Relation of Mean Right Atrial Pressure to Echocardiographic and Doppler Parameters of Right Atrial and Right Ventricular Function
Background A paucity of data exists as to the relation of mean right atrial pressure (RAP) to Doppler parameters of right atrial and ventricular filling. Furthermore, whether echocardiographic parameters of right atrial and right ventricular function and inferior vena cava improve the relation of Doppler filling dynamics with RAP has not been explored.
Methods and Results Doppler and echocardiographic studies were performed simultaneously with measurements of mean RAP in consecutive patients who either had a central venous catheter in the Intensive Care Unit or underwent catheterization of the right side of the heart. The initial population consisted of 35 patients with a mean age (±SD) of 60±15 years; 34% were on mechanical ventilation. Mean RAP averaged 9±5.7 mm Hg (range, 2 to 28 mm Hg). Among tricuspid inflow parameters, the strongest relation with RAP was observed with the ratio of early to late velocity (r=.66). For hepatic venous flow, systolic filling wave indexes had the best relation with atrial pressure, the highest being for systolic filling fraction (r=−.86). Weaker relations were noted with the use of right atrial volumes, right ventricular function, and inferior vena caval diameters. The addition of any of these variables did not improve the relation of systolic filling fraction with RAP. The regression equation (RAP=21.6−24 systolic filling fraction) was tested prospectively in the estimation of atrial pressure in 50 patients. The correlation coefficient was .89 in the prospective group and .88 in the total group of 85 patients. The mean difference between predicted and actual pressures in the whole population was −0.2±2.6 mm Hg. The sensitivity and specificity for mean RAP>8 mm Hg were 86% and 92%, respectively.
Conclusions Among echocardiographic and Doppler parameters of right atrial and right ventricular function, hepatic venous flow dynamics relate best to mean atrial pressure and can be used clinically to estimate mean RAP.
The determination of right and left ventricular filling pressures is important clinically for the diagnosis and management of various hemodynamic conditions.1 2 3 4 Furthermore, assessment of right atrial pressure (RAP) is needed in the echocardiography laboratory for the estimation of systolic right ventricular and pulmonary artery pressures.5 6 7 8 Although clinical evaluation of jugular venous pulsations is usually used to estimate RAP, it may be difficult in certain individuals because of obesity or a short neck, and such an examination may not always be possible in a busy echocardiography laboratory.6 7
The relation of Doppler and echocardiographic parameters of left-side cardiac structure and function to left ventricular filling pressure has been investigated in several studies.9 10 11 Mitral and pulmonary venous flow dynamics and left atrial size have been shown to relate to and allow estimation of left atrial pressure.9 10 11 In contrast, a few studies in selected patients have evaluated the right-side hemodynamic correlates of right atrial and right ventricular filling dynamics.12 13 14 Tricuspid and hepatic vein flow dynamics have been described in patients with restrictive cardiomyopathy13 and have been used to help differentiate constrictive pericarditis from restrictive cardiomyopathy.14 15 On the other hand, studies on the changes in the diameter of the inferior vena cava in response to negative intrathoracic pressure have demonstrated a significant relation of this index to RAP.16 17 18 19 Such an approach, however, requires the patient’s cooperation and has been limited in mechanically ventilated individuals.18 A comprehensive evaluation of the relation of echocardiographic and Doppler parameters of right ventricular function, right atrial function, and inferior vena cava to mean RAP in patients with a variety of clinical conditions has not previously been performed. Because several of these parameters can be obtained from routine cardiac ultrasound examination, it is important to evaluate which of these parameters has the best relation to RAP and whether a combination of these indexes improves the clinical estimation of right ventricular filling pressure. Accordingly, this study was undertaken to assess the relation of Doppler parameters of hepatic vein and tricuspid inflow, right ventricular, right atrial function, and inferior vena cava to mean RAP in consecutive patients undergoing measurement of RAP. The relation among these parameters was defined. The best relation with mean RAP was tested prospectively in the estimation of mean RAP.
