LV reverse remodeling imparted by aortic valve replacement for severe aortic stenosis; is it durable? A cardiovascular MRI study sponsored by the American Heart Association.

BACKGROUND: In patients with severe aortic stenosis (AS), long-term data tracking surgically induced effects of afterload reduction on reverse LV remodeling are not available. Echocardiographic data is available short term, but in limited fashion beyond one year. Cardiovascular MRI (CMR) offers the ability to serially track changes in LV metrics with small numbers due to its inherent high spatial resolution and low variability. HYPOTHESIS: We hypothesize that changes in LV structure and function following aortic valve replacement (AVR) are detectable by CMR and once triggered by AVR, continue for an extended period. METHODS: Twenty-four patients of which ten (67 ± 12 years, 6 female) with severe, but compensated AS underwent CMR pre-AVR, 6 months, 1 year and up to 4 years post-AVR. 3D LV mass index, volumetrics, LV geometry, and EF were measured. RESULTS: All patients survived AVR and underwent CMR 4 serial CMR's. LVMI markedly decreased by 6 months (157 ± 42 to 134 ± 32 g/m2, p < 0.005) and continued trending downwards through 4 years (127 ± 32 g/m2). Similarly, EF increased pre to post-AVR (55 ± 22 to 65 ± 11%,(p < 0.05)) and continued trending upwards, remaining stable through years 1-4 (66 ± 11 vs. 65 ± 9%). LVEDVI, initially high pre-AVR, decreased post-AVR (83 ± 30 to 68 ± 11 ml/m2, p < 0.05) trending even lower by year 4 (66 ± 10 ml/m2). LV stroke volume increased rapidly from pre to post-AVR (40 ± 11 to 44 ± 7 ml, p < 0.05) continuing to increase non-significantly through 4 years (49 ± 14 ml) with these LV metrics paralleling improvements in NYHA. However, LVmass/volume, a 3D measure of LV geometry, remained unchanged over 4 years. CONCLUSION: After initial beneficial effects imparted by AVR in severe AS patients, there are, as expected, marked improvements in LV reverse remodeling. Via CMR, surgically induced benefits to LV structure and function are durable and, unexpectedly express continued, albeit markedly incomplete improvement through 4 years post-AVR concordant with sustained improved clinical status. This supports down-regulation of both mRNA and MMP activity acutely with robust suppression long term.

Preoperative statin is associated with decreased operative mortality in high risk coronary artery bypass patients.

BACKGROUND: Statins are widely prescribed to patients with atherosclerosis. A retrospective database analysis was used to examine the role of preoperative statin use in hospital mortality, for patients undergoing isolated coronary artery bypass grafting (CABG.) METHODS: The study population comprised 2377 patients who had isolated CABG at Allegheny General Hospital between 2000 and 2004. Mean age of the patients was 65 +/- 11 years (range 27 to 92 years). 1594 (67%) were male, 5% had previous open heart procedures, and 4% had emergency surgery. 1004 patients (42%) were being treated with a statin at the time of admission. Univariate, bivariate (Chi2, Fisher's Exact and Student's t-tests) and multivariate (stepwise linear regression) analyses were used to evaluate the association of statin use with mortality following CABG. RESULTS: Annual prevalence of preoperative statin use was similar over the study period and averaged 40%. Preoperative clinical risk assessment demonstrated a 2% risk of mortality in both the statin and non-statin groups. Operative mortality was 2.4% for all patients, 1.7% for statin users and 2.8% for non-statin users (p < 0.07). Using multivariate analysis, lack of statin use was found to be an independent predictor of mortality in high-risk patients (n = 245, 12.9% vs. 5.6%, p < 0.05). CONCLUSIONS: Between 2000 and 2004 less than 50% of patients at this institution were receiving statins before admission for isolated CABG. A retrospective analysis of this cohort provides evidence that preoperative statin use is associated with lower operative mortality in high-risk patients.

In vivo performance of a muscle-powered drive system for implantable blood pumps.

