The pulmonary regurgitation Doppler velocity pattern is characterized by a rapid rise immediately after the closure of the pulmonary valve and a gradual deceleration until the end of diastole. The pulmonary regurgitation velocity corresponds to the diastolic pressure gradient between the PA and RV through diastole (i.e., both the velocity and gradient decline). The peak pulmonary regurgitation velocity at the start of diastole reflects the pressure gradient between the PA and RV the start of diastole. Application of the modified simplified Bernoulli equation to peak pulmonary regurgitation velocity (Fig. 11.2) with the addition of estimated RAP provides a reasonable estimate of mPAP [27]. Pulmonary regurgitation envelopes are often incomplete or absent, however, precluding this approach in many patients. Further, since the mPAP is lower than the sPAP, an inaccurate RAP estimate has a proportionally greater impact on mPAP estimates. Thus, while this method can be useful qualitatively to confirm sPAP estimates or when the TR envelope is of poor quality, it is not generally as robust or as commonly used.

The mean RV-RA systolic gradient can be derived from the Doppler profile of the tricuspid regurgitation (Fig. 11.1). The sum of this estimate and estimated RAP correlates with RHC mPAP [28]. This method depends on the presence a pristine complete tricuspid regurgitation Doppler envelope; in that situation, it has the advantage of ‘averaging the error’ of measurement over systole compared with a quadratic function a single velocity.

The third method consists of estimating the PA systolic and diastolic pressures (see below) and calculating mPAP value using a standard equation:
Where dPAP denotes diastolic PA pressure.

The end diastolic PA-RV pressure gradient can be estimated by applying the modified simplified Bernoulli equation to the end-diastolic pulmonary regurgitation velocity (Fig. 11.2). Adding the RAP to this calculation estimates dPAP; in research settings this correlates well with invasive dPAP measurements in select patients [29, 30]. As with tricuspid regurgitation, inadequate pulmonary regurgitant Doppler signals can be enhanced with the use of contrast or air-blood-saline mixture. The limitations described for mPAP Method 1 estimation apply here as well.

RV end diastolic pressure equals PA pressure at the moment of pulmonary valve opening. Given that the gradient between the RV and RA can be estimated by the velocity of the tricuspid valve regurgitant jet, applying the simplified Bernoulli at the time of pulmonary valve opening and adding the RAP to this value allows an estimation of dPAP [31]. The tricuspid valve regurgitant velocity Doppler sign needs to be aligned with the time from the QRS complex to the onset of pulmonary flow on the regurgitation velocity envelope (Fig. 11.3). While this method requires more effort and practice to allow reproducible demarcation of timing, it does correlated reasonably well with invasively measured dPAP in the research setting [31, 32].

RA dimensions, including diameter and area, measured in the four chamber apical view should be measured. RA area >20 cm² is considered abnormal and area >27 cm² portends poor prognosis among patients with pulmonary arterial hypertension [40].

It is well recognized that the RV is more challenging to assess with TTE than the LV because of its heavier trabeculations, and complicated geometry with the presence of an infundibulum. The initial physiologic response of the RV to pressure overload often includes dilatation. The apical four chamber view provides the most comprehensive view of the ventricle, though no single view is adequate (Fig. 11.4). A dilated RV often assumes a more globular shape, compared to the usual triangular appearance in the apical view. The ratio between the diastolic inflow diameters and areas of the RV and the LV correlates with the degree of RV dilatation: an area ratio ranging from 0.6–1 typically indicates mild RV dilation, while severe RV dilation is associated with a ratio > 1 [41]. Despite various attempts to quantify RV size, however, this remains a notable weakness of TTE evaluation of PHT.

