Failure of Commercially Available Chest Wall Protectors to Prevent Sudden Cardiac Death Induced by Chest Wall Blows in an Experimental Model of Commotio Cordis
OBJECTIVE. Sudden cardiac death that results from chest wall blows (commotio cordis) the second leading cause of death in young athletes. Most events are caused by blows from projectiles, such as baseballs or lacrosse balls, with a substantial proportion occurring despite the use of a chest protector. In the present experiment, we tested the effectiveness of commercially available chest protectors in preventing ventricular fibrillation (VF) that results from chest wall strikes with baseballs and lacrosse balls.
METHODS. Twelve different baseball or lacrosse chest protectors were evaluated in juvenile swines that were subjected to 40-mph baseball or lacrosse ball blows to the precordium during the vulnerable period of repolarization for VF and were compared with control impacts without chest protectors. Seven baseball chest protectors were hit by regulation baseballs, and 5 lacrosse chest protectors were tested by blows with standard lacrosse balls. Each animal received 2 chest blows for each protector and 2 control impacts without a chest protector, with the sequence of impacts assigned randomly.
RESULTS. VF was elicited by 12 (32%) of 37 strikes in control animals without baseball chest protectors. None of the baseball chest wall protectors tested were shown to decrease significantly the occurrence of VF when compared with controls. VF was elicited by 11 (46%) of 24 strikes in control animals without lacrosse chest protectors. None of the lacrosse chest wall protectors tested decreased significantly the occurrence of VF when compared with controls.
CONCLUSION. In our experimental animal model of commotio cordis, commercially available baseball and lacrosse chest wall protectors were ineffective in protecting against VF that was triggered by chest blows and, by inference, sudden cardiac death. Improvements in materials and design of chest wall barriers are necessary to reduce the occurrence of these tragic events and make the athletic field safer for youths.
Sudden death that results from chest wall blows (commotio cordis) has been recognized with increasing frequency as a devastating risk of sports and other activities.1–14 Currently, commotio cordis is the second leading cause of death in youth athletics,4 occurring most frequently between 7 and 16 years of age in baseball but increasingly also in lacrosse. Most cases that have been reported to date involved a projectile such as baseballs, lacrosse balls, and hockey pucks. In a recent report from the US Commotio Cordis Registry,1 >25% of the fatal events that occurred during organized competitive sports involved athletes who were wearing commercially available chest protectors that were intended to prevent adverse consequences of chest wall blows.
We have developed an experimental animal model of commotio cordis in which baseballs that hit the chest wall during the vulnerable period of the cardiac cycle (10–30 ms before the T-peak) produced ventricular fibrillation (VF).15–18 This model has been particularly useful in defining the determinants of commotio cordis, including the importance of precordial location,18 velocity of impact (optimal at 40 mph),16 and direct relationship of projectile hardness to the susceptibility for VF.19
Commercially available chest barriers for both baseball and lacrosse are marketed with implied or direct claims of protecting sports participants from harm that results from trauma to the chest wall. Using our established swine model of commotio cordis, we sought to test the effectiveness of a variety of such chest barriers in preventing VF that results from strikes with baseballs and lacrosse balls.
Experimental Animal Model
Six to 8-week-old domesticated juvenile swine that weighed 11 to 18 kg were used in this study. The protocol was approved by the Animal Research Committee of the Tufts-New England Medical Center in conformity with the standards of the Association for Assessment and Accreditation of Laboratory Animal Care. Animals were sedated with 12 mg/kg intramuscular ketamine, intubated, and then anesthetized using inhaled 1% to 2% isoflurane mixed with oxygen and nitrous oxide, which maintained sedation during the experiment. During the experiment, surface electrocardiogram (ECG) was recorded continuously, and left ventricular (LV) pressure data were measured by a Millar catheter (Millar Mikrotip, Houston, TX) that was positioned in the left ventricle via the femoral approach. A Power Lab (AD Instruments, Anaheim, CA) device converted signals from analog to digital, which then were displayed and stored on a personal computer (Macintosh G3; Apple Computer Inc, Cupertino, CA).
