Authors: Giulia Guglielmi, Kaushiga Krishnathasan, Andrew Constantine, Konstantinos Dimopoulos
Categories: Special issue on Pulmonary Arterial Hypertension (PAH), Cardiac catheterization, Congenital heart disease, Pulmonary arterial hypertension, Pulmonary hypertension
Source: International Journal of Cardiology Congenital Heart Disease
Cardiac catheterization (CC) is essential for the diagnosis of pulmonary hypertension (PH), and for its characterisation. It allows distinction between pre- and post-capillary PH which, when integrated with other non-invasive data, facilitates classification into one of the 5 diagnostic groups defined by international PH guidelines. CC also provides valuable information for the risk stratification of patients with PH, guiding management and the type and intensity of treatment. Right heart catheterization is usually sufficient in PH practice, yet additional information can be acquired by extending the protocol to include left heart catheterization or provocation protocols.
This review provides a detailed overview of diagnostic CC as used in PH practice, including in patients with congenital heart disease, with an emphasis on fundamental concepts, tips and tricks and potential pitfalls.
Keywords: Congenital heart disease, Pulmonary hypertension, Pulmonary arterial hypertension, Cardiac catheterization
Invasive haemodynamic assessment is one of the oldest, yet most relevant investigations in cardiology. Cardiac catheterization (CC) was in use well before the advent of echocardiography and was the means of confirming diagnoses developed through detailed history taking, careful clinical examination, electrocardiography, and chest radiography. Despite major advances in imaging and other non-invasive diagnostic modalities over recent decades, CC has several indications in contemporary clinical practice. These include the reliable assessment of intracardiac pressures, detection and quantification of haemodynamic lesions, shunts, differentiating between causes of exercise intolerance and congestive heart failure, risk stratification and quantification of the response to treatment. Diagnostic CC, and in particular right heart catheterization (RHC), has received growing attention as a result of recent advances in the care of patients with pulmonary hypertension (PH), especially those with pulmonary arterial hypertension (PAH) or chronic thromboembolic PH (CTEPH). CC also retains a prominent role in congenital heart disease (CHD) practice.
From a technical point of view, diagnostic CC may appear straightforward, especially when compared to complex modern interventional procedures, yet it is often the case that few operators have maintained the expertise and rigour to acquire reliable and easily interpretable information, upon which clinical management relies. In this paper, we provide a detailed overview of diagnostic CC as used in PH practice, including in patients with CHD, with an emphasis on fundamental concepts, tips and tricks and potential pitfalls.
The following are common indications for performing CC for a patient within a PH, a CHD, or a transplant
In PH practice, most patients only require RHC, with the addition of pulmonary angiography in CTEPH. Arterial access for left-heart catheterization is rarely required,
To promote standardization of the procedure, the RHC protocol should cover several requirements in different
The WHO checklist is a core set of safety checks, which should be carried out for all patients in surgical theatres and catheterization labs [8]. Its steps are summarized in Table 1.
Venous access depends on operator preference and anatomical characteristics, procedural needs, and anticoagulation status. Femoral, neck or arm access can be used interchangeably, with pros and cons for
Difficulties are often encountered during RHC when advancing a catheter towards the pulmonary circulation and
Pressure measurements should always be preceded by careful zeroing of transducers at mid-chest level (Fig. 1). Right atrial (RA) pressures should identify the mean and all components of the waveform, including the a and v waves (Fig. 2). A very low mean RA pressure usually reflects dehydration or an incorrect zeroing (e.g. transducer higher than the chest). Increased mean RA pressure in patients with PH typically reflects RV dysfunction and raised RV end-diastolic pressure. A prominent a wave is often present, while a prominent v wave is not uncommon when significant tricuspid regurgitation has developed. Additional information can be acquired by examining the x and y descents (e.g. RV restriction).
Fig. 1 Transducer setup. Impact of transducer height on pressure readings. The transducer should be zeroed at midthoracic level, halfway between the anterior sternum and the bed surface. If the transducer is placed too high, the pressures measured at the level of the mid-chest will be falsely lower. Conversely, if the transducer is placed too low, the pressures measured at the mid-chest level will be falsely higher.
Fig. 2 Right atrial pressure waveform. a – Atrial systole. c – Closure of tricuspid valve. x – Downward movement of the tricuspid valve. v - Rapid filling of the right atrium. y – Opening of the tricuspid valve.
