Reentrant and focal mechanisms underlying ventricular tachycardia in the human heart.

BACKGROUND: To determine the mechanisms of ventricular tachycardia (VT) in humans, three-dimensional intraoperative mapping of up to 156 intramural sites was performed in 13 patients with healed myocardial infarction and refractory VT. METHODS AND RESULTS: Mapping was of sufficient density to define the mechanism of 10 VTs in eight patients. In five of 10 cases, sustained VT was initiated in the subendocardium or epicardium by intramural reentry with marked conduction delay as well as functional and anatomic block most prominent in the subendocardium and midmyocardium. The initiating beats of reentrant VT induced by programmed electrical stimulation arose in the endocardium or midmyocardium by progressive slowing of conduction leading to unidirectional block. Multiple simultaneous reentrant circuits can be present. In contrast, five of the 10 sustained VTs were initiated by a focal mechanism as defined by the absence of electrical activity between the termination of one beat and the initiation of the next despite the presence of multiple intervening intramural electrode recording sites. Comparisons of the mapping data with results of histopathological analysis of tissue demonstrated that the location of infarction as well as that of adjacent fibrotic muscle determined sites of both fixed and functional conduction block during macroreentrant VT and that slowing of conduction occurred in a direction transverse rather than longitudinal to fiber orientation. CONCLUSIONS: Both intramural reentry and a focal mechanism underlie sustained VT in patients with healed myocardial infarction.

Cardiac myocyte interconnections at gap junctions Role in normal and abnormal electrical conduction.

Slow conduction leading to reentrant ventricular tachycardias in patients with healed myocardial infarcts appears to depend primarily on alterations in intercellular coupling at gap junctions of myocytes bordering the infarct scar. Results of correlative morphometric and electrophysiologic studies indicate that the elongated shape of individual myocytes, their complex overlapped packing in myocardium, and the number and distribution of gap junctions that electrically couple myocytes are all important structural determinants of anisotropic patterns of current spread in normal myocardium. Alterations of these structural features likely contribute to electrophysiologic derangements critical in reentrant arrhythmogenesis. Recent observations that cardiac myocytes may be coupled by multiple gap junction channel proteins having unique electrophysiologic properties provide new insights into potential mechanisms regulating intercellular current transfer in the heart.

Beta-adrenergic-mediated influences on microscopic conduction in epicardial regions overlying infarcted myocardium.

Slowing in conduction and functional block in the epicardial tissue overlying a region of transmural necrosis are pivotal in the genesis of many lethal arrhythmias during the healing phase of myocardial infarction. The mechanisms responsible for these alterations in conduction in the epicardial region have not been completely elucidated. In the present study, the microscopic abnormalities in conduction were investigated in vitro using high-density mapping with 224 bipolar electrodes (interelectrode distance, 350 microns) in isolated epicardial tissue excised from the hearts of dogs 2 weeks after anterior myocardial infarction. Seven epicardial tissue slices (group 1) exhibited uninterrupted sequential activation after stimulation at the basic cycle length as well as after premature stimuli. Four tissues (group 2) demonstrated microscopic alterations in conduction both during basic drive and premature stimuli, with the extent of conduction delay dependent on the stimulation site. In three tissues (group 3), the activation sequence was normal during the basic drive but became abnormal after premature stimulation, with the appearance of functional block in the direction transverse to fiber orientation. Superfusion with isoproterenol (10(-6) M) did not significantly modify resting membrane potential, amplitude, or maximum rate change of voltage (Vmax) of phase 0 depolarization but decreased action potential duration. Isoproterenol did not alter the activation sequence during basic drive, but it reduced the slowing in conduction elicited by premature stimulation in group 1 (p less than 0.01 in the transverse direction), alleviated microcircuitous conduction in group 2, and prevented the occurrence of functional block in tissues in group 3. Despite this salutary effect on conduction velocity, arrhythmias occurred after isoproterenol because of focal activity (three of 14, 21%) or reentry (one of 14, 7%) at very short coupling intervals of premature stimulation. These findings indicate that microscopic abnormalities in conduction occur in response to premature stimulation in epicardial regions overlying an infarct, eliciting conduction block that is more prone to occur in the direction transverse to fiber orientation and is dependent on the direction of activation. beta-Adrenergic stimulation prevents or attenuates the alterations in the pattern of activation elicited by premature stimulation. However, this potential beneficial effect of beta-adrenergic stimulation is offset by the occurrence of focal activity and reentry at short coupling intervals of premature stimulation.

Alterations of intercellular junctions induced by hypoxia in canine myocardium.

