Circulation.28



Dow

nloaded from


conditions, which, although clinically important, may not nec-essarily lead to ADHF hospitalization. Addressing these limita-tions will be a major challenge in future clinical studies evalu-ating the efficacy of new heart failure technologies.

Initial work in impedance monitoring has focused onmeasurements between the device can and right-sided leads,reflecting electric conductance across the heart, lungs, andother interpositioned tissues, thus reflective of total thoracic

Figure 5. Clinical example of a patient implanted with a thoracic impedance monitor. Top panel demonstrates the onset of atrial fibrilla-tion (red line). Bottom panel demonstrates a progressive decrease in thoracic impedance after the onset of atrial fibrillation with a cor-responding increase in the OptiVol fluid index, suggesting a significant increase in pulmonary fluid content. In this example, both theatrial fibrillation and the pulmonary edema probably will need to be addressed to restore homeostasis.



Dow

nloaded from


impedance. However, with biventricular device platforms,numerous other vectors are available for measuring imped-ance between the device can, right atrium, right and leftventricles. Studies in a dog model of pacing-induced heartfailure have suggested that although all measured vectorsdemonstrate a consistent reduction in Z as HF develops, leftventricle (LV)-dependent vectors demonstrated the fastestrate of change and the greatest magnitude of change.36

Specifically, Z measured from the LV-can vector demon-strated the closest correlation with LV end-diastolic volumeand left atrial pressure, probably reflecting a drop in imped-ance because of both increased pulmonary water content andprogressive LV cavity dilation, both of which would tend toimprove electric conductance. In other animal models, intra-cardiac impedance measured between the LV and RV leads ina CRT platform has also been closely correlated with end-di-astolic volume, end-diastolic pressure,37 stroke volume, andLV dP/dtmax.38 Furthermore, in patients undergoing CRTdevice implantation, intracardiac impedance (LV-RV) ismore closely correlated with echocardiographic changes inLV volume than intrathoracic impedance (RV-can).39 Thesefindings suggest that changes in LV volume and pressureconditions may produce a more profound change in imped-ance along the LV-RV vector, encompassing only cardiactissue, as opposed to the RV-can vector, which encompassesa larger area of noncardiac tissue and may be less sensitive tominute changes in LV loading conditions. Dynamic changesin interlead or lead-can impedance, accompanying the volu-metric changes of the LV during systole and diastole, mayalso have the potential to provide a close correlate for cardiacoutput. However, this strategy has not been tested in clinicalsettings.

Beyond trying to predict impending episodes of ADHF byserving as a surrogate for LV pressure and volume conditions,intracardiac impedance sensors may also play a role in deviceoptimization for patients undergoing CRT device implanta-tion. Measurement of intracardiac impedance has beenclosely correlated with invasive LV stroke volume measure-ments as a technique to guide LV lead position in patientsundergoing CRT device implantation.40 Similarly, intracar-diac impedance has been shown to compare favorably withboth echocardiography and invasive hemodynamic pressuremeasurement to guide atrioventricular (AV) and interventric-ular (VV) delays during CRT.41

Although the primary focus of impedance monitoring fordeveloping HF sensors has centered on detecting changes inlung water content and LV loading conditions, thoracicimpedance has also played a long-standing role in measuringminute ventilation (MV) to guide rate-responsive cardiacpacing. Changes in thoracic impedance over the course of therespiratory cycle can be used to measure tidal volume andMV42 and correlate closely with directly measured MV fromthe flowmeter of a respiratory gas analysis system.43

Impedance-based monitoring of MV has been incorporatedinto algorithms to predict the risk of impending ADHF44 andmay play a complimentary role in sensor-based heart failuremonitoring strategies.

