CRITICAL
REVIEW
Left Ventricular Function in Chronic Obstructive Pulmonary Disease* Richard G. Kachel, M.D.
The presence or absence of left ventricular dysfunction in chronic obstructive pulmonary disease has been de bated for decades. I have reviewed the following evidence pertaining to this topic: (1) left ventricular pathologic abnormalities; (2) the methods used to determine left ventricular performance; (3) specific abnormalities of left ventricular function as revealed by systolic time in tervals, left ventricular end-diastolic pressure, ejection fraction, isovolumic indices, and left ventricular function
curves; and (4) pertinent experimental data. The bulk of the evidence indicates that the clinical symptoms of left-sided failure are unreliable in those with obstructive disease of the airways and that the great majority of pa tients have normal left ventricular function, once other causes are excluded. A small group of patients have some abnormalities in left ventricular performance, but these have not been clinically significant. The ultimate impor tance of such abnormalities awaits future investigation.
r•"'hepresence of left ventricular dysfunction in •¿patients *with chronic obstructive pulmonary dis ease is controversial. One recent investigator con cluded that "sub-clinical left ventricular dysfunction is frequently present,"1 while another group stated that "there is no left ventricular involvement to be explained."2 Why is this topic so contentious? What
the usual clinical criteria for evaluating the status of the left ventricle may be misleading.
is the evidence supporting a diseased left ventricle in chronic obstructive pulmonary disease? What are the roles of hypertension, arteriosclerotic heart dis ease, pulmonary hypertension, or the severity of the lung disease itself? If left ventricular impairment is present, what is its clinical significance? These are some of the questions I will address in this review. It is my purpose not only to assess the state of knowl edge in this area but also to delineate the reasons for disagreement. The clinical evaluation of left ventricular failure in chronic obstructive pulmonary disease is inaccurate. In their pathologic study, Murphy et al3 found that clinical failure was diagnosed with almost equal frequency in those patients with and without left ventricular hypertrophy at necropsy. In one study of 28 patients with chronic obstructive pulmonary dis ease, no correlation was found between clinical, radiographic, or electrocardiographic abnormalities and increased pulmonary arterial wedge pressure.4 In another study of patients referred for symptoms of left ventricular failure, only three of 20 had ab normal wedge pressure or ejection indices.5 Clearly, "From the Division of Cardiology, Department of Medicine, Jackson Memorial Hospital and University of Miami School of Medicine, Miami, Fla. Reprint requests: Dr. Kachel, 175 Bradley Place, Palm Beach 33480
286 RICHARD G.KACHEL
PATHOLOGIC STUDIES Pathologic studies of the heart in chronic obstruc tive pulmonary disease have revealed abnormalities of the left ventricle. The first such study appeared in 1936 when Kountz et al°reported the findings in ten of 17 patients who had left ventricular hypertrophy. The status of the coronary arteries was not de scribed. Subsequent studies showed a prevalence of left ventricular hypertrophy ranging from 17.5 to 86 percent.3 Murphy et al3 reported an extensive study of 72 autopsies in patients with severe chronic bronchitis and emphysema. These investigators3 used ventricu lar weights to determine hypertrophy and post mortem angiographic studies to grade occlusive coronary arterial disease. Of 20 cases of left ventricu lar hypertrophy, ten were believed to be secondary to coronary atherosclerosis or hypertension (or both). Two patients had silent aortic stenosis. Eight cases were inadequately explained, although Mur phy et al3 believed that their study underestimated the prevalence of hypertension. There was no rela tionship of diabetes mellitus, age, duration of symp tomatic pulmonary disease, or a history of alcohol ism to the presence of left ventricular hypertrophy. The association of left ventricular hypertrophy and coronary atherosclerosis is itself controversial, and a discussion of this topic is beyond the scope of this review; however, if the association is denied, then the number of unexplained cases of left ventricular CHEST,74: 3, SEPTEMBER, 1978
hypertrophy in chronic pulmonary disease increases. In 1968, Rao et al7 described a group of eight patients with cor pulmonale and left ventricular failure. Necropsy was performed in five cases, all of which had hypertrophy (judged by the thickness of the ventricular wall) and dilatation of both ventri cles. The coronary arteries were normal in four pa tients. There was no evidence of fibrosis, cellular infiltrates, or deposits of lipids. The preceding studies demonstrate the association of chronic obstructive pulmonary disease and left ventricular hypertrophy. Nevertheless, the pathogenctic basis for this association is not always clear. METHODOLOGIC SPECIFICITY Does the left ventricle function normally in ob structive disease of the airways? To try to answer this question, investigators have used various indices to measure left ventricular performance. It is impor tant to review the factors affecting these indices, especially with coexisting cor pulmonale, in order to better understand the limitations of each mea surement. The following have been used to assess left ventricular function: (1) systolic time intervals;1-4-8(2) ejection-phase indices;2-4'9 (3) isovolumic indices;--10 (4) left ventricular end-diastolic pressure,2-7-11"14and (5) curves of