All consecutive patients undergoing catheterization of the right side of the heart or who had in-dwelling central venous catheters in the Intensive Care Units at The Methodist Hospital (Houston, Tex) during a cumulative 3-month period were eligible for this study. The investigation protocol was approved by the Institutional Review boards of The Methodist Hospital and Baylor College of Medicine. All patients gave informed consent before participation in the study. Simultaneous recordings of transthoracic echocardiographic studies and mean RAP were performed in all patients (see below). Exclusion criteria included inadequate recordings of pressure or Doppler tracings, tricuspid stenosis, severe tricuspid regurgitation assessed by color-flow Doppler,20 the presence of a prosthetic tricuspid valve, and nonsinus rhythms (atrial flutter or fibrillation, paced rhythm, or complete heart block). The echocardiographic and hemodynamic protocols and inclusion and exclusion criteria were identical in the initial and prospective populations.
Echocardiographic and Doppler Studies
The echocardiographic studies were performed by use of a Hewlett Packard ultrasound system (Sonos 1000) equipped with 2.5- and 3.5-MHz transducers with the patient in the supine position. Standard echocardiographic imaging was performed from the parasternal, apical, and subcostal windows for the evaluation of right and left ventricular function and assessment of the size of the right atrium and inferior vena cava. Color-flow Doppler was used to screen for the presence of valvular regurgitation. A pulsed Doppler recording of tricuspid inflow was obtained from the low parasternal and apical windows, with the sample volume placed at the tip of the tricuspid valve. Ten to 15 cardiac cycles from each window were recorded at a sweep speed of 100 mm/s. A recording of hepatic vein flow velocity was obtained from the subcostal window, with the sample volume placed 1 to 2 cm in the hepatic veins, close to their entrance into the inferior vena cava. Similarly, 10 to 15 cardiac cycles were recorded.
Right ventricular function was quantified by use of a modified Simpson’s rule21 from the four-chamber view in technically adequate studies and was visually assessed in all cases through changes in right ventricular dimensions.22 Maximal right atrial volume preceding tricuspid valve opening and minimal right atrial volume after atrial contraction were measured from the apical four-chamber view. Right atrial volumes and emptying fractions were calculated with modified Simpson’s rule. From the two-dimensional subcostal views, maximal inferior vena caval diameter and minimal diameter after a sniff test were measured, from which a collapse index in percent was obtained.17 In patients on mechanical ventilation, the approach used to determine maximal and minimal inferior vena caval diameters was similar to that of Jue et al.18
Tricuspid inflow measurements were performed from the window providing the highest overall velocities, implying the least angulation with flow. The following parameters were measured (Fig 1⇓) as previously described in detail23 : peak early inflow velocity, peak late velocity, and their ratio; atrial filling fraction; deceleration time; and acceleration time.
From the hepatic vein flow velocity, the following parameters were measured (Fig 1⇑): peak velocity and time-velocity integral of the systolic, diastolic, and atrial reversal waves. Ratios of systolic to diastolic peak velocity and time-velocity integral also were derived. Systolic filling fraction was calculated as the time-velocity integral of the systolic wave divided by the sum of time-velocity integrals of the systolic and diastolic waves. Velocity reversal occasionally observed at the end of the systolic wave was not included in the systolic wave or in the calculation of systolic filling fraction. Systolic filling fraction was also derived from peak velocities as peak systolic wave velocity divided by the sum of peak systolic and diastolic velocities. Duration of the atrial reversal wave was measured from beginning to end of the atrial reversal wave (Fig 1⇑).
All measurements represent an average of five to seven consecutive cardiac cycles. We have previously shown that tricuspid inflow parameters derived with this approach are almost identical to those obtained during end-expiratory apnea.23 Similar results were observed with hepatic vein flow parameters in our laboratory, with the mean percent difference between averaged values and those at end-expiratory apnea being 4±3% for peak systolic velocity, 6±5% for peak diastolic velocity, and 4.5±6% for peak velocity of atrial reversal. Quantification of Doppler and echocardiographic variables was performed by a single observer without knowledge of clinical or hemodynamic data.
Hemodynamic Recordings and Measurements
Pressure calibration was performed before and immediately after pressure measurements. All readings were referenced to midaxillary line with the patient in the supine position. Determination of mean RAP was performed with Medex reusable transducers. All patients with central venous catheters (7.9F or 9F) in the Intensive Care Units had chest radiograms identifying the position of the catheter in the superior vena cava close to its junction with the right atrium or in the high right atrium. For patients in the catheterization laboratory, the proximal port of the Swan-Ganz catheter was used for measurements. All pressure recordings were obtained simultaneously with the Doppler tracings and echocardiographic imaging. Pressure measurements were determined at end expiration, and an average of three to five cycles was obtained.
