In current practice, right atrial pressure (RAP) is an essential component in the hemodynamic assessment of patients and a requisite for the noninvasive estimation of the pulmonary artery pressures. RAP provides an estimation of intravascular volume, which is a critical component for optimal patient care and management. Increased RAP is associated with adverse outcomes and is independently related to all-cause mortality in patients with cardiovascular disease. Although the gold standard for RAP evaluation is invasive monitoring, various techniques are available for the noninvasive evaluation of RAP. Various echocardiographic methods have been suggested for the evaluation of RAP, consisting of indices obtained from the inferior vena cava, systemic and hepatic veins, tissue Doppler parameters, and right atrial dimensions. Because the noninvasive evaluation of RAP involves indirect measurements, multiple factors must be taken into account to provide the most accurate estimate of RAP. The authors review the data supporting current guidelines, identifying areas of agreement, conflict, limitation, and uncertainty.
Attention ASE Members:
ASE has gone green! Visit www.aseuniversity.org to earn free CME through an online activity related to this article. Certificates are available for immediate access upon successful completion of the activity. Non-members will need to join ASE to access this great member benefit!
Physical Examination
Jugular Venous Pressure (JVP)
Sir Thomas Lewis, in 1930, first proposed the determination of a patient’s venous pressure during the physical examination. Lewis observed that the top of the jugular veins of normal individuals and the vertical distance from the top of the fluid in the manometer (which was used at that time to measure CVP) always came to lie within 1 to 2 cm of the sternal angle. Currently, examination of JVP is the mainstay of the bedside estimation of CVP.
Evaluation of CVP using the jugular veins is preferably done using the right-sided jugular vein, which is in direct line with the right atrium, and clinical assessment of CVP on the left may be marginally higher than that on the right. Both the external jugular vein and the internal jugular vein (IJV) can be used for evaluation. Although the external jugular vein is easier to visualize, the IJV is preferred because it does not have valves and is in line with the superior vena cava (SVC) and the right atrium. The patient should be recumbent with the head elevated at 30° to 45°. The level of venous pressure is estimated by identifying the highest point of oscillation of the jugular vein (which occurs during the expiratory phase of respiration). This level is then related to the middle of the right atrium, where venous pressure is, by convention, zero. A reliable substitute is the angle of Louis at the junction of the manubrium and the body of the sternum, located 5 cm above the middle of the right atrium (in approximation for practical purposes). The vertical distance (in centimeters) from the sternal angle to the top of the jugular venous wave represents the JVP ( Figure 1 ); thus, CVP equals JVP + 5 cm. Because the clavicle lies vertically 2 cm above the sternal angle, only a CVP of 7 cm will be observed. JVP estimation may be inaccurate if the jugular vein is constricted or torturous, and it might be difficult to assess in patients with short necks, with prior neck surgery, or with prior catheter placement in the jugular vein.
The Abdominojugular Reflux
The test consists of assessing JVP before, during, and after abdominal compression. It is performed with firm pressure of 20 to 30 mm Hg applied to the midabdomen for 10 to 30 sec. The CVP of normal individuals usually remains unchanged or does not increase by >4 cm for a beat or two (usually for <10 sec), or it may fall slightly. If the CVP rises by >4 cm and stays elevated throughout the maneuver, this correlates with elevated RAP. Reports suggest that a positive result is indicative of right ventricular (RV) failure or elevated pulmonary capillary wedge pressure. However, if the patient strains (Valsalva maneuver) during the test, it may cause a false-positive result.
In general, the physical examination for the assessment of RAP has limited accuracy compared with invasive or echocardiographic studies, most commonly underestimating JVP in the setting of elevated CVP. Eisenberg et al. found that after a baseline physical examination and assessment of CVP, catheter-derived CVP measurements differed significantly from those obtained on clinical examination in 50% of patients, and in 58% of these, the treatment plan was changed on the basis of the invasive assessment of CVP. Consequently the use of the physical examination might be best for categorizing CVP as either low to normal or elevated.
