Author + information
- Received December 27, 2000
- Revision received May 25, 2001
- Accepted August 10, 2001
- Published online November 15, 2001.
- Tsu-Juey Wu, MD*,*,† (, )
- Young-Hoon Kim, MD, FACC†,
- Masaaki Yashima, MD†,
- Charles A Athill, MD†,
- Chih-Tai Ting, MD, PhD*,
- Hrayr S Karagueuzian, PhD, FACC† and
- Peng-Sheng Chen, MD, FACC†
- ↵*Reprint requests and correspondence: Dr. Tsu-Juey Wu, Division of Cardiology, Department of Medicine, Taichung Veterans General Hospital, 160, Section 3, Chung-Kang Road, Taichung, Taiwan
We sought to evaluate the effects of progressive shortening of the action potential duration (APD) on atrial wave front stability.
The mechanisms of conversion from atrial flutter to atrial fibrillation (AF) are unclear.
Isolated canine right atria were perfused with 1 to 5 μmol/l of acetylcholine (ACh). We mapped the endocardium by using 477 bipolar electrodes and simultaneously recorded transmembrane potentials from the epicardium. The APD90was measured during regular pacing (S1) with cycle lengths of 300 ms. Atrial arrhythmia was induced by a premature stimulus (S2).
At baseline, only short runs of repetitive beats (<10 cycles) were induced. After shortening the APD90from 124 ± 15 ms to 72 ± 9 ms (p < 0.01) with 1 to 2.5 μmol/l of ACh, S2pacing induced single, stable and stationary re-entrant wave fronts (307 ± 277 cycles). They either anchored to pectinate muscles (5 tissues) or used pectinate muscles as part of the re-entry (4 tissues). When ACh was raised to 2.5 to 5 μmol/l, the APD90was further shortened to 40 ± 12 ms (p < 0.01); S2pacing induced in vitro AF by two different mechanisms. In most episodes (n = 13), AF was characterized by rapid, nonstationary re-entry and multiple wave breaks. In three episodes with APD90<30 ms, AF was characterized by rapid, multiple, asynchronous, but stationary wave fronts.
Progressive APD shortening modulates atrial wave front stability and converts atrial flutter to AF by two mechanisms: 1) detachment of stationary re-entry from the pectinate muscle and the generation of multiple wave breaks; and 2) formation of multiple, isolated, stationary wave fronts with different activation cycle lengths.
Acetylcholine (ACh) shortens the atrial effective refractory period and induces re-entrant wave fronts in the atrium (1–6). Furthermore, once these re-entrant wave fronts attach to anatomic obstacles (4)or anchor to anatomic structures (such as pectinate muscle ridges) (5), the rhythm shows “flutter-like” activity. In contrast, detachment of these re-entrant wave fronts due to either spontaneous separation or outside interference leads to “fibrillation-like” activity (4,5). However, the mechanisms by which re-entrant wave fronts detach from anatomic structures remain unclear. In the present study, we applied computerized mapping techniques to study the effects of progressive shortening of the action potential duration (APD) on the wave front stability in atrial tissues. Different concentrations of ACh (1, 2.5 and 5 μmol/l) were used in isolated, perfused canine right atria. Atrial tachyarrhythmias were induced by premature stimuli. The results were used to test the following hypothesis: progressive APD shortening may lead to the detachment of re-entrant wave fronts from anatomic structures and the continuous generation of wave breaks, converting the rhythm from flutter-like to fibrillation-like activity.
The method used for tissue preparation has been described previously (5,6). Briefly, 12 mongrel dogs (weight 17 to 27 kg) were anesthetized with intravenous sodium pentobarbital (30 to 35 mg/kg), intubated and ventilated with room air by a respirator. The chest was opened through a median sternotomy, and the heart was rapidly removed. The right coronary artery (RCA) was immediately cannulated and perfused at 10 ml/min with oxygenated and warmed (36.5°C) Tyrode’s solution with a pH of 7.4 (1). The ionic composition of Tyrode’s solution has been described previously (7). The right atrial appendage and free wall were then excised along the proximal portion of the RCA. The distal portion of the RCA was ligated, and branches to the residual right ventricular tissue were cauterized. The tissue was then placed in a tissue bath and mounted on the mapping plaque with the endocardial surface down. This recording plaque (3.2 × 3.8 cm in size) was connected to a computerized mapping system (7). Data were acquired from 477 bipolar electrodes on the plaque.