ventricular function.15'16 The central problem is to separate changes in contractility from changes in preload and afterload. The velocity of circumferential fiber shortening is a good indicator of contractility but is sensitive to changes in heart rate and varies inversely with afterload.17 Short-term changes in preload and afterload may change the ejection fraction.18-19 Abnormal systolic time intervals can reflect left ventricular failure. The most useful intervals include total electromechanical systole, the left ventricular ejection time, and the preejection period. Although all of these measurements are sensitive, they are indirect and lack specificity for changes in contractil ity; for example, these measurements are affected by variations in stroke volume and heart rate.20 The pulmonary arterial wedge pressure is another measurement that reflects left ventricular contractil ity. It is pertinent to ask if the wedge pressure is valid in patients with chronic obstructive pulmonary disease. In 19 patients with chronic obstructive pul monary disease, Rice et al21 compared the actual mean wedge pressure and the mean pressure deter mined after subtracting esophageal pressures. In 16 patients the two pressures were within 3 mm Hg; in three patients with changes in intrathoracic pressure greater than 20 mm Hg, the mean wedge was falsely elevated. Although several other studies have shown good agreement between wedge and left atrial or CHEST,74: 3, SEPTEMBER, 1978
left ventricular end-diastolic pressure,7-12'15 some patients may have high values because of abnormal pulmonary mechanics. Of equal importance is the fact that changes in compliance of the right ventricle may affect the pressure-volume relationship of the left ventricle. An increased wedge or left ventricular end-diastolic pressure may then reflect a change in compliance, rather than a true change in left ventricular con tractility. This interdependence of ventricular com pliance was elaborately studied in dogs by Taylor et al.22Using increments in volume in first the beating heart and then an isolated ventricular preparation, these workers showed that the left ventricular pres sure-volume curve is influenced by the degree of right ventricular filling; for example, at a left ven tricular end-diastolic pressure of 20 mm Hg, the volume in the left ventricle was 7.1 percent larger with an empty right ventricle than when the right ventricle was filled to the pressure observed to ac company a left ventricular end-diastolic pressure of 20 mm Hg in the beating heart. Thus, states charac terized by right ventricular distention, at least on a short-term basis, may influence the relationship of left ventricular pressure and volume without changes in contractility. Conversely, the presence of a normal or slightly elevated end-diastolic pressure may not reflect a true depression of ventricular performance. This is be cause end-diastolic pressure is a poor predictor of end-diastolic volume, especially in the normal range of pressures. Because the relationship between enddiastolic pressure and volume is nonlinear, signifi cant changes in the latter can cause decisive changes in left ventricular performance while effecting small changes in pressure.23 A more dynamic means of quantifying left ven tricular performance is the curve of left ventricular function, generally obtained by plotting stroke work vs end-diastolic pressure. With this method, Ross and Braunwald,24 using an infusion of angiotensin, defined a spectrum of normal and frankly abnormal responses. Of note, all patients (except two of those more severely depressed) had normal resting car diac indices and end-diastolic pressures. With these points in mind, let us now consider the studies of left ventricular function in chronic ob structive pulmonary disease. Systolic Time Intervals Hooper and Whitcomb1 studied 28 patients with stable chronic obstructive pulmonary disease who had no history of congestive heart failure, alcohol ism, hypertension, or any evidence of cardiac dis ease. These investigators1 found statistically signifiLEFTVENTRICULAR FUNCTION IN COPD 287
cant differences in the preejection period (PEP), the left ventricular ejection time (LVET), and the ratio of PEP/LVET when comparing a control group with their more severely affected subjects. Seven of 18 patients in the less severely affected groups had a ratio of PEP/LVET that was greater than two standard deviations above the normal mean. There was no relationship between resting oxygen tension and any systolic time interval. Hooper and Whitcomb1 concluded that subclinical left ventricular impairment is common in moderate chronic obstructive pulmonary disease. Nevertheless, it is necessary to ask if right ven tricular dysfunction per se can affect left ventricular systolic time intervals. Alpert et al8 found abnormal intervals in patients with right ventricular failure from differing causes; patients without failure had normal values. Unger and his associates4 found no consistent correlation between systolic time inter vals, pulmonary arterial wedge pressures, and ejec tion-phase indices in 28 symptomatic patients with chronic pulmonary disease. Because the stroke vol ume index was low in 60 percent of their patients, Unger et al4 concluded that the abnormal systolic time intervals reflected a decreased left ven tricular preload (a consequence of diminished right ventricular output) and that left ventricular con tractility was preserved. Others supported this conclusion.13 In contradistinction, Mathay et al25 found normal basal stroke volumes, left ventricular end-diastolic volumes, and cardiac indices. Wedge or Left Ventricular End-Diustolic Pressure Several studies used left