Mechanisms of Ventricular Arrhythmias Induced by Myocardial Contusion: A High-resolution Mapping Study in Left Ventricular Rabbit Heart

The principles for the care and treatment of experimental animals complied with the national guidelines of the French Ministry of Agriculture. Forty-six New Zealand rabbits weighing a mean of 3.6 ± 0.6 kg (± SD) were anesthetized with ketamine (50 mg/kg intramuscular), the trachea was intubated, and the lungs were mechanically ventilated with 100% O²(Logic 07, ATM, Maurepas, France). The surgical excision of the heart was performed as previously described. 12,13The heart was connected to a Langendorff perfusion system using Tyrode’s solution at 37°C. The coronary arteries were perfused with a constant flow of 40–50 ml/min (Watson-Marlow 101U pump, Falmouth Cornwall, UK), resulting in a perfusion pressure of 70 ± 10 mmHg (Gould P23 Transducer, Oxnard, CA; CGR monitor, St-Cloud, France). The composition of Tyrode’s solution was NaCl 130 mM, NaHCO³20.1 mM, KCl 4.0 mM, CaCl²2.2 mM, MgCl²0.6 mM, NaH²PO⁴1.2 mM, and glucose 12 mM. The solution was saturated with a mixture of 95% O²and 5% CO², and pH was adjusted to 7.40 ± 0.02.

Forty hearts were submitted to a localized injury. The heart was placed horizontally, and a metal stem was introduced into the left ventricle through the left atrium to hold the heart during contusion. Myocardial impact was produced by a ball decreasing from a predefined height within a hollowed cylinder that was placed just above the left ventricle and inflicted a precise blow nearly on the center of the left ventricle. The height was determined to deliver four different contusion kinetic energies: 75, 100, 150, and 200 millijoules (mJ). The ball weight was 12 g for contusions of 75, 100, and 150 mJ and 21.7 g for a contusion of 200 mJ.

High-resolution mapping of epicardial excitation was performed by applying to the epicardial surface a spoon-shaped electrode containing 256 unipolar electrodes at a regular distance of 2.25 mm (fig. 1). A computerized mapping system allowed simultaneous recording, storage, and automatic analysis of all 256 electrograms and on-line presentation of color-coded activation maps (Maptech System; Maastricht, The Netherlands). 14 

Thirty hearts were used to study arrhythmogenic effects of myocardial contusion (intact hearts), and 16 hearts were submitted to an endocardial cryoprocedure to analyze the initiation mechanisms of arrhythmia (frozen hearts).

The 30 hearts in the intact heart protocol were divided in five groups of six hearts each. In one group, no contusion energy was applied, but these six hearts were placed on the contusion apparatus to serve as a control group. In the other four groups, a contusion with kinetic energy of 75, 100, 150 or 200 mJ was produced. Variables were recorded before contusion (T−1), immediately after contusion (T0), and 5 min (T5), 15 min (T15), 30 min (T30), 45 min (T45), and 60 min (T60) after contusion. At the end of each experiment, the heart was removed from the perfusion system and fixed in formaldehyde 10% for histopathologic study.

The number, nature, and time of occurrence of arrhythmias were studied. Moreover, we noticed in preliminary experiments that the impact might induce an electrically silent area or a line of fixed block. The electrically silent area was defined as an area with no significant electrical activity and where no pacing with twice the threshold intensity was possible. The occurrence of electrically silent area, the time for its disappearance, and line of fixed block occurrence were also reported.

During the experiment, effluent samples of about 3 ml were collected in tubes for creatine kinase (CK) measurement and stored at −20°C until biochemical assays were performed. The concentration of CK was measured using the immunoinhibition method at 25°C by means of the N -acetylcysteine activated test on a Cobas Bio (Roche, Diagnostic GmbH, Mannheim, Germany). 15CK concentrations were expressed in International Units per liter (IU/l). The comparison of CK release between control and contusion groups was performed using the difference value between the concentrations just after the contusion (T0) and before the contusion (T−1) for each group.

Hearts were dehydrated using graded ethanol, embedded in paraffin (Histocentre, Shandon, Runcorn, England), and cut into 5-μm-thick rings using a microtome (Reichert–Yung, Nusshloch, Germany). Horizontal sections were mounted on glass slides with albumin (0.1% m/vol, BioLyon, Lyon, France). Three pieces of each heart were stained with hematein–eosin–safran for optical microscopic examination. These three pieces were removed from the bottom, center and tip of the impact area, as perceived macroscopically. The depth and extent of tissue disruption and tissue disorganization were noted for each specimen. The pathologist was blinded according to the control group and the different kinetic energy levels used.