Other Sensor ModalitiesPhysical activity level can also be incorporated into sensormonitoring strategies by extrapolating data from accelerom-eter signals, which are already incorporated into many CIEDsfor modulating heart rate during exercise. Physical activity levelcan be estimated by integrating the accelerometer-detectedactivity counts in each minute. A minute is considered “active”if the counts exceed a threshold that corresponds to a walkingrate of approximately 70 steps per minute,10 and a low activitythreshold has been defined as less than 1 hour of activity per dayaveraged over a 1-week period.45 Physical activity level has beenincorporated into sensor-based monitoring strategies to predictshort-term risk of ADHF,45 long-term mortality,15 and clinicalresponse to CRT.46

Continuous monitoring of cardiac output (CO) via animplantable sensor can also be performed with a high degreeof accuracy. Most implantable sensors have used a techniqueof measuring oxygen saturation from a photosensitive diodeand then extrapolating CO from the venous oxygen saturationbased on the Fick equation. Several studies have demon-strated the feasibility of continuous CO monitoring based onmixed venous oxygen saturation in the RV and have demon-strated close correlations with invasive CO measurementsacross a broad range of oxygen saturation levels.47,48 Al-though combining both continuous pressure monitoring andCO monitoring from a single sensor located in the RV isfeasible,47 additional studies are needed to define the incre-mental value of continuous CO monitoring in addition topressure monitoring. In general, as clinical HF managementhas tended to focus less and less on CO as a diagnostic andtherapeutic target and more and more on ventricular fillingpressures and neurohormonal status, the importance of COmonitors as part of a sensor-based HF management strategyremains unclear.

Another modality for CIED mounted sensors is peakendocardial acceleration (PEA), which measures the maxi-mum amplitude of the vibrations produced by the first heartsound by using an implantable micro accelerometer mountedwithin the tip of a conventional pacemaker lead.30 PEA isstrongly correlated with changes in LV and RV contractility(dP/dtmax) during inotrope infusion but not during pacing-induced chronotropic stimulation,27,49 suggesting that moni-toring PEA might serve as a useful surrogate for changes inmyocardial contractility. Additionally, PEA sensors have alsobeen used to guide device optimization in patients receivingCRT, and studies have suggested that AV and VV optimiza-tion using PEA may be equivalent to echocardiography-basedalgorithms50 and LV dP/dtmax algorithms.51

Multimodality Sensor StrategiesHeart failure is a highly complex disease entity representingthe interplay of perturbations in multiple processes includingmyocardial contractility, systemic inflammation, neurohor-monal activation, and volume retention. As such, multimo-dality sensor-based strategies that are designed to monitorseveral different aspects of the HF cascade may prove moreuseful than monitoring any one individual parameter. Asproof of this concept, recent studies have demonstrated theincremental value of monitoring several different parameters



Dow

nloaded from


using CIEDs. The PARTNERS-HF study prospectively fol-lowed patients receiving CRT devices to multimodality mon-itoring including detection of atrial fibrillation and ventricularresponse rate, thoracic impedance monitoring with the Op-tiVol system, patient activity level, nocturnal heart ratetrends, heart rate variability (SDAAM), overall percentage ofCRT pacing, and defibrillator therapies.45 The multimodalityalgorithm used a diagnostic threshold of abnormal measureson at least 2 of the 7 monitored parameters during a singlemonitoring period to predict impending ADHF. Severalimportant insights emerged from this well-monitored cohort.First and perhaps most importantly, the use of a diagnosticalgorithm based on multimodality criteria significantly im-proved the ability to predict impending ADHF hospitalizationwithin the subsequent 30 days when compared with algo-rithms based on monitoring thoracic impedance alone (hazardratio, 5.5 versus 2.7, P�0.0001). Additionally, there weresignificant improvements in the predictive capability of themultimodality algorithm when the frequency of monitoringwas increased from quarterly to monthly to semimonthly,lending support to the notion that more frequent monitoring ismore effective. This capability to increase the frequency ofmonitoring is one of the major advantages of implantablesensor strategies.

Another study assessed the complementary nature of bothePAD pressure monitoring via an RV pressure sensor andthoracic impedance monitoring.52 Measurements of ePADand Z demonstrated a modest inverse correlation, particularlyin the 14 and 30 days before an HF event. Importantly,impedance remained relatively constant until ePAD crossed aparticular threshold at which point, presumably, compensa-tory mechanisms to regulate lung water accumulation wereexhausted and Z began to fall. These observations highlightthe primary role of increased cardiac filling pressures intriggering episodes of ADHF and suggest that multiple sensorstrategies might be optimally employed by using a temporalsequence whereby hemodynamic/pressure-based sensors areused for early monitoring/prevention and impedance-basedsensors are used for early detection/treatment (Figure 6). Inthis paradigm, early increases in cardiac filling pressurescould be treated with intensified neurohormonal blockadewith the goal of preventing ePAD from crossing the thresholdwhereby lung water begins to accumulate. If that threshold is

crossed, as detected by a concurrent drop in impedance, diureticsmay become necessary, in addition to neurohormonal antago-nists, with the goal of averting ADHF hospitalization.

