003 1-3998/93/3304-0359$03.00/0 PEDIATRIC RESEARCH Vol. 33. No. 1. 1993 Copyright C 1993 International Pediatric Research Foundation. Inc. Prrnrcd 1n L' S .-I.

Mechanically Increased Right Ventricular Afterload Alters Left Ventricular Configuration, Not Contractility, in Neonatal Lambs

JAY M. MILSTEIN AND STANTON A. GLANTZ Division ~f'.Yeonatolog)'.Department ofPc~diatrics. Cn'niversit~, yf Cal~fornia.Davis. California 9.5616 /J..\f..If.]: and Cardiorasc~rlarResearch Instit~tte.Department of.2fedicine. LV~i~ersit).(?fCaljfornia. Sun Francisco. Cal[fili,rnia94143 [S.A.G.]

ABSTRAn. Left ventricular dysfunction has been ob- structurally and functionally immature compared with adult served in human infants with . The myocardium (1. 2), yet the resting indices of fetal myocardial purpose of this study was to establish whether mechani- performanct. al-e ielaiii;eij;high comi;a:ed :.:i?h those nf ?he rdu!? cally increased right ventricular afterload alters left ven- (3. 4). Because the newborn is already functioning near tricular performance by altering its contractility, configu- maximum contractility with little further reserve (5-9), direct ration, or both. Six neonatal 3- to 7-d-old lambs were ventricular interaction may play a more important role in allow- acutely instrumented with micromanometer-tipped cathe- ing the left to respond to changes in loading conditions ters and two pairs of ultrasonic crystals to measure left than it does in adults. The intraventricular septum is more ventricular pressure and anterior-posterior and septal-free compliant in neonates than in fetuses. increasing the potential wall dimensions. The product of these two dimensions, for direct mechanical interaction between the right and left heart denoted left ventricular area, was used as an index of left (10). Some infants with persistent pulmonary hypertension of ventricular volume. TWOlevels of mechanically increased the newborn exhibit left ventricular dysfunction ( 1 1- 14). Simi- right ventricular afterload were induced with a closed per- larly. adults with acute pulmonary hypertension secondary to icardium under three levels of left ventricular acute respiratory failure or cor pulmonale secondary to chronic produced by whole-blood transfusions. Four brief increases lung disease have associated left ventricular dysfunction ( 15- 17). in left ventricular afterload were induced by constricting Ventricular interaction may be partially responsible for this the under each right ventricular afterload and preload association (18-21). The purpose of this study was to establish condition. Using multiple linear regression, we found that whether mechanically increased right ventricular afterload alters the slope of the end-systolic pressure-area relationship, an left ventricular performance by altering its contractility, config- index of contractility, was unchanged 10.90 + 0.1 1 mm Hg/ uration (via direct ventricular interaction). or both in the neo- mm2 (SEM)], and stroke area (65.8 f 7 mm2) and cardiac nate. output (the product of stroke area and ) (13 400 In a previous study. we demonstrated that mechanically in- f 1660 mm2/min) were maintained. However, the area creased right ventricular afterload diminishes left ventricular intercept of the pressure-area line at zero pressure (499 f performance based on simultaneous changes in systemic blood 13 mm2) shifted significantly to the left in the presence of flow and LV dP/dt,,, (22). In contrast. when comparable in- both levels of increased right ventricular afterload (by creased right ventricular afterload was induced with alveolar -39.2 f 13 and -76.2 f15 mm2, respectively). Mechani- hypoxia. systemic blood flow and LV dP/dt,,,,, increased. These cally increased right ventricular afterload alters left ven- changes could be blocked with propranolol. suggesting that they tricular configuration and causes a shift in the operating were mediated via catecholamine release (22). Inasmuch as we volume (area) range of the ventricle with no change in did not measure dimensions in our earlier work. we could not contractility in 3- to 7-d-old lambs. (Pediatr Res 33: 359- determine whether there had been a change in left ventricular 364,1993) contractility or filling. and we could not differentiate the relative roles of series and direct ventricular interaction. Abbreviations To address directly