In 16 hearts, a previously described endocardial cryoprocedure 12,16,17was used to avoid epicardial breakthrough of longitudinal wavefronts from deeper layers and to allow complete mapping of electrical activation in two dimensions. 12,13As a result of this procedure, only a thin (≈1-mm-thick) epicardial layer of the left ventricle free wall survived; the rest of myocardium was completely destroyed. 16It was previously demonstrated that in this thin surviving layer, refractoriness and conduction velocities were not affected by the procedure and remained stable for many hours, suggesting that the circulatory condition was adequate. 12,16,17At the end of the experiment, hearts were dissected to verify the efficacy of the cryoprocedure. When the freezing was not adequate, the heart was excluded from the study.

Programmed electrical stimulation was performed using a programmable constant current stimulator, which delivered square pulses of 2 ms duration at twice the diastolic threshold for both regular stimulation and induction of premature beats (Maptech System). Induction of ventricular arrhythmia was performed using a bipolar stimulation protocol in all frozen hearts and consisted of application of one, two, and three premature stimuli (S2, S3, and S4, respectively) delivered with decreasing coupling intervals after 10 basic stimuli (S1–S1) at 300-ms intervals, and application of a train of 10 stimuli at a regular cycle length that was progressively decreased in increments of 10-ms until one-to-one capture of the ventricle failed (Fmax).

Forty-five min after freezing, two groups of hearts were administered a contusion with kinetic energy of 100 mJ (n = 8) or 200 mJ (n = 8). The contusion procedure was similar to that in the intact heart protocol. Parameters were determined before contusion (T−1), immediately after contusion (T0), and 20 min (T20), 40 min (T40), and 60 min (T60) after contusion. The following electrophysiologic parameters were studied: Ventricular effective refractory period (VERP, in milliseconds) at five sites, longitudinal ventricular conduction velocity (θL, in meters/second), transverse ventricular conduction velocity (θT, in meters/second), anisotropic ratio (θL/θT), longitudinal and transverse wavelength (λL and λT, respectively, in millimeters), index of VERP dispersion (DI) and maximum VERP dispersion (Dmax, in milliseconds).

Conduction velocity was defined as the distance traveled by the wavefront per time unit. In each experiment, both longitudinal and transverse conduction velocities and anisotropic ratio were measured after 10 basic stimuli (S1–S1) at 300-ms intervals. VERP was defined as the shortest S1–S2 interval still resulting in a propagated premature impulse during regular pacing with a S1–S1 interval of 300 ms. VERP was determined by decreasing the coupling interval of premature stimulus in 1-ms increments. Measurement of VERP was performed at the same five sites throughout the experiment (fig. 1). Wavelength was defined as the product of longitudinal (λL) or transverse (λT) ventricular conduction velocity, and the VERP was measured at the center site. VERP dispersion was quantified by DI, defined as the quotient of SD and mean of VERPs for each heart, and Dmax, calculated by the difference between the maximum and the minimum values of VERP for each heart.

We defined ventricular arrhythmia as VF with sustained or nonsustained ventricular tachycardia (SVT and NSVT, respectively). NSVT was defined as a ventricular tachycardia (VT) lasting more than three successive beats but less than 30 s before spontaneous termination. SVT was defined as a VT that lasted more than 30 s. Finally, a separation into monomorphic and polymorphic ventricular tachycardia (MVT and PVT, respectively) was made. The term MVT implies a uniform beat-to-beat QRS morphology. The term PVT was defined as the occurrence of continuous change in QRS configuration. When several types of arrhythmia occurred during a given experimental step, the worst arrhythmia was used.

Except for the difference value between CK at T0 and CK at T−1, which was expressed in median and percentiles (5% and 95%), all parameters were expressed as mean ± SD. The Kruskall–Wallis test and Bonferroni correction were used for the comparison of the CK release difference between intact heart groups. To show an arrhythmogenic and gradual effect of different contusion kinetic energies, Fisher exact test followed by a test of linear trend were performed. Two-way analysis of variance for repeated measures, followed by a contrast method, Neumann-Keuls test, and Bonferroni correction were used to list time or group effect on other parameters. P < 0.05 was considered to be statistically significant.

