Gianni Plicchi

AN IMPLANTABLE INTRACARDIAC ACCELEROMETER FOR MONITORING MYOCARDIAL CONTRACTILITY

As the myocardium contracts isometrically, it generates vibrations that are transmitted throughout the heart. These vibrations can be measured with an implantable microaccelerometer located inside the tip of an otherwise conventional unipolar pacing lead. These vibrations are, in their audible component, responsible for the first heart sound. The aim of this study was to evaluate, in man, the clinical feasibility and reliability of intracavity sampling of Peak Endocardial Acceleration (PEA) of the first heart sound vibrations using an implantable tip mounted accelerometer. We used a unidirectional accelerometer located inside the stimulating tip of a standard unipolar pacing lead: the sensor has a frequency response of DC to 1 kHz and a sensitivity of 5 mV/G (G ‐ 9.81 m/s−2). The lead was connected to an external signal amplifier with a frequency range of 0.05–1,000 Hz and to a peak‐to‐peak detector synchronized with the endocardial R wave scanning the isovolumetric contraction phase. Following standard electro‐physiological studies, sensor equipped leads were temporarily inserted in the RV of 15 patients (68 ± 15 years), with normal regional and global ventricular function, to record PEA at rest, during AAI pacing, during VVI pacing, and during dobutamine infusion (up to 20 |mg/kg per min). PEA at baseline was 1.1 G ± 0.5 (heart rate = 75 ± 14 beats/rain) and increased to 1.3 G ± 0.9 (P = NS vs baseline) during AAI pacing (heart rate = 140 beats/min) and to 1.4 G ± 0.5 (P = NS vs baseline) during VVI pacing (heart rate = 140 beats/min). Dobutamine infusion increased PEA to 3.7 G ± 1.1 (P < 0.001 vs baseline), with a heart rate of 121 ± 13 beats/min. In a subset of three patients, simultaneous hemodynamic RV monitoring was performed to obtain RV dP/dtmax, whose changes during dobutamine and pacing were linearly related to changes in PEA (r = 0.9; P < 0.001). In conclusion, the PEA recording can be consistently and safely obtained with an implantable device. Pharmacological inotropic stimulation, but not pacing induced chronotropic stimulation, increases PEA amplitude, in keeping with experimental studies, suggesting that PEA is an index ofmyocardial contractility. Acute variations in PEA are closely paralleled by changes in R V dP/dtmax, but are mainly determined by LV events. The clinical applicability of the method using RV endocardial leads and an implantable device offers potential for diagnostic applications in the long‐term monitoring of myocardial function in man.

Respiratory rate as a determinant of optimal pacing rate.

Des essais ont été faits dans ľutilisation des paramètres bialogiques pour déterminer la fréquence optimale de stimulation cardiaque. Dans cette étude, le rapport entre fréquence respiratoire et fréquence cordiaque a étéétabli chez 67 patients au cours de ľexercice. Ensuite, un système de stimulation cardiaque qui repondrait àľactivation radiotélémétrique a été posé chez onze patients. Dans deux cos un système automatique a été implanté avec succès. A present, les résultats de cette stimulation pilotée par la fréquence respiratoire sont satisfaisants.

Respiration-dependent ventricular pacing compared with fixed ventricular and atrial-ventricular synchronous pacing: Aerobic and hemodynamic variables

