Method for measuring long-term function of muscle-powered implants via radiotelemetry.

Long-term remote monitoring of muscle-powered implants has been made possible with development of an adjustable workload that can be remotely monitored to assess device function. This technique obviates the need for percutaneous access lines and allows test animals to remain untethered, eliminating deleterious effects caused by infection, sedation, or animal stress. Hardware components include a latex bladder fixed within a hermetically sealed canister, multichannel implantable telemetry unit, and subcutaneous access port (for pressure charge adjustment). To validate this method, in vitro tests were performed by using a third-generation muscle energy converter designed to function as an implantable hydraulic pump. Two channels of telemetered pressure data were collected and used to calculate six indexes of device function. Calculated parameters were then compared with measured values to determine accuracy. Correlation between measured and calculated parameters was high in all instances, with most estimates yielding errors of <3%. These results demonstrate the utility of this approach and support its use as a means to monitor muscle-powered devices during long-term animal trials.

Muscle powered blood pump: design and initial test results.

A pneumatic ventricular assist device (Sarns/3M) has been redesigned for low volume hydraulic actuation to accommodate muscle powered drive systems. Design modifications include adding a bellows/piston mechanism (to compress the blood sac) and a compliance chamber for volume compensation. A simple prototype device was constructed to measure the efficacy of piston pump actuation and to validate pusher plate design. Device manufacture was affected by removing the drive line housing from the pneumatic pump and replacing it with a piston/bushing mechanism. A convex piston profile was chosen to maximize ejection fraction and minimize device size. Stroke volume was found to be a linear function of piston displacement (approximately 3 ml/mm) and reached a maximum value of 45 ml. Mean compression forces of 46-56 N acting during a 12 mm stroke (2.1 L/min at 60 cycles/min) were sufficient to generate mean afterload pressures of 70-110 mm Hg in a mock circulatory loop. Peak compression forces ranged from 72 to 86 N and work input was calculated to be 552-672 mJ/stroke. These data indicate that this method for delivering muscle power to the bloodstream is both mechanically viable and compatible with the functional capacity of conditioned latissimus dorsi muscle.

A muscle-powered energy delivery system and means for chronic in vivo testing.

Electrically stimulated skeletal muscle represents a potentially unlimited source of energy for the actuation of motor prostheses. Devices to harvest and deliver contractile power have proven mechanically feasible, but long-term efficacy has not been demonstrated. This report describes recent refinements in muscle energy converter (MEC) design and details the development of an implantable afterload chamber (IAC) designed to facilitate implant testing. The IAC comprises a fluid-filled bladder housed within a titanium cylinder that connects directly to the MEC. A vascular access port allows percutaneous measurement and adjustment of air pressure within the housing and provides a means both to monitor MEC function and to control hydraulic loading conditions. Data from in vitro tests show that IAC pressure mirrors changes in MEC-piston displacement over a wide range of actuation speeds and stroke lengths. Stroke lengths and actuation forces calculated from IAC pressure readings were typically found to be within 5% of measured values. This testing scheme may yield important information in regard to the ability to harness energy from in situ muscle over prolonged periods.

Copulsation balloon for right ventricular assistance: preliminary trials.

BACKGROUND: Options for management of acute right ventricular (RV) failure are limited. This report describes preliminary testing of a temporary RV assist device that acts by direct compression of the RV. The system comprises a pancake-shaped silicone balloon (5 cm diameter) connected to a drive console that delivers a 65-mL pneumatic pulse during cardiac systole. METHODS AND RESULTS: Initial in vivo tests were performed on 6 pigs (weight, 41+/-4 kg). RV wall motion and stroke volume were monitored via transesophageal echocardiography. Acute RV failure was created by graded right coronary ligation, which yielded a 63% reduction in RV stroke volume (39.9+/-8.2 to 14.7+/-1.9 mL; P<0.002). We secured the balloon over the RV free wall by attaching it to the edges of the opened pericardium. The sternum was then reapproximated, and data were collected with the device off and on (every beat). Device placement had no deleterious effect on RV function. Balloon activation returned RV stroke volumes to normal (37.8+/-9.2 mL) and increased mean pulmonary artery pressures from 13+/-2 to 16+/-3 mm Hg (P<0.01). RV compression did not induce or exacerbate tricuspid regurgitation. Mean aortic pressure improved from postinfarction levels but did not return to normal. CONCLUSIONS: We conclude that the pulmonary circulation can be supported in the short term via cardiac compression and that balloon copulsation techniques for short-term RV failure should be tested in long-term models.