In the control group, no arrhythmias, no electrically silent areas, and no lines of fixed block were observed. All contusion groups showed arrhythmias (VF, VT, premature ventricular contraction), electrically silent areas, lines of fixed block, or a combination of these (table 1). In 13 of 24 hearts, electrically silent areas occurred after contusion: Four lasted more than 60 min, four others completely disappeared, and five were replaced by a line of fixed block. Lines of fixed block occurred in 8 of 24 hearts, either immediately (n = 3) or after occurrence of an electrically silent area (n = 5). All lines of fixed block lasted after 60 min. Numbers of arrhythmias (P = 0.03) and electrically silent areas (P = 0.02) increased with increasing kinetic energy of contusion.

The release of CK for each group is reported in figure 2. No change occurred in the control group. In contusion groups, concentration of CK increased transiently, with a maximum at T0. The median value (T0–T−1; 5% and 95%) was 14 IU/l (1, 80), 10 IU/l (5, 129), 87 IU/l (7.4, 129), and 103.75 IU/l (21.4, 245) for 75, 100, 150, and 200 mJ, respectively. All differences (T0–T−1) between control and contusion groups were statistically significant (P = 0.02); however, the differences between different contusion groups were not significant.

The optical microscopic study of pieces of heart showed that all hearts that underwent contusion had cellular damage. In the impact area, myocardial fibers were disrupted on the surface. Vacuoles and nucleus pyknosis were observed in fiber cytoplasm. In the impact area, the fiber diameter was reduced, fibers were made round, and the extracellular medium was enlarged. The damage depth was significantly increased according to the kinetic energy impact: 45 ± 81, 325 ± 157, 517 ± 147, 600 ± 77, and 767 ± 225 μm for control, 75, 100, 150, and 200 mJ, respectively (P = 0.009).

Two of eight hearts were excluded for inadequate freezing in the 100-mJ group. Data are therefore given for six hearts in this group.

Ventricular conduction velocities were not significantly modified by contusion whatever the kinetic energy (fig. 3A). However, longitudinal conduction velocity was slightly slowed just after contusion (P = 0.09). Anisotropic ratio was not significantly modified; it changed from 2.01 ± 0.20 at T−1 to 1.87 ± 0.16 at T0 in the 100-mJ group and from 2.09 ± 0.24 at T−1 to 1.98 ± 0.43 at T0 for the 200-mJ group (P = 0.368). In contrast, VERPs were immediately shortened after contusion (P = 0.0004) and progressively lengthened in time (P = 0.004) over the two groups (fig. 3B). Longitudinal and transverse wavelengths for each group were shortened immediately after contusion (P = 0.026 for λL, and P = 0.012 for λT;fig. 3C). Both λL and λT recovered their baseline values at T20. Parameters evaluating the dispersion of VERPs were modified in the same manner in the two groups of hearts. DI and Dmax were increased at T0 (P = 0.0008 and P = 0.0052, respectively) and returned progressively to their baseline values at T60 (figs. 4A and 4B).

Contusion-induced arrhythmias are reported in table 2. Both contusions with kinetic energy of 100 mJ and 200 mJ caused arrhythmias, electrically silent areas, or lines of fixed block. These arrhythmias were mainly monomorphic, and they were induced for the majority within the first 15 min after contusion. Analysis of the arrhythmias showed that all but one were based on a reentrant mechanism. Only one nonsustained arrhythmia in the 200-mJ group was the result of ectopic activity. This initiation mechanism was illustrated in four maps (fig. 5) . Map 5A represents the epicardial activation before contusion. The pacing site was at the center, and the impulse propagated in two directions (longitudinal and transverse; θL = 0.75 m/s, θT = 0.38 m/s, θL/θT = 1.97) in an ellipsoidal manner. The next map (5B) was the epicardial depolarization after contusion and during the regular pacing cycle at 300 ms. An electrically silent area was observed at the center of the map. Thus, the impulse had to travel around it to depolarize epicardium behind the electrically silent area. Map 5C was obtained with an extrastimulus pacing (S2 = 153 ms); the propagation of electrical impulse occurred as in the previous mapping. The wavefront moved around the unexcitable area using two directions on each border of electrically silent area. The collision point of the two waves was 115 ms. On map 5D, spontaneous activity was illustrated. The ectopic site was located at the border of contusion impact area, and the propagation occurred surrounding the electrically silent area in the two directions.