Doppler Echocardiography
Doppler echocardiography is routinely used to noninvasively estimate RAP. The Doppler echocardiographic methods to estimate RAP include assessment of the inferior vena cava (IVC), SVC, and hepatic vein indices of size and flow; Doppler systemic venous flow; tricuspid valve Doppler inflow; tricuspid valve tissue Doppler; and evaluation of right atrial (RA) dimensions. However, in uncommon circumstances, the flow into the right atrium and consequently the Doppler velocities may be affected by RA compression or distortion, as well as in the setting of RV inflow obstruction due to tumor or tricuspid stenosis. In these situations, assessing the tricuspid gradient can be useful for the estimation of RAP.
IVC Indices
The IVC is a highly compliant vessel, and consequently its size and dynamics vary with changes in CVP and volume. Blood flow from the superior and IVC into the right atrium is biphasic, with the largest forward flow occurring during ventricular systole. In general, there is a reciprocal relationship between pressure and flow; when flow increases (in the IVC as well as the right atrium), pressure decreases and vice versa. During inspiration (which produces negative intrathoracic pressure), vena cava pressure decreases and flow increases ( Figures 2 A–2C). Because the vena cava acts as a capacitance reservoir, phasic increases in forward flow from the cava to the right atrium are accompanied by decreases in vena cava size. The smallest IVC dimension is seen during ventricular systole. At low or normal RAPs, there is systolic predominance in IVC flow, such that the systolic flow is greater than the diastolic flow. As RAP increases, it is transmitted to the IVC, resulting in blunting of the forward systolic flow, reduced IVC collapse with inspiration, and eventually IVC dilatation ( Figures 2 D–2F). The size and area of the IVC are affected by position as well: IVC diameter and area are consistently largest when the patient is evaluated in the right lateral position, intermediate in the supine position, and smallest in the left lateral position. Patient position, therefore, is an important factor to consider when correlating IVC size and shape with hemodynamic variables.
In 1979, Natori et al. first described measuring IVC diameter and its change during respiration. Several studies have evaluated the correlation between RAP and different IVC parameters ( Table 1 ). However, fewer studies have evaluated the validity of the suggested IVC parameters for the accuracy of the estimation of RAP. Most, but not all, studies have demonstrated good correlations between the IVC collapsibility index ([IVC max − IVC min ]/IVC max ) and RAP (0.57 < r ≤ 0.76). Although there is a correlation between IVC diameter and RAP (0.72 < r ≤ 0.86), some reports suggest that the correlation found between IVC diameter and RAP does not permit it to be used for the reliable estimation of RAP, because of the variability and overlap between patients with normal RAPs and those with elevated RAPs. It is hypothesized that an increase of RAP beyond a certain level may cause only minimal increases in IVC diameter and the degree of IVC collapsibility with inspiration. Thus, IVC dimensions and collapsibility can be used to detect elevated CVP, but they have limited utility in identifying the magnitude of CVP elevation.