Estimation of RV function may be accomplished via several methods. Percentage RV fractional area change (FAC), defined as [(end-diastolic area – end-systolic area)/end-diastolic area] × 100, is a measure of RV systolic function that has been shown to correlate with RV ejection fraction (EF) by magnetic resonance imaging (MRI) [42, 43]. RV FAC is an independent predictor of heart failure, sudden death, stroke, and/or mortality in patients after pulmonary embolism [44] and myocardial infarction [45, 46]. FAC is obtained by tracing the RV endocardium in systole and diastole from the annulus to the apex along the free wall (below the trabeculations), and then back to the annulus along the ventricular septum. The lower reference value for normal RV FAC is ~35 % [24]. Complex RV geometry, however, limits the accuracy of any 2D technique. 3 dimensional TTE is promising, but not yet widely available. Assessment of the RV myocardial performance index, also known as the Tei index, is an alternative, non-geometric approach. This index is defined as the ratio of the isovolumic contraction time plus isovolumic relaxation time to systolic ejection time [47]. With progressive ventricular dysfunction, ejection constitutes a shorter portion of this time period. In pediatric idiopathic PHT, RV performance index is closely related with invasive mPAP and that it can be used to monitor mPAP after initiation of medical therapy [48]. A smaller study conducted in patients with connective tissue disease demonstrated that a myocardial performance index >0.36 combined with sPAP ≥ 35 mmHg may improve TTE accuracy in predicting PHT. The Tei index has also been shown to have important prognostic implications: in a series of 53 PHT patients, an elevated Tei index was associated with an increased risk of clinical events and death [49]. One important caveat to use of this method in monitoring to response to therapy may be its relative independence to changes in afterload [50]. The Tei index is less commonly used in clinical practice, perhaps because of its lack of direct correspondence to any single physiologic parameter and unintuitive interpretation.

RV contraction has a predominantly longitudinal component (as opposed to torsion of the LV), and simply measuring the distance travelled by the tricuspid annulus during contraction provides a reasonable estimate of RV function [9]. This measurement is termed tricuspid annular plane systolic excursion (TAPSE) and is usually measured by M-mode echocardiography of the lateral tricuspid annulus (though it can be estimated from 2D images) (Fig. 11.5). Normal TAPSE is >1.6–1.7 cm (though most normal people have TAPSE > 2–2.2 cm) and is reduced with RV dysfunction [51]. Low TAPSE (<1.8 cm in one study) predicts increased risk of death in pulmonary arterial hypertension [52]. RV function also predicts outcome in patients with left heart failure. Important limitations to this method include its angle and load dependency, its focus on a single dimension of right ventricular motion and the fact that the tricuspid annular motion may be affected by overall heart motion [53].

The “domelike” morphology (parabolic curve) of Doppler pulmonary flow velocity in individuals with normal pulmonary pressure and resistance assumes a more “triangular” shape skewed towards late systole in patients with PHT (Fig. 11.6). Peak velocity appears during early ejection compared with mid-ejection in normal people. In some cases there is flow deceleration in mid systole followed by a second late-ejection increase in velocity, resulting in the appearance of mid-systolic notching [54, 55]. The pulmonary flow acceleration time (defined as the time interval from the onset of forward flow in the pulmonary artery to the peak velocity) is inversely related to sPAP, mPAP and PVR [56, 57]. Both notching and short acceleration time correspond to true late ejection flow deceleration secondary to reflected pressure waves; therefore, they are present in patients with elevated PVR and non-compliant pulmonary arteries but not in patients with PHT due to left heart disease.

Patients are usually asked to fast for at least 4–6 h prior to catheterization, though this is not absolutely necessary if administration of sedative agents is not planned. A peripheral intravenous line should be placed if sedation will be administered or if the patient is considered to be at elevated risk for adverse events. Most chronic medications should be continued as prescribed, including pulmonary vasodilators. It is reasonable to hold a diuretic dose on the morning of a catheterization in stable patients. There is variability in the approach to oral anticoagulant use. While continuing anticoagulants is associated with increased bleeding risk, the absolute risk of serious bleeding from venous catheterization is low even with therapeutic anticoagulation. We generally continue warfarin without change in regimen, and proceed with RHC as long as INR is <3. The INR should not be supratherapeutic, however, as this increases risk with no benefit. The risk-benefit calculus will vary depending on the indication for anticoagulation. For example, uninterrupted anticoagulation is strongly merited in the presence of a mechanical mitral valve but is less important if the indication is primary prevention of stroke in the setting atrial fibrillation. The approach to newer anticoagulants must be similarly approached, though in general holding doses prior is reasonable for most patients, with the duration depending on the specific pharmacokinetics of the drug, and will undoubtedly evolve as more experience is accrued with these agents. The timing for discontinuation of dabigatran prior to the catheterization procedure varies by kidney function and bleeding risk; for patients with glomerular filtration rate > 60 ml/min and a low bleeding risk it is reasonable to hold the medication for 24 h prior to the procedure. This time frame is also recommended for patients at low bleeding risk treated with rivaroxaban and apixaban, regardless of their kidney function. Aspirin or other oral antiplatelet agents can usually be continued without interruption, though those on dual antiplatelet therapy are at increased risk for bleeding. Given a small risk for lactic acidosis, patients taking metformin and scheduled for an elective procedure during which intravenous iodinated contrast material may be administered should hold this medication the morning of the procedure and restart after renal function is confirmed to be stable for at least 48 h after the procedure if contrast is used [64]. Other considerations include allergies, especially to heparin or latex which may require use of different techniques and equipment.

