Three-dimensional 123I-meta-iodobenzylguanidine cardiac innervation maps to assess substrate and successful ablation sites for ventricular tachycardia: feasibility study for a novel paradigm of innervation imaging.

BACKGROUND: Innervation is a critical component of arrhythmogenesis and may present an important trigger/substrate modifier not used in current ventricular tachycardia (VT) ablation strategies. METHODS AND RESULTS: Fifteen patients referred for ischemic VT ablation underwent preprocedural cardiac (123)I- meta-iodobenzylguanidine ((123)I-mIBG) imaging, which was used to create 3-dimensional (3D) innervation models and registered to high-density voltage maps. 3D (123)I-mIBG innervation maps demonstrated areas of complete denervation and (123)I-mIBG transition zone in all patients, which corresponded to 0% to 31% and 32% to 52% uptake. (123)I-mIBG denervated areas were ≈2.5-fold larger than bipolar voltage-defined scar (median, 24.6% [Q1-Q3, 18.3%-34.4%] versus 10.6% [Q1-Q3, 3.9%-16.4%]; P<0.001) and included the inferior wall in all patients, with no difference in the transition/border zone (11.4% [Q1-Q3, 9.5%-13.2%] versus 16.6% [Q1-Q3, 12.0%-18.8%]; P=0.07). Bipolar/unipolar voltages varied widely within areas of denervation (0.8 mV [Q1-Q3, 0.3-1.7 mV] and 4.0 mV [Q1-Q3, 2.9-5.6 mV]) and (123)I-mIBG transition zones (0.8 mV [Q1-Q3, 0.4-1.8 mV] and 4.6 mV [Q1-Q3, 3.2-6.3 mV]). Bipolar voltages in denervated areas and (123)I-mIBG transition zones were <0.5 mV, 0.5 to 1.5 mV, and >1.5 mV in 35%, 36%, and 29%, as well as 35%, 35%, and 30%, respectively (P>0.05). Successful ablation sites were within bipolar voltage-defined scar (7%), border zone (57%), and areas of normal voltage (36%), but all ablation sites were abnormally innervated (denervation/(123)I-mIBG transition zone in 50% each). CONCLUSIONS: (123)I-mIBG innervation defects are larger than bipolar voltage-defined scar and cannot be detected with standard voltage criteria. Thirty-six percent of successful VT ablation sites demonstrated normal voltages (>1.5 mV), but all ablation sites were within the areas of abnormal innervation. (123)I-mIBG innervation maps may provide critical information about triggers/substrate modifiers and could improve understanding of VT substrate and facilitate VT ablation. CLINICAL TRIAL REGISTRATION: URL: http://www.clinicaltrials.gov. Unique Identifier: NCT01250912.

MRI-Guided ventricular tachycardia ablation: integration of late gadolinium-enhanced 3D scar in patients with implantable cardioverter-defibrillators.

BACKGROUND: Substrate-guided ablation of ventricular tachycardia (VT) in patients with implanted cardioverter-defibrillators (ICDs) relies on voltage mapping to define the scar and border zone. An integrated 3D scar reconstruction from late gadolinium enhancement (LGE) MRI could facilitate VT ablations. METHODS AND RESULTS: Twenty-two patients with ICD underwent contrast-enhanced cardiac MRI with a specific absorption rate of <2.0 W/kg before VT ablation. Device interrogation demonstrated unchanged ICD parameters immediately before, after, or at 68±21 days follow-up (P>0.05). ICD imaging artifacts were most prominent in the anterior wall and allowed full and partial assessment of LGE in 9±4 and 12±3 of 17 segments, respectively. In 14 patients with LGE, a 3D scar model was reconstructed and successfully registered with the clinical mapping system (accuracy, 3.9±1.8 mm). Using receiver operating characteristic curves, bipolar and unipolar voltages of 1.49 and 4.46 mV correlated best with endocardial MRI scar. Scar visualization allowed the elimination of falsely low voltage recordings (suboptimal catheter contact) in 4.1±1.9% of <1.5-mV mapping points. Display of scar border zone allowed identification of excellent pace mapping sites, with only limited voltage mapping in 64% of patients. Viable endocardium of >2 mm resulted in >1.5-mV voltage recordings despite up to 63% transmural midmyocardial scar successfully ablated with MRI guidance. All successful ablation sites demonstrated LGE (transmurality, 68±26%) and were located within 10 mm of transition zones to 0% to 25% scar in 71%. CONCLUSIONS: Contrast-enhanced cardiac MRI can be safely performed in selected patients with ICDs and allows the integration of detailed 3D scar maps into clinical mapping systems, providing supplementary anatomic guidance to facilitate substrate-guided VT ablations.

