Search

Background: Late gadolinium enhanced cardiac magnetic resonance (LGE-CMR) imaging may help identify radiofrequency ablation lesions, which have been poorly described in patients with nonischemic cardiomyopathy (NICM).Objectives: The purpose of this study was to describe LGE-CMR characteristics of ablation lesions in patients with NICM and correlate them with histopathology.Methods: Twenty-six patients (24 men; ejection fraction 38% ± 14%; age 61 ± 9 years) who had undergone CMR imaging after ventricular tachycardia (VT) ablation were included. Areas of both dark and bright core lesions correlating with previous radiofrequency ablation lesions were identified. Histology was performed on an explanted heart.Results: Mean time between the ablation procedure and the LGE-CMR study was 8 [2-20] months. Twenty-three of 26 patients demonstrated dark core lesions (volume 2.16 ± 1.8 cm3; thickness 3.6 ± 1.3 mm) with transmurality of 42% ± 16% overlaying areas of intramural or transmural LGE. Fourteen of 26 patients demonstrated bright core lesions (volume 0.8 ± 0.6 cm3; depth 4.15 ± 1.76 mm) with transmurality of 34% ± 14%, which was located in areas without underlying LGE in 11 of 13 patients. Both dark and bright core lesions were visualized on standard clinical LGE-CMR imaging obtained in the acute setting and chronic settings (within 3 days and up to 2090 days postablation). Histopathologic analysis demonstrated coagulation necrosis in the area that corresponded to dark core lesions in the postablation CMR.Conclusion: Ablation lesions can be detected by LGE-CMR after VT ablation in NICM patients and have a different appearance than scar tissue. These lesions can be observed in the acute and chronic settings after ablations. [ABSTRACT FROM AUTHOR]

Within the span of a few years, watches have functionally morphed from objects that tell time to wearable minicomputers that allow real-time recording of electrocardiograms (ECGs). Considerable information can be deduced from these single lead tracings, and it is now not uncommon to see patients in whom diagnostic tracings of clinically relevant but elusive arrhythmias are captured using a smartwatch. Empowering individuals to record their own ECG tracings in scenarios such as palpitations, syncope, and for risk stratification of sudden death intuitively has considerable potential, but its value remains to be robustly demonstrated. The main objective of this review is to describe the information that can be obtained from smartwatch-based single-lead ECG recordings beyond simply differentiating between sinus rhythm and atrial fibrillation. We also review the strengths and limitations of using these devices in clinical settings and offer potential solutions to address the latter. [ABSTRACT FROM AUTHOR]

Background: Centrifugal activation is not always the origin of a focal atrial tachycardia (AT) ("true-focal"), but passive activation from the other structures ("pseudo-focal").Objective: We aimed to establish a method to differentiate true-focal from pseudo-focal.Methods: In 49 centrifugal activations in 35 patients with AT, 12-lead electrocardiogram, activation map, atrial global activation histogram (GAH), and local electrograms were analyzed. GAH demonstrates the relation between the activation area and timing through the cycle length, displayed with a normalized value, ranging from 0 (smallest activation area) to 1.0 (largest activation area).Results: Of 30 centrifugal activations observed in the septal region, 6/30 (20.0%) were true-focal. The remaining 24/60 (80.0%) were pseudo-focal, of which 23 (95.8%) were from the opposite chamber. P-wave/flutter-wave duration < 200 ms discriminated true-focal from pseudo-focal (sensitivity 100%; specificity 54.5%; positive predictive value 33.3%; negative predictive value 100%). Multiple breakthrough ruled out the possibility of a true-focal AT. Other differentiating factors were an activation area within the initial 20 ms of <5 mm2 and a typical QS pattern electrogram at the origin. Of 19 centrifugal activations observed outside the septal regions, 7 were true-focal and 12 were pseudo-focal exited from an epicardial structure: 10 of 12 (83.3%) were located around the left atrial appendage and ridge. Flutter wave, GAH score ≤ 0.05, and GAH score < 0.1 for >110 ms of cycle length differentiated true-focal from pseudo-focal with a sensitivity/negative predictive value of 100%. GAH score < 0.1 for >40% of the cycle length simply discriminated true-focal from pseudo-focal with 100% accuracy.Conclusion: Centrifugal activation is not necessarily due to a focal AT but passive activation. The activation map with GAH in addition to the 12-lead electrocardiogram and local electrograms enables an accurate differentiation. [ABSTRACT FROM AUTHOR]