Animals were placed prone in a sling to approximate physiologic blood flow and cardiac hemodynamics. The velocity,16 location,18 and timing15 of chest wall blows were designed to maximize the chance of producing VF. Chest wall impact was achieved by propelling a baseball or a lacrosse ball (depending on the protector being tested), mounted on a 20-g aluminum shaft, at 40 mph toward the center of the cardiac silhouette under echocardiographic guidance. Impact was gated to the portion of the cardiac cycle that is vulnerable for VF, 0 to 40 msec before the peak of the T wave and by use of a commercially available cardiac stimulator (EP-2; EP Medical Inc, Budd Lake, NJ), triggered by surface ECG input from the animal. Chest impacts that occurred outside this time window were excluded from the analysis. Impact velocity was measured by a chronographic instrument (Oehler Research, Austin, TX).
Chest Protectors Tested
The chest protectors that were chosen for this experiment are in common usage and are largely representative of commercially available products in both design and composition. Seven commercially available baseball chest protectors and 5 lacrosse chest protectors (Fig 1) were tested in this series of experiments. All of the chest barriers tested had a soft compliant layer that was composed of foam of varying density and thickness, and most (Brine GBP [Brine Inc, Milford, MA], Warrior Goalie Guard 5000 [Warrior Inc, Warren, MI], Rawlings Batter's Chest Protector 550 [BCP550; Rawlings Sporting Goods Company Inc, St Louis, MO], Heart-Gard [S&M Human Performance Products, Inc, Denver, CO], Louisville TPX Youth CBP [TPX-CBP; Hillerich & Bradsby Co, Louisville, KY], and Provest [Provest, Lake Bluff, IL]) had a harder plastic layer either on the external surface or embedded within the foam layer(s). The Brine Pro lacrosse protector had a unique soft layer that was composed of sleeves of expanded polypropylene beads, whereas the soft layer in all other protectors was composed of closed cell foam (Figs 2 and 3).
Because of their design, 1 of the baseball protectors (Provest) and 2 of the lacrosse protectors (Brine GBP and Warrior) had the potential to have 2 different barrier materials overlying an athlete's precordium. The Provest protector has an optional and removable plastic outer shell; therefore, it was tested with (Provest+Plastic) and without (Provest) the shell. In the case of the Brine GBP and Warrior protectors, different barrier materials could potentially overlie the precordium, depending on the position of the athlete and body motion; therefore, each of these materials was tested separately (designated Brine GBP Side and Brine GBP Center; Warrior Side and Warrior Center). Thus, a total of 15 different chest protector materials, from 12 products, were tested. Baseball and lacrosse protectors were tested in 2 independent experiments with different sets of animals.
Baseball chest protectors were hit by regulation baseballs (Rawlings Little League baseball [LLB-1]), and lacrosse chest protectors were tested by blows with standard lacrosse balls (Brine Inc).
A 4 × 4-in portion of each of the chest protectors was removed from that portion of the protector that was designed to cover an athlete's precordium. Before each strike, an echocardiogram was performed to identify the center of the LV silhouette.18 With this guidance, the detached portion of each chest protector then was placed on the chest wall and secured by use of elastic straps that were fastened around the chest of the animal. The baseball or the lacrosse ball then was directed toward and hit the previously identified target area on the chest wall. Measures were taken to ensure identical speed, timing, and location of the chest wall blows. In addition, each chest protector material was affixed in an identical manner, and any migration of the protector between chest wall blows was corrected by realignment with echocardiographic imaging before each strike.
The order in which chest protectors were applied to each animal was randomized, after which blows were delivered once to each chest protector and once to the same animal when no protector was present (which served as a control). After this initial series, a second randomization was performed and the same animal underwent another series of strikes in a different sequence. Thus, each animal received 2 impacts to each of the chest protectors tested and also 2 blows directly to the unprotected precordium. When VF did not result from a given chest blow, the subsequent impact was delivered with an interval of no less than 3 minutes. When VF was triggered, prompt defibrillation was performed, after which blood pressure, LV systolic function, and ECG pattern were assessed during an observation period of 5 minutes; when these parameters returned to normal, additional impacts were performed. When the animal remained unstable, no additional chest blows were delivered, data from that incomplete series of impacts were excluded from the analysis, and the animal was euthanized. Autopsy examination limited to the thorax was performed after completion of the protocol in each animal.