whether changes in left ventricular dimen- sions and shapes associated with changes in right ventricular LV dP/dt,.,. maximum derivative left ventricular pressure afterload could account for these changes in left ventricular Em.,, slope of end-systolic pressure-area relationship function. we instrumented neonatal lambs between 3 and 7 d of &, area-axis intercept of end-systolic pressure-area rela- age with micromanometer-tipped catheters and ultrasonic crys- tionship at zero pressure tals to measure left ventricular pressure and midventricular LVEDP. left ventricular end-diastolic pressure anterior-posterior and septal-free wall dimensions and then me- EDV. end-diastolic bolume chanically increased right ventricular afterload. We took the product of these two dimensions, denoted the left ventricular minor axis area. as a measure of left ventricular size (23). The E,,,. an index of contractility (9). did not change. However. the A,, shifted to the left in response to increased right ventricular Tremendous changes occur in the cardiovascular system dur- afterload. The stroke areas did not change. These findings are ing the transition from fetus to neonate. Fetal myocardium is compatible with a change in left ventricular configuration and a Rece~vedMa! 19. 1992: accepted December 9. 1992. shift in the operating volume range of the ventricle with no Correspondence and repnnt requests: Jay M. Milstein. Division of Neonatology. change in contractility. In the neonate. as in the adult (24). series Depanment of Pediatrics. TB 193. Un~vers~ryof California. Davis. Davis. C.4 95616. ventricular interaction plays a greater role than direct ventricular Supponed b) Nat~onalHean. Lung and Blood Institute grant HL-25869. interaction in determining left ventricular systolic ventricular 3 5 9 LNDGLANTZ function in response to changes in right ventricular loading viously isolated during the carotid artery dissection were inigated conditions. with lidocaine before data acquisition. The analog signals were recorded on a multichannel strip-chart MATERIALS AND METHODS recorder as well as an FM recorder (Honeywell 5600 C, Honey- well Inc.. Pleasantville, NY). The recordings were then digitized E.perimenta1 Preparation. Six healthy neonatal lambs be- at 250 Hz using an analog-to-digital converter and pressure-area tween 3 and 7 days of age were blindfolded for calming and plots generated on a DEC PDP- I 1/70 computer (Digital Equip- acutely anesthetized with i.v. a-chloralose (50 mg/kg in sodium ment Corporation, Maynard, MA). LV dP/dt was estimated borate solution) via a femoral venous catheter placed under local from the digitized pressure signal using central differences. End- anesthesia. After intubation, the animals were ventilated with a was defined as the time of the peak of the R-wave on Harvard animal ventilator. Arterial blood gases were maintained the electrocardiogram. and end- was defined as the time in nonhypoxic ranges [pH. 7.35 to 7.45; arterial 02 pressure. of minimum LV dP/dt. 13.06 to 33.325 kPa (98 to 250 tom); arterial CO? pressure, 4.39 We plotted pressure-area loops at several different levels of to 6.26 kPa (33 to 47 tom)] by adjusting the ventilator and afterload and used these data to define the end-systolic pressure- administering sodium bicarbonate. After a left thoracotomy, the area (i.e.volume) relationship by drawing a straight line through pericardium was opened across the base along the course of the the data by visual inspection to come as near as possible to the main to expose the great vessels and ductus upper left comer of the pressure-area loops. The slope of this line arteriosus. The ductus arteriosus was ligated. The lambs were is E,,,, and the area-axis intercept is A,. then given lidocaine systemically (1 mg/kg) to prevent arrhyth- Statistical methods. Data were analyzed using a multiple- mias, and, after gentle manual compression of the heart to reduce regression implementation of a repeated-measures analysis of filling, the apex of the heart was popped out of the pericardial variance (27). The following equation was used in the regression sac through the small opening across the base. analysis: A mid-ventricular short-axis plane was visualized approxi- mately half the distance from the apex to the base. The interven- tricular grooves were identified, and the free wall was visually