Background: The risk of heart block during radiofrequency ablation of atrioventricular (AV) nodal reentrant tachycardia and septal accessory pathways is minimized by rapidly ceasing ablation in response to markers of risk, such as atrioventricular dissociation, fast junctional rhythm, PR interval prolongation, or 2 consecutive atrial or ventricular depolarizations. Currently this is done manually.Objectives: The objectives of this study were to build and test a control system able to monitor cardiac rhythm and automatically terminate ablation energy when required.Methods: The device was built from off-shelf componentry. Preclinical testing involved real-time input of electrogram/electrocardiogram data from 209 ablation procedures (20 patients) over slow (n = 19) and fast (n = 1) AV nodal pathways. The device response speed was compared with the human response speed. The device's ability to prevent heart block was tested in 5 sheep. First-in-human testing was then performed in 12 patients undergoing AV nodal reentrant tachycardia ablation.Results: Risk conditions necessitating shutoff of ablation (200 total; 111 preclinical and 89 first-in-human) were detected by the device with 100% sensitivity and 94% specificity, automatically terminating ablation while still allowing successful ablation in all patients. Device shutoff of ablation was always faster than human response (median difference 1.24 seconds). In each of 5 sheep, 40 consecutive attempts to cause heart block by ablating over the His bundle were unsuccessful because of automatic shutoff in response to rhythm change.Conclusion: Automated shutoff of ablation close to the AV node in response to markers of the risk of heart block is feasible with high accuracy as well as faster response than human response. The system may improve the safety of ablation near the AV node by preventing heart block. [ABSTRACT FROM AUTHOR]

Background: Frequent premature ventricular complexes (PVCs) can result in PVC-induced cardiomyopathy (PICM). Scarring has been described in patients with frequent PVCs in the absence of apparent heart disease and in patients with known cardiomyopathy.Objective: The purpose of this study was to determine the impact of focal myocardial scarring as detected by cardiac magnetic resonance imaging (CMR) on PICM, procedural outcomes, and recovery of left ventricular function in patients with frequent PVCs.Methods: A total of 351 consecutive patients (181 men; age 53 ± 15 years; ejection fraction [EF] 51% ± 12%) with frequent PVCs referred for ablation were included. CMR was performed in all patients before the ablation procedure. A ≥10% increase in EF or normalization of a previously abnormal EF was defined as evidence of PICM.Results: Myocardial scarring was present in 134 of 351 patients (38%); 66 of 134 patients (49%) with scarring and 54 of 217 patients (25%) without scarring had improvement or normalization of EF after ablation. The presence of myocardial scarring, PVC burden >22%, male sex, asymptomatic status, and PVC QRS width >150 ms were associated with PICM by univariate analysis (P <.01 for all). The presence of scar was independently associated with PICM (odds ratio 2.2; 95% confidence interval 1.3-3.7; P <.005). The success rate of PVC ablation was lower in patients with scarring than in patients without focal scarring (mean 70% vs 82%; P <.01).Conclusion: Focal scar defined by CMR is independently associated with PICM. Although ablation outcomes are worse in the presence of scarring, EF recovery can occur in most of these patients after ablation. [ABSTRACT FROM AUTHOR]

Background: Patients with nonischemic cardiomyopathy (NICM) often require epicardial ventricular tachycardia (VT) ablation procedures via subxiphoid access. The coronary venous system (CVS) provides limited access to the epicardial space.Objective: The purpose of this study was to determine the value of an approach targeting the CVS in these patients.Methods: In a series of 41 consecutive patients (mean age 59.7 ± 11.5 years; 36 men [88%]; ejection fraction 34.5% ± 13.1%; 269 inducible VTs [6.6 ± 5.0 VTs per patient]) with NICM and VT, mapping and ablation were performed sequentially at the endocardium, then within the CVS, and finally within the pericardial space if required.Results: VT target sites were identified within the CVS in 15 patients and by subxiphoid access to the pericardial space in 8 patients. Ablation within the CVS eliminated VT inducibility in 9 patients without the need for epicardial ablation. Cardiac magnetic resonance imaging demonstrated that the CVS was closer to a scarred area in patients with CVS-related VT target sites than in other patients (mean 3.5 ± 3.9 mm vs 14.3 ± 8.3 mm; P < .001). A cutoff distance of ≤9 mm from the scar (area under the curve 0.91; 95% confidence interval 0.82-0.99; sensitivity 0.78; specificity 0.93) identified patients with vs patients without VT target sites within the CVS.Conclusion: A stepwise approach with mapping/ablation in the endocardium followed by ablation within the CVS can reduce the need for subxiphoid epicardial access in some patients with NICM. Proximity of the scar to the CVS detected by cardiac magnetic resonance imaging can identify the patients most likely to benefit from this approach. [ABSTRACT FROM AUTHOR]