A unique biomechanical implant has been developed to convert muscle power into hydraulic energy for the purpose of driving an implanted blood pump. This device, called a muscle energy converter (MEC), is designed to attach to the humeral insertion of the latissimus dorsi (LD) muscle, so that stimulated contractions cause a rotary cam to compress a fluid-filled bellows. Here we report results from the latest in a series of canine implant trials where the MEC was connected to an adjustable pressure load to measure power output and assess long-term function. Full-length (2 cm) actuator strokes were maintained for a period of 1 month with no discernable discomfort to the animal. Load conditions were cycled periodically to measure stroke work capacity and pressure production. The peak driveline pressure recorded in this experiment was 1743 mm Hg. Steady state power generation was measured to 478 +/- 21 mJ/stroke (mean +/- SD) with stroke work levels reaching 785 mJ in one test. Normal left and right ventricular stroke work levels in dogs this size (35 kg) are 700 and 150 mJ, respectively. These data confirm that MEC/LD power levels--maintained in tandem with an appropriate cardiac assist device--are sufficient to provide significant long-term circulatory support. Further testing, however, is still needed to demonstrate the long-term stability of this drive system.

In vitro evaluation of the TandemHeart pediatric centrifugal pump.

The pediatric TandemHeart pump is being developed for short-term circulatory support of patients varying in size from 2 to 40 kg. The pump withdraws blood from the left atrium via cannula inserted percutaneously, either through the right internal jugular vein or transhepatically, and pumps the blood back into the arterial system via the carotid or femoral artery. High resolution stereolithography (SLA) was used to create an upper housing and impeller design, which were assembled into a functional pump prototype. Pressure-flow characteristics of the pump were determined in a blood analogue solution and compared with the pressure-flow requirements of the intended cannulation. At 5,500 rpm, the pump was able to generate 0.4 L/min of flow with a pressure rise of 325 mm Hg and 2.0 L/min with a pressure rise of 250 mm Hg. The hydraulic performance of the pump will enable at least 50% of cardiac output when the arterial cannula is placed in the carotid artery. The hemolysis of the TandemHeart pediatric pump at 5,500 rpm was compared with the BP-50 pediatric centrifugal pump in vitro using bovine blood flowing at 0.4 L/min against 250 mm Hg. The TandemHeart pump produced a similar increase in plasma free hemoglobin levels during the duration of the 6 hour test.

Preoperative clinical factors are important determinants of the inflammatory state before and after heart surgery.

Postoperative inflammatory response is common in heart surgery patients, but less is known about variation in the baseline inflammatory state. This study characterizes the preoperative inflammatory profile in a group of high- and low-risk patients (n = 32; male 16, female 16; mean age, 70.3 +/- 1.8) and relates this to postoperative events. Interleukin-6 (IL-6), tumor necrosis factor (TNF)-alpha, TNF receptors (R1 and R2), and high-sensitivity C-reactive protein were measured before surgery and 4 hours after arrival in the intensive care unit. Considerable variability existed in all preoperative inflammatory mediators before surgery. Patients with an elevated baseline IL-6 level, (IL-6 >10 pg/mL) were older (73.5 +/- 2.2 vs. 67.9 +/- 2.6 years), had a lower ejection fraction (34 +/- 3.8% vs. 44 +/- 2.9%), a higher predicted risk score (10.3 +/- 1.2 vs. 5.9 +/- 1.1), and a higher baseline high-sensitivity C-reactive protein (65 +/- 10 vs. 24 +/- 6 mg/L), p < 0.05 for all. These patients had high morbidity and mortality rates after surgery. In addition, patients judged to be at high risk on clinical criteria were found to have consistent elevations in the baseline inflammatory state. All patients had a surge in inflammatory mediators after surgery, but those who started at a higher baseline reached a higher postoperative level than the others (IL-6 2023 +/- 561 vs. 361 +/- 47 pg/mL, p < 0.05). Many heart surgery patients, especially higher-risk patients, have a significant inflammatory state before surgery. These patients are at risk for high morbidity and mortality rates after surgery.

Hepatic gene expression response to acute indomethacin exposure.