The dependent variables (y) were maximum right ventricular divided into thirds. A pair of 6-mm hemispheric ultrasonic pressure, heart rate, LVEDP and maximum pressures, dP/dt,,,, piezoelectric crystals (25) were then aligned in the area of the end-diastolic and end-systolic anterior-posterior and septal-free anterior and posterior thirds to measure the anterior-posterior diameters and areas, stroke areas, cardiac outputs, Em,,, and Ad. dimension with a sonomicrometer (Triton Technology. Inc.. San Each dependent variable (y) observed in any given lamb under Diego, CA). These crystals were then sewn to their respective any set of experimental conditions is modeled as the sum of the epicardial surfaces. A 2.5-mm cylindrical crystal was tunneled average value across all lambs under control conditions (b,,), plus into the along the same minor axis plane, the effect of the volume load (V,), the kind of constriction used using a long plastic sleeve as an introducer, along a 19-gauge to acutely load the heart (C,), and different mean responses in needle track placed to the left of the interventricular groove and different lambs (L,). Specifically, to encode the differences in y below the juncture of the left anterior descending coronary artery with volume loading: and the first major diagonal branch. The final hemispheric crystal I if first volume loading was then aligned along the middle or lateral third of the free wall 0 otherwise of the left ventricle to measure the septal-free wall dimension. We took the product of the anterior-posterior and septal-free and wall dimensions as a measure of left ventricular size, denoted ventricular area. This product accurately reflects volume changes 1 if second volume loading in response to the interventions we studied (23). 0 otherwise After an additional dose of lidocaine (1 mg/kg), the heart was gently compressed again and replaced in the relatively intact and to encode differences due to direct ventricular interaction pericardial sac. The incision across the base of the pericardium secondary to acute pressure loading by pulmonary artery con- was left open, but the pericardial sac still enclosed the heart to striction: maintain direct ventricular interactions (26). Two cuffed vascular 1 if first level of pulmonary artery constriction occluders were placed around the main pulmonary artery and .I={ the descending aorta to control right and left ventricular after- 0 otherwise loads, and an electromagnetic transducer (Statham, Oxnard, CA) and was placed on the ascending aorta to measure flow. Finally, Millar micromanometer-tipped catheters (Millar Instruments 1 if second level of pulmonary artery constriction Inc., Houston, TX) were placed retrograde into the right ventricle = 0 otherwise via the main pulmonary artery and into the left ventricle and ascending aorta via the carotid artery to measure right ventricu- Finally, to allow for between-lamb differences, define the five lar, left ventricular, and aortic pressures, respectively. dummy variables, LI, . . . . L5. All animal experimentation was performed with the highest 5 standards of humane care in accord with the NIH Guidefor the 1 if lamb i (i 5) Care and Use of Laboratory Animals. Data acquisition and experimental protocol. Two levels of 0 otherwise. acutely increased right ventricular afterload were induced under The coefficient associated with each dummy variable in the baseline loading conditions as well as under two additional regression equation represents the change in y independently loading conditions generated by two transfusions of adult ewe associated with each factor. These changes were considered sta- whole blood (10 mL/kg). Four brief, sustained aortic constric- tistically significantly different from zero based on a t test for tions were induced under each right ventricular afterload and each coeficient when thc associated y was < 0.05. We aiso tested loading condition with return to baseline between each aortic whether these interventions affected the end-diastolic pressure- occlusion to generate several distinctly different end-systolic pres- volume relationship by using end-diastolic pressure as the de- sure-area points for subsequent analysis. To prevent reflex brady- pendent variable and adding end-diastolic area as an independent cardia during these occlusions, the two vagus nerve trunks pre- variable. DIRECT VENTRICULAR INTER4CTION IN NEONATAL LAMBS 36 1