Mechanical analysis of midline sternotomy wound closure.

OBJECTIVE: Unstable median sternotomy closure can lead to postoperative morbidity. This study tests the hypothesis that separation of the sternotomy site occurs when physiologic forces act on the closure. METHODS: Median sternotomy was performed in 4 human cadavers (2 male) and closed with 7 interrupted stainless steel wires. The chest wall was instrumented to apply 4 types of distracting force: (1) lateral, (2) anterior-posterior, (3) rostral-caudal, and (4) a simulated Valsalva force. Forces were applied in each direction and were limited to physiologic levels (< 400 N). Four sets of sonomicrometry crystals were placed equidistantly along the sternum to measure separation at the closure site. RESULTS: Sternal separation occurred as a result of the wires cutting through the bone. Less force was needed to achieve 2.0-mm distraction in the lateral direction (220 +/- 40 N) than in the anterior-posterior (263 +/- 74 N) and rostral-caudal (325 +/- 30 N) directions. More separation occurred at the lower end of the sternum than the upper. During lateral distraction, xiphoid and manubrial displacement averaged 1.85 +/- 0.14 and 0.35 +/- 0.12 mm, respectively. Anterior-posterior distraction caused 1.99 +/- 0.04-mm xiphoid displacement and 0.26 +/- 0.12-mm manubrial displacement. During a simulated Valsalva force, more separation occurred in the lateral (2.14 +/- 0.11 mm) than in the anterior-posterior (0.46 +/- 0.29 mm) or rostral-caudal (0.25 +/- 0.15 mm) directions. CONCLUSIONS: These data suggest that sternal dehiscence can occur under physiologic loads and that improved sternal stability may be readily achieved via mechanical reinforcement near the xiphoid. Closure techniques designed to minimize wire migration into the sternum should also be developed.

Intermittent stimulation enhances function of conditioned muscle.

Skeletal muscle is highly adaptable in that its metabolic and contractile characteristics are largely regulated by its pattern of use. It is known that muscle phenotype can be manipulated via chronic electrical stimulation to enhance fatigue resistance at the expense of contractile power. Type 2A fibers are fatigue resistant, powerful, and considered most desirable for cardiac assist purposes. We have found that 12-wk of intermittent-burst stimulation produces a high percentage of 2A fibers and increases fatigue resistance and power in rabbit latissimus dorsi muscle. Fixed-load endurance tests were used to quantify fatigue resistance among normal and trained muscle groups. Control muscles were found to fatigue completely within 10-20 min. Muscles stimulated continuously for 6 wk retained 35% (71.5 +/- 19.5 g. cm) of their initial stroke work at 40 min. Muscles stimulated 12 h/day for 12 wk had the highest initial stroke work (449.7 +/- 92.4 g. cm) and the highest remaining stroke work (234.7 +/- 50.1 g. cm) at 40 min. Results suggest that employing regular resting periods during conditioning preserves strength in fatigue-resistant muscle.

Functional properties of conditioned skeletal muscle: implications for muscle-powered cardiac assist.

Latissimus dorsi (LD) muscles of six canines were studied to assess changes induced by electrical conditioning and to quantify the capacity of these muscles to perform hemodynamic work. Muscles were conditioned using burst stimuli delivered over an 8-wk period. Contralateral LD were used as control. Muscles were tested in situ to simulate anticipated linear-pull cardiac assist conditions. This training process reduced muscle mass and cross-sectional area by 16 and 17%, respectively. Muscle phenotype shifted to a predominantly "slow" form by coordinate reduction of myosin heavy chain (MHC) 2A expression and increased expression of the MHC beta/slow form. Force generation was reduced by 54%, and contractile duration increased 13%. Fatigue resistance was markedly enhanced, and chronic stroke work increased from 0.19 to 0.72 mJ/g. The highest steady-state power output (2.06 mW/g) was obtained from one muscle fully converted to a slow phenotype. These data suggest that single LD trained via conventional techniques can provide energy sufficient for partial cardiac assistance but cannot sustain work levels needed to achieve total circulatory support.