ventricular end-diastolic pressure as the criterion of left ventricular perfor mance and found normal or slightly elevated values.2-4-1'-13Kline et al5 evaluated 20 patients for clinically suspected left ventricular failure. Of 17 with normal wedge pressures at rest and after hand grip exercise, all had normal left ventricular function using echocardiographic criteria and a radionuclide ejection fraction. Three patients did have high wedge pressures, and each showed abnormal ejec tion indices with the other two techniques. Ejection Fraction The ejection fraction is generally considered a sensitive indicator of ventricular function. A recent study used this measurement to emphasize the ef fects of coronary disease on the left ventricle in obstructive disease of the airways.9 Although the average ejection fraction of 28 patients was 52 ±2 percent, seven patients with clinical coronary dis ease had a mean value of 42 ±4 percent, compared to 55 ±2 percent in the remaining 21 (P < 0.001). 288
RICHARDG. KACHEL
In a separate group of acutely decompensated pa tients with left ventricular ejection fractions less than 40 percent, 14 died. Ten of these 14 patients were autopsied, and all showed severe coronary ar terial disease. It is important to note that five of these ten patients had no evidence of coronary dis ease before death; however, some patients with a depressed ejection fraction had no evidence of coro nary disease at necropsy. In the acutely decompen sated group, ejection fraction did not correlate with arterial oxygen tension, arterial carbon dioxide ten sion, or pH. A history of alcoholism was not men tioned. In the study by Frank et al2 of patients with cor pulmonalc, the mean ejection fraction was 37 ± 3 percent with no clinical evidence of coronary dis ease. In the study by Kline et al,5 only two of 19 patients had a depressed ejection fraction measured with a radioisotope. Curves of Left Ventricular Performance Baum et al15 studied a group of 15 patients with severe stable pulmonary disease and no obvious left ventricular impairment. There was no history of hypertension, alcoholism, or valvular disease. Using the method of Ross and Braunwald,24 curves of ventricular function were abnormal in 14 of the patients. Ten patients had coronary arteriographic and ventriculographic studies. Only two had signifi cant arterial narrowing; six had increased thickness of the left ventricular wall, and five had an increase in the size of the left ventricular chamber. Three of eight had abnormalities of contraction during ven triculographic studies; however, there was no agematched control group. Williams and his associates10 obtained different results from a group of 16 patients with no history of coronary disease or hypertension. Methoxamine was infused to increase systolic arterial pressure. Wil liams et a!1(iconcluded that the curves of ventricular function so obtained were similar to their control group and that there was no difference whether or not right ventricular failure was present; however, some patients with pulmonary hypertension did have flattened curves. Moreover, the method used by Williams et al16was different from that of Baum et al,15 and the results may not be comparable. Williams et al'G used a rise in systolic pressure as an end point for the infusion of methoxamine; Baum et al15required a rise in end-diastolic pressure in addi tion to a response in systolic pressure when they used angiotensin. EXPERIMENTAL DATA There is experimental evidence demonstrating that the left ventricle may be adversely affected by a CHEST,74: 3, SEPTEMBER, 1978
diseased right ventricle. In animals with induced right ventricular failure, decreased levels of norepinephrine and myofibrillar adenosine triphosphatase and elevations of the content of hydroxyproline have been demonstrated in the left ventricle.26'28
ventricular impairment, are a common cause of death in patients with respiratory failure.31 It is important to note, though, that there is only one study of patients with left ventricular failure in which coronary disease, hypertension, valvular dis ease, and alcoholism were clearly excluded.7
The effect of right ventricular failure on left ven tricular mechanics and contractility was studied in dogs by Kelly et al.29 Dogs with right ventricular failure demonstrated a decrease in peak left ven tricular systolic pressure, peak left ventricular wall stress, peak left ventricular rate of pressure change (dP/dt), and peak contractile element velocity at four levels of left ventricular end-diastolic pressure; however, when these values were plotted against left ventricular volume instead of pressure, they were normal. Peak dP/dt did remain abnormally de creased. In a similar study, Stool et al30 measured right and left ventricular volumes in dogs during increases of pulmonary arterial pressure induced with a balloon-tipped catheter. At a mean pulmonary arterial pressure of 40 mm Ilg, left ventricular stroke volume decreased, but ejection fraction and cardiac output were maintained. These studies show the difficulty and complexity of determining the integrity of left ventricular func tion in the presence of a diseased right ventricle,
It is apparent that the cardiac pathophysiologic findings in chronic obstructive pulmonary disease are intricate and that future studies must be de signed to measure a number of interacting variables. It was 42 years ago that the first description of left ventricular hypertrophy in chronic obstructive pul monary disease appeared. Since then, the question of left ventricular dysfunction in this pathologic state has been greatly clarified. With advancing technology and clearer insights, the remaining an swers should be forthcoming.