Opportunities for Future Sensor DevelopmentSeveral opportunities exist for enhancing the role of implant-able sensors in HF management. As discussed earlier, pertur-bations in multiple neurohormonal and inflammatory cas-cades contribute to ADHF. More sophisticated sensortechnologies may be capable of detecting these very earlypathophysiologic changes and using them to target upstreamtherapy. For instance, numerous biomarkers are significantlyaltered in both acute and chronic HF including blood chem-istries (electrolytes, glucose, pH), natriuretic peptides (ie,B-type natriuretic peptide [BNP], N-terminal proBNP, atrialnatriuretic peptide),53 inflammatory makers (ie, tumor necro-sis factor-�, interleukin-6, interleukin-1),54 oxidative stressmarkers (ie, myeloperoxidase),55 and collagen turnover/extra-cellular matrix peptides (ie, C-terminal propeptide of collagentype 1, matrix metalloproteinase, tissue inhibitor of matrixmetalloproteinases).56 Additionally, using metabolomic tech-nologies, novel pathways are being implicated in the patho-physiology of HF and may identify markers that are capableof detecting even more subtle alterations in the failing heartsuch as changes in myocardial signaling and energetics.57

These observations open the door to novel implantablesensors that are capable of measuring circulating levels ofmetabolites from multiple interacting pathways and providingan earlier, more sensitive profile of impending HF destabili-zation. Such metabolite-based sensors may also enable themonitoring and treatment of coexistent comorbidities such asischemic heart disease and diabetes in the HF patient.

Opportunities also exist for improvements in the design ofsensors. For instance, the use of biocompatible materials mayenhance the longevity and safety of implanted devices, whichcan remain in situ for decades. Similarly, the use of novelenergy scavenging technologies to power sensors usingsources such as motion or temperature gradients may over-come many of the constraints of currently available CIEDs,which are limited by finite battery lifetime and size.58

Perhaps the most important challenge for the future ofsensor-based strategies is developing tools to directly coupleimplantable sensors to equally robust effectors. To demon-

Figure 6. Theoretical paradigm for a multiplesensor strategy in which hemodynamic sensorsand impedance-based sensors are used in atemporal sequence to facilitate early preventionand treatment of HF destabilization.



Dow

nloaded from


strate an important improvement in clinical outcomes, theinformation gathered from an implantable sensor will need tobe coupled to an effector that is capable of instituting therapyin a rapid and robust manner. To date, most sensor strategieshave transmitted clinical information to a health care pro-vider, who is then responsible for “closing the loop” byimplementing an appropriate therapeutic intervention. Thisstrategy has several limitations including the potential fortime lags between the sensed event and the implementation ofthe appropriate intervention. Additionally, a health careprovider is still required to interpret the data and makeappropriate clinical decisions, which, in light of the burgeon-ing HF epidemic, may require substantial resources. Alterna-tively, information from implanted sensors may be relayeddirectly to the patient, who then serves as the primaryeffector. However, patient-driven strategies suffer from vari-ability in patient capability and compliance and also requirerelatively simple decision-making algorithms such that alarge percentage of patients are capable of interpreting theinformation from the sensor and implementing an appropriatetherapeutic strategy. Perhaps the greatest potential for CIEDsin HF management lies in the ability to directly couple bothsensor and effector functions within the device. This strategywould allow for a truly “closed-loop” system, which wouldcompletely eliminate dependence on human factors. Addi-tionally, the sensor-effector unit could work in an iterativefashion with the sensor able to measure the response to aparticular intervention and then make further changes basedon the dynamic nature of chronic HF.