A permanent prosthesis for converting in situ muscle contractions into hydraulic power for cardiac assist.

The key to utilizing muscle power for circulatory support lies with the development of a practical scheme by which contractile energy may be collected and efficiently delivered to the bloodstream. This work describes initial in vitro testing of a prototype muscle energy converter (MEC) designed to transform the power of in situ muscle contractions into hydraulic form. The MEC resembles a simple piston pump and is designed for implant beneath the humeral insertion of the latissimus dorsi muscle. Bench tests were conducted to measure component function and to characterize device performance under various hydraulic loads. Under simulated muscle-pull conditions, MEC energy transfer capacity was found to be 170 mJ/stroke while operating at peak efficiencies (i.e., > 98% of input power converted into hydraulic energy and preload work). Transfer efficiencies dropped from 96 to 38% as mean generated pressures increased from 23 to 36 N/cm2 due to metal bellows flexion. These results demonstrate that a significant amount of contractile energy can be efficiently transformed to hydraulic power via this mechanism.

Regional effects of aortomyoplasty in acute ischemia.

BACKGROUND: Aortomyoplasty is a technique for achieving autogenous diastolic counterpulsation. This experiment was designed to determine if aortomyoplasty using conditioned latissimus dorsi muscle could improve regional myocardial function during coronary ischemia. METHODS: Six mongrel dogs underwent a staged operation in which the left latissimus dorsi was conditioned in situ for 4 weeks, then wrapped around the descending aorta and stimulated during diastole with each cardiac contraction. Regional ischemia was caused by occlusion of the left anterior descending coronary artery. Regional function was measured with somomicrometry in the region of ischemia and in a control area. An intraaortic balloon pump was inserted for comparison with aortomyoplasty performance. RESULTS: Coronary artery occlusion caused a significant decrease in the percentage of regional shortening (14.2 +/- 7.9 to -2.2 +/- 4.0; p = 0.001) and thickening (11.9 +/- 4.6 to -5.8 +/- 3.3; p < 0.001). Aortomyoplasty improved regional motion in both percentage shortening (-2.2 +/- 4.0 to 2.3 +/- 3.7; p = 0.008) and thickening (-5.8 +/- 3.3 to 2.8 +/- 1.9; p < 0.001). The intraaortic balloon pump also improved percentage shortening (-3.7 +/- 2.0 to 0.7 +/- 1.9; p = 0.01) and thickening (-5.0 +/- 2.8 to 2.4 +/- 3.8; p < 0.001), and was not significantly different than aortomyoplasty. CONCLUSIONS: These data show that aortomyoplasty has beneficial effects on ischemic left ventricular contractility, and may therefore be useful for treating inoperable coronary artery disease.

Ergometric studies of untrained skeletal muscle demonstrate feasibility of muscle-powered cardiac assistance.

The feasibility of biomechanical circulatory assistance hinges on the capacity of skeletal muscle to generate significant hemodynamic work. This study quantifies linear contractile energetics via a customized hydraulic ergometer. Six normal canine latissimus dorsi (LD) muscles (200 +/- 25 g) were evaluated. The muscles were not mobilized; thereby their collateral circulation was preserved. The humeral insertion of the LD muscle was transected and connected to the ergometer. Preload was adjusted to return the LD muscle to its in situ length, and one pulse train was delivered every second. The resulting contractions generated peak pressures of 134 +/- 17 mmHg with mean pressures during shortening of 102 +/- 12 mmHg. Flow rates averaged 5.45 +/- 0.26 l/min. Mechanical work output was calculated at 1.14 +/- 0.18 J/contraction, yielding an average power production of 4.57 +/- 0.72 W during shortening. Continuous LD output power, measured at 5.76 +/- 0.90 mW/g, compares favorably with the 3.48 mW/g typically generated by a 350-g human heart. We therefore conclude that skeletal muscle of sufficient mass can sustain work rates suitable for cardiac assistance despite the 50% power losses typically experienced after muscle training.