CONCLUSIONS What conclusions can be drawn from these data? The following four points do appear valid: 1. A significant percentage of patients with chronic obstructive pulmonary disease have coro nary arterial disease,3-9 which may depress left ven tricular function. Even excluding patients with an gina or electrocardiographic patterns of infarction still leaves some with silent disease. 2. The majority of patients with chronic pulmo nary disease have normal left ventricular function once other known causes of impairment are ex cluded. This is so in spite of a clinical picture that often indicates the contrary. 3. There does seem to be a group of patients who have some left ventricular impairment, as demon strated by sophisticated means. This impairment occurs in the absence of hypertension, coronary disease, alcoholism, or valvular disease. The demon strated abnormalities have not been clinically sig nificant, although their natural history is totally un known. Whether these abnormalities are, in fact, related to the pulmonary disease or some other un suspected variable is also unknown. 4. If left ventricular failure secondary to chronic obstructive pulmonary disease occurs, it is uncom mon. One must remember that arrhythmias, not left CHEST,74: 3, SEPTEMBER, 1978
ACK NOWLEDGMENT: I wish to thank Ralph Lazzara, M.D., for his review of the manuscript and Mrs. Audrey Magid for secretarial assistance. REFERENCES0 *
1 Hooper GH, Whitcomb ME: Systolic time intervals in chronic obstructive pulmonary disease. Circulation 50: 1205-1209, 1974 2 Frank MJ, Weisse AB, Moschos CB, et al: Left ventricu lar function, metabolism, and blood flow in chronic cor pulmonale. Gircnlation 17:798-806, 1973 3 Murphy ML, Adantsun J, Hutcheson \F: Left ventricular hypertrophy in patients with chronic bronchitis and em physema. Ann Intern Med 81:307-313, 1974 4 linger K, Shaw 1), Karlincr JS, et al: Evaluation of left ventricular performance in acutely ill patients with chronic obstructive lung disease. Chest 68:135-142, 1975 5 Kline LE, Crawford MH, Mac-Donald WJ Jr, et al: \oninvasive assessmt-nt of left ventricular performance in patients with chronic obstructive pulmonary disease. Chest 72:558-564, 1977 6 Kountz. \VB, Alexander HL, Prinzmetal M: The heart in emphysema. Am Heart J 11:163-172, 19.36 7 Rao BS, Colin KE. Eldridge FL, et al: Left ventricular failure secondary to chronic pulmonary disease. Am J Med 45:229-241, 1968 8 Alpert JS, Rickman FD, Howe JP, et a!: Alterations of systolic time intervals in right ventricular faitore. Circula tion 50:317-323, 1974 9 Steele P, Ellis JH, Van Dyke D, et al: Left ventricular ejection fraction in severe chronic obstructive airways disease. Am J Med 59:21-28, 1975 10 Salel A, Mason DT, Amsterdam BA, et air Depression of left ventricular contractility in primary right ventricular overload: The "reversed Bernheim phenomenon" (ab stract ). Circulation 43 (suppl 2) :II, 220, 1971 11 Davies H, Overy HR: Left ventricular function in cor pulmonale. Chest .58:8-14, 1970 12 Burrows B, Kettel LJ, Niden AH, et al: Patterns of cardiovascular dysfunction in chronic obstructive lung disease. N Engl J Med 286:912-918, 1972 13 Khaja F, Parker JO: Right and left ventricular perfor mance in chronic obstructive lung disease. Am Heart J 82:319-327, 1971 °°A more complete list of references is available from the author on request. LEFTVENTRICULARFUNCTIONIN COPD 289
14 Lockhart A, Tzareva M, Nader F, et al: Elevated pulmo nary artery wedge pressure at rest and during exercise in chronic bronchitis: Fact or fancy? Clin Sci 37:503-517, 1969 15 Baum GL, Schwartz A, Llamas R, et al: Left ventricular function in chronic obstructive lung disease. N Engl ] Med 285:361-365, 1971 16 Williams JF, Childress RH, Boyd DL, et al: Left ven tricular function in patients with chronic obstructive pul monary disease. J Clin Invest 47:1143-1153, 1968 17 Quinones MA, Gaasch WH, Alexander JK: Influence of acute changes in preload, afterload, contractile state, and heart rate on ejection and isovolumic indices of myocardial contractility in man. Circulation 53:293-302, 1976 18 Krayenbuehl HP, Bussmann WD, Turina M, et al: Is the ejection fraction an index of myocardial contractility? Cardiology 53:1-10, 1968 19 Tsakiris AG, Vandenberg RA, Bancheo N, et al: Varia tions of left ventricular end-diastolic pressure, volume, and ejection fraction with changes in outflow resistance in anesthetized intact dogs. Circ Res 23:203-211, 1968 20 Weissler AM, Peeler RG, Rochill WH, et al: Relation ships between left ventricular ejection time, stroke vol ume, and heart rate in normal individuals and patients with cardiovascular disease. Am Heart J 62:367-378, 1961 21 Rice DL, Awe RJ, Gaasch WH, et al: Wedge pressure measurement in obstructive pulmonary disease. Chest 66:628-632, 1974 22 Taylor RR, Covell JW, Sonnenblick EH, et al: Depen dence of ventricular distensibility on filling of the opposite
290 RICHARD G.KACHEL
ventricle. Am J Physiol 213:711-718, 1967 23 Braumvald E, Ross J Jr: The ventricular cnd-diastolic pressure. Am J Med 34:147-150, 1963 24 Ross J Jr, Braunwald E: The study of left ventricular function in man by increasing resistance to ventricular ejection with angiotensin. Circulation 29:739-7 19, 1964 25 Mathay RA, Ellis JH Jr, Steele P: Effect of dextran loading on left ventricular performance in chronic ob structive pulmonary disease. Am Heart 1 92:730-736, 1976 26 Vogel JHK, Jacobowitz D, Chidsey CA: Distribution of norepinephrine in the failing bovine heart. Circ Res 24:7184, 1971 27 Buccino RA, Harris E, Spann JF, et al: Connective tissue response in the development of experimental cardiac hypertrophy (abstract). Circulation .36 (suppl 2):II, 77, 1967 28 Pool PE, Spann JF, Buccino RA, et al: Myocardial high energy phosphate stores in cardiac hypertrophy and in heart failure. Circ Res 21:365-373, 1967 29 Kelly DT, Spotnitz M, Beiser GD, et al: Effects of chronic volume and pressure loading on left ventricular perfor mance. Circulation 44:403-412, 1971 30 Stool EW, Mullins CB, Leskin SS, et al: Effect of right ventricular volume change from acute pulmonary hyper tension on left ventricular dimensions (abstract). Clin Res 20:399, 1972 31 Hudson LD, Kurt TL, Petty TL, et al: Arrhythmias associated with acute respiratory failure in patients with chronic airway obstruction. Chest 63:661-665, 1973
CHEST,74: 3, SEPTEMBER, 1978
REVIEW
Left Ventricular Function in Chronic Obstructive Pulmonary Disease* Richard G. Kachel, M.D.
