© 2002 American Thoracic Society
Echo–Doppler Demonstration of Acute Cor Pulmonale at the Bedside in the Medical Intensive Care UnitMedical Intensive Care Unit, and Department of Cardiology, University Hospital Ambroise Paré, Assistance Publique Hôpitaux de Paris, Boulogne, France Correspondence and requests for reprints should be addressed to F. Jardin, M.D., Hôpital Ambroise Paré, 9 avenue Charles de Gaulle, 92104 Boulogne Cedex, France. E-mail: francois.jardin{at}apr.ap-hop-paris.fr For a long time, the only procedure available to the intensive care unit physician to assess cardiac function at the bedside was right heart catheterization. Right heart catheterization allows measurements of pressure and flow and the computation of derived variables. But these require an appropriate integration to obtain an accurate diagnosis. In other words, the aim of the quantitative evaluation obtained by right heart catheterization is to provide an indirect qualitative estimate of cardiac function. Conversely, echocardiographic evaluation is directly qualitative: by a simple visualization in real time of the kinetics and size of cardiac cavities, an experienced critical care intensivist with a sufficient echocardiographic background can immediately establish a functional diagnosis. This clinical review aims at illustrating the valuable information provided by echocardiography for the assessment of cardiac function at the bedside. For that purpose, we have chosen a particular and well-defined clinical setting, acute cor pulmonale. ACUTE COR PULMONALE: DEFINITION Acute cor pulmonale can be defined as a clinical setting in which the right ventricle is suddenly afterloaded. Massive pulmonary embolism and acute respiratory distress syndrome are the most common clinical conditions associated with acute cor pulmonale. In both settings, right ventricular (RV) outflow impedance is suddenly increased, RV ejection is impaired, and RV size is enlarged. Thus, acute cor pulmonale combines systolic and diastolic overload. Both are integrated in its echocardiographic definition, which includes septal dyskinesia (characterizing systolic overload), associated with right ventricular enlargement (characterizing diastolic overload) (1). Acute cor pulmonale may cause or precipitate circulatory failure in a critically ill patient. VENTRICULAR INTERDEPENDENCE A brief review of this physiologic phenomenon is useful in understanding echocardiographic abnormalities observed in acute cor pulmonale. Left and right ventricular systoles occur simultaneously, with the right and left ventricles starting and ending contraction almost at the same time. When right ventricular systole is overloaded, right ventricular contraction is prolonged, so that the right ventricle can continue to push after the left ventricle systole has ended (2). At this time, and for a short instant, the pressure in the right ventricular cavity may overcome that of the left ventricle, and reversal of the trans-septal pressure gradient imposes leftward displacement of the septal wall (Figure 1) .
Left and right ventricles are both enclosed in a stiff envelope, the pericardium. Both ventricles have similar end-diastolic volumes, and there is no free space for ventricular dilatation in a normal pericardial space. Thus, when right ventricular diastole is overloaded, right ventricular enlargement can occur only at the expense of the space devoted to the left ventricle, which is thus restricted. DESCRIPTION OF MAIN ECHOCARDIOGRAPHIC VIEWS USED TO STUDY RIGHT VENTRICULAR FUNCTION IN AN INTENSIVE CARE UNIT Echocardiographic examination of the right ventricle requires a long-axis view to evaluate the size of the cavity and a short-axis view to evaluate septal kinetics (Figure 2) , a Doppler examination of blood velocity in the outflow tract, and an examination of the inflow tract, including inferior vena caval size, and blood velocity of backward flow across the tricuspid valve, if present. The interventricular septum may also be examined on an oblique plane parallel to its long axis (Figure 2).
When breathing is spontaneous, as is usually the case in patients exhibiting massive pulmonary embolism, the right ventricle can be examined via a transthoracic approach. When the patient is mechanically ventilated, which is always the case in acute respiratory distress syndrome, right ventricular evaluation should be performed via a transesophageal approach, which provides better imaging in this setting. Echocardiography is essentially a qualitative procedure. However, quantitative measurements may be useful, and examples are given in this article. We have thus tested the reproducibility of the main measurements used in our laboratory to assess right ventricular function (Table 1) .