As proof of concept, currently available rhythm manage-ment devices are capable of sensing changes and implement-ing therapy (ie, antitachycardia pacing in response to sensedventricular tachycardia) within an iterative and closed-loopsystem. Similar systems could be developed for other aspectsof HF disease management. For instance, there is greatinterest in the development of microelectromechanical sys-tems capable of controlled drug delivery via an implantabledevice.59 It is conceivable that the timely and potentiallylocalized delivery of various pharmacological agents (vaso-dilators, diuretics, antiarrhythmics) could serve as a usefultreatment modality for chronic HF management. Similarly,devices may soon be capable of modulating the AV interval,altering LV-RV pacing timings, or even delivering otherforms of treatment such as cardiac contractility modulation60

in response to a drop in myocardial contractility (detected bya PEA sensor) or an elevation in filling pressures (measuredby an LAP sensor). Ultimately, to realize their true impact,implantable sensor strategies will need to be coupled toequally robust effector strategies to truly stabilize the HFdisease process and improve patient outcomes.

The expanding role of implantable sensors is leading to aparadigm shift in HF management. Implantable sensors willbecome part of routine clinical care, and, when coupled withremote monitoring, will allow the practice of a more person-alized form of medicine. Sensor strategies will continue toevolve, and in addition to facilitating risk stratification andpredicting risk of ADHF, will enable early, automated ther-apeutic interventions and improve clinical outcomes.

DisclosuresDr Singh received research grants from St Jude Medical, MedtronicInc, Boston Scientific Corp, and Biotronik, served on advisoryboards or as a consultant for Boston Scientific Corp, St JudeMedical, and Medtronic, and received honoraria or speaker fees fromMedtronic Inc, Biotronik, Guidant Corp, St Jude Medical, andSorin Group.

References1. Fonarow GC, Heywood JT, Heidenreich PA, Lopatin M, Yancy CW.

Temporal trends in clinical characteristics, treatments, and outcomes forheart failure hospitalizations, 2002 to 2004: findings from Acute Decom-pensated Heart Failure National Registry (ADHERE). Am Heart J. 2007;153:1021–1028.

2. Krumholz HM, Parent EM, Tu N, Vaccarino V, Wang Y, Radford MJ,Hennen J. Readmission after hospitalization for congestive heart failureamong Medicare beneficiaries. Arch Intern Med. 1997;157:99–104.

3. Gheorghiade M, De Luca L, Fonarow GC, Filippatos G, Metra M, FrancisGS. Pathophysiologic targets in the early phase of acute heart failuresyndromes. Am J Cardiol. 2005;96:11G–17G.

4. Stevenson LW, Perloff JK. The limited reliability of physical signs forestimating hemodynamics in chronic heart failure. JAMA. 1989;261:884–888.

5. Chaudhry SI, Wang Y, Concato J, Gill TM, Krumholz HM. Patterns ofweight change preceding hospitalization for heart failure. Circulation.2007;116:1549–1554.

6. Klersy C, De Silvestri A, Gabutti G, Regoli F, Auricchio A. A meta-analysis of remote monitoring of heart failure patients. J Am Coll Cardiol.2009;54:1683–1694.

7. Ellery S, Pakrashi T, Paul V, Sack S. Predicting mortality and rehospi-talization in heart failure patients with home monitoring: the HomeCARE pilot study. Clin Res Cardiol. 2006;95(Suppl 3):III29–III35.

8. La Rovere MT, Pinna GD, Maestri R, Mortara A, Capomolla S, Febo O,Ferrari R, Franchini M, Gnemmi M, Opasich C, Riccardi PG, Traversi E,Cobelli F. Short-term heart rate variability strongly predicts suddencardiac death in chronic heart failure patients. Circulation. 2003;107:565–570.

9. Binder T, Frey B, Porenta G, Heinz G, Wutte M, Kreiner G, Gossinger H,Schmidinger H, Pacher R, Weber H. Prognostic value of heart rate variabilityin patients awaiting cardiac transplantation. Pacing Clin Electrophysiol.1992;15:2215–2220.

10. Adamson PB, Smith AL, Abraham WT, Kleckner KJ, Stadler RW, ShihA, Rhodes MM. Continuous autonomic assessment in patients with symp-tomatic heart failure: prognostic value of heart rate variability measuredby an implanted cardiac resynchronization device. Circulation. 2004;110:2389–2394.