All other VTs were based on a reentrant mechanism, but two different fixed obstacles were observed. In the two groups, two reentrant tachycardias were developed around the electrically silent area induced by the ball impact. One other VT in each group was due to a reentrant circuit around a line of fixed block that replaced the electrically silent area. An example of each reentrant mechanism is shown in figures 6 and 7. Figure 6illustrates a sustained monomorphic ventricular tachycardia around an electrically silent area that was induced by a contusion of 100 mJ with a programmed pacing of three extrastimuli 13 min after contusion. However, analysis showed that this tachycardia was initiated after the second extrastimulus (S3). Map 6A shows an epicardial depolarization after contusion with regular pacing (S1–S1 = 300 ms). At the map center is the electrically silent area that was produced by the impact. The impulse had to move around this region to depolarize other epicardial areas. Slowed conduction was observed close to the left anterior descending coronary artery; thus, the collision point of clockwise and counterclockwise waves was 54 ms, and this point was located above the electrically silent area. The activation of the epicardium after the second extrastimulus (S3) is represented on map 6B. The propagation occurred only in the counterclockwise direction around the electrically silent area because of the appearance of a functional conduction block in the left anterior descending coronary artery corridor close to the pacing site. Thus, the epicardium previously activated by pacing (34 ms) could be depolarized again at 127 ms. Therefore, the reentrant circuit was initiated and propagated in an counterclockwise movement around the electrically silent area (6C). Panel 6D shows electrograms at some sites around the contusion area, and these allow observation of the advancement of the tachycardia.

The second reentrant mechanism is illustrated in figure 7. Rapid pacing induced a sustained monomorphic ventricular tachycardia (Fmax, S1–S1 = 145 ms, but the tachycardia was induced before the last pacing [S1]) 16 min after a contusion impact with a kinetic energy of 100 mJ. Map 7A was the impulse propagation during the regular pacing cycle (S1–S1 = 300 ms) immediately after contusion. On this mapping, an electrically silent area was observed; therefore, the impulse had to encircle this area to depolarize the epicardium. Thereafter (16 min after contusion), the electrically silent area disappeared and was replaced by a wide line of block (i.e. , line of fixed block + functional conduction block that is located at the low part of block; 7B). The response of the epicardium to the last pacing is shown on map 7B. The wide line of conduction block delimited a zone near the left anterior descending coronary artery that was activated later. Thus, the area in front of the block, which was blocked at 28 ms, could be depolarized again at 144 ms by the influx coming from the area behind. The first VT beat was initiated in an counterclockwise movement around a fixed conduction block and propagated from apex to base (7C). The second VT beat was illustrated in panel 7D.

The current study shows that direct impact to the left ventricle can induce arrhythmias. The contusion-induced proarrhythmogenic effect increases with increasing impact energy. Because we used isolated and perfused hearts, the occurrence of ventricular arrhythmia was independent of hemodynamic or metabolic alterations induced by blunt trauma. The frozen heart study shows that the main mechanism of arrhythmia induced by contusion is based on reentry. Reentry is based on two mechanisms induced by the impact: A fixed block (electrically silent area or line of block) and a unidirectional conduction block facilitated by the shortening of wavelength and the refractoriness dispersion.

Our model differs from those previously employed in other experimental investigations of myocardial contusion. First, because of our mapping methods, we studied the electrophysiologic effect of impact solely on the left ventricle and not on the right ventricle. Second, we used the impact of a ball descending directly on the heart and not the impact of a ball on the chest with a defined velocity. 18,19A ballistic study showed that the impact of a ball with a velocity of 13 m/s delivers a kinetic energy of 33 J to the chest and leads to transient arrhythmia, whereas an impact with a velocity of 24 m/s, which corresponds to a kinetic energy of 108 J, results in VF. 20In a model of isolated rat heart, Baxter et al.  5applied a direct contusion on the heart with impacts of 78 mJ/g and 87 mJ/g. These kinetic energies were chosen to generate a 25% and 50% decrease in ventricular function, respectively. In rabbit heart, other direct kinetic energies (from 358.2 to 432 mJ) were used to induce a 40% decrease in the maximum rate of left ventricular pressure change (dP/dtmax). 21In these studies, the chosen contusion kinetic energy criterion was hemodynamic depression. The current study used kinetic energy levels 4 to 10 times less than those used by Baxter et al.  5; (delivered energies in our study were 9.4 mJ/g, 12.5 mJ/g, 18.8 mJ/g, and 25 mJ/g in the 75-mJ, 100-mJ, 150-mJ, and 200-mJ group, respectively, if we consider that a rabbit heart weighs about 8 g). However, the aim of the study was to evaluate the ability of an impact to impair electrophysiologic parameters and to analyze the occurrence of arrhythmias. All contusion kinetic energies produced a release of CK and cell damage. The histopathologic study confirmed that hearts were injured and that impact kinetic energy was sufficient to produce gradual cell damage and myocardial contusion. The current study shows that the number and the seriousness of the arrhythmias were greater with increasing power of the injury.