Background: Frequent premature ventricular complexes (PVCs) can be an indicator of structural heart disease.Objective: The purpose of this study was to determine the prevalence of scarring detected by delayed enhancement cardiac magnetic resonance (DE-CMR) imaging in patients with frequent PVCs without apparent structural heart disease and to determine the value of programmed ventricular stimulation (PVS) for risk stratification in patients with frequent PVCs and myocardial scarring.Methods: DE-CMR imaging was performed in patients without apparent heart disease who had frequent PVCs and were referred for ablation. In the presence of scarring, scar volume was measured and correlated with outcome variables. All patients underwent PVS and were monitored for the occurrence of ventricular arrhythmias. Logistic regression was used to compare imaging and procedural findings with long-term outcomes, with adjustment for postablation ejection fraction (EF).Results: The study consisted of 272 patients (135 men; mean age 52 ± 15 years; EF 52% ± 12%). DE-CMR scar was found in 67 patients (25%), and 7 (3%) were found to have inducible ventricular tachycardia (VT). The presence and amount of DE-CMR were related to the risk of long-term VT independent of EF (hazard ratio 18.8 [95% confidence interval] [2.0-176.6], P = .01; and hazard ratio 1.4 [1.1-1.7] per cm3 scar, P <.001, respectively). The positive predictive value and negative predictive value of PVS for VT during long-term follow-up were 71% and 100%, respectively.Conclusion: Preprocedural cardiac DE-CMR and PVS can be used to identify patients with frequent PVCs without apparent heart disease who are at risk for VT. [ABSTRACT FROM AUTHOR]

Background: Many centers use radiofrequency (RF) energy at 25-35 W for 30-60 seconds. There is a safety concern about using higher power, especially on the posterior wall.Objective: The purpose of this study was to examine complication rates for atrial fibrillation (AF) ablations performed with high-power, short-duration RF energy.Methods: We examined the complication rates of 4 experienced centers performing AF ablations at RF powers from 45-50 W for 2-15 seconds per lesion. In total, 13,974 ablations were performed in 10,284 patients. On the posterior wall, 11,436 ablations used 45-50 W for 2-10 seconds, and 2538 ablations used power reduced to 35 W for 20 seconds. Esophageal temperature monitoring was used in 13,858 (99.2%).Results: Demographics were age 64 ± 11 years, male 68%, left atrial size 4.4 ± 0.7 cm, paroxysmal AF 37%, persistent AF 42%, longstanding AF 20%, antiarrhythmic drugs failed 1.4 ± 0.7, hypertension 54%, diabetes 15%, previous cerebrovascular accident/transient ischemic attack 7%, and CHA2DS2-VASc score 2.1 ± 1.4. Procedural time was 116 ± 41 minutes. Complications were death in 2 (0.014%; 1 due to stroke and 1 due to atrioesophageal fistula), pericardial tamponade in 33 (0.24%; 26 tapped, 7 surgical), strokes <48 hours in 6 (0.043%), strokes 48 hours-30 days in 6 (0.043%), pulmonary vein stenosis requiring intervention in 2 (0.014%), phrenic nerve paralysis in 2 (0.014%; both resolved), steam pops 2 (0.014%) without complications, and catheter char 0 (0.00%). There was 1 atrioesophageal fistula in 11,436 ablations using power 45-50 W on the posterior wall and 3 in 2538 ablated with 35 W on the posterior wall (P = .021), although 2 of the 3 had no esophageal monitoring during a fluoroless procedure.Conclusion: AF ablations can be performed at 45-50 W for short durations with very low complication rates. High-power, short-duration ablations have the potential to shorten procedural and total RF times and create more localized and durable lesions. [ABSTRACT FROM AUTHOR]