The presence or absence of left ventricular dysfunction in chronic obstructive pulmonary disease has been de bated for decades. I have reviewed the following evidence pertaining to this topic: (1) left ventricular pathologic abnormalities; (2) the methods used to determine left ventricular performance; (3) specific abnormalities of left ventricular function as revealed by systolic time in tervals, left ventricular end-diastolic pressure, ejection fraction, isovolumic indices, and left ventricular function
curves; and (4) pertinent experimental data. The bulk of the evidence indicates that the clinical symptoms of left-sided failure are unreliable in those with obstructive disease of the airways and that the great majority of pa tients have normal left ventricular function, once other causes are excluded. A small group of patients have some abnormalities in left ventricular performance, but these have not been clinically significant. The ultimate impor tance of such abnormalities awaits future investigation.
r•"'hepresence of left ventricular dysfunction in •¿patients *with chronic obstructive pulmonary dis ease is controversial. One recent investigator con cluded that "sub-clinical left ventricular dysfunction is frequently present,"1 while another group stated that "there is no left ventricular involvement to be explained."2 Why is this topic so contentious? What
the usual clinical criteria for evaluating the status of the left ventricle may be misleading.
is the evidence supporting a diseased left ventricle in chronic obstructive pulmonary disease? What are the roles of hypertension, arteriosclerotic heart dis ease, pulmonary hypertension, or the severity of the lung disease itself? If left ventricular impairment is present, what is its clinical significance? These are some of the questions I will address in this review. It is my purpose not only to assess the state of knowl edge in this area but also to delineate the reasons for disagreement. The clinical evaluation of left ventricular failure in chronic obstructive pulmonary disease is inaccurate. In their pathologic study, Murphy et al3 found that clinical failure was diagnosed with almost equal frequency in those patients with and without left ventricular hypertrophy at necropsy. In one study of 28 patients with chronic obstructive pulmonary dis ease, no correlation was found between clinical, radiographic, or electrocardiographic abnormalities and increased pulmonary arterial wedge pressure.4 In another study of patients referred for symptoms of left ventricular failure, only three of 20 had ab normal wedge pressure or ejection indices.5 Clearly, "From the Division of Cardiology, Department of Medicine, Jackson Memorial Hospital and University of Miami School of Medicine, Miami, Fla. Reprint requests: Dr. Kachel, 175 Bradley Place, Palm Beach 33480
286 RICHARD G.KACHEL
PATHOLOGIC STUDIES Pathologic studies of the heart in chronic obstruc tive pulmonary disease have revealed abnormalities of the left ventricle. The first such study appeared in 1936 when Kountz et al°reported the findings in ten of 17 patients who had left ventricular hypertrophy. The status of the coronary arteries was not de scribed. Subsequent studies showed a prevalence of left ventricular hypertrophy ranging from 17.5 to 86 percent.3 Murphy et al3 reported an extensive study of 72 autopsies in patients with severe chronic bronchitis and emphysema. These investigators3 used ventricu lar weights to determine hypertrophy and post mortem angiographic studies to grade occlusive coronary arterial disease. Of 20 cases of left ventricu lar hypertrophy, ten were believed to be secondary to coronary atherosclerosis or hypertension (or both). Two patients had silent aortic stenosis. Eight cases were inadequately explained, although Mur phy et al3 believed that their study underestimated the prevalence of hypertension. There was no rela tionship of diabetes mellitus, age, duration of symp tomatic pulmonary disease, or a history of alcohol ism to the presence of left ventricular hypertrophy. The association of left ventricular hypertrophy and coronary atherosclerosis is itself controversial, and a discussion of this topic is beyond the scope of this review; however, if the association is denied, then the number of unexplained cases of left ventricular CHEST,74: 3, SEPTEMBER, 1978
hypertrophy in chronic pulmonary disease increases. In 1968, Rao et al7 described a group of eight patients with cor pulmonale and left ventricular failure. Necropsy was performed in five cases, all of which had hypertrophy (judged by the thickness of the ventricular wall) and dilatation of both ventri cles. The coronary arteries were normal in four pa tients. There was no evidence of fibrosis, cellular infiltrates, or deposits of lipids. The preceding studies demonstrate the association of chronic obstructive pulmonary disease and left ventricular hypertrophy. Nevertheless, the pathogenctic basis for this association is not always clear. METHODOLOGIC SPECIFICITY Does the left ventricle function normally in ob structive disease of the airways? To try to answer this question, investigators have used various indices to measure left ventricular performance. It is impor tant to review the factors affecting these indices, especially with coexisting cor pulmonale, in order to better understand the limitations of each mea surement. The following have been used to assess left ventricular function: (1) systolic time intervals;1-4-8(2) ejection-phase indices;2-4'9 (3) isovolumic indices;--10 (4) left ventricular end-diastolic pressure,2-7-11"14and (5) curves of ventricular function.15'16 The central problem is to separate changes in