Dynamic descending thoracic aortomyoplasty: comparison with intraaortic balloon pump in a model of heart failure.

Descending thoracic aortomyoplasty (DTA) uses the latissimus dorsi muscle to compress the proximal descending thoracic aorta as an autogenous diastolic counterpulsator. We studied the hypothesis that DTA could confer hemodynamic benefits equivalent to those yielded by an intraaortic balloon pump (IABP) in dogs (n = 7) with heart failure. The left latissimus dorsi muscle was wrapped around the proximal thoracic aorta and subsequently electrically conditioned to induce fatigue resistance. Heart failure was produced by rapid ventricular pacing after muscle conditioning. Data were collected under three conditions: (1) after the induction of heart failure; (2) with the 20-mL IABP at 1:1; and (3) with the DTA stimulated at 1:1. Effective diastolic counterpulsation was achieved with both the IABP and the DTA. The mean diastolic aortic pressure increased from 66 +/- 5 mm Hg at baseline to 90 +/- 4 mm Hg with the IABP and to 75 +/- 4 mm Hg with the DTA. The left ventricular peak and end-diastolic pressures decreased with IABP (95 +/- 5 mm Hg versus 88 +/- 4 mm Hg and 16 +/- 4 mm Hg versus 12 +/- 4 mm Hg, respectively; p < 0.05) and with DTA (95 +/- 5 mm Hg versus 87 +/- 4 mm Hg and 16 +/- 4 mm Hg versus 12 +/- 4 mm Hg, respectively; p < 0.05). Counterpulsation with the IABP did not change the end-systolic pressure-volume relationship or the time constant for diastolic relaxation, whereas the DTA increased the end-systolic pressure-volume relationship (3.2 +/- 0.6 mm Hg/mL versus 4.0 +/- 0.7 mm Hg/mL; p < 0.05) and decreased the time constant for diastolic relaxation (49 +/- 5 msec versus 45 +/- 6 msec; p < 0.05). These data show that DTA using conditioned skeletal muscle can provide diastolic counterpulsation in animals with compromised cardiac function.(ABSTRACT TRUNCATED AT 250 WORDS)

Experimental studies on heterotopic lung transplantation during temporary pulmonary insufficiency.

Survival from reversible forms of severe pulmonary insufficiency remains dismal despite the development of artificial oxygenators. We hypothesized that an intraabdominal heterotopic lung could help maintain adequate oxygenation during acute pulmonary insufficiency. Five mongrel dogs underwent an acute heterotopic lung transplant (HLT). The left atrial cuff was anastomosed to the inferior vena cava, and the left pulmonary artery was anastomosed to the abdominal aorta. The trachea was exteriorized, intubated, and ventilated with a volume-controlled ventilator. Ventilation to the native lungs was discontinued. The heterotopic lung was then ventilated at a rate of 20/min, tidal volume of 15 ml/kg, and inspired concentration (FIO2) of 50 percent. Partial pressure of oxygen (PO2) and mixed venous oxygen saturation (SvO2) were maintained at 53 +/- 5.2 mm Hg and 71 +/- 12 percent, respectively. Flow through the HLT was approximately 20 percent of the systemic cardiac output and did not vary with changes in FIO2, respiratory rate, or positive end-expiratory pressure (PEEP). Four separate animals underwent HLT and were studied 2 to 3 days later. The FIO2 was reduced in the native lungs to 10 percent until SaO2 was less than 90 percent. The HLT was then ventilated at a tidal volume of 300 ml, an FIO2 of 50 percent, and a respiratory rate of 10. Arterial PO2 increased from 62 +/- 4 mm Hg to 75 +/- 2 mm Hg, and SvO2 increased from 75 +/- 2 percent to 82 +/- 3 percent (p < 0.05). Flow through the HLT increased slightly to 27 percent of the systemic cardiac output. We conclude that a HLT can augment oxygenation after induction of moderate hypoxemia, but cannot serve as the sole source for gas exchange because flow through the HLT is limited to less than 30 percent of the cardiac output.

Chronic counterpulsation with descending thoracic aortomyoplasty improved cardiac function in animals with heart failure.