BACKGROUND: Rising morbidity and mortality related to the use of NSAIDs has led to the withdrawal of some of these agents and reconsideration of the adverse effects and usage paradigms of commonly available NSAIDs. Our objective in this study was to assay molecular indicators of acute hepatic injury associated with the administration of indomethacin, a prototypical NSAID, metabolized by the liver that undergoes enterohepatic circulation with associated gastrointestinal adverse effects. METHODS: Analysis of gene expression, using high-throughput, ADME (absorption, distribution, metabolism, excretion)-specific microarrays, was performed on RNA extracted from the livers of control or indomethacin treated rats, in parallel with serum enzyme tests and histological analysis of paraffin-embedded liver specimens. Male Sprague-Dawley rats (n = 45) were administered intraperitoneal injections of indomethacin for 3 days at the recommended normal dose (6.7 mg/kg), indomethacin at a high dose (20 mg/kg) or vehicle alone (controls). RESULTS: Upon termination of the study on day 4, serum gamma-glutamyl transferase activity and alkaline phosphatase/alanine aminotransferase ratios were significantly elevated in both high- and normal-dose cohorts compared with vehicle-treated animals. Diffuse microvascular steatosis was present in hepatic serial sections obtained from all animals subjected to the high-dosage regimen. High-resolution microarray analysis (six replicates/gene/animal) identified 256 genes, after outlier removal, in 17 functional classifications that were significantly altered by the high, but not by the normal dosage. These included depression of 10 of 11 cytochrome P450 genes (2B3, 2C70, 1A2-P2, 4F1, 2E1, 3A1, 2F1, 3AP7, 2C11, phenobarb-inducible P6) and 7 of 9 genes involved in the response to reactive oxygen species (e.g. glutathione reductase, glutathione transferase, and superoxide dismutase). Of 16 genes associated with toxin removal, nine exhibited significantly decreased transcripts. There was a marked shift away from lipid metabolism (decreased expression of eight genes) towards glucose utilization associated with steatosis. Despite the compromise of detoxification programs and a shift in metabolic substrate utilization, a compensatory remodeling response was activated, including genes for metalloproteases (ADAM10, MMP10, MMP11), integrins (integrin alpha-1 and alpha-E1), and extracellular matrix molecules (platelet/endothelial cell adhesion molecule-1 and heparan sulfate proteoglycan, perlecan), as well as transcripts associated with cell proliferation. The expression levels of only five genes were significantly altered among animals receiving the normal indomethacin dosage. CONCLUSION: These data confirmed that even brief exposure to indomethacin altered serum enzymatic activities and that high levels significantly altered gene expression in the liver and hepatic histology (by interfering with the clearance of toxins and xenobiotic substrates) and the regulation of basal metabolism.

Effect of a flexible ventricular restraint device on cardiac remodeling after acute myocardial infarction.

The effects of a flexible ventricular restraint device on left ventricular (LV) dilatation and hypertrophy after transmural infarction are examined in an ovine model. Left ventricular remodeling and dilatation occurs after extensive myocardial infarction. A flexible ventricular restraint made from a nitinol mesh was evaluated in adult female sheep (n=14). Cardiac magnetic resonance imaging scans and hemodynamic measurements were completed before and 6 weeks after anterior myocardial infarction. Treatment animals (n=7) received passive ventricular restraint concurrently with LV infarction; the others (n=7) served as controls. Increases in LV end-diastolic volume index were significantly less in the restraint group than in controls (0.20+/-0.41 vs 0.83+/-0.50 ml/kg, p<0.03). End-systolic volumes increased less in treatment animals (0.43+/-0.28 vs 0.90+/-0.38 ml/kg, p<0.03). Control hearts showed an increase in LV mass after infraction, whereas LV mass decreased in restrained hearts (0.14+/-0.19 vs -0.25+/-0.36 g/kg, p<0.03). Hemodynamic studies showed similar changes after infarction for the control and the device group. Gross and microscopic examination showed no device-induced epicardial injury. A flexible ventricular restraint device attenuated remodeling after acute myocardial infarction in sheep.

Upper extremity arteriovenous fistulas induce modest hemodynamic effect on the in situ internal thoracic artery.

BACKGROUND: The left internal thoracic artery is the ideal conduit for coronary artery revascularization due to superior patency compared with venous grafts. The hemodynamic effects of an arterio-venous fistula on the native in situ internal thoracic artery and the potential for coronary flow steal after revascularization with the internal thoracic artery, are not well-characterized. METHODS: Fifteen chronic hemodialysis patients with functioning left upper extremity arterio-venous fistulas were evaluated with the use of transthoracic color Doppler analysis. Flow characteristics of the left and right internal thoracic arteries at baseline, with arterio-venous fistula occluded, and during hemodialysis were recorded. Peak systolic and diastolic velocities for the right and left internal thoracic arteries were calculated for each fistula state (occluded, open, and on-dialysis). One-way analysis of variance was used to compare the different means. RESULTS: Mean flow velocity calculations failed to identify any statistically significant differences between the ipsilateral and contralateral internal thoracic artery in any fistula state. In addition, mean internal thoracic flow velocities were similar between the different fistula states. CONCLUSIONS: Changes in arteriovenous fistula flow state did not significantly alter Doppler flow hemodynamics of either the ipsilateral or contralateral in-situ internal thoracic artery. Further studies of the possible effects on internal thoracic arteries used as pedicled coronary grafts may be required.