Because severe hypovolemia may influence right ventricular size, this disorder should be corrected before echocardiographic examination. Any hypovolemia can be easily recognized during examination (3, 4) and corrected by rapid volume expansion, before pursuing measurements. ECHO–DOPPLER DEMONSTRATION OF RIGHT VENTRICULAR SYSTOLIC OVERLOAD: SEPTAL DYSKINESIA As mentioned above, right ventricular systolic overload increases the duration of right ventricular systole, although the right ventricular ejection period may be reduced. Prolongation of right ventricular systole, occurring at a moment when the left ventricle starts to relax, reverses the left-to-right pressure gradient at end systole/onset diastole, which results in septal dyskinesia (Figure 1). The interventricular septum is displaced leftward at the onset of left ventricular diastole, and suddenly returns rightward when the left ventricle resumes contraction, resulting in a paradoxic motion (5–8). If the right ventricular filling pressure is higher than the left—which is regularly the case in this setting because diastolic overload is usually associated with systolic overload—septal displacement and flattening persist through diastole, and a sudden return of the interventricular septum toward the right ventricular cavity is produced only when left ventricular pressure exceeds right ventricular pressure, at the onset of systole (7). Examination of septal shape and kinetics indicates right ventricular systolic overload (7). An accurate analysis is obtained by M-mode study in the parasternal short-axis view (Figure 3) . This analysis is not possible by transesophageal echocardiography because the M-mode cannot cross the interventricular septum. However, a frame-by-frame analysis can demonstrate septal dyskinesia (see Film 1B in the online data supplement).
In response to a persistent acute pressure overload, significant changes develop in the configuration and the functional characteristics of the right ventricle. In a long-axis view, the apical region loses its triangular shape and becomes more rounded (Figure 4 , Film 5). In the short axis, the right ventricular cavity loses its crescent shape and becomes oval by septal displacement (Figure 4, Film 5). This change in configuration makes the right ventricular cavity more like the left (Figure 4, Film 5). Doppler examination of pulmonary artery flow velocity reveals that the acceleration time/flow period ratio is reduced (Table 2) . Sudden and major pressure overload producing a low-flow state markedly reduces pulmonary artery flow velocity (Figure 5) . A biphasic pattern is also frequently observed, with a midsystolic reduction in velocity (Figure 5). Progressive or persistent acute pressure overload makes pulmonary artery flow velocity like aortic flow velocity, with an increased and premature peak velocity (Table 2 and Figure 6) .
Normal right ventricular diastolic thickness measured by a transthoracic approach has been established in one clinical study as 3.3 ± 0.6 mm (mean ± SD) (9), and in another as 2.8 ± 1 mm (mean ± SD) (10). Using the transesophageal approach, we have observed a significant increase in right ventricular diastolic thickness after 48 hours of respiratory support in patients with acute respiratory distress syndrome and exhibiting acute cor pulmonale, with an average value of 6.5 ± 1.5 mm (mean ± SD) (Figure 7) . This finding suggests that, like other mammals, humans may rapidly increase their right ventricular muscular mass in response to pressure overload. Katamaya and coworkers (11) inserted a balloon into the pulmonary artery of lambs, and they inflated the balloon for 2 hours twice daily. After 4 days of this procedure (i.e., a total of 16 hours of intermittent right ventricular afterloading), they observed a significant increase in the myocardial mass of the right ventricle. In patients with acute respiratory distress syndrome, 48 hours of mechanical ventilation with an inspiratory/expiratory ratio of 1/2 also produce 16 hours of intermittent right ventricular afterloading.