Interobserver and intraobserver reproducibilities of Doppler parameters and predicted mean RAP were performed in 10 studies chosen at random. Measurements were obtained from the same recordings (but not necessarily the same beats) by a second observer and later by the first observer. Variability was expressed as the difference between observations and the mean percent error, derived as the absolute difference between the two sets of observations divided by mean of the observations.
Results are expressed as mean±SD. An unpaired t test was used to compare variables between groups of patients. To compare qualitative ventricular function assessment and mean RAP, the χ2 test was used. Correlations between echocardiographic and Doppler parameters of systolic and diastolic function and mean RAP, age, and heart rate were performed with linear regression analysis. Stepwise multiple linear regression was subsequently performed. The best derived equation from the initial study group was then used to predict RAP in the prospective population. Statistical significance was set at P≤.05.
Of the 45 patients screened, 10 were excluded: 5 because of nonsinus rhythms, 3 for severe tricuspid regurgitation, and 2 because of inadequate recording of hepatic venous flow. The initial population therefore consisted of 35 patients, 19 women and 16 men with a mean age of 60±15 years. Fourteen patients had studies in the cardiac catheterization laboratory. Table 1⇓ shows the clinical diagnoses of the initial group. Left ventricular ejection fraction averaged 48±15% (range, 18% to 67%). Twelve patients (34%) were on mechanical ventilation. Overall right ventricular function was depressed in 10 patients (28%) and normal in 25. Right ventricular ejection fraction ranged between 25% and 80% (mean, 49±15%). Right atrial maximal volume averaged 43±21.5 cm3 (range, 21 to 142 cm3); minimal volume, 24±19 cm3 (range, 8 to 118 cm3); and atrial emptying fraction, 46.5±13.6% (range, 17 to 68%). Maximal diameter of the inferior vena cava averaged 1.4±0.52 cm (range, 0.7 to 2.6 cm), with a minimal diameter after the inspiratory effort of 0.83±0.6 cm (range, 0 to 2.2 cm) and a percent collapse of 47±22% (range, 8% to 100%). Diameter of the hepatic vein at the site of Doppler interrogation averaged 0.9±0.36 cm (range, 0.5 to 2 cm). Table 2⇓ details the results of the hemodynamic and Doppler measurements.
Relation of Echocardiographic Measurements to Mean RAP
Table 3⇓ lists the echocardiographic parameters in patients with normal mean RAP (≤8 mm Hg) and elevated pressure. Right atrial emptying fraction was significantly lower in patients with elevated mean RAP, as was the percent inferior vena caval collapse. Although absolute measurements of minimal and maximal right atrial volumes and vena caval diameters were larger in patients with elevated RAP, this difference reached statistical significance with inferior vena caval measurements. Right ventricular function tended to be lower in patients with elevated RAP but was not significantly different.
Weak relations were observed between right atrial volumes and mean RAP (r=.49 and r=−.45 for minimum volume and emptying fraction, respectively; both P<.01). The best correlations of inferior vena caval measurements with mean RAP were observed with percent collapse (r=−.63, P<.0001; Fig 2⇓) and minimum diameter (r=.5, P<.01). In the subgroup of patients with mechanical ventilation, no significant correlation was observed (r=.4, P=.23 for maximal diameter; r=.25, P=.45 for minimal diameter; r=.24, P=.48 for collapse index of the inferior vena cava; Fig 2⇓). In the group without mechanical ventilation, correlations of inferior vena caval measurements were stronger, the best being for the collapse index (collapse index: r=−.76, P<.001, mean RAP=19−0.21×caval collapse index; SEE=4.2 mm Hg; r=.4; P=.05 for maximal diameter; r=.66 for minimal diameter; Fig 2⇓).
Relation Between Doppler Parameters and Mean RAP
Table 4⇓ compares Doppler parameters in patients with normal and elevated mean RAP. Among tricuspid inflow variables, the ratio of early to late velocity was the most significant difference between the two groups, followed by atrial filling fraction. The other parameters had directional changes but did not reach statistical significance. For hepatic vein flow parameters, results with the maximal systolic, diastolic, and atrial reversal velocities were almost identical to those observed with time-velocity integral measurements. Of the hepatic vein flow parameters, the systolic filling wave (peak velocity or time-velocity integral) and parameters that incorporated this measurement were significantly different in patients with normal versus elevated RAP (Table 4⇓). A tendency for longer duration of the atrial reversal wave was seen in the group with elevated mean RAP.