11. Fantoni C, Raffa S, Regoli F, Giraldi F, La Rovere MT, Prentice J, PastoriF, Fratini S, Salerno-Uriarte JA, Klein HU, Auricchio A. Cardiac resyn-chronization therapy improves heart rate profile and heart rate variabilityof patients with moderate to severe heart failure. J Am Coll Cardiol.2005;46:1875–1882.

12. Braunschweig F, Mortensen PT, Gras D, Reiser W, Lawo T, Mansour H,Sogaard P, Stegemann B, Bruns HJ, Linde C. Monitoring of physicalactivity and heart rate variability in patients with chronic heart failureusing cardiac resynchronization devices. Am J Cardiol. 2005;95:1104–1107.

13. Adamson PB, Kleckner KJ, VanHout WL, Srinivasan S, Abraham WT.Cardiac resynchronization therapy improves heart rate variability inpatients with symptomatic heart failure. Circulation. 2003;108:266–269.

14. Gilliam FR III, Singh JP, Mullin CM, McGuire M, Chase KJ. Prognosticvalue of heart rate variability footprint and standard deviation of average5-minute intrinsic R-R intervals for mortality in cardiac resynchronizationtherapy patients. J Electrocardiol. 2007;40:336–342.

15. Singh JP, Rosenthal LS, Hranitzky PM, Berg KC, Mullin CM, ThackerayL, Kaplan A. Device diagnostics and long-term clinical outcome inpatients receiving cardiac resynchronization therapy. Europace. 2009;11:1647–1653.

16. Dokainish H, Zoghbi WA, Lakkis NM, Al-Bakshy F, Dhir M, QuinonesMA, Nagueh SF. Optimal noninvasive assessment of left ventricularfilling pressures: a comparison of tissue Doppler echocardiography andB-type natriuretic peptide in patients with pulmonary artery catheters.Circulation. 2004;109:2432–2439.

17. Lainchbury JG, Troughton RW, Strangman KM, Frampton CM, PilbrowA, Yandle TG, Hamid AK, Nicholls MG, Richards AM. N-terminal



Dow

nloaded from


pro-B-type natriuretic peptide-guided treatment for chronic heart failure:results from the BATTLESCARRED (NT-proBNP-Assisted TreatmentTo Lessen Serial Cardiac Readmissions and Death) trial. J Am CollCardiol. 2009;55:53–60.

18. Berger R, Moertl D, Peter S, Ahmadi R, Huelsmann M, Yamuti S,Wagner B, Pacher R. N-terminal pro-B-type natriuretic peptide-guided,intensive patient management in addition to multidisciplinary care inchronic heart failure a 3-arm, prospective, randomized pilot study. J AmColl Cardiol. 2010;55:645–653.

19. Pfisterer M, Buser P, Rickli H, Gutmann M, Erne P, Rickenbacher P,Vuillomenet A, Jeker U, Dubach P, Beer H, Yoon SI, Suter T, OsterhuesHH, Schieber MM, Hilti P, Schindler R, Brunner-La Rocca HP. BNP-guided vs symptom-guided heart failure therapy: the Trial of Intensifiedvs Standard Medical Therapy in Elderly Patients With Congestive HeartFailure (TIME-CHF) randomized trial. JAMA. 2009;301:383–392.

20. Lewin J, Ledwidge M, O’Loughlin C, McNally C, McDonald K. Clinicaldeterioration in established heart failure: what is the value of BNP andweight gain in aiding diagnosis? Eur J Heart Fail. 2005;7:953–957.

21. Magalski A, Adamson P, Gadler F, Boehm M, Steinhaus D, Reynolds D,Vlach K, Linde C, Cremers B, Sparks B, Bennett T. Continuous ambu-latory right heart pressure measurements with an implantable hemody-namic monitor: a multicenter, 12-month follow-up study of patients withchronic heart failure. J Card Fail. 2002;8:63–70.

22. Adamson PB, Kjellstrom B, Braunschweig F, Magalski A, Linde C,Kolodiezj A, Cremers B, Bennett T. Ambulatory hemodynamic moni-toring from an implanted device: components of continuous 24-hourpressures that correlate to supine resting conditions and acute right heartcatheterization. Congest Heart Fail. 2006;12:14–19.