Background: Voltage criteria for ventricular mapping have been obtained from small series of patients and prioritizing high specificity.Objective: The purpose of this study was to analyse the potential influence of contact force (CF) on voltage mapping and to define voltage cutoff values for right ventricular (RV) scar using the tetralogy of Fallot as a model of transmural RV scar and magnetic resonance imaging (MRI) as reference.Methods: Fourteen patients (age 32.6 ± 14.3 years; 5 female) with repaired tetralogy of Fallot underwent high-resolution cardiac MRI (1.25 × 1.25 × 2.5 mm). Scar, defined as pixels with intensity >50% maximum, was mapped over the RV geometry and merged within the CARTO system to RV endocardial voltage maps acquired using a 3.5-mm ablation catheter with CF technology (SmartTouch, Biosense Webster).Results: In total, 2446 points were analyzed, 915 within scars and 1531 in healthy tissue according to MRI. CF correlated to unipolar (ρ = 0.186; P <.001) and bipolar voltage in healthy tissue (ρ = 0.245; P <.001) and in scar tissue. Receiver operating characteristic curve analysis excluding points with very low CF (<5g) identified optimal voltage cutoffs of 5.19 mV for unipolar voltage and 1.76 mV for bipolar voltage, yielding sensitivity/specificity of 0.89/0.85 and 0.9/0.9, respectively.Conclusion: CF is an important factor to be taken into account for voltage mapping. If good CF is applied, unipolar and bipolar voltage cutoffs of 5.19 mV and 1.76 mV are optimal for identifying RV scar on endocardial mapping with the SmartTouch catheter. Data on the diagnostic accuracy of different voltage cutoff values are provided. [ABSTRACT FROM AUTHOR]

Background: Despite advancements, the goal of durable pulmonary vein isolation (PVI) in all patients undergoing ablation for atrial fibrillation (AF) remains elusive. New high-density mapping (HDM) allows detection of concealed low-voltage signals (CLVSs) that persist after PVI and may represent vulnerabilities in the lesion set.Objective: The purpose of this study was to determine the incidence of CLVSs after PVI and the effect of CLVS ablation on outcomes.Methods: We conducted a case control study comparing 150 patients undergoing HDM-guided PVI and subsequent CLVS mapping and ablation (39 redo, 111 de novo) against 452 historical controls undergoing traditional PVI alone. PVI was similarly performed and confirmed in both groups.Results: Baseline characteristics were similar, except left atrial size was larger in the HDM-guided group. Acute PVI was achieved in nearly all patients in both groups. In the HDM group, 31 of 150 patients exhibited CLVS after luminal PVI, and all were subsequently eliminated. During mean follow-up of 320 days, after controlling for baseline characteristics, the HDM-guided group exhibited a hazard ratio of 0.19 in freedom from AF (P <.001). De novo patients exhibited a hazard ratio of 0.44 relative to redo patients in the HDM-guided group. Both subgroups exhibited significantly lower event rates compared to controls in log-rank analysis (P <.001).Conclusion: CLVSs are commonly identified with HDM after PVI, likely representing vulnerabilities in antral lesion sets. Ablation of these targets seems to significantly improve freedom from AF compared to PVI alone. [ABSTRACT FROM AUTHOR]

Background: Epicardial ablation is often necessary for the treatment of complex arrhythmias refractory to endocardial ablation. Conventional needle access to the pericardial space is considered quite challenging, and it is often associated with several potential complications, particularly inadvertent right ventricular puncture. The novel EpiAccess needle tip is embedded with a pressure sensor able to report the pressure waveform in real time when used with the EpiAccess System.Objective: We prospectively evaluated the feasibility and safety of the EpiAccess System by EpiEP, Inc., with a novel epicardial access needle in a multicenter study.Methods: Twenty-five patients with a clinical need for epicardial access were enrolled. The EpiAccess needle and EpiAccess System were used for epicardial access in each case. Successful epicardial access, defined as the ability to introduce a guidewire into the epicardial space, was assessed via the device and confirmed with fluoroscopy. Significant pericardial bleeding was defined as >80 mL of blood by using peer review article definitions.Results: Patients were men (76%) with a mean age of 62 years (range 28-84 years). Epicardial access for ventricular tachycardia ablation was indicated in 80% of the patients. Successful epicardial access was obtained in all cases, with pressure monitoring guiding pericardial wire access in all cases. One delayed pericardial effusion occurred.Conclusion: Epicardial access with the novel EpiAccess needle and System with real-time pressure monitoring is feasible and safe. The pressure monitoring capability identifies successfully the epicardial space, facilitating access and potentially minimizing complications. This has relevant clinical implications. [ABSTRACT FROM AUTHOR]