In the frozen heart study, VERPs were significantly and transiently shortened by contusion. Additionally, a slight transient slowing of longitudinal ventricular conduction velocity was observed just after contusion. Consequent to the shortening of VERP and the slight slowing of conduction velocities, wavelengths were significantly shortened. Therefore, as it is well-established, the shorter the wavelength, the easier the reentry initiation. 22 

The second electrophysiologic consequence of contusion was an accentuated dispersion of VERPs. Both of the two parameters used to evaluate epicardial heterogeneity were significantly raised after contusion. The increase in refractoriness dispersion and in the maximal difference between the longest and the shortest refractory periods were reported to be associated with the creation of epicardial functional block and the occurrence of reentrant arrhythmias (dispersion-based reentry). 23–26Therefore, all of these electrophysiologic alterations could enhance contusion-induced reentrant arrhythmia.

The intact heart study confirms that myocardial contusion can induce lethal arrhythmias. Whatever the kinetic energy level used, rhythmic disturbances were observed; however, the occurrence and seriousness of arrhythmias increased with increasing kinetic energy. The impact area usually showed a transient electrically silent area, and the recovery time of electrical activity was longer when higher contusion kinetic energies were used. In some cases, the electrically silent area was replaced by a line of fixed block. One VF occurred with a contusion energy of 75 mJ, whereas none was observed at 100 mJ; however, five of six hearts presented premature ventricular contractions at 100 mJ, and none did so in the 75-mJ group. Moreover, the electrically silent area and the line of fixed block numbers are greater in the 100-mJ group than in the 75-mJ group, and we showed that these fixed blocks and areas could be responsible for arrhythmia initiation.

In the frozen heart study, two different mechanisms of arrhythmia initiation were analyzed. The first was an ectopic activity that initiated one NSVT after extrastimuli (fig. 5). The real mechanism of this spontaneous activity cannot be determined because our technique did not allow us to distinguish abnormal automatism from triggered activity. Therefore, we can only speculate that, as in acute myocardial infarction, the border area of the contusion could facilitate the occurrence of abnormal automatism. 27 

The frozen heart study clearly shows that the main mechanism of arrhythmia due to myocardial contusion is based on reentry developed around a fixed obstacle. Two kinds of fixed obstacles were observed: An electrically silent area and a line of fixed block. Although the partial (i.e. , area becomes line) or complete excitability recovery of the electrically silent area remains unclear, the initiation of reentrant arrhythmias does not seem to differ from the model used by Brugada et al.  28Using high-resolution epicardial mapping in rabbit left ventricular epicardium, these authors studied reentrant tachycardia initiated around a fixed obstacle. 28,29In their model, the impulse used a circuit around a fixed obstacle leading to travel parallel, oblique, and transverse to the fiber direction; therefore, conduction velocities varied in the circuit according to the direction considered. Brugada et al.  28observed that unidirectional block typically occurred in the longitudinal direction because of the lower safety factor in the longitudinal direction than in the transverse. 30In our study, the unidirectional conduction block could therefore be due to the anisotropy of conduction around the fixed obstacle; however, the dispersion of refractoriness and shortened wavelengths were also observed. Spach et al.  31showed, in dog atria, that functional conduction block could result from a combination of repolarization heterogeneity and anisotropic propagation. We can therefore postulate that all these factors are simultaneously involved in the facilitation of reentrant arrhythmias induced by myocardial contusion.

In a recent study using a similar cryotechnique, Haberl and Allessie 32showed that a high concentration of potassium could terminate VT, whereas heptanol, which induced a similar depressed conduction, could reset arrhythmia. These different effects are explained by the fact that during potassium concentration increase the number of available sodium channels is reduced; thus, conduction block occurred and stopped VT. In our study, cellular lysis leads to a release of potassium in the extracellular medium. This increase in potassium level could explain the occurrence of functional conduction block, the slight slowing of longitudinal conduction velocity, or the transient electrically silent area and fixed block. However, in our Langendorff model, adequate perfusion of heart was maintained during the contusion experiment, inducing a quick wash-out of potassium and other substances released during cellular lysis. Further studies are necessary to determine the mechanism of block initiation and influx slowing.

