AJP - Heart Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 287: H823-H832, 2004; doi:10.1152/ajpheart.00795.2003
0363-6135/04 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Taneja, T.
Right arrow Articles by Kadish, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Taneja, T.
Right arrow Articles by Kadish, A.

Is there a correlation between ventricular fibrillation cycle length and electrophysiological and anatomic properties of the canine left ventricle?

Taresh Taneja, George Horvath, Darlene K. Racker, David Johnson, Jeffrey Goldberger, and Alan Kadish

Division of Cardiology, Department of Medicine, Feinberg Cardiovascular Research Institute, Northwestern University, Chicago, Illinois 60611

Submitted 18 August 2003 ; accepted in final form 11 March 2004


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
We hypothesized that myocardial infarction-related alterations in ventricular fibrillation (VF) cycle length (VFCL) would correlate with changes in local cardiac electrophysiological and anatomic properties. An electrophysiological study was performed in normal, subacute, and chronic infarction mongrel dogs. VF was induced by programmed electrical stimulation and mean and minimum early and late VFCL was determined and correlated with local electrophysiological and anatomic properties. Effective refractory period (ERP), activation recovery time (ART), ERP/ART ratio, threshold, and ERP and ART dispersion were determined at 112 sites on the anterior left ventricle. Wave front progression was analyzed over a 2-s period. The extent of local tissue necrosis and of myocardial fiber disarray was also evaluated. The early mean VFCL was significantly longer in the subacute infarction (149 ± 35 ms) and chronic infarction dogs (129 ± 18 ms) compared with control dogs (102 ± 15 ms; P < 0.0001 for both comparisons) as was the early minimum VFCL with similar trends seen during late VF. Complete epicardial reentrant circuits were significantly more common in normal dogs (4.3 ± 2.4, 22.4% of cycles) than in subacute (0.75 ± 0.96, 5.3% of cycles, P < 0.05 vs. normal) and chronic infarction dogs (1.3 ± 1.3, 7.5% of cycles, P < 0.05 vs. normal). There was a poor correlation between the mean and minimum early and late VFCL and local electrophysiological and anatomic properties (R2 < 0.2 for all comparisons) with a much better correlation between average mean and minimum VFCL (over the entire plaque) and global ERP and ART dispersion during early and late VF. In conclusion, VFCL in normal and infarcted myocardium shows a poor correlation with local ventricular electrophysiological and anatomic properties measured in sinus rhythm. However, there was a much better correlation between the average VFCL with global dispersion of repolarization. The lack of correlation between local VFCL and refractoriness and the infrequent occurrence of epicardial reentry suggests that intramural reentry may be the primary mechanism of VF in this model.

VENTRICULAR FIBRILLATION (VF) is the most common terminal arrhythmia in patients experiencing a cardiac arrest (32). Although VF may occur in patients with structurally normal hearts, its incidence is increased in the presence of ischemic heart disease. Reentry during VF has been demonstrated to be due to functional reentrant circuits that have been modeled as rotors or spiral waves. Although such rotors can occur in normal myocardium, anatomic or structural abnormalities may cause anchoring of these rotors, and local electrophysiological heterogeneity may alter the shape and drifting pattern of rotors in in vitro or Langendorff preparations (6, 7, 15, 29). If such anchoring of rotors is present in vivo or if local electrophysiological and anatomic properties are important in the maintenance of VF in the presence of infarction, alterations in local electrophysiology and anatomic characteristics should alter VF cycle length (VFCL).

We (9) have reported that in the presence of infarction there is a prolongation of effective refractory periods (ERP) and activation recovery times (ART) that is most prominent in subacute infarction. In addition, VFCL is longer in animal models of subacute infarctions than that in chronic infarctions (4). We examined whether the electrophysiological properties determined in sinus rhythm would correlate with VFCL in an infarction model based on the following criteria: 1) our findings, which suggested changes in local cardiac electrophysiology with evolving canine myocardial infarction, and 2) the findings of earlier studies (13, 14, 19, 21, 22) on atrial and ventricular fibrillation, which demonstrated that local fibrillation cycle length is highly correlated with local refractory period. In a prior study (26), we described the relationship between the excitable gap during VF and VFCL in this same group of experiments. An additional goal of the present study was to determine the extent to which local electrophysiological and anatomic properties alter local VFCL.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Three groups of mongrel dogs were studied. Group 1 consisted of seven control dogs without infarction. One dog was excluded because metabolic abnormalities developed during the protocol. Thus six group 1 dogs were analyzed. Group 2 consisted of six dogs studied 5 days postmyocardial infarction, and group 3 of six dogs studied 8 wk postinfarction. Data on electrophysiological properties of these animals during sinus rhythm have been previously published (9).

Experimental preparation. Myocardial infarction was created using a two-step ligation procedure of the left anterior descending artery (8). The animals were anesthetized with 0.1 mg/kg iv acetyl promethazine and 25 mg/kg iv thiopental sodium, and mechanically ventilated with a mixture of room air and 1–1.5% halothane. Under sterile conditions, a left thoracotomy was performed at the fifth intercostal space, the pericardium was opened, and the left anterior descending artery was isolated proximal to the origin of the first major diagonal artery. Myocardial infarction was created with the use of partial occlusion of the left anterior descending artery for 20 min, followed by total occlusion. Intravenous lidocaine (1 mg/kg bolus) was administered if ventricular arrhythmias occurred after infarction; all animals received prophylactic antibiotics, and analgesics were administered as required. The study protocols conformed to the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, Revised 1996) and were approved by the Animal Care and Use Committee at Northwestern University.