Descending thoracic aortomyoplasty uses latissimus dorsi muscle for extraaortic diastolic counterpulsation. We hypothesized that descending thoracic aortomyoplasty could improve ventricular function in dogs (n = 5) with heart failure. The left latissimus dorsi muscle was wrapped around the descending aorta and conditioned for 4 weeks with a burst stimulator (five pulses, 33 Hz, 28 bursts/min). Heart failure was induced by rapid ventricular pacing after conditioning. Left ventricular volume was measured with a conductance catheter. Left ventricular and aortic pressures were measured with a micromanometer. Mean diastolic blood pressure, endocardial viability ratio, left ventricular peak pressure, left ventricular end-diastolic pressure, stroke work, isovolumic relaxation time constant, and the end-systolic pressure volume relation were measured at baseline (after heart failure) and with the descending thoracic aortomyoplasty stimulated at 1:1. Contraction of the descending thoracic aortomyoplasty augmented mean diastolic blood pressure (62 +/- 4 to 71 +/- 3 mm Hg) and endocardial viability ratio (1.0 +/- 0.30 to 1.5 +/- 0.13) (p < 0.05). Left ventricular peak pressure (98 +/- 4 to 88 +/- 3 mm Hg), left ventricular end-diastolic pressure (19 +/- 4 to 14 +/- 4 mm Hg), and stroke work (1048 +/- 124 to 743 +/- 80 mm Hg.cm3) (p < 0.05) were reduced. The end-systolic pressure volume relation increased with descending thoracic aortomyoplasty stimulation (3.7 +/- 0.7 to 4.5 +/- 0.8 mm Hg/mL), and the isovolumic diastolic relaxation time constant significantly decreased (54 +/- 6 to 49 +/- 7 msec) (p < 0.05). We conclude that descending thoracic aortomyoplasty can provide diastolic counterpulsation and reduce stroke work in animals with heart failure.(ABSTRACT TRUNCATED AT 250 WORDS)

Autogenous cardiac assist with chronic descending thoracic aortomyoplasty.

Alternative surgical treatments to orthotopic cardiac transplantation are needed for patients with heart failure. We hypothesized that descending thoracic aortomyoplasty with conditioned (fatigue-resistant) latissimus dorsi muscle could provide diastolic augmentation that would improve left ventricular function. Six mongrel dogs were studied. The left latissimus dorsi muscle was wrapped clockwise around the descending thoracic aorta. Left ventricular volume was measured with a conductance catheter. Aortic and left ventricular pressures were measured with a micromanometer. The following were measured after descending thoracic aortomyoplasty at baseline and with the descending thoracic aortomyoplasty stimulated 1:1 with the heart rate: stroke work, stroke volume, left ventricular peak pressure, maximum rate of increase of left ventricular pressure, diastolic relaxation time constant, peak rate of pressure decay, left ventricular end-diastolic pressure, endocardial viability ratio, mean diastolic aortic pressure, peak diastolic aortic pressure, and time-averaged aortic diastolic velocity. Before data collection, the latissimus dorsi was stimulated (5 pulses delivered at 33 Hz at a rate of 28 per minute for 4 weeks) with burst stimulation to induce fatigue resistance. Results (expressed as the mean +/- the standard error of the mean) showed significant improvement in the indices of ventricular contractility (maximum rate of increase of left ventricular pressure, 1,217 +/- 83 to 1,414 +/- 91 mm Hg/s) and diastolic relaxation mechanics (peak rate of pressure decay, 1,152 +/- 92 to 1,282 +/- 79 mm Hg/s; diastolic relaxation time constant, 43 +/- 2 to 38 +/- 2 ms). Significant differences were noted with stimulation at 1:1 in the endocardial viability ratio (0.90 +/- 0.05 to 1.14 +/- 0.04), an index of myocardial oxygen supply. Systemic diastolic pressures (peak diastolic aortic pressure, 95 +/- 6 to 107 +/- 5 mm Hg; mean diastolic aortic pressure, 92 +/- 6 to 102 +/- 6 mm Hg) and the time-averaged aortic diastolic velocity (1.5 +/- 0.6 to 3.3 +/- 1.0 m/s) increased significantly. We conclude that descending thoracic aortomyoplasty stimulation with conditioned latissimus dorsi muscle can improve indices of ventricular contractility, diastolic relaxation mechanics, diastolic pressures, and diastolic aortic velocity in the nonfailed canine heart. Further studies with the chronic failed heart model are required.