A bipolar electrode with an interpolar distance of 0.5 mm was used to record bipolar electrograms from the epicardium to document the tissue response to pacing and premature stimuli. A pseudo-electrocardiogram (ECG) was also registered with widely spaced bipoles, one at each end of the tissue preparation. The data were acquired by Axon TL-1-40 A/D acquisition hardware and Axoclamp-2A software (Axon Instrument, Inc., Foster City, California) and were digitized at 1 kHz with 12 bits of accuracy (5,6).
Transmembrane potential recordings
As described previously (4,6), transmembrane potentials (TMPs) were recorded from the epicardial surface with conventional machine-pulled, glass capillary electrodes filled with 3 mol/l of potassium chloride, with a tip resistance of ∼20 MΩ. The electrodes were coupled by silver-silver chloride wire in a right-angle micropipette holder leading to a high-input impedance and variable-capacitance neutralization amplifier (IE-251, Warner Instrument Corp., Hamden, Connecticut) (4,6). The sites of TMP recordings were always located at the central regions of these tissues. The data were acquired by the same system for pseudo-ECG recordings, as mentioned previously.
As shown in Figure 1, a bipolar stimulating electrode was placed at either the left edge or the bottom of the epicardial surface to deliver baseline pacing (S1) with twice the diastolic threshold current at cycle lengths of 300 ms. Another pair of epicardial stimulation electrodes were placed 1.5 cm away from the S1site to give premature stimulation (S2) to induce re-entry (5,6). The initial strength of S2was 5 mA. If repetitive activations were not induced, the strength of S2was increased in 5-mA increments until the induction of re-entry or until 20 mA was reached. If the arrhythmia was not induced at baseline, 1, 2.5 or 5 μmol/l of ACh was added to the perfusate, and the same induction protocol was repeated. In the last three tissues, three concentrations (from 1 to 2.5 to 5 μmmol/l) of ACh were perfused sequentially. Atrial arrhythmia was induced in each concentration. Once arrhythmia was induced, endocardial mapping was performed.
The method for selecting the time of activation has been reported in detail previously (8). Once the times of activation in each electrode were determined, they were displayed dynamically on the computer (3,8). Briefly, when an electrode site became activated, that site first became illuminated as red, then yellow, green, light blue and, finally, dark blue before fading away. The persistence of each color was for 10 ms. The patterns of activation were then studied. The activation times were also used to construct conventional isochronal activation maps.
The APD90was measured at baseline (no ACh) and with 1 to 5 μmol/l of ACh during S1pacing. With or without ACh (as demonstrated in Fig. 1A), the conduction velocity (CV) of these atrial tissues was not uniform during S1pacing. However, uniform propagation was always present along a central line in the mapped tissue, starting from the S1site. The propagation velocity along this central line was used to represent the CV during S1pacing. The wavelength during S1pacing was obtained by using the following formula:
A re-entrant wave frontwas defined as a wave front that propagated around a central area (core) and reentered the site of origin. The location of the corewas identified by dynamic display as the area encircled by the path of the tip of the re-entrant wave front (3,8). The tipof the re-entry was defined as the innermost edge of the re-entrant wave front (3). If the tip was found to deviate grossly from its initial wave tip path (i.e., by more than one interelectrode distance [1.6 mm] for roughly >75% of its path), this re-entrant wave front was considered as nonstationary. Otherwise, it was considered as stationary(5).
Histologic examination and anatomic correlation
At the conclusion of each study, the preparation was photographed before removal from the tissue bath. The tissue was then fixed in 10% neutral buffered formalin and processed routinely. The areas of slow conduction, conduction block and the core of the re-entrant wave front were correlated with anatomic macroscopic and histologic findings. Cross sections were performed from the epicardium to endocardium and were stained with hematoxylin-eosin.
The results are expressed as the mean value ± SD. Bonferroni ttests were used to determine whether there were differences in the APD90during S1pacing with cycle lengths of 300 ms, when short runs of repetitive beats (no ACh) and flutter-like (1 to 2.5 μmol/l of ACh) and fibrillation-like (2.5 to 5 μmol/l of ACh) activities occurred. Bonferroni-adjusted p values were used to determine the significance. Paired Wilcoxon signed rank tests were used to compare CV, wavelength and TMP characteristics (such as APD90and action potential amplitude) between baseline and ACh infusion. A p value ≤0.05 was considered significant.