Correlations between mean RAP and Doppler parameters of tricuspid and hepatic vein flow velocities are shown in Table 5⇓. Among tricuspid inflow parameters, the most significant relation with mean RAP was observed with the ratio of early to late velocity (r=.66; Fig 3⇓). Weak relations were observed with early and late velocities and atrial filling fraction, whereas no significant relation was seen with acceleration or deceleration time. On the other hand, stronger relations were seen between Doppler parameters of hepatic vein flow and RAP, particularly with parameters that included the systolic filling wave alone or in combination with diastolic filling wave (r range, −.76 to −.86; Table 5⇓). The highest correlation was between mean RAP and systolic filling fraction (r=−.85 for velocities and r=−.86 for time-velocity integrals; Fig 4⇓). Excluding patients with heart rates >90 beats per minute (n=7) resulted in a correlation coefficient between systolic filling fraction and mean RAP of .92. The duration of atrial reversal wave related significantly to RAP (r=.63).
Significant interrelations were observed among tricuspid Doppler parameters and those of hepatic vein flows. A significant correlation was noted between tricuspid early velocity and hepatic vein diastolic velocity (r=.47, P=.03) and between the ratio of early to late velocity and systolic filling fraction (r=−.61, P=.002).
Relation of Doppler Variables to Right Ventricular Systolic Function
In the group with depressed right ventricular systolic function (n=10), deceleration time was shorter compared with the group with normal function (154±82 versus 257±82 milliseconds, P<.01). A trend for a higher early velocity and ratio of early to late velocity along with shorter acceleration time in patients with depressed right ventricular function was also seen. For hepatic flow variables, the only significant difference was observed with systolic filling fraction (0.34±0.28 and 0.59±0.14 in depressed and normal function, respectively; P<.05). The group with depressed right ventricular systolic function tended to have lower peak systolic velocity, higher peak diastolic and atrial reversal velocities, and a longer duration of atrial reversal wave. A statistically significant but weak relation was noted between right ventricular ejection fraction and systolic filling fraction (r=.38, P<.01).
Relation of Doppler Variables to Right Atrial Volumes and Inferior Vena Caval Dimensions
The most significant relation observed between right atrial function and tricuspid inflow dynamics was that of right atrial emptying fraction to tricuspid ratio of early to late velocity (r=−.66, P=.003). Trends for a higher ratio of early to late velocity with larger atrial maximal and minimal volumes were noted but did not reach statistical significance. Other tricuspid inflow parameters did not relate significantly to atrial volumes or emptying fraction.
For hepatic venous flow, significant inverse correlations were observed between systolic filling fraction and right atrial maximal volume (r=−.45) and minimal volume (r=−.51), and a direct correlation was noted with right atrial emptying fraction (r=.57). A significant relation (r=−.48) was observed between atrial emptying fraction and the difference in duration of the atrial reversal wave and tricuspid A wave. Other hepatic vein flow variables related poorly to right atrial volumes and function. A significant correlation was noted between inferior vena caval collapse index and systolic filling fraction (r=.64, P=.01).
Relation of Doppler Variables to Pulmonary Artery Pressure, Age, and Heart Rate
A trend was noted in patients with higher pulmonary artery systolic pressure to have a higher ratio of early to late velocity, shorter deceleration time, and lower hepatic vein systolic filling fraction. However, none reached statistical significance. No significant relation was observed between age or heart rate and Doppler parameters of tricuspid or hepatic vein flows.
Prediction of Mean RAP in a Prospective Population
Table 6⇓ shows the sensitivity and specificity of the best Doppler and two-dimensional variables for separating normal from elevated (>8 mm Hg) mean RAP in the initial population. The best separation was seen with systolic filling fraction. With stepwise multiple linear regression analysis, the relation of systolic filling fraction to mean RAP was not improved by the addition of tricuspid or hepatic vein Doppler parameters or echocardiographic measurements of right ventricular and atrial function or inferior vena cava. The best model in the prediction of RAP was Mean RAP=21.6−24×SFF, where RAP is in millimeters of mercury and SFF is systolic filling fraction, derived as hepatic vein systolic time-velocity integral divided by the sum of systolic and diastolic time-velocity integrals. This equation was also similar to that derived with maximal velocity measurements.