The most relevant RV pressure values are the peak systolic and end-diastolic (Table 2). The former should be similar to systolic PA pressure (in the absence of pulmonary stenosis) and the latter should be similar to mean RA pressure (Fig. 3). The mean RA pressure and RV end-diastolic pressure are important parameters in PH, especially when associated with severe heart failure, as they provide important information on RV preload.
Fig. 3 Right ventricular and pulmonary artery pressure waveforms. IVC= Isovolumic contraction. IVR = Isovolumic relaxation.
PAWP is a surrogate of left atrial (LA) pressure. Direct LA pressure cannot usually be obtained unless an ASD is present. PAWP involves the use of a balloon-tipped end-hole catheter (Swan-Ganz). With the balloon inflated in a large PA, the catheter is advanced until it wedges into a smaller PA branch. This generates a “static column of blood” in that vessel, which extends across the respective arterioles, capillaries and venules into the larger pulmonary veins where blood ceases to be static thanks to the return from other lung segments. As the pressure inside a static fluid is the same in all directions, the pressure measured at the tip of the catheter reflects the pressure in the pulmonary vein draining that lung segment. Thus, the pressure measured at the tip of the catheter is equivalent to the LA pressure, unless a pulmonary venous stenosis is present (Fig. 4).
Fig. 4 Measurement of wedge pressure and impact of obstruction. Impact of fixed obstruction and stenosis on wedge pressure measurement. The inflated balloon into a small PA branch generates a static column of blood across arterioles, capillaries and venules. Stenoses within the static column of blood (wide white arrow) cannot be detected by the PAWP. Conversely, stenoses in larger pulmonary veins (thin black arrow) do have an impact on wedge pressure.
Understanding the principle behind PAWP is important for avoiding pitfalls. Stenoses within the static column of blood (e.g. at the level of the pulmonary venules, as in pulmonary veno-occlusive disease) cannot be detected by the PAWP, which only reflects pressure in the larger pulmonary veins. Partial occlusion (a partially wedged balloon) will allow pressure from the PA to reach the tip of the catheter, overestimating LA pressure. It is important to observe the pressure waveform change as the catheter is moved from the PA to the PAWP position, obtaining the expected waveforms described for the RA.
Over-wedging is a phenomenon commonly observed, where the pressure rises when the balloon wedges, often to pressures higher than those of the PA. In such cases, partial release of the balloon may help obtain an adequate waveform.
It is important never to pull the catheter back while the balloon is inflated, especially in the wedge position where damage to the vessel can be catastrophic (e.g. causing rupture and haemoptysis). The same is true for the tricuspid valve, which can be damaged by forceful withdrawal of an inflated balloon.
Significant respiratory variation can be observed in both the RA and LA pressures, and occasionally in pulmonary arterial pressure (PAP) in patients with significant lung disease or obesity. Guidelines recommend measurement of pressures, especially PAWP, at end expiration without breath-holding, ideally averaging over ≥4 respiratory cycles. When excessive variability is observed or values appear implausible, it is recommended to measure LV end-diastolic pressure (if left heart access possible). Many centres do, in such cases, use the electronic mean (e.g. ignoring the respiratory cycle), though there is a risk of underestimating PAWP [1,9].
It is recommended that non-invasive aortic pressure and saturation (pulse oximetry) are documented since they cannot be measured invasively unless arterial access (or puncture) is obtained.
Serial oximetry is recommended during RHC and should typically include measurements from the high and low SVC, IVC, mid-RA, RV, and PA. Oxygen saturation is used
where AoO2= systemic arterial saturation, PAO2 = saturation in the PA, and PVO2 = saturation in the pulmonary veins.
where VO2 = (resting) oxygen consumption, Ca = systemic arterial oxygen content, and C~v=~mixed venous oxygen content.
Ca is calculated
Cv is calculated
where PaO2 = partial pressure of oxygen in arterial blood (mmHg), SaO2 = arterial oxygen saturation (%), PvO2 = mixed venous oxygen tension (mmHg).
Aortic saturations can be invasively measured through an arterial blood gas (ABG) analysis; in practice, however, non-invasive methods such as plethysmography are often preferred.