contractility from changes in preload and afterload. The velocity of circumferential fiber shortening is a good indicator of contractility but is sensitive to changes in heart rate and varies inversely with afterload.17 Short-term changes in preload and afterload may change the ejection fraction.18-19 Abnormal systolic time intervals can reflect left ventricular failure. The most useful intervals include total electromechanical systole, the left ventricular ejection time, and the preejection period. Although all of these measurements are sensitive, they are indirect and lack specificity for changes in contractil ity; for example, these measurements are affected by variations in stroke volume and heart rate.20 The pulmonary arterial wedge pressure is another measurement that reflects left ventricular contractil ity. It is pertinent to ask if the wedge pressure is valid in patients with chronic obstructive pulmonary disease. In 19 patients with chronic obstructive pul monary disease, Rice et al21 compared the actual mean wedge pressure and the mean pressure deter mined after subtracting esophageal pressures. In 16 patients the two pressures were within 3 mm Hg; in three patients with changes in intrathoracic pressure greater than 20 mm Hg, the mean wedge was falsely elevated. Although several other studies have shown good agreement between wedge and left atrial or CHEST,74: 3, SEPTEMBER, 1978
left ventricular end-diastolic pressure,7-12'15 some patients may have high values because of abnormal pulmonary mechanics. Of equal importance is the fact that changes in compliance of the right ventricle may affect the pressure-volume relationship of the left ventricle. An increased wedge or left ventricular end-diastolic pressure may then reflect a change in compliance, rather than a true change in left ventricular con tractility. This interdependence of ventricular com pliance was elaborately studied in dogs by Taylor et al.22Using increments in volume in first the beating heart and then an isolated ventricular preparation, these workers showed that the left ventricular pres sure-volume curve is influenced by the degree of right ventricular filling; for example, at a left ven tricular end-diastolic pressure of 20 mm Hg, the volume in the left ventricle was 7.1 percent larger with an empty right ventricle than when the right ventricle was filled to the pressure observed to ac company a left ventricular end-diastolic pressure of 20 mm Hg in the beating heart. Thus, states charac terized by right ventricular distention, at least on a short-term basis, may influence the relationship of left ventricular pressure and volume without changes in contractility. Conversely, the presence of a normal or slightly elevated end-diastolic pressure may not reflect a true depression of ventricular performance. This is be cause end-diastolic pressure is a poor predictor of end-diastolic volume, especially in the normal range of pressures. Because the relationship between enddiastolic pressure and volume is nonlinear, signifi cant changes in the latter can cause decisive changes in left ventricular performance while effecting small changes in pressure.23 A more dynamic means of quantifying left ven tricular performance is the curve of left ventricular function, generally obtained by plotting stroke work vs end-diastolic pressure. With this method, Ross and Braunwald,24 using an infusion of angiotensin, defined a spectrum of normal and frankly abnormal responses. Of note, all patients (except two of those more severely depressed) had normal resting car diac indices and end-diastolic pressures. With these points in mind, let us now consider the studies of left ventricular function in chronic ob structive pulmonary disease. Systolic Time Intervals Hooper and Whitcomb1 studied 28 patients with stable chronic obstructive pulmonary disease who had no history of congestive heart failure, alcohol ism, hypertension, or any evidence of cardiac dis ease. These investigators1 found statistically signifiLEFTVENTRICULAR FUNCTION IN COPD 287
cant differences in the preejection period (PEP), the left ventricular ejection time (LVET), and the ratio of PEP/LVET when comparing a control group with their more severely affected subjects. Seven of 18 patients in the less severely affected groups had a ratio of PEP/LVET that was greater than two standard deviations above the normal mean. There was no relationship between resting oxygen tension and any systolic time interval. Hooper and Whitcomb1 concluded that subclinical left ventricular impairment is common in moderate chronic obstructive pulmonary disease. Nevertheless, it is necessary to ask if right ven tricular dysfunction per se can affect left ventricular systolic time intervals. Alpert et al8 found abnormal intervals in patients with right ventricular failure from differing causes; patients without failure had normal values. Unger and his associates4 found no consistent correlation between systolic time inter vals, pulmonary arterial wedge pressures, and ejec tion-phase indices in 28 symptomatic patients with chronic pulmonary disease. Because the stroke vol ume index was low in 60 percent of their patients, Unger et al4 concluded that the abnormal systolic time intervals reflected a decreased left ven tricular preload (a consequence of diminished right ventricular output) and that left ventricular con tractility was preserved. Others supported this conclusion.13 In contradistinction, Mathay et al25 found normal basal stroke volumes, left ventricular end-diastolic volumes, and cardiac indices. Wedge or Left Ventricular End-Diustolic Pressure Several studies used left ventricular end-diastolic pressure as