The above relation was tested prospectively in a separate patient population for the estimation of mean RAP. Of 58 patients screened, 8 were excluded. Reasons for exclusion were severe tricuspid regurgitation in 2 patients, inability to record hepatic venous flow in 3, and nonsinus rhythm in 3. The prospective population therefore consisted of 50 patients, 30 men and 20 women with a mean age of 60±17 years (range, 17 to 85 years); 4 patients were younger than 40 years of age. Left ventricular ejection fraction averaged 51±16.6% (range, 16% to 75%). Ten patients had depressed right ventricular systolic function. Fifty percent were on mechanical ventilation. Tables 2⇑ and 7⇓ give the hemodynamic data and clinical diagnoses, respectively, of this group. Mean RAP ranged between 1 and 22 mm Hg (mean, 9.16±5.33 mm Hg), and systolic filling fraction ranged between 0% and 0.85% (mean, 0.55±0.2%). Predicted RAP with the regression equation correlated well with observed RAP (r=.89) with a regression equation close to the identity line (Fig 5⇓).
In the whole group of 85 patients, 20 had depressed right ventricular systolic function. Of 49 patients with normal RAP, 6 had depressed right ventricular function. Of 36 patients with elevated RAP, 22 had normal right ventricular systolic function. Fig 5⇑ shows the correlation between Doppler-predicted and observed mean RAP in the total population. The relation was similar in patients on mechanical ventilation (r=.91) and in ambulatory patients in the catheterization laboratory (r=.85). The difference between predicted and observed pressure averaged −0.31±2.41 mm Hg in the prospective group (range, −6 to 6 mm Hg) and −0.2±2.6 mm Hg in the 85 patients (Fig 6⇓). Sensitivity and specificity for a mean RAP of >10 mm Hg were 82% and 96%, respectively, for the 85 patients; with 8 mm Hg as the cutoff, sensitivity and specificity were 86% and 92%, respectively.
Table 8⇓ shows intraobserver and interobserver reproducibilities of Doppler parameters of hepatic vein flow and derived mean RAP with the regression equation. Intraobserver and interobserver derivations of mean RAP were highly correlated (r=.93 and .95, respectively). Reproducibility of tricuspid inflow variables has been previously reported from our laboratory.23
This study demonstrates that among various echocardiographic and Doppler variables of the right side of the heart, Doppler parameters of hepatic vein flow, particularly those including the systolic filling wave, had the strongest relation to mean RAP. The interrelation among echocardiographic and Doppler parameters was evaluated. The strong relation observed between systolic filling fraction and mean RAP was maintained in a prospective patient population with a variety of cardiac and noncardiac clinical conditions.
Relation of Tricuspid Inflow Variables to RAP
Tricuspid diastolic inflow reflects the effect of filling pressure, right ventricular relaxation, and net AV compliance. Similar to mitral inflow, the pattern of tricuspid inflow in normal individuals has been shown to depend on age among other factors, with the older age group having a reduced early diastolic filling.23 24 In the present study, the confounding effects of filling pressures and right ventricular disease have altered the above relation with age, with a tendency for only a lower ratio of early to late velocity in older individuals. The effect of disease states on the relation of mitral inflow dynamics with age also was demonstrated previously in our laboratory.25 The dependence of tricuspid filling dynamics on loading conditions was described previously.26 A higher mean RAP results in a greater initial transvalvular gradient and therefore in a higher early velocity and ratio of early to late velocity. Deceleration time is likely to be affected predominantly by net AV compliance and tricuspid valve area rather than actual pressures,27 accounting for the weak relation observed between this parameter and mean RAP. Overall, the multifactorial determinants of tricuspid inflow variables most likely precluded a stronger relation of tricuspid inflow dynamics with mean RAP.
Relation of Hepatic Venous Flow to RAP
The phases of hepatic venous flow in normal individuals were described recently.28 Determinants of the systolic forward flow include atrial relaxation, descent of the tricuspid annular plane toward the ventricular apex, and RAP. The higher the RAP, the lower the pressure gradient between the hepatic veins and the right atrium and thus the lower the forward systolic flow. This observation was described previously in patients with restrictive heart disease and elevated filling pressures.12 13 In the present study, systolic forward flow parameters, particularly systolic filling fraction, derived with either time-velocity integrals or maximal velocities, had the best relation to mean RAP and allowed a good estimation of atrial pressure in patients with a variety of underlying clinical conditions. Although right ventricular function was also a determinant of systolic filling fraction, this relation was much weaker and failed to contribute significantly in the multivariate model. Similarly, other echocardiographic and Doppler parameters that related significantly to mean RAP failed to add to the already strong relation of systolic filling fraction with mean RAP.