RESULTS end systole. Pulmonary artery constriction had minimal effects on left ventricular anterior-posterior dimension, but significantly Basic hemodj-namic reslrlts. The basic hemodynamic results decreased the septal-free wall dimension at both end diastole and are shown in Table 1. Volume loading and pulmonary artery end systole. Thus, increasing right ventricular afterload decreases constriction increased maximum right ventricular pressure sig- left ventricular end-diastolic area. nificantly. Heart rate did not change. Volume loading increased The changes in area in response to volume loading and right LVEDP significantly and shifted the end-diastolic pressure-area ventricular afterload are shown in Table 1. The major changes curve upward but did not affect the other variables. The two of interest are that both end-diastolic and end-systolic areas levels of pulmonary artery constriction decreased maximum left decreased significantly in the presence of pulmonary artery con- ventricular pressure significantly. Similarly. LVEDP decreased striction. but stroke area did not change significantly. As a result, during pulmonary artery constriction as did LV dP/dt,,,. These , which equals the product of stroke area and heart results are similar to our previous findings (22). rate, did not change significantly. Systolic pressure-area loops. Figure 1 presents the hemody- Table 1 shows the changes in Em,, and Ad. Emaxdoes not namic responses to one level of pulmonary artery constriction in change significantly in the presence of volume loading or pul- one lamb. Constricting the pulmonary artery increased right monary artery constriction. However, Ad decreases significantly ventricular systolic pressure and reduced left ventricular systolic with pulmonary artery constriction. These results combine to pressure. Both left ventricular dimensions fell. but the dimension indicate that the pulmonary artery constriction reduced the changes during systole were maintained, so left ventricular stroke operating volume of the left ventricle without changing its ino- area (and aortic flow) did not change significantly with pulmo- tropic state. nary artery constriction. By transiently changing afterload. multiple end-systolic pres- sure-area points can be generated to determine the end-systolic DISCUSSION elastance. (Note that the areas are based on external measure- We hypothesized that acute pulmonary hypertension in a ments and include myocardial mass.) The area intercept at zero neonatal animal model may alter left ventricular performance ventricular pressure is an indication of the operating area (vol- by altering its contractility, configuration, or both. These exper- ume) range for the ventricle. It does not shift with changes in iments demonstrate that left ventricular configuration, and not inotropic state (28). Figure 2A represents a series of pressure-area contractility. changes in response to increases in right ventricular loops generated under baseline load conditions. Figure 2B rep- afterload. resents a similar series under the same loading conditions in the Mechanically increased right ventricular afterload decreased same lamb when the first level of pulmonary artery constriction LVEDP and LV dP/dt,,,. which could both be related to changes was imposed. Figure 2C shows these two series of loops super- in preload. It also reduced the septal-free wall dimensions at both imposed to reveal that the slope of the end-systolic pressure-area end systole and end diastole with no significant changes in relationship is similar in these two different right ventricular anterior-posterior dimensions. Both end-systolic and end-dia- loading conditions, whereas the area intercept shifts to the left stolic areas decreased so that stroke area and cardiac output were with increased right ventricular afterload. The stroke areas (anal- maintained. Ad decreased. suggesting that the operating range of ogous to ) did not change. areas (or volumes) of these ventricles in the presence of increased ConfiRurational changes. Table 1 addresses the configurational right ventricular afterload decreased. Right ventricular afterload changes in response to volume loading or pulmonary artery altered left ventricular configuration. particularly in the septal- constriction. Volume loading