In the baseball protocol, 20 animals received a total of 365 chest impacts, and in the lacrosse experiment, 12 animals had 193 chest blows. Of these 558 blows, 33 occurred either outside the 0- to 40-msec interval of vulnerability for VF or after hemodynamic deterioration and, as prospectively determined, therefore were excluded from analysis. Consequently, a total of 333 baseball impacts and 192 lacrosse impacts compose the present data set.
The primary endpoint of this study was the occurrence of VF. Data first were analyzed using a χ2 test to compare the response to each protector versus controls (no protector). Second, the McNemar's test was used to control for possible confounding variables among the animals, such as body weight, and chest wall characteristics, such as different rib structure and compliance.
Assuming an incidence of VF in control strikes of 50%, using a 2-sided test with an α level of .05 and a power level of 80%, detection of a reduction in VF from 50% to 10% would require 20 impacts per chest protector. In this experiment, each baseball protector underwent 40 chest wall blows, and each lacrosse protector was exposed to 24 chest wall blows.
Secondary endpoints include peak intracavitary LV pressure measured at chest impact, as well as ST segment elevation and the presence of bundle branch block (BBB; defined as ≥100% increase in QRS duration). Peak LV pressure was measured from the digitally recorded pressure from the LV catheter and compared with control impacts with a paired t test. ST segment elevation and BBB were assessed only in strikes that did not result in VF. The degree of ST elevation was measured from the digitally recorded surface ECG, 80 ms after the J point of the first technically satisfactory beat immediately after chest impact. Differences in the incidence of BBB between animals with chest protectors and controls were assessed using McNemar's test. For all analyses, P < .05 was considered statistically significant.
Baseball Chest Protectors
VF was elicited by 12 (32%) of 37 strikes as a result of control impacts in animals without baseball chest protectors. None of the baseball chest wall protectors tested significantly decreased the occurrence of VF when compared with controls (Fig 4). With chest protectors, VF occurred in the range of 22% to 49% of chest blows: Provest, 18 (49%) of 37 (P = .16); Provest with plastic, 8 (22%) of 37 (P = .30); BCP550, 15 (41%) of 37 (P = .47); Heart-Gard, 15 (41%) of 37 (P = .47); Cooper Pro BPX (BPX; Bauer Inc, Toronto, Ontario, Canada), 10 (27%) of 37 (P = .61); TPX-CPB (Bauer Inc), 14 (38%) of 37 (P = .63); Cooper BP-40 (BP-40; Bauer Inc), 13 (35%) of 37 (P = .81); and LBP-1, 13 (35%) of 37 (P = .81).
The peak LV pressure that was produced by the blow correlated linearly with the probability of VF. Mean peak LV pressure in controls was 709 ± 114 mm Hg. Three chest protectors showed significantly lower mean LV pressures than controls: BCP550 (676 mm Hg; P = .014), BPX (648 mm Hg; P = .012), and TPX-CBP (675 mm Hg; P = .027).
Mean ST segment elevation in control strikes that did not result in VF was 214.7 ± 132.3 μv. Two chest protectors were associated with significantly less ST segment elevation compared with controls: BPX (150.9 μv; P = .025) and Provest with plastic (120.1 μv; P = .001).
Among controls, BBB was observed in 16 (64%) of 25 strikes. No significant reduction in the occurrence of BBB was attributable to any of the baseball chest protectors.
Lacrosse Chest Protectors
VF was elicited by 11 (46%) of 24 strikes as a result of control impacts in animals without lacrosse chest protectors. None of the lacrosse chest wall protectors tested significantly decreased the occurrence of VF when compared with controls (Fig 5). With chest protectors, VF occurred in the range of 21% to 50% of chest blows: Brine GBP Center, 8 (33%) of 24 (P = .37); Warrior Center, 8 (33%) of 24 (P = .37); deBeer Icon (deBeer Inc, Albany, NY), 9 (38%) of 24 (P = .55); Brine GBP Side, 12 (50%) of 24 (P = .77); STX Aegis (STX, Baltimore, MD), 12 (50%) of 24 (P = .77); and Warrior Side, 12 (50%) of 24 (P = .77). The Brine Pro demonstrated a trend toward lower probability of VF when compared with control strikes (21% vs 49%; P = .07).