The relations described here for atrial filling dynamics are similar overall to those described previously for the left side of the heart. The high correlations of hepatic vein systolic filling fraction observed in this study are similar to those observed between pulmonary venous systolic filling fraction and left atrial pressure by use of transesophageal echocardiography (r=.88).9 The weak but significant relation found between the duration of atrial reversal wave and RAP also was reported previously.29 However, the lack of a significant relation between velocity of atrial reversal wave and mean atrial pressure in the present study has not been consistently demonstrated for the left atrium.9 11 29 The lack of correlation between diastolic and atrial reversal velocities and mean RAP may reflect that each relates best to atrial v and a pressure waves, respectively, rather than mean atrial pressure.12
With respiratory variability present in any right-side Doppler velocity recording, measurements were performed as an average of several consecutive beats, which yields results similar to those obtained during end-expiratory apnea.23 This allowed assessment of RAP in patients with dyspnea or on mechanical ventilation in whom the relation of systolic filling fraction with RAP proved to be similar to those without assisted ventilation. Exclusions related mostly to nonsinus rhythms and occasional technical difficulties in patients in the Intensive Care Unit. Unlike recordings of pulmonary vein flow,30 Doppler assessment of hepatic vein flow with the transthoracic approach is simpler and feasible in the majority of patients even in Intensive Care Units, as shown in this study, thus allowing assessment of RAP in most patients. Furthermore, transesophageal echocardiography was shown to offer no advantage over the transthoracic approach in obtaining the above measurements.31
Relation Between Tricuspid and Hepatic Venous Flow
The higher the early velocity across the tricuspid valve, the smaller the blood volume in the right atrium and hence the greater the forward flow from the hepatic veins in diastole. Such a relation was described previously for left-side inflow patterns.29 The difference in duration of hepatic venous atrial reversal and tricuspid A wave was related only weakly to mean RAP (r=.27). A similar weak relation was described previously for this variable derived from left-side Doppler measurements and pulmonary capillary wedge pressure. This index has been shown in the left ventricle to relate well with the left ventricular A wave and end-diastolic pressure.11 In this study, measurements of right ventricular diastolic pressures were not available to assess this relation.
Systolic Right Ventricular Function and RAP
Patients with depressed right ventricular function were observed to have a shorter tricuspid deceleration time, predominant diastolic flow in the hepatic veins, and a tendency for higher early velocity and ratio of early to late velocity. A significant overlap in these Doppler variables, however, was observed among patients with normal and abnormal right ventricular systolic function, which stems from the several determinants of ventricular filling dynamics.2 32 Right ventricular systolic function was a poor discriminator of mean RAP, which can be explained predominantly by the wide range of loading conditions in the present population. Right ventricular systolic function related weakly to hepatic vein systolic filling fraction in the present study, a finding consistent with previous observations by Basnight et al,33 who demonstrated no relation between systolic pulmonary venous flow and left ventricular ejection fraction. Similarly, in a study by Keren et al,34 10 of 28 patients with dilated cardiomyopathy had preserved systolic and diastolic pulmonary veins flows despite significantly reduced mitral annular descent. In addition, the presence of reduced systolic venous flow in patients with restrictive cardiomyopathy, elevated atrial pressure, and preserved ventricular systolic function13 35 further support that ventricular systolic function, through descent of the cardiac base, is a minor determinant of atrial filling dynamics compared with atrial pressure. However, in patients with pericardial disease such as constrictive pericarditis or cardiac tamponade, assessment of mean atrial pressure with systolic filling fraction is limited because the systolic venous wave usually is preserved despite elevated filling pressures.35 36 In these cases, usually suspected with echocardiographic findings of a plethora of the inferior vena cava, exaggerated respiratory variability in ventricular inflow dynamics, septal bounce, or pericardial effusion, determination of atrial pressure with the proposed Doppler method should not be performed.
Right Atrial Function and RAP
Measurements of right atrial volumes and function related significantly to RAP. With the relation of atrial pressure to volume, larger atrial volumes at end systole and end diastole and lower emptying fractions were found to be associated with higher mean RAP. However, because atrial compliance and function are also significant modifiers of this relation, the weak yet significant relations observed between atrial volumes and mean RAP may be accounted for on this basis. The presence of a larger atrial volume at the onset of ventricular systole (minimal atrial volume) is likely to reduce forward systolic flow from the hepatic veins. A larger atrial volume at the end of ventricular systole (maximal atrial volume) is also likely to cause similar findings, hence the negative correlations observed between right atrial volumes and hepatic vein systolic filling fraction. The addition of atrial volumes and function in the multivariate model did not improve the already strong relation of systolic filling fraction to mean RAP.