Attention is needed in the interpretation of saturations in patients receiving supplemental oxygen during RHC. An FiO2 >0.3 is likely to result in oxygen being dissolved in the blood, which should be taken into account in the Fick calculations by measuring PaO2. Even though it is recommended that PaO2 is measured in such cases, it is recognized that the above formula gives a low relative importance to it (i.e. multiplying PaO2 measured in mmHg by 0.003). Perhaps more importantly, supplemental oxygen typically causes an artificial rise in oxygen saturations “across the board”, making interpretation of shunt fraction difficult.
A fundamental component of the RHC is the calculation of CO. The term CO should be clarified in patients with intracardiac shunts, in whom the systemic (Qs) and pulmonary blood flow (Qp) differ. Moreover, it is essential to remember that, in the presence of shunts, the methods for calculating “CO” described below, measure Qp, not Qs. Finally, Qp, not Qs is the denominator in the formula for calculating pulmonary vascular resistance (PVR), an obvious but often confusing point for inexperienced operators.
The following methods can be used in the catheter lab to calculate Qp:
The calculation of PVR is one of the most important targets of invasive haemodynamic assessment. This is calculated
where mPAP = mean PAP, WU= Wood Units.
It is integral to differentiating between pre-capillary and post-capillary PH, grading the severity of pulmonary vascular disease, and deciding on the operability of CHD with a L-R shunt. The definition of a normal PVR has changed over time, and so has the PVR cutoff for defining operability. Currently, a normal PVR is defined as ≤2 WU, while patients with a PVR <5 WU and a L-R shunt causing volume overload should be considered for repair [1,13].
Indexed PVR (PVRI) is commonly used to account for body habitus and is particularly relevant to paediatric practice. A common pitfall in calculating PVRI when PVR and BSA are known, is to divide PVR by BSA instead of multiplying it. Indeed, PVRI is measured as WU x m^2^.
By using PVRI, we multiply PVR by the BSA, thus giving a greater weight to a raised PVR in larger individuals. A smaller individual is expected to have a smaller total cross-sectional area of their pulmonary vascular bed. For example, an estimated PVR of 5 WU in a large individual is more likely to reflect pulmonary vascular disease than in a small child in whom the PVR is expected to be higher due to their small size. Debate is ongoing between experts on whether PVR or PVRI should be used in the adult population, and whether indexing provides additional clinical value. The latest ACHD ESC guidelines have removed PVRI from the definition of operability [13].
It is noteworthy that PVR was previously not part of the diagnosis of pre-capillary PH in the international guidelines but was reintroduced in 2015. This is likely due to concerns regarding accuracy in calculating PVR, mainly its denominator i.e. CO (Qp), which remains an important source of error.
Beyond the baseline haemodynamics measured during RHC, provocation tests are often used and provide a more comprehensive understanding of the condition.
Pulmonary vasoreactivity testing is a crucial diagnostic tool used to determine the nature of the increased PVR observed at baseline. By using pulmonary vasodilators in the catheter lab, one can differentiate between vasoconstriction versus fixed/structural obstruction. Inhaled NO and inhaled iloprost are the most commonly used vasodilators for this purpose. Although more technically demanding, intravenous epoprostenol can also be used [1].
Vasoreactivity testing is mandated in idiopathic, heritable, or drug-induced PAH. A positive vasoreactive response in this setting is defined as a reduction in mPAP by at least 10 mmHg, leading to an absolute value of ≤40 mmHg, with a stable or increasing CO. Patients who show a positive vasoreactivity response should be offered treatment with calcium channel blockers (CCBs) at large doses, followed by repeated RHC [14]. Unfortunately, such a favourable response to CCBs is only observed in a minority of patients, with even fewer exhibiting a satisfactory long-term response [15]. Vasoreactivity testing is not recommended in PAH other than in idiopathic, heritable, and drug-induced PAH, and should be avoided in cases of suspected or confirmed pulmonary veno-occlusive disease or group 2 PH (risk of acute pulmonary oedema) [1].
In the past, vasoreactivity testing was a standard component of the operability assessment in CHD-associated PH (CHD-PH). However, current guidelines do not routinely recommend acute vasoreactivity testing in this population, due to the lack of prospective data on its utility for evaluating operability and the normalization of PVR after repair [13]. A vasoreactive response also appears to hold prognostic value in patients with Eisenmenger syndrome receiving PAH targeted therapy, but has not entered clinical practice [16].