Severe right ventricular pressure overload causes a drop in cardiac stroke output, which may be measured by both aortic and pulmonary artery velocity–time integral. Whereas the normal value of pulmonary artery velocity–time integral by pulsed Doppler in a group of 24 normal volunteers was found to be 18 ± 3 cm (mean ± SD) (12), we found markedly reduced values in acute cor pulmonale complicating both acute respiratory distress syndrome and massive pulmonary embolism (Table 2). By analogy with the left side, it is attractive to evaluate right ventricular systolic function by measuring the fractional area contraction in a long-axis view, which is easily obtained from an apical or a transesophageal approach. Right ventricular fractional area contraction is calculated as right ventricular end-diastolic area minus right ventricular end-systolic area, divided by right ventricular end-diastolic area. However, the large range of normal values (40 to 74% in our laboratory [13], 30 to 60% in Weyman's laboratory [14]), and the lack of correlation with pulmonary artery systolic pressure or angiographic obstruction index in pathologic conditions (15, 16), make the fractional area contraction of little value in clinical practice. ECHO–DOPPLER DEMONSTRATION OF RIGHT VENTRICULAR DIASTOLIC OVERLOAD: RIGHT VENTRICULAR ENLARGEMENT A normal right ventricle has an end-diastolic volume similar to that of the left ventricle (17, 18). Because of its regular shape, left ventricular volume can easily be measured by two-dimensional echocardiography, even with a monoplane formula (19), and a normal left ventricular end-diastolic volume is close to 70 cm3/m2 (19, 20). Conversely, its irregular shape makes measurements of right ventricle volume via echocardiography cumbersome. Despite several attempts using two planes, this evaluation has not gained acceptance in clinical practice. However, an accurate measurement of the absolute value of right ventricular volume is not necessary in clinical settings, whereas an accurate diagnosis of dilatation and a quantitative approach to monitoring its evolution are essential. Right ventricular diastolic dimensions can be obtained by measuring right ventricular end-diastolic area in the long axis, from an apical four-chamber view, or by a transesophageal approach (1, 10, 21–23). Because pericardial constraint necessarily results in left ventricular restriction when the right ventricle acutely dilates, the best way to quantify right ventricular dilatation is to measure the ratio between the right and left ventricular end-diastolic areas, which circumvents individual variations in cardiac size. We found this diastolic ventricular ratio to be 0.48 ± 0.12 (mean ± SD) in a group of normal volunteers (16). We have thus defined moderate RV dilatation as a diastolic ventricular ratio greater than 0.6, and major right ventricular dilatation as a diastolic ventricular ratio greater than or equal to 1 (1). Significant right ventricular enlargement is observed in acute cor pulmonale complicating both acute respiratory distress syndrome and massive pulmonary embolism (Tables 3 and 4) (15, 16).
Right ventricular diastolic enlargement is usually associated with right atrial dilatation (Figure 4), inferior vena caval dilatation, and tricuspid regurgitation. When pressure in the right atrium exceeds that in the left atrium, the foramen ovale may open. Detection of a patent foramen ovale requires contrast echocardiography, and has been greatly improved by transesophageal echocardiography (24). Measurement of the maximal velocity of tricuspid backward flow when regurgitation is present allows estimation of pulmonary artery systolic pressure (Figure 8) . A moderate elevation is observed in acute cor pulmonale complicating both acute respiratory distress syndrome and massive pulmonary embolism (Tables 3 and 4) (15, 16). However, in an individual patient, this measurement is of limited value because systolic pulmonary artery pressure also depends on right ventricular stroke output.
LEFT VENTRICULAR CONSEQUENCES OF ACUTE COR PULMONALE Sudden right ventricular enlargement in the stiff pericardial space causes left ventricular compression. Whereas the normal left ventricular end-diastolic volume according to the area–length formula is close to 70 cm3/m2 by an apical window (19, 20), and nearly 62 cm3/m2 by a transesophageal approach (20), we have observed severe reduction in acute cor pulmonale (Tables 3 and 4) (15, 16). This acute preload deficit, which is more sudden and more marked in massive pulmonary embolism, often produces acute circulatory failure in this setting (16). Even if more progressive, it may also precipitate circulatory failure in acute respiratory distress syndrome (Films 2, 3, and 6A). Left ventricular filling impairment in acute cor pulmonale results from combined pulmonary vascular occlusion and septal displacement (6). The latter produces abnormal left ventricular relaxation, which is also evidenced by an abnormal mitral flow profile, with a preeminent peak velocity during atrial systole (Figure 6). When calculating the ratio of peak velocity during early diastolic filling (E wave) for peak velocity of the atrial systole (A wave), we found an inverted ratio in acute cor pulmonale complicating both acute respiratory distress syndrome and massive pulmonary embolism (Tables 3 and 4) (15, 16). LIMITATIONS OF DIASTOLIC VENTRICULAR RATIO IN GRADING RIGHT VENTRICULAR DILATATION Cor pulmonale was first described as a long-term complication of chronic obstructive pulmonary