LABORATORY INVESTIGATION LEFT VENTRICULAR PERFORMANCE The effects of pressure-induced right ventricular hypertrophy

To determine whether chronic pressure overload and hypertrophy of the right ventricle alter the diastolic properties of the left ventricle, six adult dogs underwent banding of the pulmonary artery and were instrumented for studies 8 months later. Fourteen control dogs were also studied. Pressure and dimension data were collected from the dogs while they were awake and unsedated. The anterior-posterior, septal-free wall, and base-apex axis diameters of the left ventricle were measured with ultrasonic dimension transducers. Right and left ventricular pressures were measured with micromanometers. Pulmonary arterial banding resulted in increased right ventricular/body mass ratios (2.70 + 0.36 g/kg vs 1.52 + 0.15 g/kg control; p ' .05) and increased left ventricular/body mass ratios (4.84 ± 0.64 g/kg vs 4.21 ± 0.49 g/kg control; p ' .05). Right ventricular peak systolic and enddiastolic pressures were higher among the banded dogs (50 ± 20/7 + 5 mm Hg vs 31 + 6/3 ± 2 mm Hg control; p ' .05). A rearrangement in the three-dimensional geometry of diastolic filling occurred in the banded dogs. Extension from unstressed diastolic dimension (strain) in the base-apex axis was significantly larger in the banded dogs at left ventricular transmural pressures of 12, 8, and 4 mm Hg; strains in the septal-free wall axis were significantly smaller at transmural pressures of 12 and 8 mm Hg. Normalized diastolic left ventricular pressure-volume data and midwall circumferential stressstrain data were fit to the Kelvin viscoelastic equation. The normalized pressure-volume relationships of the banded dogs lay significantly to the left of those of the controls, indicating a loss of left ventricular chamber compliance. The midwall circumferential stress-strain relationships of the banded dogs were also shifted to the left, indicating a loss of intrinsic myocardial compliance. Thus, during the course of right ventricular hypertrophy caused by right ventricular pressure overload, alterations in the mass, geometry, and material properties of the left ventricle occur. At 8 months the chamber compliance of the left ventricle is compromised by these changes. Circulation 74, No. 2, 410-419, 1986. RECENT CLINICAL DATA suggest that in some forms of cyanotic congenital heart disease with pressure or volume overload of the pulmonary ventricle, the systemic ventricle may function abnormally and continue to function abnormally after operative correction. 1-3 When dysfunction of the systemic ventricle is noted during the course of these diseases, it would be useful to identify a component of this dysfunction that is strictly attributable to elevated pressures in the pulFrom the Department of Surgery, University of Minnesota, Minneapolis. Supported by NIH grants HL-22152 and HL-20598 from the USPHS, NIH Research Fellowships HL-05704, HL-05759, HL-06001, and HL06143, and a grant from the American Heart Association, Minnesota Affiliate. Address for correspondence: Marc S. Visner, M.D., Department of Surgery, Georgetown University Hospital, 3800 Reservoir Road, N.W., Washington, DC 20007. Received July 1, 1985; revision accepted April 24, 1986. monary ventricle, since with surgical correction and restoration of normal pressures in the pulmonary ventricle, this portion of the abnormality may relent. In contrast, dysfunction of the systemic ventricle that is related to a change in the intrinsic material properties of the myocardium may represent a slowly reversible or irreversible process, and its onset during the course of congenital heart disease may be an indication for immediate operative intervention. The results of clinical studies suggest that in the setting of acyanotic congenital lesions, chronic right ventricular hypertension is the cause of reversible abnormalities in left ventricular dynamic geometry and systolic function.4'In patients with an atrial septal defect, the interventricular septum is flattened and displaced toward the left ventricle at end-diastole. There is brisk motion of the septum toward the right ventricle CIRCULATION 410 by gest on A ril 4, 2017 http://ciajournals.org/ D ow nladed from LABORATORY INVESTIGATION-LEFT VENTRICULAR PERFORMANCE during early systole, and only during late ejection is there effective left ventricular contraction in the septal-free wall axis. After closure of the atrial septal defect and normalization of right ventricular pressures, septal geometry and left ventricular systolic function usually become normal.7' 8 Similar changes in septal position and motion have been demonstrated during acute right ventricular hypertension induced by pulmonary arterial constriction in conscious dogs.9' 10 The reversibility of these abnormalities in the setting of chronic right ventricular hypertension and their presence during acute right ventricular hypertension suggest that they are the direct result of elevated rightsided pressures rather than the byproduct of some change in the intrinsic properties of the left ventricular myocardium. The present study was undertaken to document whether the same or similar rearrangements in left ventricular geometry occur in a preparation of chronic right ventricular pressure overload, and to determine to what extent in the intact circulation left ventricular chamber compliance is altered by these rearrangements or by other concomitants of pressureinduced right ventricular hypertrophy. Materials and methods Experimental preparation. Six healthy adult dogs (20 to 30 kg) underwent banding of the pulmonary artery through a right thoracotomy. Right ventricular pressures were monitored during the procedure. The band was slowly