In this protocol, the peak LV pressure that was produced by the chest blow also correlated linearly with the probability of VF. Mean peak LV pressure that was produced by chest wall blows in controls was 641 ± 116 mm Hg. Significantly lower mean maximum LV pressure was observed during lacrosse ball impacts with 3 protectors: the Brine Pro (598 mm Hg; P = .007), Brine GBP Center (597 mm Hg; P = .006), and Brine GBP Side (611 mm Hg; P = .040).
Mean ST segment elevation with chest impact in controls was 63.2 ± 30.3 μv. Three chest protectors showed significantly less ST elevation when compared with control chest wall blows: the Brine Pro (19.8 μv; P = .001), deBeer Icon (35.5 μv; P = .016), and the Brine GBP Center (19.6 μv; P = .001).
Among controls without lacrosse chest protectors, BBB was observed after all 13 (100%) strikes. When compared with controls, 6 of the 7 materials tested had significantly lower frequency of BBB: Brine Pro (42%; P = .016), deBeer Icon (46%; P = .031), Brine GBP Center (18%; P = .004), Brine GBP Side (40%; P = .031), Warrior Center (46%; P = .031), and Warrior Side (30%; P = .016).
Commotio cordis is an increasingly recognized cause of sudden death on the athletic field during organized, competitive sports1–7 and of mounting concern among the general public and those who are involved in youth sports.6,7 Previous efforts to reduce the risk for these events have focused on the use of softer-than-normal (safety) baseballs that have been shown to reduce (in our experimental model)15,19 but not abolish the likelihood of VF during play.1
In our experimental animal model of commotio cordis, commercially available baseball and lacrosse chest wall protectors failed to protect against VF that was triggered by chest blows. These findings are consistent with the clinical observation that commotio cordis not infrequently occurs despite the use of chest wall barriers that are believed to be protective. Indeed, in a recent report from the US Commotio Cordis Registry (Minneapolis, MN), almost 30% of fatal commotio cordis events occurred in athletes who were nevertheless afforded some form of chest protection.1 In some cases, the lack of protection by the chest barriers was apparently attributable to migration of the equipment during physical activity, exposing the precordium to direct impact. In other cases, however, projectiles were known to strike directly on the chest protector, including several lacrosse goalies and baseball catchers.1 Indeed, in the present animal model, with a representative portion of the chest protector material securely affixed directly over the heart, VF nevertheless occurred frequently with each of the barriers tested and no less frequently than in controls without chest protectors.
The velocity, location, and timing of chest wall impacts in our model were designed to maximize the likelihood of VF. Although chest wall impacts during athletic training and competition do not always share these characteristics, testing protectors under laboratory circumstances allows for assessment of their ability to protect against a worst-case scenario, the standard against which protector design should be measured.
All but 1 of the chest protectors evaluated in this experiment were composed of a compliant layer(s) of closed cell foam of varying thickness and density, which is intended to dissipate the energy of an impact, and did or did not have a harder plastic shell material either covering or embedded within the foam, which is intended to diffuse the force of impact over a greater surface area. Given the presence of such soft, presumably energy-absorbing material, we intuitively expected these chest barriers to afford a large measure of protection against commotio cordis; however, this did not occur under our experimental conditions. Furthermore, there is no evidence that the presence of a hard plastic shell acted to reduce significantly the occurrence of VF. Only the Brine Pro protector exhibited a trend toward protection from VF as induced by chest impacts with lacrosse balls. Of note, this particular chest barrier (which does not have a hard plastic shell) is composed of a layer that contains sleeves of expanded polypropylene beads, a unique material composition that is not found in other available products.
Only 1 previous study has assessed the protective effect of chest barriers against baseball impacts.20 These investigators used a nonbiological 3-rib structure to assess responses to a baseball projected at speeds that ranged from 40 to 70 mph. Four of the 5 chest protectors tested reduced the viscous response, which is a measure of deformation and compression of the experimental model over the duration of the impact. However, such a nonbiological model does not test or produce VF or mimic the complex phenomenon of commotio cordis15,17; therefore, its relevance to this clinical problem may be limited.