Right latissimus dorsi cardiomyoplasty augments left ventricular systolic performance.

We hypothesized that the right latissimus dorsi cardiomyoplasty augments left ventricular performance. Five dogs underwent staged right latissimus dorsi cardiomyoplasty. Ventricular function was studied 1 to 3 weeks later. Left ventricular pressure was measured with a micromanometer and left ventricular dimensions with piezoelectric crystals. Inferior vena caval occlusion was used to vary preload. Pressure-volume data were collected with the muscle unstimulated and stimulated at 1:2 and 1:1 muscle/heart ratios. The end-systolic pressure-volume relation (mm Hg/mL), stroke work, preload recruitable stroke work, left ventricular end-diastolic volume, and the diastolic relaxation constant were calculated and expressed as mean +/- standard deviation. Stimulated beats at a 1:2 ratio showed an increase in stroke work of 42.1% (978 +/- 381 to 1,390 +/- 449 g.cm; p < 0.01) and preload recruitable stroke work of 28.8% (59.4 +/- 20.7 to 76.6 +/- 11.0 g.cm/cm3; p = 0.05) compared with the unstimulated beats. With the stimulator on at 1:1, smaller changes occurred: stroke work increased 9% (1,167 +/- 390 to 1,273 +/- 363 g.cm; not significant) and preload recruitable stroke work increased 27% (63.9 +/- 22.7 to 80.9 +/- 23.1 g.cm/cm3; p = 0.05). There were no significant changes in the end-systolic pressure-volume relation. The diastolic relaxation constant did not change at 1:1 (36 +/- 9.7 to 37 +/- 6.4 ms; not significant) or 1:2 (36 +/- 9.3 to 39 +/- 8.2 ms; not significant). Left ventricular end-diastolic volume was unchanged at 1:1 (34 +/- 10.7 to 32 +/- 10.3 mL) and at 1:2 (31 +/- 9.0 to 32 +/- 8.7 mL).(ABSTRACT TRUNCATED AT 250 WORDS)

Early effects of right latissimus dorsi cardiomyoplasty on left ventricular function.

BACKGROUND: We hypothesized that left ventricular function could be improved with cardiomyoplasty using the right latissimus dorsi. METHODS AND RESULTS: Five dogs underwent cardiomyoplasty using the right latissimus dorsi. Left ventricular volume and pressure were measured using sonomicrometry and a micromanometer catheter, respectively. Pressure volume loops were recorded with the muscle stimulated at 1:2 and with transient caval occlusion. During stimulated beats, there were significant increases in stroke work (13.90 +/- 4.49 vs 9.78 +/- 3.81 g/m, P < .01), preload recruitable stroke work (0.766 +/- 0.110 vs 0.594 +/- 0.207 g.m-1 x m-3, P < .05), and stroke volume (15 +/- 4 vs 10 +/- 3 mL, P < .05) when compared with unstimulated beats. There were no changes in diastolic filling. This operation was done in 11 patients, with no operative deaths. Six weeks after surgery, resting left ventricular ejection fraction (LVEF) increased from 25 +/- 1.6% to 35 +/- 3% (P < .05), and left ventricular end-diastolic volume (LVEDV) decreased from 365 +/- 18 to 307 +/- 24 mL, (P < .05). Nine patients were alive at 6 months. Preoperative and 6-month LVEF and LVEDV for those 9 patients were 26 +/- 2% and 29 +/- 2% (P = NS) and 316 +/- 23 and 261 +/- 22 mL (P < .05), respectively. CONCLUSIONS: Long-term studies are needed to determine if these changes will improve patient survival.

Static left latissimus dorsi cardiomyoplasty: effect on left ventricular function.