| Study | Number of patients | Timing of invasive and noninvasive evaluation | Major findings |
|---|---|---|---|
| Natori et al. | 14 | NA | IVCCI was inversely proportional to CVP when >10 cm H 2 O |
| Mintz et al. | 111 | NA | IVC EDD normalized to BSA was correlated with RAP ( r = 0.72) |
| IVC EDD > 10 mm/m 2 predicted elevated RAP (predictive accuracy, 94%) | |||
| IVCCI was not correlated with RAP | |||
| Moreno et al. | 175 | <24 h | IVC diameter was poorly correlated with RAP ( r ≤ 0.4) |
| IVCCI was correlated with RAP ( r = 0.71) | |||
| Nakao et al. | 83 | <24 h | IVC diameter ( r = 0.85) and area ( r = 0.89) were correlated with RAP |
| IVC diameter > 10 mm predicted elevated RAP (>8 mm Hg) (sensitivity, 84%; specificity, 96%; predictive accuracy, 95%) when measured in the left lateral position | |||
| IVCCI was not correlated well with RAP and was more affected by IVC size than by RAP | |||
| Simonson et al. | 27 | Simultaneous | Poor association between maximal IVC diameter and RAP ( r = 0.35) |
| Good association between minimal IVC diameter and RAP ( r = 0.56) | |||
| IVCCI was inversely correlated with RAP ( r = −0.57) | |||
| Kircher et al. | 83 ∗ | <24 h | IVCCI was correlated with RAP ( r = 0.75); best sensitivity, specificity, and predictive accuracy for RAP < 10 and > 10 mm Hg when IVCCI was ≥50% |
| IVC EDD was correlated with RAP ( r = 0.71) | |||
| Jue et al. | 49 † | Simultaneous | IVC diameter at expiration was poorly correlated with RAP ( r = 0.58) |
| No correlation between IVC change and RAP ( r = 0.13) | |||
| IVC diameter ≤ 12 mm predicted RAP ≤ 10 mm Hg (sensitivity, 25%; specificity, 100%) | |||
| Nagueh et al. | 35 | Simultaneous | IVCCI was correlated with RAP ( r = 0.76) |
| No good correlation for IVCCI with mechanically ventilated patients ( r = 0.4) | |||
| Ommen et al. | 71 | Simultaneous | IVC dimension was correlated directly with RAP ( r = 0.74) |
| Brennan et al. | 102 ‡ | ±1 h of invasive evaluation | Five different classifications on the basis of IVC size and collapsibility (see Table 2 ) |
| Using traditional ASE 2005 criteria, only 43% of patients were classified adequately | |||
| Indexing IVC to BSA did not improve accuracy |
∗ Patients on mechanical ventilation excluded; retrospective analysis.
† All patients on mechanical ventilation.
‡ Thirty percent of patients had undergone cardiac transplantation.
On the basis of several studies, the American Society of Echocardiography (ASE) in 2005 recommended using maximal IVC diameter 1 to 2 cm from the junction of the right atrium and the IVC at end-expiration and the IVC collapsibility index to give an estimate of RAP. This measurement is best obtained from the subcostal view, with the IVC viewed in its long axis ( Figure 2 ). Measurements should be made with the patient lying supine (because the left lateral position may underestimate maximal IVC diameter). However, Brennan et al. evaluated the 2005 ASE recommendations for assessing RAP and found that only 43% of patients were correctly classified. On the basis of the results of their study, they suggested a new cutoff range that gives an estimation with range limits of 5—10 mm Hg ( Table 2 ). The newer 2010 ASE guidelines have been revised as noted in Table 2 . These parameters yield more accurate results when estimating low or high RAP: IVC diameter < 2.1 cm and collapse > 50% correlates with a normal RAP of 0 to 5 mm Hg. IVC diameter < 2.1 cm with < 50% collapse and IVC diameter > 2.1 cm with > 50% collapse correspond to an intermediate RAP of 5 to 10 mm Hg. IVC diameter > 2.1 cm with < 50% collapse suggests a high RAP of 15 mm Hg. The guidelines recommend using midrange values of 3 mm Hg for normal and 8 mm Hg for intermediate RAP. If there is minimal collapse of the IVC (<35%) and/or secondary indices of elevated RAP are present ( Table 3 , discussed next), the guidelines recommend upgrading to the higher pressure limit (i.e., 5 and 10 mm Hg in the cases of normal and intermediate RAPs, respectively). Patients with low compliance with deep inspirations may have diminished IVC collapse, and a “sniff” maneuver causing a sudden decrease in intrathoracic pressure and by that accentuating the normal inspiratory response might be required to differentiate those with true diminished IVC collapsibility from those with normal collapsibility.