Inferior Vena Cava and RAP
Measurement of inferior vena caval diameter and its change during an inspiratory effort, commonly a sniff test, is used frequently in the echocardiography laboratory to estimate RAP.17 Its use is limited, however, in tachypneic patients and those on mechanical ventilation.18 Another limitation is the difficulty of standardizing the inspiratory effort. Previous studies have demonstrated a good relation between the collapse index of the inferior vena cava and RAP in patients without assisted ventilation.17 In patients with mechanical ventilation, however, inferior vena caval collapse has been shown to relate poorly to mean RAP (r=.13), whereas maximal caval diameter showed only a weak relation in these patients (r=.58).18 Our findings are similar to those previously reported. Because the present population included patients with and without assisted ventilation, overall weaker relations with the caval collapse index were demonstrated for the whole population. When patients were separated into those with and without mechanical ventilation, the relations observed were nearly identical to those previously reported.16 17 Even in patients without assisted ventilation, the relation of caval collapse index with mean RAP was weaker than that for systolic filling fraction, with a larger spread and standard error of estimate. Similar to other echocardiographic indexes, the addition of parameters derived from the inferior vena cava did not improve the relation of hepatic vein systolic filling fraction to mean RAP.
Estimation of RAP is helpful in the overall management of patients with hemodynamic disorders and in the derivation of pulmonary artery pressure with Doppler echocardiography. Among all parameters tested, hepatic vein systolic filling fraction, inferior vena caval collapse index, and the ratio of early to late velocity of tricuspid inflow have provided useful yet simple indexes for the assessment of mean RAP. Determination of systolic filling fraction provided a reasonable estimate of mean RAP, with a 95% CI of ±5 mm Hg, smaller than that observed with the other parameters tested, and is in general the preferred method for estimating mean RAP. Doppler recordings of tricuspid inflow and echocardiographic imaging of the vena cava during an inspiratory effort, however, are also important in the overall evaluation of right-side hemodynamics and, in addition to hepatic venous flow, are already acquired routinely in several laboratories. Assessment of these parameters corroborates findings of atrial filling dynamics and is important when derivation of mean RAP cannot be performed with systolic filling fraction. The latter was feasible in 82% of the total patients screened (n=103), the majority of whom were in an intensive care setting, and therefore would be applicable to an even higher percentage of the general population.
It is noteworthy to mention the characteristics of patients in whom the observed relations in this study may not be applicable. These include patients with nonsinus rhythm or pericardial diseases in whom parameters derived from the inferior vena cava would be more accurate for estimation of RAP. Application of the current Doppler relations also should be reserved for middle-aged and older individuals (age group of present study) because of the effect of age on filling dynamics. Patients with severe tricuspid regurgitation, who usually have reversal of the systolic hepatic vein flow, also were not included in the present study and await evaluation. On the other hand, in patients on mechanical ventilation, inferior vena caval collapse index should not be used, and estimation of mean RAP with systolic filling fraction is the preferred method.
The effect of respiration on right-side inflow has been well described.23 Although simultaneous recording of respiration was not performed in this study, we have shown previously that averaging consecutive cardiac cycles yields almost identical results to those during end-expiration apnea. This allows evaluation and application of the current findings in the general population, including dyspneic patients and those on mechanical ventilation, without the need for special devices for recording of simultaneous respiration.
In the present study, we measured mean RAP and did not quantify phasic RAP. This was performed because pressure was measured with clinically available fluid-filled catheters rather than high-fidelity micromanometers. The fluid-filled catheters used may have had some damping effect on the phasic pressure waveforms. The use of mean RAP is much less affected by damping,37 is used clinically in hemodynamic management, and has been advocated in previous studies for the assessment of pulmonary artery pressure.6 Furthermore, in the estimation of pulmonary artery pressures, the use of peak atrial v wave may result in overestimation of pulmonary pressure.6 7
Right ventricular isovolumic relaxation time was not measured and may have helped improve the relation of tricuspid inflow dynamics in the evaluation of RAP in a manner analogous to left-side pressures.10 11 30 In contrast to the left ventricle, however, measurement of right ventricular isovolumic relaxation time requires phonocardiography for optimum accuracy, which adds to the complexity of the evaluation and renders it less clinically applicable.