An important step in the haemodynamic assessment of PH is distinguishing between pre- and postcapillary physiology. Small deviations in PAWP can make the difference between a pre or postcapillary diagnosis and substantially influence management decisions. In this setting, euvolemia is desirable, yet patients undergoing CC are often dehydrated, having been kept nil by mouth for several hours, leading to a falsely low measurement of PAWP. It is, therefore, important to unmask postcapillary PH using fluid challenge in patients with a relevant risk factor, such as older age, type 2 diabetes, hypertension, obesity, etc. [17,18].
Fluid challenge is performed through infusion of normal saline, for a total of 500 mL (7–10 mL/kg) over 5–10 min. It is considered “positive” if PAWP rises to ≥18 mmHg. This can have therapeutic implications, as PAH therapies can be detrimental by markedly increasing PAWP and causing pulmonary congestion in patients with occult diastolic dysfunction of the LV.
The primary indication for exercise testing during CC is the investigation of dyspnoea of unknown aetiology, or of symptoms that are disproportionate to the severity of the underlying cardiac or pulmonary condition. In fact, physical activity may worsen or unmask underlying pathologies (e.g. mitral insufficiency, diastolic LV dysfunction, respiratory conditions), causing a pathological rise in PAP or PAWP.
Exercise RHC uses stationary cycle ergometers set on the catheterization lab table. While standardized protocols for this procedure are lacking, a dynamic form of exercise is recommended, avoiding isotonic arm exercises. For achieving a steady-state oxygen uptake, it is advisable to maintain a duration of 3 min per stage. Parameters such as PAP, PVR, and cardiac index (CI) should be assessed at each exercise stage, though accurate measurements require expertise. RAP, MVO2, SaO2, and arteriovenous oxygen difference should be measured at rest and at peak exercise. Subsequently, calculations for mPAP/CO and PAWP/CO slopes can be derived [19]. The physiological augmentation of CO and mPAP during exercise is reflected in the slope of mPAP/CO that should not exceed 3 mmHg/L/min. A higher slope delineates exercise-induced PH [1]. A PAWP/CO slope >2 mmHg/L/min during exercise CC is useful for unmasking postcapillary PH [20].
In skilled hands, exercise RHC poses no additional risk of complications compared to resting RHC and cardiopulmonary exercise testing, and should, therefore, be integrated into routine clinical practice for evaluating patients suspected of having exercise-induced PH or occult post-capillary PH. Expertise is, however, required for acquiring high-quality, interpretable data during exercise.
RV-PA coupling highlights the RV's capacity to adapt to the increased afterload [21]. In PH, the RV initially adapts to increased afterload following a “homeometric adaptation” with concentric hypertrophy, which allows CO, ejection fraction and exercise capacity to be maintained. If the afterload increases further, there is a transition towards a maladaptive state, causing RV dilatation, and eventually dysfunction (“heterometric adaptation”) [22].
Standard CC is unable to provide reliable information regarding RV-PA coupling. This requires special catheters for recording pressure-volume (PV) loops and measuring the ratio of end-systolic elastance/arterial elastance (Ees/Ea). Ees represents RV contractility, whereas Ea is a surrogate of afterload (end-systolic pressure/stroke volume). The ideal Ees/Ea ratio is thought to be between 1.5 and 2.0 [23,24]. Description of PV loops and elastance, which remain research tools, are beyond of the scope of this paper.
Recently, the ratio of TAPSE/systolic PAP on echocardiography has been shown to be associated with invasively measured Ees/Ea [25].
CTEPH is one of the most common types of PH, characterized by the persistence of organised thromboembolic material obstructing the PAs several months or years after acute pulmonary embolism. CTEPH often coexists with group 2 PH, i.e. a postcapillary PH component, which is associated to a worse prognosis [26].
When CTEPH is suspected on a ventilation/perfusion (V/Q) scan, pulmonary angiography should be considered during RHC, and interpreted in conjunction with data from non-invasive modalities.
Digital subtraction angiography is preferred in this setting, as it requires less contrast than conventional angiography and allows better detection of stenotic or obstructive lesions in the pulmonary arterial tree. Typical angiographic findings associated with CTEPH result from the organization and recanalization processes of vessels with persistent clots, and include irregularities in vessel wall contour, constricted vessel bands, web-like formations within vessel lumens, early vanishing of vessels, and ‘pouch’ defects [27]. While non-invasive techniques may provide valuable information, pulmonary angiography remains essential when contemplating surgical (pulmonary endarterectomy) or percutaneous (BPA) treatment.