had limited effects on both ante- free wall dimension. Based on the end-systolic pressure-volume rior-posterior and septal-free wall dimensions at end diastole and relationship. however. changes in right ventricular afterload did Table I. Resnonses to chanacs in loadina conditions* ~- Variable Baseline Vol B Vol C Con I Con I1 R' Hemodynamic results RVP,,. (mm Hg) 39.2 + 2.5 6.5 + 3t 9.2 * 3t 18.7 + 3t 29.9 + 3t 0.83 HR (bpm) 193.7 + 7 15.4 f 8 13.2 + 8 5.2 + 7 1.8 t 8.7 0.52 LVEDP (mm Hg) 14.3 + 0.8 3 + 0.9t 5.5 + 0.9t -4.1 + 0.8t -4.7 + 0.9t 0.82 LVPmax(nun Hg) 106.5 + 5 -2.5 + 6 11.7 + 6 -15.8 + 5t -10.7 + 6 0.6 1 dP/dt,,, (mm Hg)/s 3262.6 + 268 -90.4 f 3 12 572.0 + 312 -580 + 274t -673.3 + 323t 0.64 Dimensional results ED AP (mm) 26.27 + 0. I 0.2 + 0.1 0.57 + It -0.16 +O.lt -0.27 + 0. l t 0.95 ES AP (mm) 25.1 + 0. l 0.03 + 0. l 0.20 + 0.1 0.08 + 0. l 0.06 + 0. l 0.95 ED SF (mm) 26.29 + 0.5 1.04 + 0.5 -0.27 + 0.5 -2.31 + 0.5t -2.87 + 0.6t 0.86 ES SF (mm) 24.88 + 0.3 -0.52 + 0.4 -0.47 + 0.4 -2.19 + 0.3t -3.03 + 0.4t 0.92 Area results EDA (mm') 687.8 + 13 -20.9 + 15 7.1 + 15 -64.4 + 13t -81.1 + 16t 0.80 ESA (mm') 621.9 + 8 -1 1.1 + 10 17.9 + 10 -52.8 + 9t -74.9 k lot 0.90 SA (mm') 65.8 + 7 9.9 + 8 -10.9 + 8 -1 1.5 + 7 -6.2 + 9 0.44 CO (SA X HR) 13429+1659 -l997+1928 -2184+1928 -1 907 + 1697 -861 * 1 995 0.42 (mm2/min) Contractility results Em, (mm Hg/mm2) 0.901 + 0. l I 0.026 + 0.13 -0.072 t 0.13 -0.084 + 0.1 I -0.109 + 0.13 0.44 Ad (mm2) 498.6 + 13 5.8 + 15 13.1 + 15 -39.2 + 13t -76.2 + 15t 0.78 * All changes are in absolute (not percentage) units. Vol B. second loading condition: Vol C. third loading condition: Con I. first level pulmonary artery constriction: Con 11. second level pulmonary artery constriction: R'. correlation coefficient: RVP,,,. peak right ventricular pressure: HR. heart rate: LVP,,,. peak left ventricular pressure: dP/dt,,,. maximum derivative left ventricular pressure: ED AP. end-diastolic anterior-posterior: ES AP. end-systolic anterior-posterior: ED SF. end-diastolic septal-free: ES SF. end-systolic septal-free: EDA. enddiastolic area: ESA. end-systolic area: SA. stroke area: CO. cardiac output. t p < 0.05: dependent variables determined from multiple linear regression. Values are coefficient + SEM. n = 46. 362 MILSTEIN AND GLANTZ

0.00 1.00 2.00 3.00 0.00 2.00 3.00 TINE Fig. 1. Digitized tracings depicting the hemodynamic response before and after the first level of pulmonary artery constriction. In response to the constriction, right ventricular pressure (RI'P) increases. and left ventricular pressure (LC'P) and LV dP/dt decrease. The ventricular dimensions decrease during pulmonary artery constriction throughout the , maintaining the stroke area. (Data from lamb 815.) LI' AP. left ventricular anterior-posterior; LV St.; left ventricular septal-free; AoP, .

AREA (mw) AREA (mw) AREA (mw) Fig. 2. A, A series of pressure-area loops recorded during brief, sustained aortic occlusions under baseline loading conditions. The Em,, represents the end-systolic elastance. B, A series of pressure-area loops recorded during brief, sustained aortic occlusions when the first level of pulmonary artery constriction was imposed under baseline loading conditions. C, The two series of pressure-area loops under baseline loading conditions with and without the first level of pulmonary artery constriction are superimposed. The maximum elastance, Em,,, does not change: however, the A,, decreases in the presence of pulmonary artery constriction. (Data from lamb 816.) not alter left ventricular contractility. Thus, in spite of configu- sistent with findings reported by Olsen et a[. (21) and Feneley et rational changes, normal left ventricular systolic function was al. (29) in adult dogs. inaiiltained without a change in inotropic state. To maintain The fact that the ventncle operates at a lower Ad after pul- stroke area and cardiac output in the presence of a shift of the monary artery constriction than before constriction is entirely area intercept and change in configuration, the left ventricle predictable based on simple geometry. Because the end-systolic would have to increase its . This result is con- force-length relationship is a property of the myocardial fiber, DIRECT VENTRICULAR INTERACTION IN NEOY4T4L L4MBS 363 anything that alters the relationship between fiber length and measures of