Activation pattern and APD at baseline
Figure 1Ashows an isochronal activation map during S1pacing in a representative isolated canine right atrial tissue (Fig. 1B). During S1pacing at cycle lengths of 300 ms, no area of conduction block could be detected in this or any of the other atrial tissues studied. However, the CV was not uniform (Fig. 1A). The nonuniformity of the CV was caused by the presence of gross endocardial structures, such as the crista terminalis and pectinate muscle bundles (Fig. 1B). Before ACh administration, the mean APD90value recorded at the central regions of these tissues during S1pacing at cycle lengths of 300 ms was 124 ± 15 ms (n = 12, one from each tissue; range 105 to 144 ms). At baseline (no ACh), only short runs of repetitive beats (<10 beats) could be induced in each tissue.
Activation patterns and APDs at different concentrations of ACh
In the initial nine tissues, different ACh concentrations were used (1 μmol/l in three, 2.5 μmol/l in seven and 5 μmol/l in one; two of them perfused sequentially with two different concentrations). In the remaining three tissues, three concentrations (from 1 to 2.5 to 5 μmol/l) of ACh were perfused sequentially.
After shortening the APD90to 72 ± 9 ms (n = 10, data from nine tissues; range 56 to 84 ms; p < 0.01) during S1pacing at cycle lengths of 300 ms with 1 to 2.5 μmol/l of ACh, S2pacing induced single, stable and stationary re-entrant wave fronts. They either anchored to large pectinate muscles (five tissues) or used pectinate muscle bundles as part of the re-entrant circuit (four tissues). The mean cycle length of these re-entrant wave fronts (25 episodes) was 127 ± 22 ms, and the mean life span was 307 ± 277 cycles. Figure 2shows an example of stationary re-entry with a life span of 27 rotations. Verified anatomically, this re-entry rotated around the insertion site of a large pectinate muscle to the underlying atrial tissue (“bridge-like” structure, as shown in Fig. 1B).
When ACh was raised to 2.5 to 5 μmol/l, the APD90was further shortened to 40 ± 12 ms (n = 10, data from seven tissues; range 22 to 56 ms; p < 0.01) during S1pacing at cycle lengths of 300 ms; S2pacing consistently induced multiple wave fronts (16 episodes; maximal wave fronts: 4.5 ± 1.1, range 3 to 7) lasting 167 ± 135 s. In most episodes (n = 13), these wave fronts included rapid, nonstationary re-entrant wave fronts with moving cores and “daughter” wave fronts arising from continuous wave breaks (Figs. 3 and 4). ⇓They converted the flutter-like to fibrillation-like activity on the pseudo-ECG (Figs. 2H, 3I and 4J). Note that, as demonstrated in Figures 3G and 3H, adaughter wave front (marked by an asterisk in Fig. 3G) spontaneously detached from the anchoring site and propagated outside the mapped area. In the remaining three episodes, when the APD90during S1pacing was <30 ms (22, 24 and 27 ms, respectively), rapid, multiple, but stationary wave fronts were induced (Fig. 5A to 5D). No obvious wave break was observed. The activation cycle lengths were different among these stationary wave fronts (Fig. 5E), and the pseudo-ECG showed fibrillation-like activity (Fig. 5F). These findings indicate that a different form of in vitro fibrillation was present during high concentrations of ACh, with an extremely short APD.
In three tissues, ACh washout converted the fibrillation-like pattern with multiple wave fronts to flutter-like activity sustained by single re-entrant wave fronts. All of them anchored to large pectinate muscles.
Effects of ACh on CV, wavelength and TMP characteristics during pacing
In 8 of 12 tissues, we compared CV, wavelength and TMP characteristics, including APD90and action potential amplitude between baseline and ACh infusion (2.5 μmol/l in four and 5 μmol/l in four) when S1pacing at cycle lengths of 300 ms was performed.