To delineate potential structural mechanisms of impaired cell coupling induced by hypoxia in canine myocardium, we characterized derangements in intercellular junctions and alterations in the space constant in strips of ventricular epimyocardium before and after selected intervals of hypoxia in vitro. Tissue samples were analyzed morphometrically with transmission and freeze-fracture electron microscopy. Space constants in control tissues averaged 1.61 +/- 0.47 mm (mean +/- SD). After 30 and 60 min of hypoxia, space constants declined by 0.24 +/- 0.22 and 0.32 +/- 0.17 mm, respectively (P less than 0.05 vs. control in each case). Impaired coupling was not reversible with reoxygenation. Focal pathological separation of intercalated disk membranes was observed after 30 min of hypoxia, but morphometric analysis demonstrated no reduction in gap junction surface density to account for uncoupling after 30 min of hypoxia. However, after 60 min of hypoxia, gap junction surface density was reduced by 45%. Quantitative analysis of freeze-fractured gap junction replicas after 30 min of hypoxia revealed a significant decrease in P-face particle diameter from 8.51 +/- 1.64 nm in control tissues to 7.25 +/- 1.33 nm (P less than 0.01) with no further change at 60 min. Thus impaired coupling at 30 min is likely related to a change in the gap junction particle. Further uncoupling after 60 min of hypoxia is likely related, in addition, to reduced gap junction surface density. These results suggest that alterations in P-face particles and gap junction surface density are important determinants of progressive cellular uncoupling induced by hypoxia.

Quantitative analysis of intercellular connections by immunohistochemistry of the cardiac gap junction protein connexin43.

Polyclonal antisera directed against epitopes in the cytoplasmic domain of rat connexin43, the predominant cardiac gap junction protein, were used to delineate immunohistochemically the distribution of gap junctions in sections of canine left ventricle. Antigen-antibody binding and tissue structure were preserved after paraformaldehyde fixation and paraffin embedment of canine myocardium. Specific binding of antibody to the cytoplasmic surfaces of ultrastructurally identified gap junctions was confirmed with electron microscopy. Light microscopic morphometric analysis of immunostained sections in five separate experiments revealed a mean gap junction surface density of 0.0052 micron2/micron3 myocyte volume, which is consistent with previously reported values determined by use of quantitative electron microscopy. This new method permits quantitative determinations of gap junction surface density and distribution in relatively large heterogeneous areas of myocardium in which ultrastructural morphometry would be impractical. This approach should facilitate analysis of the relation between potential alterations in electrical coupling of myocytes and abnormalities of myocardial conduction occurring at the macroscopic scale in regions such as structurally heterogeneous infarct border zones.

A high density in vitro extracellular electrode array: description and implementation.

The detailed activation sequence in myocardium provides information critical to the understanding of the mechanisms of cardiac arrhythmias and the influence of interventions. Despite the pivotal role of activation mapping, the interpretation of isochronic maps and the correlation to detailed tissue morphology may be limited when the interelectrode distances are large with respect to cell dimensions. Additionally, dynamic beat-to-beat variations in the activation pattern or the effect of interventions such as single extra stimuli cannot be assessed adequately without recording from all sites simultaneously. To surmount these limitations, we have fabricated and tested an extracellular recording array consisting of 224 bipolar tungsten wire electrodes with a 350-microns interelectrode distance (140 microns edge-to-edge distance), and used signal processing equipment to record from all electrodes simultaneously at a 2-kHz sample rate. Stimulation can be performed sequentially from 12 different sites at 30 degree angles around the periphery of the recording array. Transarray bipoles can be recorded from any combination of eight radially oriented sites. Activation maps recorded in normal tissue after programmed stimulation and activation maps from an area of fixed anatomic block in the epicardial border zone of infarcted tissue are presented. The results demonstrate a lack of influence of the recording array on the electrophysiological properties of the tissue as verified with transmembrane action potential recordings and sequential extracellular maps. This electrode permits precise assessment of transient details of the activation sequence with unparalleled anatomic resolution.

Pseudothrombocytopenia induced by a monoclonal IgM kappa platelet agglutinin.

Pseudothrombocytopenia is an in vitro phenomenon usually associated with anticoagulant (EDTA)-dependent IgG platelet agglutinins. A low Coulter platelet count was investigated in a 63-year-old woman with multiple sclerosis and a monoclonal IgM kappa gammopathy. An unusual type of EDTA- and temperature-independent IgM platelet agglutinin was identified by in vitro agglutination of donor platelets and was removed by immunoabsorption.

Distribution and three-dimensional structure of intercellular junctions in canine myocardium.