inotropic state and the operating area range of the area will alter the force-area relationship. Contained area is ventricle. maximal for a circular shape. For any given circumference. the Other methods have been proposed to obtain quantitative contained area will fall if the chamber becomes noncircular. This measures of inotropic state. A second method is the relationship is a mechanical. geometric effect. between left ventricular stroke work and EDV at different EDV Human adults as well as adult animal models with acute (41, 42): this linear relationship is reasonably independent of pulmonary hypertension exhibit impaired left ventricular per- load and the slope changes in inotropic state. The stroke-work- formance ( 15- 17. 30. 3 1 ). Changes in left ventricular geometry EDV relationship is essentially a restatement of the original or configuration rather than contractility appear to be responsible Frank-Starling Law. This relationship is defined by plotting the for this impaired performance. with the effects being primarily areas of the loops in Figure I against the EDV volume of each on the left ventricular diastolic pressure-volume relationship beat. We did not find significant correlation between stroke work rather than on left ventricular systolic function (21. 32). and end-diastolic area during intervention runs in our neonatal Lirnitations of'thisstlrdj.. We used the product of two perpen- lambs. Finally. Little rt 01. (43. 44) proposed examining the dicular left ventricular equatorial dimensions as a substitute for relationship between dP/dtm,, and EDV as a measure of inotropic left ventricular volume. We used these dimensions because we state. This relationship is also reasonably linear, independent of anticipated greater changes in the septal-free wall dimension than loading. and has a slope that changes in response to inotropic in the anterior-posterior dimension because of direct ventricular agents. All three methods are thought to provide different views interaction accompanying the increase in right ventricular after- of the same phenomena: we do not know why the stroke work load (21. 33-35). We confirmed more striking changes in the was not correlated with end-diastolic area in the neonatal lambs septal-free wall dimension. as have other investigators (2 1. 33). we studied. The base-apex dimension. which makes a relatively minor con- Sign~ficuncec?f olrrtinding~.Despite the relative structural and tribution to changes in volume compared with the minor axis functional immaturity (1-4. 6-9), only left ventricular diastolic area, would be essential to assess absolute volume. but its exciu- funciion (p~t-~siiis-vo:iiiiicrc!a:ionship) -as 2fTected hv ir?creases sion from our measurements does not impede us from addressing in right ventricular afterload without changes in left Gentricular the relative effects of increased right ventricular afterload on contractility. Thus, if the human neonatal heart is functionally global left ventricular configuration and contractility (33). We similar to the heart of the ovine neonate. the impaired left did not measure and subtract myocardial area to assess chamber ventricular systolic function observed in some infants with per- area. Because the myocardial area should be constant or change sistent pulmonary hypertension of the newborn may be due to minimally under the conditions in this study. little information altered preload mediated by mechanical ventricular interaction, of value would be added by measuring the ventricular axis in without alterations in contractility. terms of the underlying hypothesis being tested (33). We used moderate changes in left ventricular afterload to .-lclino~~kd~~rnen~.The authors thank James Stoughton for his generate multiple pressure-area loops. This procedure may result excellent technical assistance. in baroreceptor reflexes (8). which may change end-systolic elast- ance. To avoid such changes. we blocked these reflexes by sprinkling the vagal trunks with lidocaine. so that we could better REFERENCES define contractility changes associated with our interventions I. Friedman WF 1973 The tntnnsic properties of the de\eloping heart. In: (36). Inotropic blockade was not used for two reasons. In one Friedman WF. Lesch M. Sonnenblick EH (eds) Neonatal Heart Disease. previous study (22). the effects of increased right ventricular