A pacemaker that adapts heart rate in response to the patients metabolic requirements has been developed. The pacemaker uses breathing frequency and tidal volume as the indicators of physiologic demand. Maximal physical work capacity, anaerobic threshold, oxygen uptake (16 patients) and hemodynamic variables (9 patients) were assessed with fixed rate (VVI), atrial synchronous (VDT/I) and respiration-dependent ventricular (VVI-RD) pacing. All subjects attained their anaerobic threshold in stress tests with VVI pacing. The maximal physical capacity (p less than 0.001), work time to attain the anaerobic threshold (p less than 0.01) and oxygen uptake (p less than 0.001) were significantly greater with VVI-RD than with VVI pacing. The transition from the supine to the standing position was characterized by a significant increase of cardiac index at rest with both VDT/I and VVI-RD pacing as compared with VVI pacing. Progressive increments in the cardiac index and average left ventricular stroke work index were significantly different at submaximal and maximal exercise when VVI and VVI-RD were compared. At maximal exercise, mean cardiac output was also significantly different: 10.21 +/- 2.5 (SD) liters/min with VVI, 11.2 +/- 0.8 liters/min with VDT/I (p less than 0.05) and 12.65 +/- 3.1 liters/min with VVI-RD (p less than 0.05) pacing. Maximal oxygen extraction values were greater with VVI and VVI-RD pacing than with VDT/I pacing. Pulmonary artery end-diastolic pressures at maximal exercise were within the normal range with the three different modes of pacing. In conclusion, there is a significant (25%) improvement in exercise performance with VVI-RD pacing as compared with VVI pacing.(ABSTRACT TRUNCATED AT 250 WORDS)

Validation of a peak endocardial acceleration-based algorithm to optimize cardiac resynchronization: early clinical results

Aims Cardiac resynchronization therapy (CRT) involves time-consuming procedures to achieve an optimal programming of the system, at implant as well as during follow-up, when remodelling occurs. A device equipped with an implantable sensor able to measure peak endocardial acceleration (PEA) has been recently developed to monitor cardiac function and to guide CRT programming. During scanning of the atrioventricular delay (AVD), PEA reflects both left ventricle (LV) contractility (LV dP/dtmax) and transmitral flow. A new CRT optimization algorithm, based on recording of PEA (PEAarea method) was developed, and compared with measurements of LV dP/dtmax, to identify an optimal CRT configuration. Methods and results We studied 15 patients in New York Heart Association classes II–IV and with a QRS duration >130 ms, who had undergone implantation of a biventricular (BiV) pulse generator connected to a right ventricular (RV) PEA sensor. At a mean of 39 ± 15 days after implantation of the CRT system, the patients underwent cardiac catheterization. During single-chamber LV or during BiV stimulation, with initial RV or LV stimulation, and at settings of interventricular intervals between 0 and 40 ms, the AVD was scanned between 60 and 220 ms, while LV dP/dtmax and PEA were measured. The area of PEA curve (PEAarea method) was estimated as the average of PEA values measured during AVD scanning. A ≥10% increase in LV dP/dtmax was observed in 12 of 15 patients (80%), who were classified as responders to CRT. In nine of 12 responders (75%), the optimal pacing configuration identified by the PEAarea method was associated with the greatest LV dP/dtmax. Conclusion The concordance of the PEAarea method with measurements of LV dP/dtmax suggests that this new, operator-independent algorithm is a reliable means of CRT optimization.

First experimental evaluation of cardiac apex rotation with an epicardial coriolis force sensor.

Cardiac apex rotation, quantified by sophisticated techniques (radiopaque markers and tagged magnetic resonance), has been shown to provide a sensitive index of left ventricle (LV) dynamics. The authors describe the first experimental assessment of cardiac apex rotation using a gyroscopic sensor based on Coriolis force, epicardially glued on the apex. Dynamics of apex rotation were evaluated in a sheep at baseline, after a positive inotropic drug infusion, and after impairment of cardiac function induced by coronary ligation. To evaluate the efficacy of the sensor to monitor cardiac function, results were compared to contractility variations expressed by the maximum value of the first derivative of LV pressure (LVdP/dtMAX). After inotropic drug infusion, a parallel increasing trend resulted for LVdP/dtMAX, for the maximum value of angular velocity measured by the sensor, and for apex rotation angle derived from velocity signal (+146%, +155%, and +11% from baseline, respectively), whereas a decreasing trend of all three parameters resulted after coronary ligation (–35%, –31%, and –65%). The twist pattern also was altered from baseline. These initial results suggest that the use of an implantable rotation sensor based on Coriolis force can be an efficient and effective tool to assess LV torsional deformation both in normal and failing hearts.