Development of a percutaneous pediatric ventricular assist device.

Remarkable progress has been made on ventricular assist devices for adult patients. Unfortunately, similar devices are not yet available in the United States for pediatric heart patients. The goal of this project is to demonstrate the feasibility of a percutaneous ventricular assist device for pediatric patients above 2 kg. The proposed system consists of an extracorporeal centrifugal blood pump, a transseptal venous cannula that takes blood from the left atrium to the pump, an arterial cannula that returns the blood to the arterial system, and a controller to adjust pump speed/flow. Using an ad hoc pediatric pump prototype and a spectrum of specially designed cannulae of various sizes, benchtop studies showed that the proposed system could deliver blood flow in a range of 0.3 to 3.0 l/min. For smaller patients (2-35 kg), the transseptal cannula was designed to be placed in the internal jugular vein and the arterial cannula in the internal carotid artery. For larger patients (> 35 kg), the femoral vein and artery would be used. Further development effort will be focused on reducing the hemolysis of the pump design, refining the cannula design, and demonstrating the safety and functionality in animal studies.

Improved mechanism for capturing muscle power for circulatory support.

Although it is now understood that trained skeletal muscle can generate enough steady-state power to provide significant circulatory support, there are currently no means by which to tap this endogenous energy source to aid the failing heart. To that end, an implantable muscle energy converter (MEC) has been constructed and its function has been improved to optimize durability, anatomic fit, and mechanical efficiency. Bench tests show that MEC transmission losses average less than 10% of total work input and that about 85% of this muscle power is successfully transferred to the working fluid of the pump. Results from canine implant trials confirm excellent biocompatibility and demonstrate that contractile work of the latissimus dorsi muscle-measured to 290 mJ/stroke in one dog-can be transmitted within the body at levels consistent with cardiac assist requirements. These findings suggest that muscle-powered cardiac assist devices are feasible and that efforts to further develop this technology are warranted.

Physiologic compensation is supranormal in compensated aortic stenosis: does it return to normal after aortic valve replacement or is it blunted by coexistent coronary artery disease? An intramyocardial magnetic resonance imaging study.

BACKGROUND: In compensated aortic stenosis (AS), cardiac performance measured at the ventricular chamber is typically supranormal, whereas measurements at the myocardium are often impaired. We investigated intramyocardial mechanics after aortic valve replacement (AVR) and the effects relative to the presence or absence of coronary artery disease (CAD+ or CAD-), respectively. METHODS AND RESULTS: Twenty-nine patients (46 to 91 years, 10 female) with late but not decompensated AS underwent cardiovascular MRI before AVR (PRE), with follow-up at 6+/-1 (EARLY) and 13+/-2 months (LATE) to determine radiofrequency tissue-tagged left ventricle (LV) transmural circumferential strain, torsion, structure, and function. At the myocardial level, concentric LV hypertrophy regressed 18% LATE (93+/-22 versus 77+/-17 g/m2; P<0.0001), whereas at the LV chamber level, ejection fraction was supranormal PRE, 67+/-6% (ranging as high as 83%) decreasing to 59+/-6% LATE (P<0.05), representing not dysfunction but a return to more normal LV physiology. Between the CAD+ and CAD- groups, intramyocardial strain was similar PRE (19+/-10 versus 20+/-10) but different LATE, with dichotomization specifically related to the CAD state. In the CAD- patients, strain increased to 23+/-10% (+20%), whereas in CAD+ patients it fell to 16+/-11% (-26%), representing a nearly 50% decline after AVR (P<0.05). This was particularly evident at the apex, where CAD- strain LATE improved 17%, whereas for CAD+ it decreased 2.5-fold. Transmural strain and myocardial torsion followed a similar pattern, critically dependent on CAD. AVR impacted LV geometry and mitral apparatus, resulting in decreased mitral regurgitation, negating the double valve consideration. CONCLUSIONS: In AS patients after AVR, reverse remodeling of the supranormal systolic function parallels improvement in cardiovascular MRI-derived regression of LV hypertrophy and LV intramyocardial strain. However, discordant effects are evident after AVR, driven by CAD status, suggesting that the typical AVR benefits are experienced disproportionately by those without CAD and not by those obliged to undergo concomitant coronary artery bypass grafting/AVR.