23. Reynolds DW, Bartelt N, Taepke R, Bennett TD. Measurement of pul-monary artery diastolic pressure from the right ventricle. J Am CollCardiol. 1995;25:1176–1182.

24. Adamson PB, Magalski A, Braunschweig F, Bohm M, Reynolds D,Steinhaus D, Luby A, Linde C, Ryden L, Cremers B, Takle T, Bennett T.Ongoing right ventricular hemodynamics in heart failure: clinical value ofmeasurements derived from an implantable monitoring system. J Am CollCardiol. 2003;41:565–571.

25. Bourge RC, Abraham WT, Adamson PB, Aaron MF, Aranda JM Jr,Magalski A, Zile MR, Smith AL, Smart FW, O’Shaughnessy MA, JessupML, Sparks B, Naftel DL, Stevenson LW. Randomized controlled trial ofan implantable continuous hemodynamic monitor in patients withadvanced heart failure: the COMPASS-HF study. J Am Coll Cardiol.2008;51:1073–1079.

26. Zile MR, Bennett TD, St John Sutton M, Cho YK, Adamson PB, AaronMF, Aranda JM Jr, Abraham WT, Smart FW, Stevenson LW, Kueffer FJ,Bourge RC. Transition from chronic compensated to acute decom-pensated heart failure: pathophysiological insights obtained from con-tinuous monitoring of intracardiac pressures. Circulation. 2008;118:1433–1441.

27. Plicchi G, Marcelli E, Parlapiano M, Bombardini T. PEA I and PEA IIbased implantable haemodynamic monitor: pre clinical studies in sheep.Europace. 2002;4:49–54.

28. Ritzema J, Troughton R, Melton I, Crozier I, Doughty R, Krum H,Walton A, Adamson P, Kar S, Shah PK, Richards M, Eigler NL, WhitingJS, Haas GJ, Heywood JT, Frampton CM, Abraham WT. Physician-directed patient self-management of left atrial pressure in advancedchronic heart failure. Circulation. 2010;121:1086–1095.

29. Abraham WT, Adamson P. Primary results of the CardioMEMS heartsensor allows monitoring of pressure to improve outcomes in NYHAclass III heart failure patients (CHAMPION) trial ESC Heart FailureCongress 2010. Berlin, Germany, 2010.

30. Braunschweig F. Therapeutic and diagnostic role of electrical devices inacute heart failure. Heart Fail Rev. 2007;12:157–166.

31. Packer M, Abraham WT, Mehra MR, Yancy CW, Lawless CE, MitchellJE, Smart FW, Bijou R, O’Connor CM, Massie BM, Pina IL, GreenbergBH, Young JB, Fishbein DP, Hauptman PJ, Bourge RC, Strobeck JE,Murali S, Schocken D, Teerlink JR, Levy WC, Trupp RJ, Silver MA.Utility of impedance cardiography for the identification of short-term riskof clinical decompensation in stable patients with chronic heart failure.J Am Coll Cardiol. 2006;47:2245–2252.

32. Luthje L, Drescher T, Zenker D, Vollmann D. Detection of heart failuredecompensation using intrathoracic impedance monitoring by a triple-chamber implantable defibrillator. Heart Rhythm. 2005;2:997–999.

33. Yu CM, Wang L, Chau E, Chan RH, Kong SL, Tang MO, Christensen J,Stadler RW, Lau CP. Intrathoracic impedance monitoring in patients with

heart failure: correlation with fluid status and feasibility of early warningpreceding hospitalization. Circulation. 2005;112:841–848.

34. Catanzariti D, Lunati M, Landolina M, Zanotto G, Lonardi G, Iacopino S,Oliva F, Perego GB, Varbaro A, Denaro A, Valsecchi S, Vergara G. Mon-itoring intrathoracic impedance with an implantable defibrillator reduceshospitalizations in patients with heart failure. Pacing Clin Electrophysiol.2009;32:363–370.

35. Vollmann D, Nagele H, Schauerte P, Wiegand U, Butter C, Zanotto G,Quesada A, Guthmann A, Hill MR, Lamp B. Clinical utility ofintrathoracic impedance monitoring to alert patients with an implanteddevice of deteriorating chronic heart failure. Eur Heart J. 2007;28:1835–1840.