| ASE 2010 recommendations | Brennan et al. | ||
|---|---|---|---|
| IVC diameter (cm) and collapse (%) | RAP range (mean) (mm Hg) | IVC diameter (cm) and collapse (%) | RAP range (mm Hg) |
| Normal: ≤2.1 and >50 | 0–5 (3) | ≤2.1 and >55 | <5 |
| Indeterminate ∗ | 5–10 (8) | ≤2.1 and 35–55 or >2.1 and >55 | 0–10 |
| High: >2.1 and <50 | 10–20 (15) | ||
| >2.1 and 35–55 | 10–15 | ||
| >2.1 and <35 | 10–20 | ||
| ≤2.1 and <35 | Undetermined | ||
∗ In cases which the IVC diameter and collapse do not fit the normal or high criteria. Preferably secondary indices of elevated RAP should be integrated (see Table 3 ).
| • Dilated IVC with diminished respiratory collapse |
| • Tricuspid E/e′ ratio > 6 |
| • Diastolic flow predominance in the SVC, jugular vein, or hepatic veins |
| • Bulging interatrial septum to the left atrium |
| • Dilated right atrium |
The validity of adjusting the evaluation of IVC size to body surface area (BSA) is controversial. Data correlating IVC size indexed mainly to BSA are inconsistent and limited to only a few studies. Mintz et al. indexed IVC size measurements to patients’ BSAs and found a correlation between IVC index and RAP, but they still concluded that the degree of correlation did not permit the use of the IVC for an accurate estimation of RAP. Moreno et al. found no correlation between RAP and IVC parameters, whether indexed to BSA or not ( r < 0.25), whereas Cheriex et al. showed a good correlation ( r = 0.92) between IVC index and RAP for evaluating hydration status in hemodialysis patients. Kosiak et al. developed an alternative index consisting of the IVC/aorta ratio for evaluating patients’ volume status in the emergency department. They found in 52 healthy volunteers that IVC and aorta diameters significantly increased after fluid intake and proposed an IVC/aorta index of 1.2 ± 2 SD for SD = 0.17 as a reference value for the healthy population aged 20 to 30 years; however, this finding needs confirmation. Brennan et al. did not find that indexing the IVC to BSA improved accuracy over the standard IVC measurement for RAP estimation, and the ASE 2010 recommendations do not recommend an IVC index related to BSA.
The published data indicate that IVC size and collapsibility indices are appropriate to define RAP as high or low and are not a method for providing a precise numeric value. It should be noted that the IVC can be dilated in individuals with normal RAPs; Table 4 lists the common causes of a dilated IVC in the setting of normal RAP. To overcome some of the limitations of RAP estimation through IVC indices, additional Doppler echocardiographic parameters have been evaluated and proposed to better quantify RAP ( Table 5 ), which are further discussed.
| • Prominent Eustachian valve |
| • Athletic training |
| • Large BSA |
| • Mechanical ventilation |
| • Narrowing of the IVC-RA junction |
| • Web or tissue present in the IVC |
| Method | Number of patients | Timing of invasive and noninvasive evaluation | Major findings |
|---|---|---|---|
| Systemic veins | |||
| Jugular vein pattern and collapsibility | 34 | Simultaneous | CVP categorized as high (≥10 mm Hg) or low (<10 mm Hg) was correlated with IJV diameter ( r = 0.82) ∗ |