PoPH is a pulmonary vascular complication related to advanced liver disease. In case of clinical and echocardiographic suspicion of PH and a clinical history or suspicion of portal hypertension, RHC should be performed and should include measurement of WHVP [1].
WHVP is an indicator of portal venous pressure and is measured by advancing a Swan-Ganz catheter into the hepatic vein with the balloon inflated, similar to the PAWP recording [28]. The free hepatic venous pressure (FHVP) is also recorded with the tip of the catheter into the hepatic vein (balloon deflated), 2–3 cm from the IVC. The difference between WHVP and FHVP is called “hepatic venous pressure gradient” (HVPG) and, when increased (>5 mmHg, but clinically significant when ≥10 mmHg) confirms the presence of portal hypertension (Fig. 5) [29].
Fig. 5 Example of hepatic venous pressures measurement. Free and wedged hepatic venous pressures in a patient with liver cirrhosis. Note, the hepatic venous pressure gradient exceeding 10 mmHg.
Assuming that RAP is freely transmitted to the hepatic venous system, four scenarios are possible (Fig. 6):
Fig. 6 Impact of congestive heart failure and liver disease on transhepatic gradient. Heart failure and/or liver cirrhosis affect free hepatic venous pressure and wedged hepatic venous pressure to a different extent, altering the hepatic venous pressure gradient.
In patients with advanced liver disease and increased intra-abdominal pressure, there is a discrepancy between RAP and FHVP. In this setting, the HVPG has shown superior clinical prognostic value than WHVP-RAP and is, therefore, preferred [29,30].
The latest International PH guidelines recommend the use of a three-strata model to predict mortality in patients with PAH. This model contains clinical, laboratory, functional, imaging, and invasive haemodynamic parameters, including RAP, CI, stroke volume index (SVI), MVO2. Patients with RAP >14 mmHg, CI < 2.0 L/min/m2, SVI <31 mL/m2, or MVO2<60 % are deemed at a high mortality risk (>20 % in one year). The guidelines also advise repeating the RHC at 3–6 months after changes in therapy, especially in patients with inadequate response to treatment or evidence of clinical deterioration [31].
Prognostic information can also be acquired during RHC by using provocation tests in specific subgroups of PAH. A reduction in PVR or mPAP during acute administration of inhaled NO has been shown to predict long-term outcome in patients with PAH [15]. Vasoreactivity testing is particularly important for the risk stratification of patients with idiopathic PAH, as previously described, with emphasis on repeating the RHC after establishing treatment with high-dose CCB, with responders exhibiting a far better prognosis that non-responders with similar baseline hemodynamics [32].
CC in CHD presents several challenges and requires expertise due to the heterogeneous underlying anatomical and physiological conditions,
Catheterization in complex CHD, such as patients with univentricular heart and a Fontan-type circulation, presents several
Despite its invasive nature and major improvements in imaging, CC remains fundamental to cardiology and PH practice. It is the gold standard for the assessment of intracardiac and pulmonary pressures, detection and quantification of shunts and other hemodynamic lesions, and is integral in the diagnosis and classification of PH. It also provides valuable prognostic information. Expertise and careful quality control are required to minimize complications and obtain reliable data, especially in special cohorts, such as those with CHD.
None.
Giulia Guglielmi: Writing – review & editing, Writing – original draft. Kaushiga Krishnathasan: Writing – review & editing, Writing – original draft. Andrew Constantine: Writing – review & editing, Supervision. Konstantinos Dimopoulos: Writing – review & editing, Writing – original draft, Validation, Supervision, Conceptualization.
The authors declare the following financial interests/personal relationships which may be considered as potential competing Konstantinos Dimopoulos reports a relationship with Janssen-Cilag Ltd that speaking and lecture fees. Kaushiga Krishnathasan reports a relationship with Janssen-Cilag Ltd that speaking and lecture fees. Andrew Constantine reports a relationship with Janssen-Cilag Ltd that speaking and lecture fees. IJCCHD Editorial Board membership, Konstantinos Dimopoulos If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
None.