Electrophysiological studies were performed in normal dogs without prior surgery. Subacute and chronic infarction dogs were studied 5.0 ± 1 days and 8.1 ± 0.5 wk postinfarction, respectively. The dogs were anesthetized with pentobarbital (30 mg/kg iv initial dose and 4 mg·kg–1·h–1 subsequently), intubated, and mechanically ventilated. A rectangular 8 x 14 electrode array (Bard; Billerica, MA) was sutured to the epicardial surface of the left ventricle over the region of infarction with its long axis parallel to the left anterior descending artery. The plaque array has an edge-to-edge interelectrode distance of 2.5–2.75 mm and an electrode diameter of 0.25 mm. Pacing electrodes were sutured to the epicardial surface of the right ventricle, basal left ventricle, and the right atrium. Recordings from the 112 unipolar leads of the anterior plaque electrode and the surface electrocardiographic limb leads were acquired and stored in digitized form on a videotape with the use of a cardiac mapping system (Map Tech; Maastricht, The Netherlands), and a custom-designed software program was run on a computer system (Dell; Dallas, TX). The signals were acquired with 8-bit resolution at a sampling frequency of 1,000 Hz.

Electrophysiological study. The methodology of the electrophysiological study has been described previously (9). After instrumentation, recordings were made during sinus rhythm. VF was induced by burst pacing. A 15-beat train was used at 5 mA beginning at a cycle length of 200 ms. If VF was not induced, the cycle length was decreased in 10-ms decrements until fibrillation occurred. After being recorded for ~20 s, ventricular fibrillation was terminated by a 10 J shock administered by epicardial paddles.

Electrophysiological properties, fibrillation cycle length, and the anatomic properties were determined at each recording electrode. After a 10-min recovery period, bipolar pacing was performed at twice diastolic threshold using a 2-ms pulse width from two sites: the right ventricle (septal margin of the plaque) and the center of the recording plaque for the measurement of ART and conduction velocities. At each site, pacing was performed at cycle lengths of 300 and 220 ms for 15 beats each.

Unipolar pacing was performed from each of the 112 sites of the anterior electrode array. The indifferent electrode was a skin patch. At each site, the late diastolic pacing threshold was determined. Sites at which capture could not be obtained using current strengths of 5 mA were defined as inexcitable. ERP determinations were performed at each of the 112 sites during pacing at twice diastolic threshold and 2-ms pulse width, using an 8-beat drive train with a cycle length of 300 ms and a single extrastimulus with 4-s intertrain pauses (Fig. 1). Atrial pacing was performed during the drive train to eliminate a variable AV interval (11). The S1-S2 interval was progressively increased by 5-ms increments until capture occurred. Before the ERP determinations began, a 5-min conditioning period of atrial pacing was carried out with the use of 8-beat drives at a pacing cycle length of 300 ms and 4-s intertrain pauses (18). Pacing at a cycle length of 300 ms was maintained using an 8-beat train and 4-s intertrain pause during the ~120 min required to determine ERP at each of the 112 sites to ensure that a steady state was maintained. The ERP was defined as the longest premature interval (S1S2) that failed to capture.



View larger version (41K):
[in this window]
[in a new window]
  		Fig. 1. Example of the surface ECG and epicardial recordings during ventricular fibrillation (VF) with local VF cycle length (VFCL) and effective refractory period (ERP) measurements in a normal dog. A: recordings of early VF in a control dog from surface lead II and epicardial channel 72. B: designated epicardial activations during VF in channel 72 in a 640-ms window. C: mean VFCL at each of the 112 sites from recordings shown above. *Site of the local electrogram recording in B. Missing sites were due to noisy channel recordings. D: ERP determined at a pacing cycle length of 300 ms at each of the 112 sites. The correlation coefficient (R2) was 0.017 between the local mean VFCL and ERP at 300 ms, illustrating the poor correlation between VFCL and local electrophysiology.

 
After the ERP determinations, pacing from the right ventricle at cycle lengths of 300 and 220 ms was repeated to assess the stability of the preparations (Fig. 1). ART and conduction velocities obtained before and after determination of the ERP at all 112 sites were compared. They varied by a mean of <5% in all but one dog, and this animal was excluded from further analysis.

Electrophysiological data analysis. Two separate time windows of VF were analyzed; one window beginning 4 s after VF initiation and the other beginning 18 s after VF initiation. Activations at each of the 112 sites were marked using an automated system (Map Tech) and manually over-read. The criterion for local activation on unipolar electrograms was a peak negative dV/dt of at least 0.9 V/s, based on prior studies in our laboratory (4). We also followed the criteria of other investigators, for which local activation on unipolar electrograms have ranged from –0.2 to –2.5 V/s (1), and when the dV/dt was between 0.5 and 1.5 V/s, the presence of an activation on reconstructed bipolar recordings was made from the unipolar sites (12). If two activations were within 50 ms of each other, the one that was larger and had a greater dV/dt was included (Fig. 2). R-R intervals were determined, with those intervals spanning failed activations excluded. VF was divided into "cycles" as previously described (4). The distance between successive conducted R-R intervals at each site (in ms) was taken as the VFCL at the site for that cycle (Fig. 1). To determine failed activations, if an entire cycle was present without activation at a site, i.e., an absence of a maximal negative dV/dt of ≥0.9 mV/ms on the unipolar electrogram or an absent recording on a bipolar electrogram, it was defined as a failed activation (Fig. 2). In that instance, a VFCL was not calculated. The mean and minimum R-R interval and the frequency of failed activation were noted at each site.