| 67 | Within 1 h | An increase of >17% IJV CSA during Valsalva maneuver ruled out elevated RAP (>12 mm Hg) (sensitivity, 90%; specificity, 74%; NPV, 90%; PPV, 60%) | |
| 38 † , | Simultaneous | Ultrasound compression was correlated with JVP but underestimated CVP ( r = 0.81–0.91 for JVP, r = 0.62–0.5 for CVP) | |
| Jugular venous flow ‡ | 82 | Within 1 d | Vs > Vd was associated with normal RAP |
| Vs ≤ Vd was associated with elevated RAP | |||
| Single diastolic flow was associated with elevated RAP | |||
| SVC flow | 40 | NA | Defined the normal pattern of SVC flow (Vs > Vd) |
| SVC flow § | 120 | Within 1 h | 2 ≥ Vs/Vd ≥ 1 (normal) = RAP ≤ 5 mm Hg (sensitivity, 86%; specificity, 78%) |
| Vs/Vd >2 (predominant systolic) = RAP ≤ 8 mm Hg (sensitivity, 69%; specificity, 81%) | |||
| Vs/Vd < 1 (predominant diastolic) = RAP > 8 mm Hg (sensitivity, 52%; specificity, 95%) | |||
| Hepatic veins | |||
| Hepatic venous flow | 40 | Within 12 hours | Defined normal hepatic venous flow (Vs > Vd) |
| 14 | Vs/Vd ≤ 1 was associated with elevated RAP | ||
| Left hepatic vein diameter || | 32 | Simultaneous | Left hepatic vein diameter was correlated well with RAP ( r = 0.81) |
| HVFF | 35 | Simultaneous | >55% suggested RAP > 8 mm Hg (sensitivity, 86%; specificity, 90%) ¶ |
| 50 | Simultaneous | RAP = 21–24 × HVFF ( r = 0.89) | |
| 71 | Simultaneous | Modest correlation with RAP ( r = −0.5) # | |
| DTI | |||
| Acceleration rate of RV early filling | 77 | Simultaneous | Acceleration rate was a predictor of RAP ( r = 0.98) |
| Acceleration rate > 560 cm/sec predicted RAP > 5 mm Hg (sensitivity, 100%; specificity, 99%) | |||
| Tricuspid E/e′ ratio | 62 | Simultaneous | E/e′ ratio > 6 suggested RAP > 10 mm Hg ( r = 0.75) ∗∗ |
| 89 | Simultaneous | RAP = 1.62 E/e′ +2.13 ( r = 0.7) †† | |
| 40 | Immediately before obtaining invasive data | No good correlation ( r = 0.09) ‡‡ | |
| Tricuspid E/A ratio | 40 ‡‡ , | Immediately before obtaining invasive data | Good correlation with RAP ( r = 0.57) §§ |
| RV rIVRT | 45 | Within 45 min | RV rIVRT< 59 ms was a predictor of RAP > 8 mm Hg (sensitivity, 80%; specificity, 88%; AUC, 0.994) |
| RA dimensions | |||
| 3D RAVi | 80 ‡‡ | Immediately before obtaining invasive data | 3D RAVi ≥ 35 mL/m 2 + IVC ≥ 2 cm was accurate for identifying RAP > 10 mm Hg (sensitivity, 89%; specificity, 92%; accuracy, 88%) |||| |
∗ Not applicable to mechanically ventilated patients.
† Twenty-nine percent of patients were mechanically ventilated.
‡ Not applicable to patients who had undergone cardiac surgery and patients with severe left ventricular volume overload.
§ Only patients with ejection fractions < 35% were included. Patients with severe tricuspid regurgitation were excluded.
|| Imaging was obtainable in 90% of patients. Underestimation occurred when RAP was >12 mm Hg.
¶ Also validated for mechanically ventilated patients. Patients with severe tricuspid regurgitation were excluded.
# Patients with mechanical ventilation or severe tricuspid regurgitation were not included.
∗∗ Also validated for mechanically ventilated patients.
†† Not applicable to patients who had undergone cardiac surgery.
‡‡ Only patients with acute decompensated heart failure were included.
§§ Inflow parameters could be determined in only 43% of patients.