the criterion of left ventricular perfor mance and found normal or slightly elevated values.2-4-1'-13Kline et al5 evaluated 20 patients for clinically suspected left ventricular failure. Of 17 with normal wedge pressures at rest and after hand grip exercise, all had normal left ventricular function using echocardiographic criteria and a radionuclide ejection fraction. Three patients did have high wedge pressures, and each showed abnormal ejec tion indices with the other two techniques. Ejection Fraction The ejection fraction is generally considered a sensitive indicator of ventricular function. A recent study used this measurement to emphasize the ef fects of coronary disease on the left ventricle in obstructive disease of the airways.9 Although the average ejection fraction of 28 patients was 52 ±2 percent, seven patients with clinical coronary dis ease had a mean value of 42 ±4 percent, compared to 55 ±2 percent in the remaining 21 (P < 0.001). 288
RICHARDG. KACHEL
In a separate group of acutely decompensated pa tients with left ventricular ejection fractions less than 40 percent, 14 died. Ten of these 14 patients were autopsied, and all showed severe coronary ar terial disease. It is important to note that five of these ten patients had no evidence of coronary dis ease before death; however, some patients with a depressed ejection fraction had no evidence of coro nary disease at necropsy. In the acutely decompen sated group, ejection fraction did not correlate with arterial oxygen tension, arterial carbon dioxide ten sion, or pH. A history of alcoholism was not men tioned. In the study by Frank et al2 of patients with cor pulmonalc, the mean ejection fraction was 37 ± 3 percent with no clinical evidence of coronary dis ease. In the study by Kline et al,5 only two of 19 patients had a depressed ejection fraction measured with a radioisotope. Curves of Left Ventricular Performance Baum et al15 studied a group of 15 patients with severe stable pulmonary disease and no obvious left ventricular impairment. There was no history of hypertension, alcoholism, or valvular disease. Using the method of Ross and Braunwald,24 curves of ventricular function were abnormal in 14 of the patients. Ten patients had coronary arteriographic and ventriculographic studies. Only two had signifi cant arterial narrowing; six had increased thickness of the left ventricular wall, and five had an increase in the size of the left ventricular chamber. Three of eight had abnormalities of contraction during ven triculographic studies; however, there was no agematched control group. Williams and his associates10 obtained different results from a group of 16 patients with no history of coronary disease or hypertension. Methoxamine was infused to increase systolic arterial pressure. Wil liams et a!1(iconcluded that the curves of ventricular function so obtained were similar to their control group and that there was no difference whether or not right ventricular failure was present; however, some patients with pulmonary hypertension did have flattened curves. Moreover, the method used by Williams et al16was different from that of Baum et al,15 and the results may not be comparable. Williams et al'G used a rise in systolic pressure as an end point for the infusion of methoxamine; Baum et al15required a rise in end-diastolic pressure in addi tion to a response in systolic pressure when they used angiotensin. EXPERIMENTAL DATA There is experimental evidence demonstrating that the left ventricle may be adversely affected by a CHEST,74: 3, SEPTEMBER, 1978
diseased right ventricle. In animals with induced right ventricular failure, decreased levels of norepinephrine and myofibrillar adenosine triphosphatase and elevations of the content of hydroxyproline have been demonstrated in the left ventricle.26'28
ventricular impairment, are a common cause of death in patients with respiratory failure.31 It is important to note, though, that there is only one study of patients with left ventricular failure in which coronary disease, hypertension, valvular dis ease, and alcoholism were clearly excluded.7
The effect of right ventricular failure on left ven tricular mechanics and contractility was studied in dogs by Kelly et al.29 Dogs with right ventricular failure demonstrated a decrease in peak left ven tricular systolic pressure, peak left ventricular wall stress, peak left ventricular rate of pressure change (dP/dt), and peak contractile element velocity at four levels of left ventricular end-diastolic pressure; however, when these values were plotted against left ventricular volume instead of pressure, they were normal. Peak dP/dt did remain abnormally de creased. In a similar study, Stool et al30 measured right and left ventricular volumes in dogs during increases of pulmonary arterial pressure induced with a balloon-tipped catheter. At a mean pulmonary arterial pressure of 40 mm Ilg, left ventricular stroke volume decreased, but ejection fraction and cardiac output were maintained. These studies show the difficulty and complexity of determining the integrity of left ventricular func tion in the presence of a diseased right ventricle,
It is apparent that the cardiac pathophysiologic findings in chronic obstructive pulmonary disease are intricate and that future studies must be de signed to measure a number of interacting variables. It was 42 years ago that the first description of left ventricular hypertrophy in chronic obstructive pul monary disease appeared. Since then, the question of left ventricular dysfunction in this pathologic state has been greatly clarified. With advancing technology and clearer insights, the remaining an swers should be forthcoming.