View larger version (31K):
[in this window]
[in a new window]
  		Fig. 2. Demonstration of the method to determine the frequency of blocked impulses. A: two successive 640-ms windows of early VF from an animal with a subacute infarction. Window 1 shows relatively uniform conduction across a vertical column of 8 channels, whereas window 2 shows intermittent functional block across a horizontal row of 8 channels (* in B). The activation in channel 66 at 244 ms in window 1 had a dV/dt = –1.17 V/s, as opposed to the prior activation (not marked) at 204 ms with a dV/dt = –0.94 V/s. Similarly, in channel 78 in window 1 the first activation (not marked) at 138 ms had a dV/dt = –1.41 V/s as opposed to the first marked activation at 178 ms with a dV/dt = –1.64 V/s. Because these successive activations were <50 ms apart the greater activation of the two was taken. B: isochronal maps of the first and second beats in window 2. The first isochrone shows the impulse originating at 90-ms encounters a zone of functional block and then conducts homogeneously whereas the second isochrone shows the wave front originating at 300 ms with rapid homogeneous conduction in one direction (straight arrow) and curving around a line of functional block in the other direction (curved arrow). Top, the area shown in the shaded polygon was considered to represent a cycle with a blocked impulse.

 
Functional block (Fig. 2) was distinguished from anatomic block by the presence of activation in sinus rhythm and evidence of excitability during unipolar pacing. If there was a failure of activation during VF in a site that was activated in sinus rhythm as well as evidence of excitability during pacing, then the failure of activation during VF was attributed to functional block as opposed to anatomic block.

ART in sinus rhythm were measured using a custom-designed software system. Activation time was measured at the maximal negative dV/dt of the unipolar electrogram recorded by each electrode. The recovery time was measured at the maximal positive or negative dV/dt of the terminal portion of the T wave (17). The ART at each site was defined as the interval between activation and recovery and calculated as the average over the last three beats of the drive train. Local dispersion in ERP, ART, and ERP/ART ratio [an index of postrepolarization refractoriness which predisposes to conduction block and reentry (10)], and conduction velocity was evaluated using the maximum difference (largest – smallest value) among each group of four adjacent sites, so that 91 data points were available for each experiment (9).

Isochronal and vector maps of 2 s of VF beginning 4 s after initiation were analyzed to examine wave front progression and reentry. A complete reentrant wave front was defined as the complete propagation along a point or line of block and the duration of reentry was determined based on the number of cycles. Wave fronts were characterized as epicardial breakthroughs (originated within the plaque) and entries (wave fronts entering from the plaque edge). The results were corrected for the percentage of available sites on the plaque.

Histological methods. After the electrophysiological studies, the animals were euthanized by induction of VF and the corners of the plaque were marked on the hearts with sutures and the hearts were fixed in 10% buffered formalin for at least 1 wk. The methodology for histological analysis has been described in detail previously (9). In brief, a block of tissue correlating with the plaque was removed using a template and the tissue was divided into 10 blocks, four vertical blocks (containing the full thickness of the ventricular wall and 2.5 mm in thickness), and six horizontal blocks (made parallel to the epicardial surface and ~3 mm in thickness) (Fig. 3). The blocks were embedded in paraffin and sectioned at 5-µm thickness, and the first full section from each block was stained with hematoxylin and eosin (9).



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 3. Diagram of the template used for blocking and histological analysis. Vertical (V1-V4) and horizontal (H1-H6) were examined. The horizontal sections were used for myocardial viability and the vertical sections to assess depth of infarction. Figure 3 has been modified from Fig. 1 in Ref. 6.

 
The viability of the myocardium and myofiber orientation was evaluated in the horizontal sections and using a template showing electrode positions and fields (Fig. 3). Myofiber viability was graded semiquantitatively as follows: fields containing no viable or all necrotic myocardium, <50% viable fibers, >50% viable fibers and containing normal myofibers, or having 100% viability. Normal myocardium is defined as uniformly stained with regular sarcoplasmic content, sarcolemma contours, and myofiber size. Fields containing only epicardial fat and blood vessels were excluded from analysis.

Myocardial fiber orientation with respect to the electrodes and the long axis of the plaque was determined at a magnification of x40 using an ocular reticule marked in degrees with the reticule's 0 to 180° axis parallel to the length of the plaque. Four measurements were made randomly from each field and averaged to determine the mean fiber angle, from which local and regional measurements of dispersion were calculated (9).

The transmurality of pathological alterations was evaluated in vertical sections. The presence of infarct or healed scar was ranked 1 when localized to the subendocardium, 2 when extending into the midwall, 3 when extending from the subendocardium to the subepicardium, and 0 in the absence of infarction.

Statistical analysis. Comparisons among groups were performed using one-way analysis of variance. Post hoc analyses were performed using Fisher's protected least-significant difference method. Site-by-site correlations between VFCL and electrophysiological-anatomic properties were performed using simple linear regression and stepwise linear regression analyses. A P value of ≤0.05 was considered significant. These categories were then compared using factorial ANOVA to the mean and minimum VFCL.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Differences in electrophysiological and anatomic properties among the different animal groups have been previously reported (9) (see Fig. 3).