Systemic Venous Indices
The central venous flow pattern seen in the vena cava, jugular, and hepatic veins is characterized as seen in Figure 3 A by three distinct waveforms when evaluated by Doppler. The first is the systolic wave (Vs), caused by RA relaxation and descent of the tricuspid ring associated with RV systole. The second is the diastolic wave (Vd), which occurs during rapid ventricular filling when the tricuspid valve is open. The third is a positive A wave, which occurs with RA contraction and represents reverse flow. The A wave is small and might not be present in normal individuals. In the majority of normal adults, inspiration increases the magnitude of Vs and Vd, whereas the A wave, if present, decreases in size. At low or normal RAPs, there is systolic predominant venous flow, such that the velocity of Vs is greater than the velocity of Vd ( Figure 3 A). As demonstrated in Figure 3 B, with elevation of RAP, the systolic flow predominance is lost, such that Vs is substantially decreased, and Vs/Vd is <1. The higher the RAP, the lower the pressure gradient between these veins and the right atrium, causing diminished forward systolic flow. This blunted gradient is present in patients with restrictive heart disease and elevated right-sided filling pressures. Sivaciyan and Ranganathan demonstrated that the jugular venous flow pattern and its variation are closely related to RAP. Although Vs/Vd > 1 was associated with normal RAP, those who demonstrated ratios ≤ 1 had elevated RAPs.
Appleton et al. found in healthy individuals that SVC flow was obtainable in all subjects, but hepatic vein flow imaging was adequate in only 40% of patients. A subsequent study by the same group in 14 patients with restrictive ventricular physiology also showed that in patients with elevated RAP, Vs/Vd was ≤1. A study by Ghio et al. also confirmed that SVC flow was closely related to RAP, with a normal Doppler pattern identifying patients with normal RAPs, a predominant Vs pattern identifying those with slightly elevated RAPs (≤8 mm Hg) (sensitivity, 69%; specificity, 81%), and a predominant Vd identifying those with elevated RAPs (>8 mm Hg) (sensitivity, 52%; specificity, 95%). Consistent with the initial study by Appleton et al. , the success rate reported in obtaining flow velocity curves from the SVC was 100% compared with 76% in the hepatic veins. Patients with severe tricuspid regurgitation were excluded from this study because this may cause altered flow patterns in the systemic veins.
Jugular Venous Flow and JVP
Several investigators have used ultrasonography of the jugular veins to predict CVP. Although the external jugular vein is easier to visualize, its tortuous course, competent valves, and small size might not accurately reflect the transmission of pressure from the right atrium. The right IJV is a large vessel that directly connects to the SVC, but when evaluated clinically, the IJV may be visible only 20% of the time. Lipton observed and described sonographic patterns within the IJV determining venous collapse between the supine, semiupright, and upright positions and concluded that patients can be differentiated as those with low (<10 mm Hg), high (>10 mm Hg), and extremely high (>20 mm Hg) CVPs. Donahue et al. demonstrated a good correlation ( r = 0.82) between IJV end-expiratory diameter and CVP in 34 supine patients. They determined with good accuracy whether a patient had low or high CVP. Simon et al. assessed right IJV vascular compliance and concluded that an increase in the cross-sectional area of the right IJV of >17% during the Valsalva maneuver ruled out elevated RAP. Deol et al. compared ultrasound collapse pressure with JVP obtained clinically and with CVP in a group of 38 patients, of whom 11 (29%) were mechanically ventilated. Although the ultrasound collapse pressure was capable of accurately measuring JVP, it underestimated CVP, especially when CVP levels were high, concluding that measuring the ultrasound collapse point (which reflects JVP) does not reflect true CVP.
Hepatic Vein Dimensions, Flow Patterns, and Collapsibility
The right and left hepatic veins empty into the IVC at the level of the diaphragm and are best imaged by echocardiography from the subcostal view ( Figure 4 ). The hepatic venous flow pattern is closely related to the central venous flow pattern ( Figure 3 B). Like the IVC, as RAP rises, the hepatic veins dilate and collapse less with inspiration. In one study, left hepatic vein diameter correlated well ( r = 0.81) with the percentage increase in CVP. Imaging was obtained in 90% of patients, yet underestimation occurred when RAP values exceeded 12 mm Hg.