The early repolarization (ER) pattern has historically been regarded as a benign ECG variant. However, in recent years this view has been challenged based on multiple reports linking the ER pattern with an increased risk of sudden cardiac death. The mechanistic basis of ventricular arrhythmogenesis in ER syndrome is presently incompletely understood. Furthermore, strategies for risk stratification and therapy for ER syndrome remain suboptimal. The recent emergence of novel mapping techniques for cardiac arrhythmia has ushered a new era of research into the mechanistic basis of ER syndrome. This review provides an overview of current evidence relating to ER and risk of ventricular arrhythmias and discusses potential future areas of research to elucidate the mechanisms of ventricular arrhythmogenesis. [ABSTRACT FROM AUTHOR]

Background: During radiofrequency ablation, effective contact is crucial in determining lesions efficacy. Objective: The purpose of this study was to compare operators’ ability to assess contact pressure using visual and tactile feedbacks together or alone in an experimental model. Methods: In a in vitro experimental setup replicating manual catheter manipulation and recording the applied force, evaluators were asked to identify three levels of force (first, ablation, and maximum contact) as the catheter contacted the tissue model using (1) visual feedback only by fluoroscopy, “blinded” to touch; (2) tactile feedback only, blinded to fluoroscopy; and (3) both tactile and visual feedback together. The latter was regarded as reference. The experiment was repeated using a catheter force sensing technology during robotic navigation. Results: During manual navigation, tighter association was shown for the visual method than for the tactile method: median difference with reference: first contact –1 (P = .97) vs –2 (P = .90); ablation contact 2 (P = .1) vs –7 (P = .03); maximum contact 2 (P = .06) vs –28 (P = .02). Bland-Altman plot and Deming regression confirmed for the visual method the good agreement with reference and the absence of bias at any level and showed for the tactile higher values and proportional bias that reached statistical significance at ablation and maximum contact. During robotic navigation, agreement was higher for the tactile than for the visual only method. Conclusion: During manual navigation, visual feedback alone is in better agreement with the reference compared to the tactile only approach. During robotic navigation, agreement is looser for the visual only approach. More objective feedback of contact pressure during ablation procedures is desirable. [Copyright &y& Elsevier]

Background: A pacemaker recipient may be hospitalized recurrently with an infection of unknown origin despite detailed investigations. Objective: The purpose of this study was to investigate whether 18F-fluorodeoxyglucose positron emission tomography/computerized tomography (FDG-PET/CT) scanning has a role in identifying pacing material infection in these difficult cases. Methods: Ten patients who presented with fever of unknown origin despite detailed investigations including transesophageal echocardiography underwent FDG-PET/CT scanning. Identification of increased FDG uptake along a pacing lead prompted the removal of the entire pacing system, whereas in the absence of increased FDG uptake the pacing material was left in place. Forty control pacemaker recipients underwent FDG-PET/CT scanning as part of investigation of malignancy. Results: Among the 40 patients in the control group, FDG-PET/CT scanning was normal in 37 (92.5%) patients. Among the 10 patients who presented with suspected pacing system infections, FDG-PET/CT scanning showed increased FDG uptake along a lead in six patients; as a result of this finding, these patients subsequently underwent complete removal of the implanted material. Cultures of the leads were positive in all six patients, confirming involvement of the leads in the infectious process. In the other four patients, the pacing system was left in place without objective signs of active lead endocarditis during follow-up. Conclusion: This study demonstrates the potential value of FDG-PET/CT scanning in the diagnosis of pacing lead endocarditis in difficult cases. Increased FDG uptake along a lead in this clinical context appears to be a reliable sign of active infection. [ABSTRACT FROM AUTHOR]