Effects of myocardial infarction on VFCL. Figure 1 shows VFCL determined in a single control animal during early VF at each of the 112 sites. The mean early VFCL was significantly longer in the subacute infarction dogs (149 ± 35 ms) and chronic infarction dogs (129 ± 18 ms) than in control dogs (102 ± 15 ms) (P < 0.0001 for both comparisons). The difference between the subacute and chronic infarction dogs was also significant (Fig. 4). The mean minimum early VFCL was also significantly longer in the subacute and chronic infarction (110 ± 36 and 93 ± 22 ms, respectively) compared with the control animals (73 ± 16 ms; P < 0.0001 vs. each of the groups) (Fig. 4). During late VF, the mean VFCL was shortest in control animals (116 ± 21 ms; P < 0.0001 vs. chronic and subacute infarction animals), followed by chronic infarction dogs (126 ± 21 ms, P < 0.0001 vs. subacute infarction animals) and subacute infarction dogs (153 ± 33 ms). The mean minimum VFCL was 76 ± 20 ms in controls (P < 0.0001 vs. chronic and subacute infarction animals), 109 ± 47 in subacute (P < 0.0001 vs. chronic infarction animals), and 83 ± 24 ms in chronic infarction dogs (P < 0.05 vs. controls) (Fig. 4).



View larger version (53K):
[in this window]
[in a new window]
  		Fig. 4. Mean and minimum early and late VFCL in different groups of dogs. *P ≤ 0.0001 vs. other groups in each of the four subgroups.

 
Anatomic effects on VF cycle length and block during VF. Histology could not be determined because of serial sectioning difficulties in one dog each in the subacute and chronic infarction groups. Thus a total of 16 canine hearts were studied. In the subacute infarction dogs, 90% of slices showed patchy transmural infarctions and 10% showed infarcts extending to the midmyocardium. In the chronic infarction dogs, 10% of slices had evidence of transmural healed scar, 25% had scars extending to the midmyocardium, 30% had subendocardial scars, whereas 35% had no scar, changes consistent with postinfarct remodeling.

Most subepicardial fields had some degree of viability. In the subacute infarction dogs 37.2% of fields had <50% viability, 24.4% of the fields had complete viability, 21.8% had >50% viability, and 16.6% had no apparent viability. In contrast, in chronic infarction dogs, 48.2% had 100% viability, 25.9% had <50% viable myofibrils, 22.2% had >50% viability, and 3.7% had no viability (P < 0.0001; subacute vs. chronic infarctions). Because only 3.7% of sites had no viability in chronic infarctions, they were excluded from further analysis. The greater extent of viability in chronic as opposed to subacute infarctions was consistent with the remodeling that occurs following acute infarction (5).

Eighty-six (4.3%) sites showing no activation in sinus rhythm and VF and were excluded from analysis. These constituted 9.23% (62/432) of sites in subacute infarction dogs, 3.42% (23/432) in chronic infarction dogs, and one site in control dogs. The percentage of failed activation during VF was significantly greater in the subacute infarctions (59 ± 34%) versus control dogs (21 ± 13%) and chronic infarction dogs (26 ± 19%) (P < 0.05 for each comparison). In subacute infarctions, the percentage of failed activation was significantly greater in sites with no apparent viability (80 ± 25%) or <50% viability (79 ± 30%) than sites with >50% viability (58 ± 33%) and areas with 100% viability (44 ± 31) (P ≤ 0.001 for each comparison). In the chronic infarction group the percentage of failed activation was greatest in areas with <50% or >50% viability (23 ± 17% and 24 ± 17%, respectively) compared with areas with 100% viability (19 ± 9%, P < 0.05 vs. >50% viability, P = 0.06 vs. <50% viability) (Fig. 5). Even in fields with no viability, activation was occasionally noted likely because there were scattered surviving myocardial cells that exhibited residual activation.



View larger version (35K):
[in this window]
[in a new window]
  		Fig. 5. Percentage of failed activation in subacute and chronic infarction dogs. Subacute infarctions had a greater percentage of failed activation than chronic infarctions. *P < 0.001 vs. >50% and 100% viability; {dagger}P < 0.05 vs. 100% viability; {ddagger}P < 0.05 vs. >50% viability; §P = 0.06 vs. <50% viability.

 
In the subacute infarction group, the mean early VFCL was significantly longer in areas with >50% or complete viability (161 ± 34 and 166 ± 19 ms, respectively) compared with areas with no or <50% viability (145 ± 26 and 149 ± 38 ms, respectively; P < 0.05). However, substantial overlap was noted among the early VFCL in each group. In the chronic infarction group, the mean early VFCL was significantly longer in areas with <50% viability (135 ± 14 ms) and areas with >50% viability (132 ± 20 ms) than areas with 100% viability (124 ± 18 ms).

There was a poor correlation of ERP, ART, and ERP/ART with the extent of transmural infarction (R2 of 0.06, 0.05, and 0.25, respectively). In contrast, there was a much better correlation between the extent of infarction with the threshold, maximum ERP dispersion, and maximum ART dispersion with R2 of 0.45, 0.62, 0.56, respectively (P < 0.001 for each comparison). The correlation of percent failed activation to local electrophysiological properties was poor, with R2 values of 0.005, 0.06, 0.08, 0.08, 0.20, and 0.09 with ERP, ART, ERP/ART, threshold, maximum ERP dispersion, and maximum ART dispersion, respectively (P < 0.0001 for each).