36. Khoury DS, Naware M, Siou J, Blomqvist A, Mathuria NS, Wang J, ShihHT, Nagueh SF, Panescu D. Ambulatory monitoring of congestive heartfailure by multiple bioelectric impedance vectors. J Am Coll Cardiol.2009;53:1075–1081.

37. Stahl C, Beierlein W, Walker T, Straub A, Nagy Z, Knubben K, GreinerTO, Lippert M, Czygan G, Paule S, Schweika O, Kuhlkamp V. Intra-cardiac impedance monitors hemodynamic deterioration in a chronicheart failure pig model. J Cardiovasc Electrophysiol. 2007;18:985–990.

38. Stahl C, Walker T, Straub A, Kettering K, Knubben K, Greiner TO, PauleS, Lippert M, Czygan G, Schweika O, Kuhlkamp V. Assessing acuteventricular volume changes by intracardiac impedance in a chronic heartfailure animal model. Pacing Clin Electrophysiol. 2009;32:1395–1401.

39. Maines M, Landolina M, Lunati M, Lonardi G, Pappone A, Proclemer A,Zanotto G, Santini M, Varbaro A, Vimercati M, Valsecchi S.Intrathoracic and ventricular impedances are associated with changes inventricular volume in patients receiving defibrillators for CRT. PacingClin Electrophysiol. 2010;33:64–73.

40. Bocchiardo M, Meyer ZU, Vilsendorf D, Militello C, Lippert M, CzyganG, Gaita F, Schauerte P, Stellbrink C. Intracardiac impedance monitorsstroke volume in resynchronization therapy patients. Europace. 2010;12:702–707.

41. Bocchiardo M, Militello C, Azzaro G, Miceli S, Lercari F, Lippert M,Czygan G, Gaita F. Intracardiac impedance optimizes AV- and VV-delayin heart failure patients. Heart Rhythm. 2007;4:S397.

42. Hamilton LH, Beard JD, Kory RC. Impedance measurement of tidalvolume and ventilation. J Appl Physiol. 1965;20:565–568.

43. Cole CR, Jensen DN, Cho Y, Portzline G, Candinas R, Duru F, Adler S,Nelson L, Condie C, Wilkoff BL. Correlation of impedance minuteventilation with measured minute ventilation in a rate responsivepacemaker. Pacing Clin Electrophysiol. 2001;24:989–993.

44. Page E, Cazeau S, Ritter P, Galley D, Casset C. Physiological approachto monitor patients in congestive heart failure: application of a newimplantable device-based system to monitor daily life activity and ven-tilation. Europace. 2007;9:687–693.

45. Whellan DJ, Ousdigian KT, Al-Khatib SM, Pu W, Sarkar S, Porter CB,Pavri BB, O’Connor CM. Combined heart failure device diagnosticsidentify patients at higher risk of subsequent heart failure hospitalizations:results from PARTNERS HF (Program to Access and Review TrendingInformation and Evaluate Correlation to Symptoms in Patients WithHeart Failure) study. J Am Coll Cardiol. 2010;55:1803–1810.

46. Kadhiresan VA, Pastore J, Auricchio A, Sack S, Doelger A, Girouard S,Spinelli JC. A novel method–the activity log index–for monitoringphysical activity of patients with heart failure. Am J Cardiol. 2002;89:1435–1437.

47. Ohlsson A, Kubo SH, Steinhaus D, Connelly DT, Adler S, Bitkover C,Nordlander R, Ryden L, Bennett T. Continuous ambulatory monitoring ofabsolute right ventricular pressure and mixed venous oxygen saturation inpatients with heart failure using an implantable haemodynamic monitor:results of a 1 year multicentre feasibility study. Eur Heart J. 2001;22:942–954.

48. Kjellstrom B, Linde C, Bennett T, Ohlsson A, Ryden L. Six yearsfollow-up of an implanted SvO(2) sensor in the right ventricle. EurJ Heart Fail. 2004;6:627–634.

49. Rickards AF, Bombardini T, Corbucci G, Plicchi G. An implantableintracardiac accelerometer for monitoring myocardial contractility: theMulticenter PEA Study Group. Pacing Clin Electrophysiol. 1996;19:2066–2071.