disease. Echocardiographic abnormalities associated with chronic cor pulmonale share many similarities with those of acute cor pulmonale (25), but also two prominent differences: first, there is a marked increase in right ventricular diastolic wall thickness (which is usually greater than 9 mm at end diastole), associated with marked intracavity muscle trabeculations (25). Second, there is left ventricular hypertrophy, involving both the interventricular septum and free wall (26). Right ventricular cavity enlargement may also be observed in more than 50% of patients exhibiting right ventricular infarction, and be graded by diastolic ventricular area ratio measurement. While the clinical context is usually clearly different, the echocardiographic pattern might be confused with that of acute cor pulmonale. In this setting, the presence of left ventricular posterior or inferior free wall abnormalities usually allows the two conditions to be distinguished (27). Finally, gradation of right ventricular dilatation by ventricular diastolic area ratio may be inoperative when chronic left ventricular dilatation is present, because of valvular disease or dilated cardiomyopathy. ACUTE COR PULMONALE IN MASSIVE PULMONARY EMBOLISM Massive pulmonary embolism, that is, pulmonary embolism involving at least two lobar arteries, is the first cause of acute cor pulmonale. The invaluable aid provided by echocardiography in the evaluation of massive pulmonary embolism was first described by Kasper and coworkers (28), and several authors have described measurements different from those proposed here, to quantify right ventricular dysfunction. These authors have proposed the diastolic ventricular diameter ratio (28), right ventricular systolic dysfunction (29), or hypokinesia (30) to quantify right ventricular impairment in massive pulmonary embolism. However, as mentioned above, right ventricular enlargement is usually accompanied by a change in apical shape and is not accurately evaluated by diameter measurement. The quantification of systolic dysfunction as proposed by Ribeiro and coworkers is not standardized (29), and hypokinesia (30) is a qualitative parameter that may be somewhat subjective. Moreover, measurement of right ventricular fractional area contraction, as quantitatively evaluated by echocardiography, is weakly related to the severity of circulatory failure, as mentioned above. We thus believe that our echocardiographic definitions, combining an increased diastolic ventricular ratio with septal dyskinesia, is more reliable. Using our definition, we found a 61% incidence of acute cor pulmonale in a group of 161 patients with massive pulmonary embolism proven by angiography or enhanced contrast helicoidal computed tomography (16). Most patients presenting with massive pulmonary embolism do not require mechanical ventilation, and echocardiographic examination is usually performed via a transthoracic approach. However, in some patients, acute cor pulmonale is diagnosed via a transesophageal approach, after an unexplained cardiac arrest. In such patients, the pulmonary artery thrombus often can be visualized during echocardiographic study (31) (see Film 4). Acute circulatory failure, defined as the requirement for vasopressor support to maintain arterial systolic pressure above 90 mm Hg, is associated with acute cor pulmonale complicating massive pulmonary embolism in two-thirds of cases (16). Several authors have emphasized the poor prognosis of this hemodynamic instability (30, 32). In fact, we have observed in this setting that only patients with metabolic acidosis, defined by a base deficit greater than 5 mMole/L, had a poor prognosis (59% mortality rate), whereas 97% of patients with acute cor pulmonale and no metabolic acidosis eventually recovered, sometimes after several hours of vasopressor support (16). CAN BEDSIDE DEMONSTRATION OF ACUTE COR PULMONALE BY ECHOCARDIOGRAPHY AID CLINICAL DECISION-MAKING IN PATIENTS WITH MASSIVE PULMONARY EMBOLISM? Contrast angiography has long remained the "gold" standard for the diagnosis of pulmonary embolism. However, this invasive procedure is not always immediately available and may carry some risks in a patient exhibiting circulatory failure. Bedside echocardiography, by demonstrating the presence of acute cor pulmonale in a suggestive clinical context, markedly increased the probability of a pulmonary embolism and allowed us to initiate specific treatment before definitive confirmation of massive pulmonary embolism by angiography. Enhanced contrast helicoidal computed tomography has currently replaced contrast angiography to diagnose pulmonary embolism (33). This less invasive procedure is immediately available in modern hospitals and is usually performed before echocardiography. Thus, when the intensivist performs echocardiography in this context, he or she is usually aware of the diagnosis. The demonstration of acute cor pulmonale would then help assess the hemodynamic tolerance of massive pulmonary embolism and guide subsequent monitoring. At this time, there is no consensus on the precise indications for thrombolytic therapy in acute cor pulmonale complicating massive pulmonary embolism. Although recommended by some authors (32), and considered to offer no advantages by others (34), thrombolytic therapy was not associated with improved prognosis in our patients presenting with the most severe form of acute cor pulmonale (16). However, the total resolution of RV dilatation, which usually takes between 10 and 20 days (Film 5) (16), is hastened by thrombolytic