A femoral artery cannula that allows distal blood flow.

OBJECTIVE: A femoral artery cannula is used for certain types of circulatory support but can cause ischemia, especially during prolonged perfusion. This study tests the function of a femoral cannula designed to allow proximal and distal blood flow. METHODS: Five pigs were used in the study. In each animal a distal-flow cannula was implanted in the femoral artery of one leg, and the same-sized standard cannula was implanted in the other. Blood was drained from the left atrium and delivered to the femoral artery through the distal-flow cannula or standard cannula by using a centrifugal pump. An ultrasonic flow probe and microspheres were used to quantify flow and perfusion distal to the cannula. RESULTS: Distal femoral flow and tissue perfusion were present in all animals (5/5) with the distal-flow cannula but only in 1 of 5 animals with the standard cannula (P < .048). Distal flow did not change with pump flow. Mean distal flow at each level of pump flow was higher with the distal-flow cannula (P < .05). Tissue perfusion was also higher with the distal-flow cannula (0.052 +/- 0.028 vs 0.010 +/- 0.022 mL x min(-1) x g(-1), P < .03). CONCLUSIONS: In the swine model the distal-flow cannula allowed greater and more consistent distal flow than the standard cannula. The use of a distal-flow cannula for circulatory support might reduce the risk of distal limb ischemia.

Experimental and clinical studies with the Paracor cardiac restraint device.

Progressive cardiac enlargement is a consequence of congestive heart failure (CHF) and a cause of further deterioration. Cardiac restraint devices are intended to interrupt and reverse this process, thereby improving the natural history of CHF. The Paracor restraint device is made from a superelastic nitinol mesh that exerts a small but continuous epicardial force to relieve left ventricular (LV) wall stress and to limit LV dilation. Animal studies have used various heart failure models. The data show the device reduces LV dilation after infarction, preserves LV contractility with rapid pacing, and avoids diastolic constriction in the embolization model. Ten patients in Europe and 10 in the United States have been implanted through a small thoracotomy incision without the use of cardiopulmonary bypass. Changes in LV size, functional status, and exercise performance have been measured, but the detailed data have not yet been presented or published. The device remains experimental, pending completion of additional clinical trials.

Comparison of dog and pig models for testing substernal cardiac compression devices.

Subtle anatomic differences between species can be a critical consideration when determining whether a given animal model is appropriate for surgical research purposes, especially when testing biomechanical implants. This study compares the effectiveness of two common animal models (dogs and pigs) in testing a balloon based cardiac compression device designed for substernal placement. Pigs were used in acute studies using an infarction model of heart failure, whereas dogs were used in chronic experiments in which heart failure was induced via rapid pacing. Systolic cardiac compression was accomplished in both species using identical balloons inflated between the sternum and right ventricle with every heartbeat. Results showed the device to be much more effective in pigs, where cardiac stroke volumes returned to normal with balloon assistance (14.7 +/- 1.9 to 37.8 +/- 9.2 mL, p < 0.005). Stroke volumes in dogs, however, remained essentially unaltered by balloon activation (28.1 +/- 14.1 to 29.6 +/- 14.7 mL, p = NS). Retrospective comparisons showed pig models to be a much closer approximation to the human anatomy because of a more similar thoracic cavity shape and heart orientation. These findings suggest that certain large animal models should not be used in research in which chest wall shape or cardiac orientation within the thoracic cavity may influence outcomes.

Improvement of sternal closure stability with reinforced steel wires.

BACKGROUND: Sternal dehiscence occurs when steel wires pull through sternal bone. This study tests the hypothesis that closure stability can be improved by jacketing sternal wires with stainless steel coils, which distribute the force exerted on the bone over a larger area. METHODS: Midline sternotomies were performed in 6 human cadavers (4 male). Two sternal closure techniques were tested: (1) approximation with six interrupted wires, and (2) the same closure technique reinforced with 3.0-mm-diameter stainless steel coils that jacket wires at the lateral and posterior aspects of the sternum. Intrathoracic pressure was increased with an inflatable rubber bladder placed beneath the anterior chest wall, and sternal separation was measured by means of sonomicrometry crystals. In each trial, intrathoracic pressure was increased until 2.0 mm of motion was detected. Differences in displacement pressures between groups were examined at 0.25-mm intervals using the paired Student's t test. RESULTS: The use of coil-reinforced closures produced significant improvement in sternal stability at all eight displacement levels examined (p < 0.03). Mean pressure required to cause displacement increased 140% (15.5 to 37.3 mm Hg) at 0.25 mm of separation, 103% (34.3 to 69.8 mm Hg) at 1.0 mm of separation, and 122% (46.8 to 103.8 mm Hg) at 2.0 mm of separation. CONCLUSIONS: Reinforcement of sternal wires with stainless steel coils substantially improves stability of sternotomy closure in a human cadaver model.