50. Dupuis JM, Kobeissi A, Vitali L, Gaggini G, Merheb M, Rouleau F,Leftheriotis G, Ritter P, Victor J. Programming optimal atrioventriculardelay in dual chamber pacing using peak endocardial acceleration:comparison with a standard echocardiographic procedure. Pacing ClinElectrophysiol. 2003;26:210 –213.



Dow

nloaded from


51. Delnoy PP, Marcelli E, Oudeluttikhuis H, Nicastia D, Renesto F, Cer-cenelli L, Plicchi G. Validation of a peak endocardial acceleration-basedalgorithm to optimize cardiac resynchronization: early clinical results.Europace. 2008;10:801–808.

52. Vanderheyden M, Houben R, Verstreken S, Stahlberg M, Reiters P,Kessels R, Braunschweig F. Continuous monitoring of intrathoracicimpedance and right ventricular pressures in patients with heart failure.Circ Heart Fail. 2010;3:370–377.

53. O’Donoghue M, Braunwald E. Natriuretic peptides in heart failure:should therapy be guided by BNP levels? Nat Rev Cardiol. 2010;7:13–20.

54. von Haehling S, Schefold JC, Lainscak M, Doehner W, Anker SD.Inflammatory biomarkers in heart failure revisited: much more thaninnocent bystanders. Heart Fail Clin. 2009;5:549–560.

55. Reichlin T, Socrates T, Egli P, Potocki M, Breidthardt T, Arenja N,Meissner J, Noveanu M, Reiter M, Twerenbold R, Schaub N, Buser A,Mueller C. Use of myeloperoxidase for risk stratification in acute heartfailure. Clin Chem. 2010;56:944–951.

56. Kanoupakis EM, Manios EG, Kallergis EM, Mavrakis HE, GoudisCA, Saloustros IG, Milathianaki ME, Chlouverakis GI, Vardas PE.Serum markers of collagen turnover predict future shocks inimplantable cardioverter-defibrillator recipients with dilated cardio-myopathy on optimal treatment. J Am Coll Cardiol. 2010;55:2753–2759.

57. Isserlin R, Merico D, Alikhani-Koupaei R, Gramolini A, Bader GD, EmiliA. Pathway analysis of dilated cardiomyopathy using global proteomicprofiling and enrichment maps. Proteomics. 2010;10:1316–1327.

58. Romero E, Warrington RO, Neuman MR. Energy scavenging sources forbiomedical sensors. Physiol Meas. 2009;30:R35–R62.

59. Nuxoll EE, Siegel RA. BioMEMS devices for drug delivery. IEEE EngMed Biol Mag. 2009;28:31–39.

60. Salazar C, Abraham WT. Biventricular and novel pacing mechanisms inheart failure. Curr Heart Fail Rep. 2009;6:14–18.

KEY WORDS: heart failure, congestive � treatment, pacemaker � diagnostictesting, other � contractile function



Dow

nloaded from


Faisal M. Merchant, G. William Dec and Jagmeet P. SinghImplantable Sensors for Heart Failure

Print ISSN: 1941-3149. Online ISSN: 1941-3084 Copyright © 2010 American Heart Association, Inc. All rights reserved.

Avenue, Dallas, TX 75231is published by the American Heart Association, 7272 GreenvilleCirculation: Arrhythmia and Electrophysiology

doi: 10.1161/CIRCEP.110.9595022010;3:657-667Circ Arrhythm Electrophysiol.

http://circep.ahajournals.org/content/3/6/657World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circep.ahajournals.org//subscriptions/

is online at: Circulation: Arrhythmia and Electrophysiology Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Answer

Permissions and Rights Question andunder Services. Further information about this process is available in thepermission is being requested is located, click Request Permissions in the middle column of the Web pageClearance Center, not the Editorial Office. Once the online version of the published article for which

can be obtained via RightsLink, a service of the CopyrightCirculation: Arrhythmia and Electrophysiologyin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:


Dow

nloaded from

http://circep.ahajournals.org/content/3/6/657

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circep.ahajournals.org//subscriptions/


advances in arrhythmia and electrophysiology implantable sensors

Documents