tightened until peak systolic right ventricular pressure exceeded 80 mm Hg or until frequent ventricular extrasystoles developed. The azygos vein was ligated during this initial procedure to ensure inflow stasis during vena caval occlusions performed during subsequent studies. Eight months later, these six animals and 14 controls of comparable size were subjected to a left thoracotomy under general anesthesia (pentobarbital 30 mg/kg) for the implantation of instrumentation to collect pressure and dimension data. The experimental preparation has been described previously.9' 10 Briefly, three pairs of ultrasonic dimension transducers were implanted to measure the anterior-posterior minor axis, septalfree wall minor axis, and base-apex major axis of the left ventricle. The anterior-posterior minor axis and base-apex major axis transducers were sewn to the epicardium to measure external diameters. The septal-free wall diameter was measured from the septal midwall to the epicardium of the lateral free wall. Silicone rubber catheters were inserted into the right ventricle and left atrium so that during subsequent studies micromanometers could be introduced to measure right and left ventricular pressures. A third silicone rubber catheter closed at its distal end by a compliant silicone rubber balloon (%/oo inch thickness) was positioned in the left pleural cavity at the level of the aortic arch to measure intrapleural pressure. A fluid-filled polyvinyl chloride catheter was inserted through the left internal mammary artery into the aortic arch to measure aortic pressure. Inflatable silicone rubber occluders were placed around the two venae cavae. The electrical leads, catheters, and occluders were all exteriorized dorsal to the thoracotomy incision. The pericardium was left open and the chest closed. Postoperatively the dogs received intramuscular injections of dihydrostreptomycin (0.75 g) and penicillin (6 x 10 U) for 3 days. Instrumentation and data acquisition. Each dog was allowed to recover from surgery for a week before being studied. Data were collected while the dogs were awake and unsedated. Left and right ventricular pressures were measured with catheter-tipped micromanometers (Millar PC-350). They were driven with Hewlett-Packard 8805C carrier preamplifiers and were zeroed and balanced to atmospheric pressure at 380 C. Zero drift of each transducer did not exceed 0.5 mm Hg during any study. The sonomicrometer used in these experiments was constructed in this laboratory. It converts the transit time of a pulse of ultrasound between two piezoelectric crystals into an analog signal."1 Since the velocity of sound is constant in tissues and blood, the measured transit time is directly proportional to the distance between the transducers. The sampling rate of the device is 1 kHz, and its resolution is 0.05 mm. Pleural and aortic pressures were measured with external transducers (Statham P23Db) connected to fluid-filled catheters. A No. 6F catheter (U.S.C.I.) was inserted through the previously implanted pleural catheter to measure pleural pressure. The catheter-balloon system was filled with between 1.0 and 2.0 ml of saline before insertion of the No. 6F catheter. This volume was enough to fill the system with the No. 6F catheter in place and to measure 0.0 mm Hg pressure during testing before implantation. Aortic pressure was monitored with the polyvinyl chloride catheter previously implanted in the aortic arch. The external transducers were zeroed and balanced to atmospheric pressure at the mid chest level. The zero drift of the transducer measuring pleural pressure did not exceed 0.5 mm Hg during the course of any study. Analog data representing pressures and dimensions were recorded onto magnetic tape with a Hewlett-Packard 3968A tape recorder. Data were collected from each dog during a baseline period and during a vena caval occlusion. To perform the vena caval occlusion, the vena caval occluders were inflated so that peak systolic left ventricular pressure was gradually reduced to 25 mm Hg over a 30 sec period. Each dog was killed and autopsies were performed at the conclusion of the experiment. The position of the septal transducer was ascertained to be within 2mm of the midwall position in each dog. The mass of the right ventricular free wall and the mass of the left ventricle including the interventricular septum were measured. The left ventricular papillary muscles were then removed and left ventricular mass again determined. Data analysis. The analog signals representing pressures and dimensions were digitized at 5 msec intervals on a PDP 11/34 computer (Digital Equipment Corp.). The left ventricle was modeled as a generalized ellipsoidal shell with three orthogonal axes. The measured base-apex axis was considered the external major axis diameter of the shell. The measured anterior-posterior dimension was considered one of the two external equatorial minor axis diameters. The measured septal-free wall dimension was set equal to the external equatorial septal-free wall diameter minus one-half of a calculated equatorial wall thickness. The dimension of left ventricular equatorial wall thickness was calculated from the three measured left ventricular axis dimensions and the postmortem mass of the left ventricle without its papillary muscles. The external volume (Ve) of the shell is equal to: Ve = 7r/6(a)(b)(c + 0.5h) (1) where a, b, and c are the measured major axis, anterior-posterior minor axis, and septal-free wall minor axis, respectively, and his equatorial wall thickness. The inner volume (Vi) of the shell is equal to: Vi = 7T/6(a 1.lh)(b 2h)(c 1.5h) (2) Vol. 74, No. 2, August 1986 411 by gest on A ril 4, 2017 http://ciajournals.org/ D ow nladed from