Three-dimensional contrast-enhanced multidetector CT for anatomic, dynamic, and perfusion characterization of abnormal myocardium to guide ventricular tachycardia ablations.

BACKGROUND: Advances in contrast-enhanced multidetector CT enable detailed characterization of the left ventricular myocardium. Myocardial scar and border zone (BZ), as the target of ventricular tachycardia ablations, displays abnormal anatomic, dynamic, and perfusion characteristics during first-pass CT. This study assessed how contrast-enhanced CT can predict voltage-defined scar and BZ and integrate its scar reconstructions into clinical mapping systems to guide ventricular tachycardia ablations. METHODS AND RESULTS: Eleven patients with ischemic cardiomyopathy underwent contrast-enhanced CT before ventricular tachycardia ablation. Segmental anatomic (end-systolic and end-diastolic wall thickness), dynamic (wall thickening, wall motion), and perfusion (hypoenhancement) characteristics were evaluated. Receiver operating characteristic curves assessed the ability of CT to determine voltage-defined scar and BZ segments. Three-dimensional epi- and endocardial surfaces and scar borders were reconstructed, coregistered, and compared to voltages using a 17-segment model. Abnormal anatomic, dynamic, and perfusion data correlated well with abnormal (<1.5 mV) endocardial voltages (r=0.77). Three-dimensional reconstruction integrated into the clinical mapping system (registration accuracy, 3.31±0.52 mm) allowed prediction of homogenous abnormal voltage (<1.5 mV) in 81.7% of analyzed segments and correctly displayed transmural extent and intramural scar location. CT hypoperfusion correlated best with scar and BZ areas and encompassed curative ablations in 82% cases. CONCLUSIONS: Anatomic, dynamic, and perfusion imaging using contrast-enhanced CT allows characterization of left ventricular anatomy and 3D scar and BZ substrate. Integration of reconstructed 3D data sets into clinical mapping systems supplements information of voltage mapping and may enable new image approaches for substrate-guided ventricular tachycardia ablation.

Clinical application of PET/CT fusion imaging for three-dimensional myocardial scar and left ventricular anatomy during ventricular tachycardia ablation.

Integration of 3D PET with Voltage Map for VT Ablation. Background: Image integration has the potential to display three-dimensional (3D) scar anatomy and facilitate substrate characterization for ventricular tachycardia (VT) ablation. However, the current generation of clinical mapping systems cannot display 3D left ventricle (LV) anatomy with embedded 3D scar reconstructions or allow display of border zone and high-resolution anatomic scar features. Objective: This study reports the first clinical experience with a mapping system allowing an integrated display of 3D LV anatomy with detailed 2D/3D scar and border zone reconstruction. Methods: Ten patients scheduled for VT ablation underwent contrast-enhanced computed tomography (CT) and Rubidium-82 perfusion/F-18 Fluorodeoxyglucose metabolic Positron Emission Tomography (PET) imaging to reconstruct 3D LV and scar anatomy. LV and scar models were co-registered using a 3D mapping system and analyzed with a 17-segment model. Metabolic thresholding was used to reconstruct the 3D border zone. Real-time display of CT images was performed during ablation. Results: Co-registration (error 4.3 +/- 0.7 mm) allowed simultaneous visualization of 3D LV anatomy and embedded scar and guided additional voltage mapping. Segments containing homogenous or partial scar correlated in 94.4% and 85.7% between voltage maps and 3D PET scar reconstructions, respectively. Voltage-defined scar and normal myocardium had relative FDG uptakes of 40 +/- 13% and 89 +/- 30% (P < 0.05). The 3D border zone correlated best with a 46% metabolic threshold. Real-time display of registered high-resolution CT images allowed the simultaneous characterization of scar-related anatomic changes. Conclusion: Integration of PET/CT reconstruction allows simultaneous 3D display of myocardial scar and border zone embedded into the LV anatomy as well as the display of detailed scar anatomy. Multimodality imaging may enable a new image-guided approach to substrate-guided VT ablation.