DETECTING INCIPIENT VASOVAGAL SYNCOPE : INTRAVENTRICULAR ACCELERATION

The peak endocardial acceleration (PEA) caused by ventricular isometric contraction can be measured with an implantable microaccelerometer located inside the tip of a normal unipolar pacing lead. It has been shown that PEA correlates with myocardiai contractility and the maximum rate of rise of ventricular pressure (peak dP/dt) of the left ventricle. A PEA measuring system was temporarily inserted into the apex of the right ventricle in seven patients affected by syncope of uncertain origin. Each patient subsequently underwent 60 tilt testing with three different protocols: without pharmacological challenge (baseline); potentiated with sublingual trinitroglycerin (at a dose of 0.3 mg); and with isoproterenol infusion (at a dose of 3 μg/min). Each phase lasted 20 minutes. Syncope was induced in 1 patient during the baseline phase, in 3 patients during the trinitrin phase, and in 4 patients during the isoproterenol phase. Six patients had a negative response during the baseline phase and served as a control group. From the beginning of upright posture to the time of maximum heart rate, PEA increased by about the same amount in both positive and negative patients, but absolute values were from two‐ to three fold higher with isoproterenol (from 1.2 ± 0.5 G to 1.6 ± 0.8 G, from 0.8 ± 0.2 G to 1.2 ± 0.4 G, and from 2.8 ± 1.8 G to 3.6 ± 1.8 G, respectively, for negative, positive baseline or trinitrin, and positive isoproterenol tests). At the time of syncope, PEA values fell to baseline values. PEA changes were inversely correlated with blood pressure changes and directly correlated with heart rate changes. Thus, tilt induced syncope occurred both at low and high levels of left ventricular contractility. Whether spontaneous syncopes occur at low or high PEA behavior remains to be established. Since heart rate correlates well with changes in PEA and is far easier to measure, it is unlikely that a PEA measurement system or, in general, a contractility‐based system, might become an ideal sensing parameter for the introduction of devices to combat vasovagal syncope.

Peak Endocardial Acceleration Reflects Heart Contractility Also in Atrial Fibrillation

Previous studies demonstrated that peak endocardial acceleration (PEA) in sinus rhythm is related to LV dP/dtmax. Until now, PEA was never evaluated during R‐R interval variations in AF. The aim of this study was to establish the behavior of PEA in AF and the relationship of PEA versus LV dP/dtmax. Six sheep (65 ± 6 kg) were instrumented with a LV Millar catheter and with an accelerometer lead. AF was induced and PEA, LV dP/dtmax, and ECG were monitored. AF persisted for 5 ± 1.3 minutes. From sinus rhythm to AF, the heart rate went from 92 ± 3 to 130 ± 35 beats/mm (P < 0.05), LV dP/dtmax from 684 ± 18 to 956 ± 344 mmHg/s (P = NS) and PEA from 0.82 ± 0.06 to 0.94 ± 0.33 g (P = NS). The correlation between PEA and LV dP/dtmax was significative in sinus rhythm (r = 0.7, P < 0.05) and in AF (r = 0.8, P < 0.05). A positive relationship was found between the preceding interval and PEA (r = 0.4 ± 0,07, P < 0.05) and LV dP/dtmax (r = 0.61 ± 0.08, P < 0.05), while a negative one was found between the prepreceding interval and both PEA (r =− 0.39 ± O.11.P < 0.05) and LV dP/dtmax (r =− 0.64 ± 0.05, P < 0.05). At the onset of AF, LV dP/dtmax and PEA showed similar changes: beat‐to‐beat correlation between PEA and LV dP/dtmax was high. As for LV dP/dtmax, PEA is positively related to the preceding interval and negatively related to the prepreceding interval. These data confirm that PEA reflects heart contractility also during AF and hold promise for the use of this sensor in therapeutic implantable devices.

Initial experience with a telerobotic system to remotely navigate and automatically reposition standard steerable EP catheters.