Correlation of VFCL to baseline electrophysiological and anatomic properties: individual site analysis. The correlation between the mean and minimum early and late VFCL to baseline electrophysiological properties in all animals is shown in Tables 1 and 2. The mean and minimum VFCL during early and late VF correlated only weakly with ERP, ART, ERP/ART, threshold, and fiber angle. Data on early VFCL and ERP in one experiment are shown in Fig. 1. The correlation of the mean and minimum early and late VFCL to measures of local dispersion, i.e., ERP, ART, ERP/ART, threshold, and conduction velocity and maximum ERP and ART dispersion though significant were weak (Table 1 and 2). Subgroup analyses showed similar weak correlations (Tables 1 and 2; Fig. 6). On stepwise regression of mean and VFCL with ART at 300 ms, threshold, maximum surrounding ERP, and ART dispersion, the R2 value increased to 0.34. Similar stepwise regression of minimum early VFCL resulted in R2 values of 0.23. During late VF, on stepwise regression of mean and minimum VFCL with ERP at 300 ms, threshold, maximum surrounding ERP, and ART dispersion, the R2 value increased to 0.276 and 0.149, respectively. There was also poor correlation between early and late VFCL and the percentage of failed activation (Tables 1 and 2).


View this table:
[in this window]
[in a new window]
  		Table 1. R2 of early VFCL to baseline electrophysiological properties and measures of local dispersion

 

View this table:
[in this window]
[in a new window]
  		Table 2. R2 of late VFCL to baseline electrophysiological properties and measures of local dispersion

 


View larger version (35K):
[in this window]
[in a new window]
  		Fig. 6. Scatterplots of mean early VFCL and ERP at individual sites in control dogs (A), subacute infarction dogs (B), and chronic infarction dogs (C). Note the lack of correlation in all three sets of animals.

 
Correlation of mean VFCL to global electrophysiological properties. When average values of early and late VFCL and cardiac electrophysiological properties across the entire recording plaque in each animal were evaluated, the correlation increased dramatically (Tables 2 and 4). For all animals, the mean and minimum early VFCL correlated best with the threshold (R2 = 0.29 and 0.25, respectively; P ≤ 0.01), maximum surrounding ERP (R2 = 0.38 and 0.46, respectively; P ≤ 0.001), and maximum surrounding ART dispersion (R2 = 0.27 and 0.24, respectively; P ≤ 0.05). On subgroup analysis (Table 3), in control dogs, the mean and minimum early VFCL correlated best with maximum surrounding ART dispersion (R2 = 0.80 and 0.64, respectively; P ≤ 0.05). In subacute infarction dogs, the mean and minimum early VFCL correlated best with ERP/ART, a measure of repolarization (R2 = 0.77 and 0.84, respectively; P < 0.05). The correlation between early VFCL and ERP was still poor (Table 2). Also, the correlation of early VFCL to global failed activation continued to be poor.


View this table:
[in this window]
[in a new window]
  		Table 4. R2 of average late VFCL to global electrophysiological properties

 

View this table:
[in this window]
[in a new window]
  		Table 3. R2 of average early VFCL to global electrophysiological properties

 
During late VF, for all animals the mean and minimum VFCL correlated best with maximum ART dispersion (R2 = 0.82 and 0.80, respectively; P < 0.05), followed by threshold (R2 = 0.34 and 0.35, respectively; P < 0.05 for each) and ERP/ART ratio (R2 = 0.32 and 0.30, respectively; P < 0.05 for each). In control dogs, the minimum VFCL correlated best with the ART and ERP at 300 ms (P ≤ 0.09), whereas the mean VFCL had no significant correlations (Tables 2 and 4). In subacute infarction dogs, the mean and minimum late VFCL correlated best with maximum ART dispersion with R2 of 0.80 and 0.92, respectively (P < 0.05), with the minimum VFCL also correlating with the ERP/ART (R2 = 0.84; P < 0.05) and ERP at 300 ms (R2 = 0.63; P = 0.06). The late VFCL in chronic infarction dogs had the weakest correlations (Tables 2 and 4).

Propagation during VF. Two dogs in the subacute infarct group had too few active sites during VF to accurately analyze propagation during VF. Thus four dogs were analyzed from each group. In normal dogs, 99% of sites were analyzed compared with 80% and 78% of sites in subacute and chronic infarction dogs. Nineteen cycles of VF were analyzed in normal dogs as opposed to 14.7 VF cycles in subacute infarction dogs (P < 0.05 vs. normal) and 16 VF cycles in chronic infarction dogs. Complete epicardial reentrant circuits, while infrequent, were significantly more common in normal dogs (4.3 ± 2.4) than in subacute infarction dogs (0.75 ± 0.96; P < 0.05 vs. normal dogs) and chronic infarction dogs (1.3 ± 1.3; P < 0.05 vs. normal dogs). There were no significant differences in the duration of the reentrant wave front (1.4 ± 0.3, 1.8 ± 0.4, and 1.7 ± 0.6 cycles in normal, subacute, and chronic infarction dogs). Complete reentrant circuits accounted for 22.4 ± 11.7% of VF cycles in normal dogs compared with 5.3 ± 7.3% of VF cycles in subacute infarction dogs (P < 0.05 vs. normal dogs) and 7.5 ± 7.9% of VF cycles in chronic infarction dogs (P < 0.5 vs. normal dogs). Epicardial breakthroughs were more common in subcaute (6.6 ± 8.4) and chronic infarction dogs (6.8 ± 6.1) than in normal dogs (0.8 ± 1.0). The predominant mode of activation of the left ventricular epicardium during VF was wave front entry from the plaque margins (20.0 ± 3.5 in normal dogs, 29.0 ± 11.8 in subacute infarction, and 26.0 ± 9.4 in chronic infarction dogs). The total number of wave fronts over 2 s of VF was similar in the groups: 26.6 ± 6.0 in normal dogs, 37.3 ± 7.3 in subacute infarction dogs, and 35.6 ± 8.1 in chronic infarction dogs.