The current experimental study confirms that myocardial contusion per se  induces lethal arrhythmias. However, a distinction must be made between ventricular arrhythmias induced by commotio cordis and that induced by myocardial contusion. Maron et al.  7reported 25 sudden deaths in young athletes after blunt impact. There were no cardiac structural abnormalities; only small contusions on the left side of the chest were observed in 12 of 22 autopsies. These authors hypothesized that sudden death might be the result of apnea, vasovagal reflex, or ventricular arrhythmia. More recently, Link et al.  8demonstrated in a model of commotio cordis in anesthetized pigs that low-energy chest wall impact induces VF. VF occurs only when the impact happens within a window of 15–30 ms before the peak of the T-wave of the electrocardiogram. Because no significant cardiac injury was observed during autopsy, these authors concluded that commotio cordis induced by low-energy chest trauma (e.g. , a baseball) could be responsible for immediate sudden death due to VF depending on the precise timing of impact on the chest. Clinical reports of these catastrophic events also described immediate sudden death after impact. 7Although an impact sufficient to cause a myocardial contusion could also induce immediate VF, probably due to the same mechanism, the fact remains that experimental clinical studies and the current study also reported delayed arrhythmias. In an open-chest dog model, Stein et al.  33observed that fatal arrhythmia did not occur immediately, but appeared within 2 or 3 min after the impact. Fabian et al.  10reported that the major portion of patients with myocardial contusion developed significant arrhythmias within 24 h of admission. Baxter et al.  3observed that patients had arrhythmias and conduction defects more than 12 h after myocardial injury. In our study, the main cause of reentry initiation was the impact-induced occurrence of fixed conduction block (electrically silent area or line) and functional conduction block. These alterations could last for more than 60 min. Additionally, the current study and another experimental study of myocardial contusion showed that the occurrence and the seriousness of arrhythmias depend on the kinetic energy power. 34All of these facts seem to justify early and postoperative cardiac monitoring, even in patients suffering from only moderate blunt chest trauma. Finally, Link et al.  8did not study various kinetic energies because they specifically studied commotio cordis. Our model was a myocardial contusion model, and the impact was not timed according to the cardiac cycle. Nevertheless, because the incidence of VF increases with increasing contusion power, and because the incidence of arrhythmia seems to depend especially on the appearance of a fixed obstacle, we can speculate that the more powerful the chest trauma, the more likely the occurrence of arrhythmia, whatever the cardiac cycle phase at impact.

Care must be taken before extrapolating our results to the clinical setting. This study shows that contusion was responsible for arrhythmia, independent of hemodynamic or metabolic conditions. It is well-established that blunt chest trauma can cause hypoxia and a decrease in cardiac contractility and in output. 21,33,35These hemodynamic conditions could facilitate dysrhythmia. Furthermore, in the current study, hearts were denervated, and sympathetic nervous system stimulation induced by trauma is known to enhance the occurrence of cardiac arrhythmia. 36One could argue that in the intact heart study, ventricular arrhythmia spontaneously occurred, whereas in the frozen heart study, VTs were induced by pacing. We can therefore only speculate that reentry facilitated by block, wavelength shortening, and repolarization dispersion as the main mechanism of contusion-induced arrhythmia. One could also argue that in the intact heart study, VFs were observed, whereas in the frozen heart study, there were especially MVTs. This difference may be related to the fact that elimination of four fifths of the left ventricular wall by the cryoprocedure diminished the ventricular mass to such an extent that multiple reentering wavelets could no longer persist. 37Because the impact was directly applied on the left ventricle, we cannot directly extrapolate our findings to blunt chest trauma. Finally, we studied only the left ventricle, yet the right ventricle is also widely involved in myocardial contusion. 9,35Nevertheless, the fact remains that impact on left ventricle can induce lethal arrhythmias that have been described in blunt chest trauma patients.

This study, performed in rabbit left ventricular epicardium, shows that myocardial contusion has direct arrhythmogenic effects that are independent from hemodynamic and metabolic conditions. The incidence and the seriousness of arrhythmias increase with increasing power of impact kinetic energy. Using a high-resolution mapping system, the analysis of arrhythmias in frozen heart shows that VTs are based mainly on reentry. Reentry is based on two mechanisms induced by the impact: A fixed block (electrically silent area or line of block) and a unidirectional conduction block facilitated by the shortening of the wavelength and an increase in the dispersion of refractoriness.

The authors thank the entire Pharmacy Staff of the University Hospital of Nîmes; Patrick Chartreux, Christian Francou, and the entire personnel of the Medical School Laboratory, especially Robert Sempèrè, for their skillful technical assistance; and Margaret Manson for her English assistance.