Our animal model of commotio cordis, out of necessity, is not identical to the human condition. For example, swine chest walls are more ovoid than humans; however, at the site of impact in both swine and humans, the precordial strike occurs perpendicular to the chest wall, minimizing the differences in the chest wall anatomy. Our experimental design used a 4 × 4-in material that was cut from the commercially available chest protector that was affixed over the swine precordium, thereby also minimizing the potential effects of different chest contours in humans and swine.
Chest wall protectors have been widely marketed and promoted as providing protection against injury in both baseball and lacrosse, particularly in organized youth sports, for catchers and goalies. Manufacturers have been (and may continue to be) unaware of commotio cordis and its consequences and therefore have designed chest barriers to protect athletes primarily against soft tissue and bone injury but not against VF, potentially a more lethal potential outcome of chest wall trauma. When commotio cordis occurs despite the use of a chest protector, the event is especially troubling, given the false sense of security that is conveyed unintentionally by wearing such a commercially available barrier.
The present experimental data, showing that none of the commercial chest barriers tested in our swine model provided significant protection against commotio cordis, has stimulated our laboratory to undertake a systematic effort to determine, under experimental conditions with the swine model, the most optimal chest barrier material composition and design with the potential ultimately to offer virtually absolute protection against chest blow–induced VF. Novel materials and design that are effective in dissipating impact energy in time and space differently from barriers that are designed to protect against tissue injury may be necessary to achieve this goal. The findings of this experiment provide a starting point for effective chest protector design by exposing the ineffectiveness of the available protectors, as well as the potential implied by a unique material such as expanded polypropylene beads in contrast to the more standard (closed cell foam) materials.
Some of the chest protectors tested here seemed to blunt some of the previously reported facets of commotio cordis, such as ST segment elevation, BBB, and LV systolic pressure. This may be evidence that the chest wall barriers that we tested do, in fact, mitigate in some respect the commotio cordis phenomenon. However, the failure of chest barriers to abolish VF in our experiment represents the most clinically relevant evidence of their ineffectiveness in preventing commotio cordis and emphasizes the inadequacy of the materials that currently are in use. It is our expectation that by using the present experimental data, improved chest barriers that ultimately will prevent this increasingly recognized cause of sudden cardiac death during sports activities can be designed and produced.
This study was supported by grants from Kiwanis Pediatric Trauma Institute, Tufts-New England Medical Center (Boston, MA), National Operating Committee on Standards for Athletic Equipment (Overland Park, KS), and Louis J. Acompora Foundation (Northport, NY).
We are indebted to Stacey E. Supran, Division of Clinical Care Research, New England Medical Center, for assistance with statistical analysis.
- Accepted November 11, 2005.
- Address correspondence to Mark S. Link, MD, Tufts-New England Medical Center, Box 197, 750 Washington St, Boston, MA 02111. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
The opinions expressed herein are the opinion of the authors and do not necessarily represent the opinions of the funding organizations.
- ↵Zrebiec J. Lacrosse player dies after chest hit. Baltimore Sun. March 19, 2004:1C
- ↵Freidlander BP. CU mourns death of men's lacrosse team leader George Boiardi. Cornell Chronicle.2005;35 :1– 2
- Dickman GL, Hassan A, Luckstead EF. Ventricular fibrillation following baseball injury. Phys Sport Med.1978;6 :85– 86
- Frazer M, Mirchandani H. Commotio cordis, revisited. Am J Forensic Med Pathol.1984;5 :249251
- ↵Link MS, Wang PJ, VanderBrink BA, et al. Selective activation of the K+ATP channel is a mechanism by which sudden death is produced by low-energy chest-wall impact (commotio cordis). Circulation.1999;100 :413– 418
- ↵Link MS, Maron BJ, Wang PJ, Pandian NG, VanderBrink BA, Estes NAM III. Reduced risk of sudden death from chest wall blows (commotio cordis) with safety baseballs. Pediatrics.2002;109 :873877
- Copyright © 2006 by the American Academy of Pediatrics