The number of epicardial reentrant circuits correlated poorly with ART, ERP, and ART dispersion, with an R2 of 0.16, 0.24, and 0.13, respectively. There was a moderately good inverse correlation between epicardial circuit number and mean and minimum VFCL (R2 of –0.43 for each correlation; P < 0.01). The correlation of reentrant circuit number with the threshold and ERP dispersion was better with R2 of 0.39 and 0.44 (P < 0.004).


   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
Main findings. The present study demonstrates that no single or group of local electrophysiological property correlates well with the mean or minimum VFCL or with the presence of failed impulse propagation during early VF. This finding was present both in control animals as well as in animals with varying phases of experimental myocardial infarction. A longer VFCL was associated with fewer epicardial reentrant circuits. However, a greater extent of histological necrosis was associated with a higher percentage of impulse blocks during VF and possibly a longer VFCL in subacute infarcts. In addition, when electrophysiological properties were evaluated over a large region of the epicardium, a much higher degree of correlation between indexes of dispersion of repolarization and VFCL was seen. Thus the VFCL in this model is influenced by electrophysiological properties over a large region of myocardium.

Anatomic and electrophysiological effects on VFCL. Reentrant circuits underlying VF usually are secondary to functional reentry that may be epicardial, transmural, and intramural. In a recent report by Valderrabano et al. (27), intramural reentry in the left ventricle was demonstrated in all 27 VF episodes studied, and transmural reentry was present in 24 of 28 right VF episodes in an in vitro swine heart preparation. Several other recent studies have suggested that three-dimensional reentrant waves or rotors may underlie VF (6, 15, 30) and atrial fibrillation (AF) (16, 25), although one study (23) has suggested that in normal pig hearts epicardial reentry may be demonstrated. These rotors may either drift through the heart producing beat-to-beat changes or remain stationary anchored by anatomic barriers. The presence of such drifting and stationery reentrant circuits in fibrillation had led us to suggest that local anatomic and electrophysiological properties influence VF and would correlate to VFCL.

In evaluating VFCL, we divided episodes of fibrillation into individual cycles and found that in the presence of myocardial infarction, conduction block was frequent, presumably due to prolongation of refractoriness. The presence of local tissue necrosis was associated with a high frequency of functional impulse block, especially in subacute infarction presumably due to a prolongation of refractoriness or a lower safety factor for conduction (9). Conduction block complicates the interpretation of the relationship between VFCL and electrophysiological properties. Even in chronic myocardial infarction, where impulse block was encountered less frequently, a significant number of sites demonstrated conduction block. Thus the weak relationship between VFCL and electrophysiological properties could have related in part to the influence of conduction block. However, when the frequency of conduction block was correlated with local electrophysiological properties, only weak correlations were noted. Thus the lack of a good correlation between VFCL and electrophysiological properties was not secondary to the extent of block.

On the basis of our observations and prior studies (31), epicardial wave fronts formed only a portion of reentrant circuits or represented passive activation. Thus electrophysiological and anatomic properties elsewhere in the circuit could have contributed to alterations in VFCL. Even in normal dogs with more epicardial circuits, there was not a good correlation between local EP properties and VFCL. Although a poor correlation was seen between ERP and many local electrophysiological properties, a higher correlation was noted when global VFCL and electrophysiological properties were compared.

The unexpected lack of relationship between VFCL to local electrophysiological properties may result mainly from several factors, including the rate dependence of refractoriness in our model of infarcted myocardium, the type of reentrant circuits seen, or the transmural nature of reentry during VF. The results of the present study and past studies (23, 27, 31) suggest that intramural reentry is an important component of VF. Because we sampled refractory periods only in a portion of the reentrant circuit, the VFCL may be influenced by intramural reentrant circuits that were not sampled in our determination of refractory periods.

Although we determined activation recovery times as cycle lengths of 300 and 220 ms, the mean VFCL was just over 100 ms in the present study. Thus we cannot exclude the possibility that the determination of refractory periods in faster cycle lengths, if feasible, would have resulted in a better correlation between VFCLs and refractory period. Finally, the study by Zaitsev et al. (31) and their model of VF has shown how a continuous distribution of refractory periods can result in different refractory periods-VFCL relationships and thus leading to lines of functional block.

Potential explanations for the better relationship between global dispersion of refractoriness and VFCL in the present study are less clear. Although there are several theoretical constructs that could potentially explain this observation, the mechanism for this association cannot be determined with certainty from the results of the present study. Further work will be required to determine the mechanism for this association.

Prior studies. VFCL has been considered a surrogate for refractory period duration based on prior studies (20, 21). Opthof et al. (21) induced VF by direct current in a single dog with no structural heart disease and found a very good correlation (R = 0.95, P < 0.0005) between VF interval and ERP at 300 ms recorded from 11 sites. Murakawa et al. (20), using a pair of unipolar electrodes on the right ventricle, found a correlation coefficient (R) of 0.609 between right ventricular ERP at 250 ms and VF interval 8–10 s into VF. A recent study by Choi et al. (3) showed a significant correlation (R = 0.70) between activation intervals during VF 2 min after onset and action potential duration at 300 ms.

The markedly different results between our study and the earlier report by Opthof et al. (21) can be attributed to several factors: namely, the different time epochs analyzed, the use of R versus R2 to report the results, and the limited number of sites analyzed by Opthof et al. (11 epicardial sites from a single dog as opposed to 1,638 sites from 18 dogs in the present study). Opthof et al. studied VF 10 min after its onset, whereas we studied early VF (4–8 and 18–20 s after its onset). These two different times may represent very different types of fibrillation (24, 28). Murakawa et al. (20) also examined VF soon after its onset but studied only two epicardial sites. Similarly, different results have been reported by Choi et al. (3). In that report, a moderate correlation between VFCL and action potential duration was noted. However, the period of VF analyzed was 2 min after onset and an in vitro preparation was used, which could have accounted for the different results. Also, different studies could have analyzed different types of VF, which may account for the diverse results (2). It is important to point out that we too found significant correlations between average VFCL and ART dispersion with R as high as 0.89 that were better than the correlation with action potential duration.