Electrotonic coupling of cardiac myocytes at gap junctions may influence patterns of conduction in myocardium. To delineate the three-dimensional structure and distribution of intercellular junctions, we analyzed serial ultrathin sections of canine myocardium with transmission electron microscopy and disaggregated myocytes with scanning electron microscopy. Morphometric analysis of left ventricular myocardium sectioned in three orthogonal planes revealed that 80% of total gap junctional membrane occurred in large, ribbon-like gap junctions oriented transversely at cell end processes. The remaining 20% of gap junctional membrane was contained in small gap junctions located within plicate segments (interdigitating regions of cell-to-cell adhesion) of intercalated disks. In serial ultrathin sections, all gap junctions were contiguous with plicate segments. Thus, true "lateral" gap junctions do not exist in working ventricular myocytes and would not likely be able to withstand shear forces created by laterally sliding cells. Examination of serial plastic sections with light microscopy revealed complex overlapping of myocytes such that individual myocytes were connected at intercalated disks to an average of 9.1 +/- 2.2 other myocytes. These observations provide an improved understanding of the extent and distribution of cell junctions and should facilitate experimental and model studies of conduction in myocardium.

Transcatheter desiccation of the canine left ventricle using radiofrequency energy: a pilot study.

Catheter ablation of cardiac tissue by means of direct-current electrical energy is associated with several complications. We assessed the efficacy and safety of closed-chest catheter desiccation of the left ventricular myocardium with microbipolar radiofrequency (RF) energy (750 kHz) in five dogs. The unipolar configuration was used with RF energy delivered between the tip electrode of a standard No. 7F tripolar catheter in the left ventricle and an external patch electrode on the left lateral chest wall. A single application with different RF energy settings (100 J, 200 J, and 300 J) was delivered to three individual endocardial sites of the left ventricle. Ventricular tachycardia or fibrillation was not observed during energy application and 24 hours after ablation, as assessed by a Holter recording. There was no damage to the electrode catheter. Dogs were killed on the fifth day. Pathology showed well-delineated ovoid or round-shaped coagulation necrosis at the ablation sites. Microscopic findings consisted of circumscribed areas of necrosis surrounded by a zone of fibroblastic and mononuclear proliferation. In conclusion, catheter ablation of the ventricular myocardium with RF energy is an apparently safe procedure and can effectively produce discrete areas of injury without destruction of surrounding uninvolved myocardium. This method offers potential clinical utility for catheter ablation of refractory sustained ventricular tachycardia.

Assessment of fibrosis in infarcted human hearts by analysis of ultrasonic backscatter.

In animal hearts, the magnitude of integrated ultrasonic backscatter is increased in fibrotic myocardium. Our purpose in this study was to quantitate the relationship between ultrasonic backscatter and collagen deposition in 10 excised human hearts with old infarcts. A 2.25 MHz, 50% fractional bandwidth transducer was positioned at the transducer focal distance from the epicardium of each specimen. The radio frequency backscatter signal was digitized, squared, and integrated to yield the integrated ultrasonic backscatter, which was referenced to the backscatter from a water/steel interface. The interrogated myocardium was then excised and divided into two portions. One portion was assayed for hydroxyproline, a marker for collagen. A second portion was sectioned, stained with Masson's trichrome, and studied with the use of a computer-assisted image analysis system. There was a linear correlation between the magnitude of integrated backscatter and myocardial collagen content estimated by hydroxyproline assay (r = .78). Quantitative histologic analysis revealed a variable relationship between the transmural distribution of collagen and the corresponding transmural pattern of the backscatter signal. In two specimens exhibiting a discrete layer of subendocardial fibrosis, the backscatter amplitude was also increased in the subendocardial region. In specimens with other patterns of fibrosis, the local backscatter amplitude did not correspond to the transmural pattern of collagen distribution. We conclude that the quantitative analysis of ultrasonic backscatter shows promise for the noninvasive evaluation of myocardial fibrosis after infarction.

Computer-assisted quantitation of myocardial fibrosis in histologic sections.

Histologic characterization of the myocardium may require estimation of the collagen content of the specimen. A computer-assisted method can be used to determine collagen content from histologic sections. Full-thickness segments of normal or fibrotic left ventricular myocardium from ten human hearts were divided into portions for hydroxyproline assay and for trichrome staining of histologic sections. The sections were digitized using a computerized video analysis system and were segmented into collagen and muscle regions using a gray-level threshold technique that yielded an estimate of the percent of collagen. A wide range of collagen content, from normal (5 micrograms of hydroxyproline per milligram of dry weight) to fibrotic (38 micrograms/mg) myocardium, was studied. The percentage of collagen from analysis of histologic sections strongly correlated with the hydroxyproline level, with excellent intraobserver and interobserver concordance. Results were unaffected by variations in histochemical staining.