Abstract: Background: Rotational angiography with three-dimensional reconstruction (3DRA) is a new imaging tool recently introduced to guide mapping and ablation of the left atrium. Objective: The purpose of this study was to determine the utility of 3DRA for imaging the ventricles and guiding ventricular tachycardia (VT) ablation. Methods: Using the Philips Allura Xper FD10 system, 3DRA was performed in eight patients referred for right ventricular outflow tract (RVOT) VT ablation. The imaging protocol for right ventricular (RV) injection is described. IV contrast was injected at the RA/IVC junction over 4 sec and 3DRA was obtained immediately. Images were segmented manually on the EP Navigator workstation and registered on live fluoroscopy. Intracardiac electrograms were superimposed on 3DRA creating a true electroanatomic map (ElectroNav). CARTO mapping and echocardiograms were performed on all patients, cardiac computed tomography (CT) in 4, and magnetic resonance imaging (MRI) in 1. Results: Three-dimensional rotational angiography was successful in 7 of 8 patients. Image interpretation was unsuccessful in one patient due to poor isocentering. RV imaging was performed with 82 ± 18 mL of contrast. RV image segmentation required 19 ± 5 minutes. CARTO maps of the RVOT required 43 ± 12 minutes and additional fluoroscopy. Three-dimensional rotational angiography was used to guide VT ablation by providing realistic anatomic images of the pulmonary valve plane, endo-views of the ventricle, and ablation point tagging. Anatomic detail provided by 3DRA was qualitatively superior to CARTO. VT ablation was acutely successful in all patients. Close concordance between echocardiographic, CT/MRI, and 3DRA measurements of the RVOT was observed (r = 0.9, P <.01). Conclusion: Three-dimensional rotational angiography of the RV and RVOT is a feasible imaging technique that utilizes a protocol of timed angiography, manual segmentation, image registration, and superimposition of intracardiac electrograms to create an angiogram-based electroanatomic model of these structures. [Copyright &y& Elsevier]

Background: Optimal treatment of right ventricular (RV) dysfunction observed in patients after tetralogy of Fallot (TOF) repair is unclear. Studies of biventricular (BiV) stimulation in patients with congenital heart disease have been retrospective or have included patients with heterogeneous disorders. Objective: The purpose of this study was to determine the effects on cardiac function of stimulating at various cardiac sites in an animal model of RV dysfunction and dyssynchrony and in eight symptomatic adults with repaired TOF. Methods: Pulmonary stenosis and regurgitation as well as RV scars were induced in 15 piglets to mimic repaired TOF. The hemodynamic effects of various configurations of RV and BiV stimulation were compared with sinus rhythm (SR) 4 months after surgery. In eight adults with repaired TOF, RV and left ventricular (LV) dP/dtmax were measured invasively during SR, apical RV stimulation, and BiV stimulation. Results: At 4 months, RV dilation, dysfunction, and dyssynchrony were present in all piglets. RV stimulation caused a decrease in LV function but no change in RV function. In contrast, BiV stimulation significantly improved LV and RV function (P < .05). Echocardiography and epicardial electrical mapping showed activation consistent with right bundle branch block during SR and marked resynchronization during BiV stimulation. In patients with repaired TOF, BiV stimulation increased significantly RV and LV dP/dtmax (P < .05). Conclusion: In this swine model of RV dysfunction and in adults with repaired TOF, BiV stimulation significantly improved RV and LV function by alleviating electromechanical dyssynchrony. [Copyright &y& Elsevier]

Background: It is unclear whether, in recipients of cardiac resynchronization therapy (CRT) systems, the optimal AV delay should be the same, shorter, or longer during exercise than at rest. Objective: This study sought to examine the effects of atrioventricular (AV) delay optimization at rest and during exercise in 50 recipients of CRT systems. Methods: We measured left ventricular (LV) outflow tract velocity time integral (OT-VTI) and LV filling time (FT) echocardiographically, at rest and during exercise to 60% of the maximal predicted heart rate, with the sensed AV delay set at 40, 70, 100, 120, 150, and 200 ms. The measurements made at rest versus those made during exercise, and among the several programmed AV delays, were compared. Results: The optimal AV delay based on LVOT-VTI was shorter during exercise than at rest in 37%, unchanged in 37%, and longer in 26% of patients. The optimal AV delay based on LVFT was shorter during exercise than at rest in 27%, unchanged in 23%, and longer in 50% of patients. Optimization of the AV delay during exercise increased LVFT and LVOT-VTI significantly (P < .05) compared with (1) any other arbitrarily chosen AV delay, (2) the optimal AV delay at rest, (3) an AV delay systematically shortened from rest to exercise. Conclusion: Optimization of the AV delay had a positive effect on echocardiographic indices of LV function. The systematic shortening of the AV delay during exercise is not recommended because, in a high proportion of patients, the optimal AV delay was longer during exercise than at rest. [Copyright &y& Elsevier]