CONCLUSIONS What conclusions can be drawn from these data? The following four points do appear valid: 1. A significant percentage of patients with chronic obstructive pulmonary disease have coro nary arterial disease,3-9 which may depress left ven tricular function. Even excluding patients with an gina or electrocardiographic patterns of infarction still leaves some with silent disease. 2. The majority of patients with chronic pulmo nary disease have normal left ventricular function once other known causes of impairment are ex cluded. This is so in spite of a clinical picture that often indicates the contrary. 3. There does seem to be a group of patients who have some left ventricular impairment, as demon strated by sophisticated means. This impairment occurs in the absence of hypertension, coronary disease, alcoholism, or valvular disease. The demon strated abnormalities have not been clinically sig nificant, although their natural history is totally un known. Whether these abnormalities are, in fact, related to the pulmonary disease or some other un suspected variable is also unknown. 4. If left ventricular failure secondary to chronic obstructive pulmonary disease occurs, it is uncom mon. One must remember that arrhythmias, not left CHEST,74: 3, SEPTEMBER, 1978
ACK NOWLEDGMENT: I wish to thank Ralph Lazzara, M.D., for his review of the manuscript and Mrs. Audrey Magid for secretarial assistance. REFERENCES0 *
1 Hooper GH, Whitcomb ME: Systolic time intervals in chronic obstructive pulmonary disease. Circulation 50: 1205-1209, 1974 2 Frank MJ, Weisse AB, Moschos CB, et al: Left ventricu lar function, metabolism, and blood flow in chronic cor pulmonale. Gircnlation 17:798-806, 1973 3 Murphy ML, Adantsun J, Hutcheson \F: Left ventricular hypertrophy in patients with chronic bronchitis and em physema. Ann Intern Med 81:307-313, 1974 4 linger K, Shaw 1), Karlincr JS, et al: Evaluation of left ventricular performance in acutely ill patients with chronic obstructive lung disease. Chest 68:135-142, 1975 5 Kline LE, Crawford MH, Mac-Donald WJ Jr, et al: \oninvasive assessmt-nt of left ventricular performance in patients with chronic obstructive pulmonary disease. Chest 72:558-564, 1977 6 Kountz. \VB, Alexander HL, Prinzmetal M: The heart in emphysema. Am Heart J 11:163-172, 19.36 7 Rao BS, Colin KE. Eldridge FL, et al: Left ventricular failure secondary to chronic pulmonary disease. Am J Med 45:229-241, 1968 8 Alpert JS, Rickman FD, Howe JP, et a!: Alterations of systolic time intervals in right ventricular faitore. Circula tion 50:317-323, 1974 9 Steele P, Ellis JH, Van Dyke D, et al: Left ventricular ejection fraction in severe chronic obstructive airways disease. Am J Med 59:21-28, 1975 10 Salel A, Mason DT, Amsterdam BA, et air Depression of left ventricular contractility in primary right ventricular overload: The "reversed Bernheim phenomenon" (ab stract ). Circulation 43 (suppl 2) :II, 220, 1971 11 Davies H, Overy HR: Left ventricular function in cor pulmonale. Chest .58:8-14, 1970 12 Burrows B, Kettel LJ, Niden AH, et al: Patterns of cardiovascular dysfunction in chronic obstructive lung disease. N Engl J Med 286:912-918, 1972 13 Khaja F, Parker JO: Right and left ventricular perfor mance in chronic obstructive lung disease. Am Heart J 82:319-327, 1971 °°A more complete list of references is available from the author on request. LEFTVENTRICULARFUNCTIONIN COPD 289
14 Lockhart A, Tzareva M, Nader F, et al: Elevated pulmo nary artery wedge pressure at rest and during exercise in chronic bronchitis: Fact or fancy? Clin Sci 37:503-517, 1969 15 Baum GL, Schwartz A, Llamas R, et al: Left ventricular function in chronic obstructive lung disease. N Engl ] Med 285:361-365, 1971 16 Williams JF, Childress RH, Boyd DL, et al: Left ven tricular function in patients with chronic obstructive pul monary disease. J Clin Invest 47:1143-1153, 1968 17 Quinones MA, Gaasch WH, Alexander JK: Influence of acute changes in preload, afterload, contractile state, and heart rate on ejection and isovolumic indices of myocardial contractility in man. Circulation 53:293-302, 1976 18 Krayenbuehl HP, Bussmann WD, Turina M, et al: Is the ejection fraction an index of myocardial contractility? Cardiology 53:1-10, 1968 19 Tsakiris AG, Vandenberg RA, Bancheo N, et al: Varia tions of left ventricular end-diastolic pressure, volume, and ejection fraction with changes in outflow resistance in anesthetized intact dogs. Circ Res 23:203-211, 1968 20 Weissler AM, Peeler RG, Rochill WH, et al: Relation ships between left ventricular ejection time, stroke vol ume, and heart rate in normal individuals and patients with cardiovascular disease. Am Heart J 62:367-378, 1961 21 Rice DL, Awe RJ, Gaasch WH, et al: Wedge pressure measurement in obstructive pulmonary disease. Chest 66:628-632, 1974 22 Taylor RR, Covell JW, Sonnenblick EH, et al: Depen dence of ventricular distensibility on filling of the opposite
290 RICHARD G.KACHEL
ventricle. Am J Physiol 213:711-718, 1967 23 Braumvald E, Ross J Jr: The ventricular cnd-diastolic pressure. Am J Med 34:147-150, 1963 24 Ross J Jr, Braunwald E: The study of left ventricular function in man by increasing resistance to ventricular ejection with angiotensin. Circulation 29:739-7 19, 1964 25 Mathay RA, Ellis JH Jr, Steele P: Effect of dextran loading on left ventricular performance in chronic ob structive pulmonary disease. Am Heart 1 92:730-736, 1976 26 Vogel JHK, Jacobowitz D, Chidsey CA: Distribution of norepinephrine in the failing bovine heart. Circ Res 24:7184, 1971 27 Buccino RA, Harris E, Spann JF, et al: Connective tissue response in the development of experimental cardiac hypertrophy (abstract). Circulation .36 (suppl 2):II, 77, 1967 28 Pool PE, Spann JF, Buccino RA, et al: Myocardial high energy phosphate stores in cardiac hypertrophy and in heart failure. Circ Res 21:365-373, 1967 29 Kelly DT, Spotnitz M, Beiser GD, et al: Effects of chronic volume and pressure loading on left ventricular perfor mance. Circulation 44:403-412, 1971 30 Stool EW, Mullins CB, Leskin SS, et al: Effect of right ventricular volume change from acute pulmonary hyper tension on left ventricular dimensions (abstract). Clin Res 20:399, 1972 31 Hudson LD, Kurt TL, Petty TL, et al: Arrhythmias associated with acute respiratory failure in patients with chronic airway obstruction. Chest 63:661-665, 1973
CHEST,74: 3, SEPTEMBER, 1978