The use of robotic systems in cardiac interventional procedures is growing. The insertion and maneuvering in the human body of electrophysiology (EP) catheters is currently carried out manually under fluoroscopic guidance, resulting in operator fatigue and prolonged x-ray exposure. We report our initial animal experience with a novel telerobotic system (TS) to remotely navigate and automatically reposition standard steerable EP catheters within the heart. We developed a TS able to guide, as a “robotic hand,” EP catheters without the need of dedicated catheters and cumbersome devices. During tests on three sheep, catheter navigation and repositioning to 12 predefined endocardial targets were previously performed by conventional manual procedure and then using the TS implemented with an automatic catheter repositioning function. All the predefined targets were reached under fluoroscopy visualization, and procedural times were measured during catheter navigation and repositioning. The use of the TS slightly reduced the time necessary for catheter navigation compared with the conventional manual procedure (13.0 ± 5.6 vs 16.1 ± 6.4 seconds, p < 0.001) and significantly decreased the time for a precise catheter repositioning (9.2 ± 2.5 vs 17.8 ± 7.1 sec, p < 0.001). The TS proved to be a promising tool for remote navigation of standard EP catheters reducing the time necessary for catheter repositioning.

Effect of Right Ventricular Pacing on Cardiac Apex Rotation Assessed by a Gyroscopic Sensor

To quantify cardiac apex rotation (CAR), the authors recently proposed the use of a Coriolis force sensor (gyroscope) as an alternative to other complex techniques. The aim of this study was to evaluate the effects of right ventricular (RV) pacing on CAR. A sheep heart was initially paced from the right atrium to induce a normal activation sequence at a fixed heart rate (AAI mode) and then an atrioventricular pacing was performed (DOO mode, AV delay = 60 ms). A small gyroscope was epicardially glued on the cardiac apex to measure the angular velocity (Ang V). From AAI to DOO pacing mode, an increase (+9.2%, p < 0.05) of the maximum systolic twisting velocity (Ang VMAX) and a marked decrease (–19.9%, p < 0.05) of the maximum diastolic untwisting velocity (Ang VMIN) resulted. RV pacing had negligible effects (–3.1%, p = 0.09) on the maximum angle of CAR, obtained by integrating Ang V. The hemodynamic parameters of systolic (LVdP/dtMAX) and diastolic (LVdP/dtMIN) cardiac function showed slight variations (–3.8%, p < 0.05 and +3.9%, p < 0.05, respectively). Results suggest that cardiac dyssynchrony induced by RV pacing can alter the normal physiological ventricular twist patterns, particularly affecting diastolic untwisting velocity.

Computational finite element model of cardiac torsion.

Purpose A novel finite-element model of ventricular torsion for the analysis of the twisting behavior of the left human ventricle was developed, in order to investigate the influence of various biomechanical parameters on cardiac kinematics. Methods The ventricle was simulated as a thick-walled ellipsoid composed of nine concentric layers. Arrays of reinforcement bars were embedded in each layer to mimic physiological myocardial anisotropy. The reinforcement bars were activated through an artificial combination of thermal and mechanical effects in order to obtain a contractile behavior which is similar to that of myocardial fibers. The presence of an incompressible fluid inside the ventricular cavity was also simulated and the ventricle was combined with simple lumped-parameter hydraulic circuits reproducing preload and afterload. Changes to a number of cardiac parameters, such as preload, afterload and fiber angle orientation were introduced, in order to study the effects of these changes on cardiac torsion. Results The model is able to reproduce a similar torsional behavior to that of a physiological heart. The results of the simulations showed that there was sound correspondence between the model outcomes and available data from the literature. Results confirmed the importance of symmetric transmural patterns for fiber orientation. Conclusions This model represents an important step on the path towards unveiling the complexity of cardiac torsion. It proves to be a practical and versatile tool which could assist clinicians and researchers by providing them with easily-accessible, detailed data on cardiac kinematics for future diagnostic and surgical purposes.