The poor correlation between local VFCL and ERP was also unexpected given the good correlation between ERP and AF cycle length in studies involving AF (13, 14, 19, 22). There are several possible explanations for this finding, including differences in wave front characteristics between the two fibrillatory rhythms, the presence of a larger excitable gap in certain models of VF, the three-dimensional nature of activation during VF, or a difference in rate adaptation of ERP and ART among different regions of ventricular muscle. Mandapati et al. (16), using optical mapping studied AF and localized rotors in the left atrium with a mean core area of 3.8 ± 2.8 mm2. In experiments in isolated rabbit hearts, the core area of the rotors in VF was 5.3 ± 0.7 mm2 in control animals and 13.6 ± 1.7 mm2 during global ischemia (15). Thus in AF, because of the smaller size of the rotors, small reentrant circuits are more likely and any one region may be activated from more directions at a wider range of time intervals and this would increase the probability that a region would activate near its ERP. Conversely, in VF, because of the presence of fewer and larger reentrant circuits, a region with a shorter ERP may less commonly be stimulated by a wave front at a time near its ERP thus limiting the ERP-VFCL relationship (15, 16). Finally, electrophysiological properties were measured at a pacing cycle length of 300 and 220 ms. Differing rate adaptations of local electrophysiological parameters at the rates at which VF occurred (VFCL 100–159 ms) may be responsible for the lack of correlation between ERP and local electrophysiological properties.

Limitations. This study was an in vivo model of canine VF. The measurements in cycle length and electrophysiological properties could have been influenced by intrinsic neuronal reflexes, including changes in autonomic tone. Thus the VFCL could reflect both the intrinsic ERP, rate-dependent changes, and changes in autonomic influences on electrophysiological properties. Electrophysiological measurements were made only on the epicardial surface and did not account for transmural differences in electrophysiological properties that may be important in arrhythmogenesis. Measurements were not made from the noninfarcted region of the left or the right ventricle although the changes in local electrophysiology would probably have been greatest in the region of infarction. We did not do three-dimensional mapping in this study and hence could not confirm the presence of rotors and show reentry or differentiate type 1 from type 2 VF, which could potentially have accounted for the diverse results from other studies (2). Also, as noted above, the electrophysiological properties were determined at a pacing cycle length of 300 ms. If the ERP had been determined during VF, the level of correlation with VFCL may have been better.


   GRANTS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This study was supported in part by National Heart, Lung, and Blood Institute Grant HL-40667, a grant from the Fannie Penikoff Trust, the Feinberg Cardiovascular Research Institute and Department of Medicine, Division of Cardiology, Northwestern University (Chicago, IL).