Capturing in situ skeletal muscle power for circulatory support: a new approach to device design.

Efforts to harness in situ skeletal muscle for circulatory support have been extensive, but implants designed to tap this power source have yet to meet the strict performance standards incumbent upon such devices. A fourth generation muscle energy converter (MEC4) is described that represents a significant departure from previous hydraulic muscle pump designs, all of which have assumed a long cylindrical profile. The MEC4, in contrast, features a puck shaped metallic bellows oriented so that its end fittings lie parallel to the chest wall. The fixed end is centered over a fluid port that passes into the thoracic cavity across one resected rib. The opposite end of the bellows supports a roller bearing that moves beneath a linear cam fixed to a reciprocating shaft. The shaft exits the housing through a spring loaded seal and is attached to a sintered anchor pad for muscle tendon fixation. This configuration was chosen to improve bellows durability, lower device profile, and reduce tissue encumbrance to actuator recoil. Bench tests show that modest actuation forces can effect full actuator displacement in 0.25 seconds against high pressure loads, transmitting up to 0.9 J/stroke at 60% efficiency. In vitro tests also confirm that key device performance parameters can be computed from pressure readings transmitted via radiotelemetry, clearing the way for long-term implant studies in conscious animals.

Lower sternal reinforcement improves the stability of sternal closure.

BACKGROUND: This study uses a mechanical testing system to evaluate three methods of sternal closure. METHODS: Twelve sternal replicas composed of a polyurethane foam bone analogue were divided in the midline and reapproximated using three stainless steel wire techniques: six simple wires (6S), six figure-of-eight wires (6F8), or seven simple wires (7S), which included an extra wire at the lower sternum. The closures were subjected to increasing lateral distraction from 0 to 400 Newtons (N) (1 N = 0.224 lbs), and motion was measured using transducers stationed across the manubrium, midsternum, and lower sternum. RESULTS: With each method of closure, the manubrium was the most stable, the lower sternum the least stable, and the midsternum intermediate between the other two. There were also differences between sternal closure methods, but only at the lower sternum. Less sternal distraction was measured with the 7S than the 6S and 6F8 methods, starting at 100 N (0.20 +/- 0.06 mm vs 0.48 +/- 0.19 and 0.39 +/- 0.10, p = 0.003), and progressively increasing until the study was stopped at 400 N (1.64 +/- 0.39 mm vs 4.92 +/- 1.73 and 5.1 +/- 1.43 mm, p = 0.003). CONCLUSIONS: These data show that the lower sternum is the site of greatest instability and that reinforcement of this area with an additional wire effectively stabilizes the closure. Figure-of-eight wires are not superior to simple wires.

Validation of a bone analog model for studies of sternal closure.

BACKGROUND: The incidence of serious sternal wound complications may be reduced with improvements in closure methods. Biomechanical testing of median sternotomy closures in cadavers has proven useful but is limited by availability, high cost, and wide variations in the material properties of the sterna. This study tests whether artificial sterna can be used to replace whole cadavers in sternal closure testing. METHODS: Two common wire closure techniques were tested using both whole cadavers and artificial sternal models formed from bone analogue material. Sternal models were molded from polyurethane foam (20 lbs/ft3) to simulate the mechanical properties observed in human cadaveric sterna. The force vector previously identified as the most detrimental to sternal cohesion (lateral traction) was used to stress the closures. Separation of the incision site was measured at the manubrium, midsternum, and xiphoid and data were compared between cadaver and bench test groups. RESULTS: Sternal separations recorded in cadavers were found to be similar to bench test results for both closure types. Data variability within test groups was found to be consistently lower using artificial sterna, where peak standard deviations for sternal motion averaged less than half that measured in cadavers. CONCLUSIONS: Results suggest that anatomic sternal models formed from solid polyurethane foam can be used to approximate the biomechanical properties of cadaveric sterna and that reliable information regarding sternal closure stability can be secured through this means. Moreover, bench test data were shown to be less variable than cadaveric results, thus enhancing the power to detect small differences in sternal fixation stability.