Acetylcholine significantly shortened the APD90(124 ± 17 ms vs. 42 ± 18 ms; p < 0.01) during S1pacing. However, as shown in Figure 6A and 6B, ACh had no effect on CV (85 ± 20 cm/s vs. 85 ± 21 cm/s; p = 0.47) during S1pacing, consistent with previous reports (3,9). Therefore, ACh significantly shortened the wavelength (11 ± 2.8 cm vs. 3.5 ±1.5 cm; p < 0.01) during S1pacing. Acetylcholine had no significant effect on the action potential amplitude (77 ± 10 mV vs. 81 ± 11 mV; p = 0.38) during S1pacing.
Characteristics of TMP during fibrillation-like activity
Figure 6C to 6Fshows an example of a series of TMP recordings (from the same tissue) during different concentrations of ACh. When nonstationary re-entry and continuous wave breaks occurred (as shown in Figs. 3 and 4), the simultaneous TMP recordings always showed variable APD90values and small amplitudes (marked by the vertical arrows in Fig. 6E). However, once multiple, but stationary wave fronts appeared (as shown in Fig. 5), the APD90and amplitude became relatively constant (Fig. 6F).
In this study, we found that progressive APD shortening modulates atrial wave front stability and converts atrial flutter to atrial fibrillation (AF) by two mechanisms: 1) detachment of stationary re-entry from the pectinate muscle and generation of multiple wave breaks; and 2) formation of multiple, isolated, stationary wave fronts with different activation cycle lengths.
When the mean APD90during regular pacing at cycle lengths of 300 ms was 72 ms, pectinate muscles may either serve as part of a re-entrant circuit (bridge) or provide a natural anchor (ridge) to the re-entrant wave fronts. However, once the mean APD90was further shortened to 40 ms, detachment of re-entrant wave fronts from these structures and continuous generation of wave breaks occurred in most episodes, thus converting flutter-like to fibrillation-like activity. In three episodes, however, multiple, stationary wave fronts were observed. Co-existence of multiple, stationary wave fronts were also associated with AF.
Effects of progressive APD shortening
As reported previously (4,5), anatomic obstacles and structures in the atrium are important for the formation of intra-atrial re-entry. Because ACh is required for the induction of re-entrant wave fronts, the functional characteristics of the tissue (primarily shortening of refractoriness and APD) are also important for re-entry formation (10). However, as demonstrated in this study, extreme shortening of APD90(40 ms) not only increases the atrial vulnerability and number of wave fronts during atrial tachyarrhythmias (1,2), but also facilitates the detachment of re-entrant wave fronts and promotes the continuous generation of wave breaks, leading to the wave front instability. Both effects (detachment and wave break) are important consequences of progressive APD shortening.
Continuous wave breaks in atrial tissues
The safety factor of impulse propagation in cardiac tissue depends on the relationship between the source (amount of current available upstream) and sink (the structure that determines the current density downstream) (11–14). In atrial tissues, there is a source-sink mismatch caused by uneven thickness when the impulse propagates from the atrial free wall into the complicated pectinate muscle structures. In this study, the data show that ACh significantly shortened both the APD90and wavelength during S1pacing (small source). Furthermore, TMP recordings with variable APD90values and small amplitudes were frequently observed during ACh-induced AF (variable sources). These findings suggest that ACh might enhance the preexisting source-sink mismatch in atrial tissues, thus facilitating the generation of wave breaks (Figs. 3 and 4).
Detachment of re-entrant wave fronts from anatomic structures
We have previously reported that at low concentrations of ACh, stationary re-entry sometimes detached from the pectinate muscle due to either outside interference or spontaneous separation, owing to cycle length oscillation (5). The separation occurred when the re-entrant wave front was making an abrupt turn at the end of the pectinate muscle ridge and when the cycle length oscillated to a short cycle (5). These findings support the notion that a small action potential arising from premature activation may not have the sufficient source/sink ratio required to complete the abrupt turn, leading to spontaneous separation (4,5). Similarly, in the canine atrial tissues with fixed sizes of pectinate muscles, progressive APD shortening and continuous wave breaks, with the formation of small daughter wave fronts (small source) induced by increasing ACh concentrations, might create an extremely low source/sink ratio, resulting in the detachment (Fig. 3).