Among echocardiographic parameters of right-side cardiac structure and function and Doppler variables of right atrial and ventricular filling dynamics, parameters that included hepatic systolic wave had the strongest relation to mean RAP. The use of hepatic vein systolic filling fraction provided a good prediction of mean RAP in a population with a variety of underlying clinical conditions. Thus, Doppler echocardiography provides an assessment of RAP that can be used clinically in the overall hemodynamic evaluation of right atrial and ventricular function.
Computational assistance was provided by the CLINFO Project, funded by grant RR-00350 from the Division of Research Resources, NIH, Bethesda, Md. We wish to thank Eula Landry for her assistance in the preparation of this manuscript.
Guest editor for this article was Bernadine Healy, MD, Cleveland (Ohio) Clinic Foundation.
Presented in part at the 44th Annual Scientific Session of the American College of Cardiology, New Orleans, La, March 20, 1995.
- Received May 22, 1995.
- Revision received October 4, 1995.
- Accepted October 15, 1995.
- Copyright © 1996 by American Heart Association
Miyatake K, Okamoto M, Kinoshita N, Matsuhisa M, Nagata S, Beppu S, Park YD, Sakakibara H, Nimura Y. Pulmonary regurgitation studied with the ultrasonic pulsed Doppler technique. Circulation. 1982;65:969-976.
Yock PG, Popp RL. Noninvasive estimation of right ventricular systolic pressure by Doppler ultrasound in patients with tricuspid regurgitation. Circulation. 1984;70:657-662.
Kuecherer HF, Muhiudeen IA, Kusumoto FM, Lee E, Moulinier LE, Cahalan MK, Schiller NB. Estimation of mean left atrial pressure from transesophageal pulsed Doppler echocardiography of pulmonary venous flow. Circulation. 1990;82:1127-1139.
Sivaciyan V, Ranganathan N. Transcutaneous Doppler jugular venous flow velocity recording. Circulation. 1978;57:930-939.
Hatle LK, Appleton CP, Popp RL. Differentiation of constrictive pericarditis and restrictive cardiomyopathy by Doppler echocardiography. Circulation. 1989;79:357-370.
Suzuki Y, Kambara H, Kadota K, Tamaki S, Yamazato A, Nohara R, Osakada G, Kawai C. Detection and evaluation of tricuspid regurgitation using a real-time, two-dimensional, color-coded, Doppler flow imaging system: comparison with contrast two-dimensional echocardiography and right ventriculography. Am J Cardiol. 1986;57:811-815.
Jiang L, Wiegers SE, Weyman A. Right ventricle—principles and practice of echocardiography. In: Weyman A, ed. Principles and Practice of Echocardiography. Philadelphia, Pa: Lea & Febiger; 1994:901-906.
Zoghbi WA, Habib JB, Quiñones MA. Doppler assessment of right ventricular filling in a normal population: comparison with left ventricular filling dynamics. Circulation. 1990;82:1316-1324.
Nishimura RA, Abel MD, Hatle LK, Tajik AJ. Relation of pulmonary vein to mitral flow velocities by transesophageal Doppler echocardiography: effect of different loading conditions. Circulation. 1990;81:1488-1497.
Stoddard MF, Pearson AC, Kern MJ, Ratcliff J, Mrosek DG, Labovitz AJ. Influence of alteration in preload on the pattern of left ventricular diastolic filling as assessed by Doppler echocardiography in humans. Circulation. 1989;79:1226-1236.
Keren G, Sonnenblick EH, LeJemtel TH. Mitral anulus motion: relation to pulmonary venous and transmitral flows in normal subjects and in patients with dilated cardiomyopathy. Circulation. 1988;78:621-629.
Klein AL, Cohen GI, Pietrolungo JF, White RD, Bailey A, Pearce GL, Stewart WJ, Salcedo EE. Differentiation of constrictive pericarditis from restrictive cardiomyopathy by Doppler transesophageal echocardiographic measurements of respiratory variations in pulmonary venous flow. J Am Coll Cardiol. 1993;22:1935-1943.
Gabe IT. Pressure measurement in experimental physiology. In: Bergel DH, ed. Cardiovascular Fluid Dynamics. New York, NY: Academic Press Inc; 1972:11-15.