We believe that conscious sedation (e.g., intravenous benzodiazepines or opiates) prior to or during the catheterization procedure should not be routine but rather should be considered on an individual basis. While providers vary in their use of sedation, in the absence of severe anxiety diagnostic RHC is usually tolerated without subjective distress. Conversely, although some prior studies have raised the concern that administrating systemic sedation to patients with PAH may be associated with adverse outcomes [65], the risk is likely small in hemodynamically stable patients unless liver or kidney function is impaired. An additional downside to sedation is that patients must be monitored for a more extended period after the procedure if sedation is administered; this is inconvenient for patients and requires additional resources. Another reason for avoiding systemic sedation is the potential effect on the patient’s hemodynamic profile during catheterization. There are two potential mechanisms: first, administration may result in systemic hypotension [66] leading to unrepresentative (low relative to the unsedated state) readings of certain variables, such as LV end diastolic pressure; second, benzodiazepines have a variable negative inotropic effect with resulting worsening RV function, an effect that has been shown in patients with pre-existing LV diastolic dysfunction [67]. These effects are usually not clinically apparent or relevant but the lack of benefit from routine sedation makes the potential for even minor or rare adverse effects unacceptable. Indirect effects via respiratory inhibition with subsequent hypercapnia can also affect hemodyanamic values.

If mechanical ventilation is clinically indicated, the potential associated hemodynamic effects must be considered. There is often an increase in RV afterload and decline in RV contractility, along with a number of other effects. With the use of positive end expiratory pressure (PEEP), positive intrathoracic pressure is variably transmitted to the central circulation, thus affecting right-sided pressures (especially during forced inspiration) [68–70]. The extent to which PEEP affects right-sided pressures is unpredictable and depends on various variables: baseline filling pressure, compliance of the chest wall, lung and of cardiac chamber compliance, and intravascular/cardiac volume status. There is no universal adjustment to account for the effect of PEEP on the PAWP, but PEEP < 10 cm H₂O does not importantly affect PAWP [70]. For values above that, one suggested correction is that PAWP rises ~2–3 cm for every 5 cm H₂O increment in PEEP [71]. To avoid unnecessary uncertainty, PEEP should be maintained as low as clinically reasonable.

A second alteration to be considered is an inaccurately elevated PAWP reading due to unexpected lung zone position of the catheter tip. In the upright position, lung zone 3 is located at the base of the lung, where alveolar pressure is lower than both pulmonary arterial and pulmonary venous pressure; position of the catheter tip in zone 3 allows pressure transmission directly from the left atrium to the wedged catheter tip. Under normal conditions, lung zone 3 is where the PAWP accurately reflects left atrial pressure [70]. With mechanical ventilation, the catheter tip is less likely to be located in zone 3, since alveolar pressure is increased and zone 3 accounts for a smaller portion of the lung. Use of PEEP and hypovolemia exacerbate this. Demonstrating that the catheter tip is below the level of the left atrium suggests zone 3 location [72]. Note that when the patient is supine, one cannot confirm position with AP imaging, since zone 3 is posterior to, not inferior/caudal to, the left atrium. It is reasonable to obtain hemodynamic measurements in various pulmonary segments, preferably in both the right and left lungs, to support the assertion that any single measurement is representative of underlying hemodynamics.

All roads lead to Rome and all systemic veins lead to the heart; therefore, any systemic vein could be used for venous access. Considerations as to optimal access sites include size of the vein, ease of vascular access, directness of course from access site to right heart, and requirements for monitoring after sheath removal.