A different form of in vitro fibrillation during high concentrations of ACh
Schuessler et al. (2)previously reported that high concentrations of ACh can facilitate the induction of a small, single, rapid and relatively stable re-entrant circuit in isolated canine atria. In the present study, we found that high-dose ACh associated with an extremely short APD (<30 ms) can induce multiple, stationary wave fronts in a similar fashion. Although the activation patterns (Figs. 5A to 5D)did not show wandering wavelets or frequent generation of new wave breaks, the pseudo-ECG nevertheless showed fibrillation-like activity (Fig. 5F).
The mere presence of multiple stationary wave fronts does not necessarily produce fibrillation. Gray et al. (15)reported that multiple re-entrant wave fronts rotating at the same cycle length resulted in a periodic and regular rhythm, not fibrillation. In contrast, a computer simulation study by Xie et al. (16)demonstrated that co-existence of multiple spiral waves with different periods might result in fibrillation-like activity. The latter study showed that different periods were possible only if barriers separated these wave fronts. Skanes et al. (17)also reported that in an animal model of AF, two re-entrant circuits with different periods might occur simultaneously and separately in the left and right atria. In this study with isolated canine right atria, pectinate muscles could serve as barriers. Wave fronts separated from each other by pectinate muscles can activate at different rates (Fig. 5E), creating fibrillation-like activity.
A limitation of this study was that the CV during S2stimulation was not routinely obtained. Furthermore, it was difficult to evaluate the CV during AF. Therefore, the wavelength (source) during S2stimulation and AF could not be determined accurately. Also, the possible effect of ACh on the sink (i.e., change of the critical curvature for impulse propagation induced by ACh) remains unclear and deserves further investigation.
The authors thank Hsun-Lun A. Huang, Avile McCullen and Meiling Yuan for their technical assistance, as well as Elaine Lebowitz for her secretarial assistance.
☆ This study was performed in part during the tenure of a National Institutes of Health (NIH, Bethesda, Maryland) Fellowship Grant to Dr. Athill, a Fellowship Grant from the Department of Medicine of Korea University to Dr. Kim, a Cedars-Sinai ECHO Foundation Award to Dr. Karagueuzian and a Pauline and Harold Price Endowment to Dr. Chen, and it was supported in part by NIH SCOR Grant in Sudden Death no. P50-HL52319, NIH Grant R01-HL66389 (National Heart, Lung and Blood Institute), University of California Tobacco-Related Disease Research Program 9RT-0041, American Heart Association (National Center) Grants-in-Aid 9750623N and 9950464N, the Ralph M. Parsons Foundation (Los Angeles, California) and the Yen Tjing Ling Medical Foundation (CI-89-7-3, Taipei, Taiwan).
- atrial fibrillation
- action potential duration
- conduction velocity
- right coronary artery
- transmembrane potential
- Received December 27, 2000.
- Revision received May 25, 2001.
- Accepted August 10, 2001.
- American College of Cardiology
- Schuessler R.B,
- Rosenshtraukh L.V,
- Boineau J.P,
- Bromberg B.I,
- Cox J.L
- Schuessler R.B,
- Grayson T.M,
- Bromberg B.I,
- Cox J.L,
- Boineau J.P
- Ikeda T,
- Wu T.-J,
- Uchida T,
- et al.
- Ikeda T,
- Yashima M,
- Uchida T,
- et al.
- Wu T.-J,
- Yashima M,
- Xie F,
- et al.
- Athill C.A,
- Ikeda T,
- Kim Y.-H,
- et al.
- Ikeda T,
- Uchida T,
- Hough D,
- et al.
- Lee J.J,
- Kamjoo K,
- Hough D,
- et al.
- Rensma P.L,
- Allessie M.A,
- Lammers W.J.E.P,
- et al.
- Moe G.K,
- Rheinboldt W.L,
- Abildskov J.A
- Mendez C,
- Mueller W.J,
- Urguiaga X
- Spach M.S,
- Miller W.T,
- Geselowitz D.B,
- Barr R.C,
- Kootsey J.M,
- Johnson E.A
- Cabo C,
- Pertsov A.M,
- Baxter W.T,
- Davidenko J.M,
- Gray R.A,
- Jalife J
- Gray R.A,
- Jalife J,
- Panfilov A.V,
- et al.
- Xie F,
- Qu Z,
- Weiss J.N,
- Garfinkel A
- Skanes A.C,
- Mandapati R,
- Berenfeld O,
- et al.