Grune and Stratton. New York. pp 21-49 1. Friedman WF 1972 The intrtnsic physiologic properties of the developing afterload induced by either mechanical constriction of the main heart. Prog Cardiobasc Dis 1537-1 1 I pulmonary artery or alveolar hypoxia on left ventricular function 3. Berman Jr W. Musselman J 1979 Myocardial performance in the newborn were compared. Significant decreases in systemic blood flow and lamb. Am J Phys~ol237:H66-H70 dP/dtm,, developed with no changes in heart rate or aortic blood 4. Riemenschneider TA. Brenner RA. Mason DT 1981 Maturational changes in myocardial contractile state of newborn lambs. Pediatr Res 15:349-356 pressure in the presence of mechanically increased right ventric- 5. Ktrkpatnck SE. Pitlick PT. Nal~boffJ. Friedman WF 1976 Frank-Starling ular afterload. When the afterload was induced with alveolar relationsh~pas an important determinant of fetal cardiac output. Am J hypoxia, these four variables increased. These increases could be Physiol 131495-500 prevented with B-adrenergic blockade. Because the lambs in this 6. Haaktns J. Van Hare GF. Schmidt KG. Rudolph AM 1989 Effects of increas- ing ali~rloadon left ventricular output in fetal lambs. Circ Res 65:127-134 study were not hypoxemic and their heart rates did not increase 7. Van Hare GF. Hawkins JA. Schmidt KG. Rudolph AM 1990 The effects of in the presence of pulmonary artery constriction. an increase in increastng on left \entncular output in newborn inotropic background was unlikely. In another study (37). me- lambs. Circ Res 67:78-83 chanically increased right ventricular afterload was induced by 8. Teitel DF. Sidi D. Chtn T. Brett C. Heymann M.4. Rudolph 4M 1985 De\elopmental changes in myocardial contractile reserve in the lamb. Pediatr balloon distension of the left pulmonary artery resulting in small Res 19:948-955 increases in systemic and mean aortic pres- 9. Teitel DF. Klautz R. Steendijk P. Van Der Velde ET. Van Bel F. Baan J 1991 sure. Although these increases could be prevented with cr-adre- The end-s)stolic pressure-\olume relationship in the neuborn lamb: effects of loading and inotroptc intementions. Pediatr Res 29:473-482 nergic blockade. blockade was not used in this study because the 10. M~nczakBM. Wolfson MR. Santamore WP. Shaffer TH 1988 Developmental magnitude of the increases would be unlikely to alter left ven- changes in diastolic \entricular Interaction. Pediatr Res 23:466-469 tricular function and the site of the pulmonary artery constriction I I. Milste~nJM. Goet7man BW. Riemenschneider T.4. Wennberg RP 1979 In- was proximal to the site of the stretch receptors responsible for creased s!stemlc bascular reststance in neonates wtth pulmonar) h!perten- sion. 4m J Cardiol 44: 1 159- 1 162 changes in systemic vascular resistance (37). 12. Riemenschneider T4. Nielson HC. Ruttenberg HD. JatTe RB 1976 Disturb- We used Em,, as an index of contractility relatively independ- ances of the transitional circulation: spectrum of pulmonary hypertension ent of preload and afterload in neonates (8). In adult animals. and m)ocardial d!sfunction. J Pediatr X9:622-625 there appears to be contractility-dependent curvilinearity to the 13. Riggs T. Hirschfeld S. Fanaroff A. Liebman J. Fletcher. Meyer R. Bormuth C 1977 Pers~stenceof fetal circulation s)ndrome: an echocardiographic study. end-systolic pressure-volume relationship (38). Thus. Ad deter- J Ped~atr9 1 :626-63 1 mined by extrapolating multiple end-systolic pressure-volume 14. Rowe RD. Hoffman T 1972 Transient mbocardial ischemia of the newborn data may be shifted with changes in load with inotropic state infant: a form of severe cardiorespiraton distress in full-term infants. J (39. 30). Teitel et (11. (9). demonstrated that. although there may Pediatr 8 I :243-10 15. Zapol WM. Sn~derMT 1977 Pulmonar) hypertension in severe acute respi- be some cunilinearity to the end-systolic pressure-volume rela- rator) failure. N En@ J Jed 796:476-480 tionship of neonates. this cunilinearity is small. so the linear 16. Baum GL. Schwartz A. Llamas R. Castillo C 197 1 Left ventricular function in points we used provide reasonable estimates of Em,, and Ad as chronic obstructi\e lung disease. N Engl J Med 285:361-365 364 MILSTEIN AND GLANTZ