   FOOTNOTES
 

Address for reprint requests and other correspondence: A. Kadish, Feinberg Pavilion, Suite 8-536, 251 E. Huron, Chicago, IL 60611 (E-mail: a-kadish{at}northwestern.edu).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Biermann M, Shenasa M, Borggrefe M, Hindricks G, Haverkamp W, and Breithardt G. The interpretation of cardiac electrograms. In: Cardiac Mapping, edited by Shenasa M, Borggrefe M, and Breihardt G. New York: Futura, 1993, p. 11–34.
  2. Chen PS, Wu TJ, Ting CT, Karagueuzian HS, Garfinkel A, Lin SF, and Weiss JN. A tale of two fibrillations. Circulation 108: 2298–2303, 2003.[Free Full Text]
  3. Choi BR, Liu T, and Salama G. The distribution of refractory periods influences the dynamics of ventricular fibrillation. Circ Res 88: e49-e58, 2001.
  4. Damle RS, Robinson NS, Ye D, Roth SI, Greene R, Goldberger J, and Kadish AH. Electrical activation during ventricular fibrillation in the subacute and chronic phases of healing canine myocardial infarction. Circulation 92: 535–545, 1995.[Abstract/Free Full Text]
  5. Fishbein MC, Maclean D, and Maroko PR. The histopathological evolution of myocardial infarction. Chest 73: 843–849, 1978.[Free Full Text]
  6. Gray R, Jalife J, Pafilov A, Baxter W, Davidenko J, and Pertsov A. Mechanisms of cardiac fibrillation. Science 270: 1222–1223, 1995.[Abstract/Free Full Text]
  7. Gray R, Pertsov A, Arkady M, and Jalife J. Spatial and temporal organization during cardiac fibrillation. Nature 392: 75–78, 1998.[CrossRef][Medline]
  8. Harris A. Delayed development of ventricular ectopic rhythms following experimental coronary artery occlusion. Circulation 1: 1318–1328, 1950.[ISI][Medline]
  9. Horvath G, Racker D, Goldberger J, Johnson D, Jain S, and Kadish A. Electrophysiologic and anatomic heterogeneity in evolving canine myocardial infarction. Pacing Clin Electrophysiol 23: 1068–1079, 2000.[CrossRef][Medline]
  10. Janse M and Opthof T. Mechanisms of Ischemia-Induced Arrhythmias. Philadelphia, PA: Saunders, 1995, p. 489–496.
  11. Kadish A, Kou WH, Schmaltz S, and Morady F. Effect of atrioventricular relationship on ventricular refractoriness in humans. Am J Physiol Heart Circ Physiol 259: H1463–H1470, 1990.[Abstract/Free Full Text]
  12. Kadish A, Spear J, Levine J, Hanich R, Prood C, and Moore E. Vector mapping of myocardial activation. Circulation 74: 603–615, 1986.[Abstract/Free Full Text]
  13. Kim K, Rodefeld M, Schuessler R, Cox J, and Boineau J. Relationship between local atrial fibrillation interval and refractory period in the isolated canine atrium. Circulation 94: 2961–2967, 1996.[Abstract/Free Full Text]
  14. Lammers W, Allessie M, Rensma P, and Schalij M. The use of fibrillation cycle length to determine spatial dispersion in electrophysiological properties and to characterize the underlying mechanism of fibrillation. New Trend Arrhyth 2: 109–112, 1986.
  15. Mandapati R, Asano Y, Baxter W, Gray R, Davidenko J, and Jalife J. Quantification of the effects of global ischemia on dynamics of ventricular fibrillation in isolated rabbit hearts. Circulation 98: 1688–1696, 1998.[Abstract/Free Full Text]
  16. Mandapati R, Skanes A, Chen J, Berenfeld O, and Jalife J. Stable microreentrant sources as a mechanism of atrial fibrillation in the isolated sheep heart. Circulation 101: 194–199, 2000.[Abstract/Free Full Text]
  17. Millar CK, Kralios FA, and Lux RL. Correlation between refractory periods and activation-recovery intervals from electrograms: effects of rate and adrenergic interventions. Circulation 72: 1372–1379, 1985.[Abstract/Free Full Text]
  18. Morady F, Kadish A, Rosenheck S, Schmaltz S, and Buitleir MD. Effects of the intertrain pause on the ventricular effective refractory period measured by the extrastimulus technique. Pacing Clin Electrophysiol 13: 405–409, 1990.[CrossRef][Medline]
  19. Morillo C, Klein G, Jones D, and Guiraudon C. Chronic rapid atrial pacing: structural, functional, and electrophysiologic characteristics of a new model of sustained atrial fibrillation. Circulation 91: 1588–1595, 1995.[Abstract/Free Full Text]
  20. Murakawa Y, Yamashita T, Kanese Y, Kazunori S, and Omata M. Is ventricular fibrillation interval an indicator of electrical defibrillation threshold? Pacing Clin Electrophysiol 22: 302–306, 1999.[CrossRef][Medline]
  21. Opthof T, Ramdat Misier A, Coronael R, Vermeulen J, Verberne H, Frank R, Moulijn A, van Capelle F, and Janse M. Dispersion of refractoriness in canine ventricular myocardium: effects of sympathetic stimulation. Circ Res 68: 1204–1215, 1991.[Abstract/Free Full Text]
  22. Ramdat Misier A, Opthof T, van Hemel N, Defauw J, de Bakker J, Janse M, and van Capelle F. Increased dispersion of refractoriness in patients with idiopathic paroxysmal atrial fibrillation. J Am Coll Cardiol 19: 1531–1535, 1992.[Abstract]
  23. Rogers J, Huang J, Melnick S, and Ideker R. Sustained reentry in the left ventricle of fibrillating pig hearts. Circ Res 92: 539–545, 2003.[Abstract/Free Full Text]
  24. Rogers JM, Huang J, Smith WM, and Ideker RE. Incidence, evolution, and spatial distribution of functional reentry during ventricular fibrillation in pigs. Circ Res 84: 945–954, 1999.[Abstract/Free Full Text]
  25. Skanes A, Mandapati R, Bo DJ, and Jalife J. Spatiotemporal periodicity during atrial fibrillation in the isolated sheep heart. Circulation 98: 1236–1248, 1998.[Abstract/Free Full Text]
  26. Taneja T, Horvath G, Racker D, Goldberger J, and Kadish A. Excitable gap in canine fibrillating ventricular myocardium: effect of subacute and chronic myocardial infarction. J Cardiovasc Electrophysiol 12: 708–715, 2001.[CrossRef][ISI][Medline]
  27. Valderrabano M, Lee MH, Ohara T, Lai A, Fishbein M, Lin AF, Karagueuzian H, and Chen PS. Dynamics of intramural and transmural reentry during ventricular fibrillation in isolated swine ventricles. Circ Res 88: 839–848, 2001.[Abstract/Free Full Text]
  28. Wiggers C. Studies of ventricular fibrillation caused by electrical shock: II. Cinematographic and electrocardiographic observations of the natural process in the dogs heart. Its intitution by potassium and the revival of coordinated beats by calcium. Am Heart J 5: 351–365, 1930.[CrossRef][ISI]
  29. Winfree A. Electrical turbulence in three-dimensionless heart muscle. Science 266: 1003–1006, 1994.[Abstract/Free Full Text]
  30. Witkowski F, Leon L, Penkoske P, Giles W, Spano M, Ditto W, and Winfree A. Spatiotemporal evolution of ventricular fibrillation. Nature 392: 78–81, 1998.[CrossRef][Medline]
  31. Zaitsev AV, Berenfeld O, Mironov SF, Jalife J, and Pertsov AM. Distribution of excitation frequencies on the epicardial and endocardial surfaces of fibrillating ventricular wall of the sheep heart. Circ Res 86: 408–417, 2000.[Abstract/Free Full Text]
  32. Zipes D and Wellens H. Sudden cardiac death. Circulation 98: 2334–2351, 1998.[Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Taneja, T.
Right arrow Articles by Kadish, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Taneja, T.
Right arrow Articles by Kadish, A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2004 by the American Physiological Society.