therapy (Film 6) (16). ACUTE COR PULMONALE IN ACUTE RESPIRATORY DISTRESS SYNDROME In the acute respiratory distress syndrome, two factors combine to produce right ventricular systolic overload: the pathologic features of the syndrome per se, which can be associated with distal occlusion of the pulmonary arterial bed (35), and mechanical ventilation, which increases right ventricular outflow impedance (36). Compared with that required for a normal lung, mechanical ventilation of a diseased lung requires a higher transpulmonary pressure (i.e., alveolar pressure minus pleural pressure), a factor limiting blood flow through the pulmonary capillary bed (37). A normal right ventricle may easily develop a systolic pressure of 25–30 mm Hg. During tidal ventilation, this forward pressure must overcome a backward pressure, that is, the transpulmonary pressure (37), a trivial task when ventilating a normal lung. However, this backward pressure becomes substantial when the lung is damaged. In this setting, persistently high right ventricular afterload induced by mechanical ventilation may precipitate acute cor pulmonale. In addition, the application of external positive end-expiratory pressure (PEEP), which increases functional residual capacity by distending undamaged areas, produces increased pulmonary vascular resistance when used above a given level (38). Applying high PEEP levels thus represents another potential factor for increased right ventricular outflow impedance (39, 40). The occurrence of acute cor pulmonale as a complication of acute respiratory distress syndrome is more progressive and less sudden than in massive pulmonary embolism. It usually requires at least 48 hours of respiratory support (15) (Film 7). We described in 1985 the echocardiographic pattern of acute cor pulmonale complicating acute respiratory distress syndrome (41). At this time, high tidal volumes were used (13 ml/kg) and the incidence of acute cor pulmonale was high (61%) (41). In the most severe forms, where the diastolic ventricular ratio was greater than or equal to 1, we observed a 100% mortality (41). More recently, with airway pressure limitation, the incidence of acute cor pulmonale in acute respiratory distress syndrome has declined to 25% (15). Moreover, with adequate ventilatory management, acute cor pulmonale is not significantly associated with increased mortality (15). Even patients presenting with the most severe form (i.e., with a diastolic ventricular ratio of 1 or more), may recover (15). CAN BEDSIDE DEMONSTRATION OF ACUTE COR PULMONALE BY ECHOCARDIOGRAPHY AID CLINICAL DECISION-MAKING IN PATIENTS WITH ACUTE RESPIRATORY DISTRESS SYNDROME? In the past, acute cor pulmonale during adult respiratory distress syndrome was a dreaded complication. Now, with the widespread acceptance of a low-stretch ventilation strategy, the prognosis has improved (42). In our opinion, demonstration of acute cor pulmonale in a patient with adult respiratory distress syndrome should lead the clinician to ask two questions: 1. Is the lung stretch really sufficiently low? All pressure settings should be reexamined, particularly plateau pressure, which should be maintained below 30 cm H2O or less if possible. Also PEEP, and particularly intrinsic PEEP, should be reduced. The latter is often caused by an excessive respiratory rate, an insidious cause of right ventricular afterloading (43). If PEEP reduction does not appear possible in the supine position without worsening hypoxemia, prone positioning should be considered. 2. Is it possible to reduce hypercapnia, without increasing tidal volume? In the experimental animal, hypercapnic acidosis has been shown to worsen right ventricular performance (44). Increasing respiratory rate, while reducing the ventilator circuit dead space, may be appropriate, provided it does not produce intrinsic PEEP. However, in some cases, acute cor pulmonale reflects only the severity of lung injury and may be marginally influenced by the specific maneuvers described above, until respiratory failure improves. At this stage, adding inhaled NO may buy time by temporarily reversing acute cor pulmonale (Film 8). CONCLUSION When the circulatory status of a critically ill patient is impaired, clinical evaluation alone is of limited value, and it has long been accepted that additional measurements are often needed. Bedside right heart catheterization using a balloon floating catheter has been available since 1970 and has been largely used in intensive care units during the following 20 years. Over the past 10 years, the routine practice of this procedure has been questioned. Echocardiography, an alternative minimally invasive procedure, which provides a dynamic view and can be repeated at the bedside as needed, allows a reasonably accurate hemodynamic evaluation of right ventricular function in patients at risk for acute cor pulmonale in the intensive care unit. Echocardiography can provide useful information for monitoring such patients and their response to therapeutic management. Acknowledgments The authors gratefully acknowledge Christian Brun-Buisson, M.D., for helping with the presentation of this manuscript. FOOTNOTES This article has an online data supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Received in original form February 25, 2002; accepted in final form September 23, 2002 REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ur pulmonaire chronique. In: Jardin F, Dubourg O, editors. L'exploration echocardiographique en médecine d'urgence. Paris: Masson; 1986. p. 125–134.