17. Rao BS. Cohn KE. Eldridge FL. Hancock EW 1968 Left ventricular failure the left \entricle during acute pulmonar) arterial hypenens~onIn dogs. 4m secondar) to chronic pulmonar) dixax. Am J Med 45:229-241 J Cardlol 33:RhX-87.5 18. Elzinga G. Van Grondelle R. Westerhof N. Van DenBos GC 1974 Ventricular Cal\.ln JE. Baer RW. Glantz SA 1986 Pulmonar) iniur) depresses cardiac interference. Am J Physiol 226:941-947 s!stolic function through Starling mechanism. 4m J Phys~olll:H722- 19. Elzinga G. Piene H. De Jong JP 1980 Left and right ventricular pump function H733 and consequences of having two pumps in one heart. Circ Res 46564-574 Slinker BK. Glantz S.4 198 The accuracy of Inferring left ventricular volume from dlmenslon depends on the frequency of information needed to answer 20. Glantz SA. Misbach GA. Moores WY. Mathey DG. Lekven J. Stowe DF. a given question. Clrc Res 56: 161-1 74 Parmley WM. Tyherg JV 1978 The pericardium substantially affects the left Slinker BK. Chagas ACP. Glantz SA 1987 Chronic pressure overload hypertro- ventricular diastolic pressure-volume relationships In the dog. Circ Res phy decreases dlrect ventricular interaction. Am J Physiol 253:H347-H357 42:433-44 1 Sagawa K 1981 Editorial: the end-systolic pressure-volume relation of the 21. Olsen CO. Tyson GS. Maier GW. Spratt JA. Davis JW. Rankln JS 1983 ~entncle:definition. modificauons and clinical use. Clrculat~on631223- Dvnamic ventricular interaction in the conscious don. Circ Res 52:85-I04 1227 22. ~iliteinJM, Bennett SH 1987 Increased right ventricilar afterload alters left 36. Sagawa K. Maughan L. Suga H. Sunagawa K 1988 Cardiac Contraction and ventricular function in newborn lambs. Am Heart J 114:369-377 the Pressure-Volume Relationship. Oxford Universit? Press. Oxford. UK pp 23. Appleyard RF. Glantz SA 1990 Two dimensions describe left ventricular 139-143 volume change during hemodynamic transients. Am J Physiol 258:H277- 37. Milste~nJM. Goetzman BW 1985 Effect of pulmonap arten constriction on HZ84 systemic vascular resistance In newhorn lambs. Am J Card10156:681-684 24. Slinker BK. Glantz SA 1986 End-systolic and endd~astollcventricular inter- 38. Burkhoff D. Sugiura S. Yue DT. Sagawa K 1987 Contractilitydependent action. Am J Physiol 25 1 :H 1062-H 1075 curv~linearityof end-systolic pressure-volume relat~ons. Am J Phys~ol 252:H1218-HI227 25. Van Trigt P. Bauer BJ. Olsen CO. Rankin JS. Wechsler AS 198 1 An improved 39. van der Velde ET. Burkhoff D. Steendl~kP. Karsdon J. Sagawa K. Baan J transducer for measurement of cardlac dimensions and sonomicrometry. 199 1 Nonllneanty and load sensitivity of end-systolic pressure-volume rela- Am J Physiol 240:H664-H668 tion of canine left ventricle rn v1\,o. Circulation 83:3 15-327 26. Minczak BM. Wolfson MR. Santamore WP. Shaffer TH 1990 Pericardial 40. Kass DA. Beyar R. Lankford E. Heard M. Maughan WL. Sagawa K 1989 effects on diastolic ventricular interaction during development. Pediatr Res Influence of contractile state on cuw~l~nearityof ~n sirli end-svnolic presurr- ?7:547-551 volume relations. Circulat~on79: 167- 178 27. Glant7 SA. Slinker BK 1990 Primer of Applied Regression and Analysis of 41. Misbach GA. Glantz SA 1979 Changes In the diastolic pressure-diameter Variance. McGraw-Hill, New York. pp 390-400 relation alter ventricular function curves. Am J Physiol 237:H644-H648 28. Sagawa K 1978 The ventricular pressure-volume diagram revisited. Circ Res 41. Glower DD. Spratt JA. Snow ND. Kahas JS. Davis JW. Olsen CO. Tyson GS. 43577-687 Sabiston Jr DC. Rankln JS 1985 Linearity of the Frank-Starling relationsh~p 29. Feneley MP. Olsen CO. Glower DD. Rankin JS 1989 Effect ofacutely increased In the intact heart: the concept of preload recmltable stroke work. Circulation 7 1 :994- 1009 right ventricularafterload on work output from the left ventricle In conscious 43. Llttle WC I985 The left ventricular dP/dt,..-end-diastolic volume relation in dogs. Circ Res 65: 135-145 closed chest dogs. Circ Res 56:808-8 I5 30. Kelly DT, Spotnitz HM, Beiser GD, Pierce JE. Epstein SE 1971 Effects of 44. Little WC. Cheng C. Mumma M, Igarashi Y. Vinten-Johansen J. Johnston chronic right ventricular volume and pressure loading on left ventricular WE 1989 Comparison of measures of left ventricular contractile performance performance. Circulation 44:403-4 I2 derived from pressure-volume loops in consclousdogs. Circulation 80: 1378- 31. Stool EW. Mullins CB. Leshin SJ. Mitchell JH 1974 Dimensional changes of 1387