When the latissimus dorsi is used for ventricular augmentation in cardiomyoplasty, a delay of several weeks occurs before the muscle revascularizes, adheres to the heart, and is transformed to fatigue-resistant status. This study analyzes the effect of static (unstimulated) cardiomyoplasty on left ventricular function. Four mongrel dogs underwent staged left latissimus dorsi cardiomyoplasty. Left ventricular pressure was measured with a micromanometer catheter. Left ventricular volume was measured by sonomicrometry. Cardiac output, heart rate, preload recruitable stroke work, maximum elastance, left ventricular end-diastolic volume, left ventricular end-diastolic pressure, stroke work, and the diastolic relaxation constant were measured before and immediately after cardiomyoplasty with the myoplasty static. Results, expressed as mean +/- standard error of the mean, showed no significant differences in indexes of systolic function (stroke work, 1017 +/- 223 gm.cm to 984 +/- 403 gm.cm; preload recruitable stroke work, 110 +/- 13 gm.cm/cm3 to 115 +/- 19.8 gm.cm/cm3; maximum elastance, 10.38 +/- 5.6 mm Hg/ml to 13.59 +/- 6.5 mm Hg/ml; cardiac output 4.51 +/- 0.43 L/min to 4.21 +/- 0.34 L/min) or diastolic function (left ventricular end-diastolic volume, 21 +/- 5.2 ml to 20 +/- 5.3 ml; left ventricular end-diastolic pressure, 13 +/- 3.5 mm Hg to 15 +/- 3 mm Hg; diastolic relaxation constant 42.8 +/- 5.2 msec to 42.5 +/- 4.5 msec). Heart rate also remained unchanged (131 +/- 8.9 beats/min to 140 +/- 9.8 beats/min). The static (unstimulated) left latissimus dorsi cardioplasty can be done with little effect on left ventricular systolic or diastolic function in the normal canine heart.

Component analysis of bilateral anterior cardiomyoplasty.

This experiment was designed to analyze the mechanism of ventricular augmentation generated with bilateral anterior cardiomyoplasty by comparing it in an acute setting with its components, the left anterior cardiomyoplasty and the right anterior cardiomyoplasty. Hemodynamic variables were measured in 8 dogs before and after each flap was positioned around the heart. Stimulation was achieved by R-wave synchronous latissimus burst pacing at a ratio of 1:4, an amplitude of 5 V, a frequency of 33 Hz, and a duration of 30% of the R-R interval. Hemodynamic changes were again recorded during latissimus stimulation. Construction of the bilateral anterior wrap (static cardiomyoplasty) caused some depression of baseline hemodynamic function, which was greater than that caused by either the static right or left anterior cardiomyoplasty. With stimulation of the muscles (dynamic cardiomyoplasty), the bilateral wrap caused significant biventricular augmentation. Evaluation of the components of the bilateral wrap demonstrated that dynamic right anterior cardiomyoplasty also provided significant biventricular augmentation, but the dynamic left anterior cardiomyoplasty augmented only right-sided variables. The mechanism of biventricular compression by the bilateral procedure is due mostly to the right wrap. The right anterior cardiomyoplasty may provide significant biventricular compression for treatment of heart failure, without the complexity associated with bilateral anterior cardiomyoplasty.

Development of an implantable centrifugal blood pump.

The efficacy of centrifugal pumps for short-term (0-30 days) ventricular support has been widely reported and favorably compared with pulsatile systems. A small, durable, implantable centrifugal blood pump is being developed for medium-term use (up to 6 months). The pump is based on the Medtronic Hemadyne system that has existed in multiple forms over the past 30 years. The pump is approximately the size of a tennis ball, weighs 240 g, and is comprised of a 2.5 cm plastic impeller driven by a radially coupled brushless DC motor. In vitro hydraulic performance was recorded over a wide range of flow conditions on a mock circulatory loop. The pump generated 7 L/min flow against an afterload of 100 mmHg pressure, with a maximum power draw of 10.4 watts. Pulsatile flow was preserved when placed in conjunction with a simulated left ventricle. In vivo testing was performed in 10 healthy sheep for 10-292 hr. Heparin was used to facilitate cannulation, and no anticoagulation was administered after pump implantation. Blood chemistries reflecting hematologic, pulmonary, renal, and hepatic functions were recorded and demonstrated no adverse effects with normal pump operation. Complications were related to kinking of blood conduits and thrombus formation within the cannulae. These results are encouraging and warrant further studies to prove feasibility of this pump as a medium-term implantable ventricular assist device.