The most common access sites used in catheterization for hemodynamic assessment of PHT are the femoral vein and the right internal jugular vein (IJ). The right IJ is preferred unless there is planned concomitant femoral arterial access or specific procedures (e.g., probing for PFO) are expected. Access via the left internal jugular vein or left subclavian vein does not provide as direct a catheter course, but is reasonable when right IJ access is not a good option (e,g., thrombosis). Proximal arm veins can also be used for venous access. This is commonly the favored approach when radial arterial access is used for a simultaneous left heart catheterization, but it limits catheter size and manipulation. Several groups have reported on their experience with the IJ access technique in large groups of patients; when performed by experienced personnel, IJ access is quick (<15 min) and associated with a very low complication rate (1–1.7 %); hemostasis can be rapidly achieved and patients (in the absence of sedation) can be discharged directly following the procedure in most cases [73, 74].To facilitate both cannulation success and speed, 2D ultrasound guidance may be used to locate the IJ and adjacent carotid artery. Ultrasound guidance is especially encouraged when technical difficulties are suspected (i.e. patients with short or thick neck) or when difficulties are encountered at the time of the procedure. Potential complications with IJ access are uncommon and include pulmonary hemorrhage, pneumothorax, carotid puncture, sinus bradycardia (vasovagal, usually transient and responsive to fluid; rarely requiring atropine administration) and complete heart block (especially in setting of pre-existing left bundle branch block). When considering catheter advancement to the pulmonary artery, one should keep in mind that IJ access will usually lead the catheter to the right PA, whereas femoral access will more commonly lead to catheter position in the left PA. When femoral venous catheterization is performed, patients are usually observed in the supine position for at least 2 h following removal of the sheath to ensure adequate venous hemostasis prior to ambulation. In the setting of active anticoagulation or a larger sheath, observation may be extended. Vascular closure devices, such as the Perclose ProGlide Suture-Mediated Closure System (Abbot Vascular, USA) can decrease bleeding time and access site complications. Sheath size is typically required to be 5 French or larger when using closure devices, and other contraindications should be checked with the manufacturer prior to usage. In the authors opinion, the internal jugular approach is preferable when no other interventions are planned, as it allows easier catheter manipulation through the right ventricular outflow tract and into the pulmonary arteries and, barring complications, the patient can be discharged home soon after sheath removal.

When deciding upon access site in a given patient, historical data regarding prior venous and arterial access and potential obstructions, compressions, complex anatomical courses and prior procedural complications should be obtained. Patients with kidney disease on hemodialysis, congenital heart disease, pacemakers, or indwelling chronic venous access for other reasons (e.g., intravenous prostanoids) are more likely to have had multiple prior procedures and vascular injury or thrombosis. Imaging of the vascular bed may be established by various modalities including ultrasound, computed tomography and magnetic resonance angiography.

The quality of hemodynamic data depends on an array of variables. First, the quality of tracings depends on the frequency response of the fluid filled catheter system. Frequency response is directly proportional to the size of the lumen and inversely related to length of tubing. Thus, optimal tracings are obtained with catheters with large lumens and minimal excess tubing.

Of note, the external diameter of the catheter is not always indicative of lumen size. A large catheter with multiple ports (e.g., 8 F catheter with RA, thermodilution and PA ports) will often have slower frequency response than a smaller single port catheter (e.g., 6 F single lumen catheter). Frequency response is generally <12 Hz [75]. Selection of catheter balances the risk associated with larger access sheath (albeit a very small risk), benefits of additional ports or technologies (e.g., ability to perform thermodilution cardiac output estimates) and quality of hemodynamic tracings.

Catheters with end holes are necessary to measure occlusion (wedge) pressures and for sampling pressures within small chambers or when discerning pressure gradients over short distances. While catheters with side holes are preferable to measure intra-cardiac pressures to prevent erroneous readings due to damping effect resulting from obstruction of the catheter tip by cardiac tissue this is a modest concern and issues are recognizable (blunted waveforms) and can be addressed with catheter repositioning. Generally only in the context of congenital heart disease are side-ports truly needed for this reason.

A “Swan-Ganz” catheter is commonly used for measuring right-heart pressures when thermodilution is considered for evaluation of the cardiac output (see below). In addition to the balloon at the tip for flotation, it consists of an end-hole port, a side-hole port 30 cm from the catheter tip, and a thermistor for measurement of cardiac output by the thermodilution method. Other balloon flotation catheters include: the Berman catheter, which includes multiple side-holes near the tip and no end hole or thermistor and is used mainly for angiography; catheters with an RV side port to allow a pacing via a temporary pacemaker wire; and the balloon-wedge catheter, which contains an end-hole similar to the Swan-Ganz catheter but no thermistor for cardiac output measurement or additional infusion or pressure monitoring ports [70].