Prognostic Significance of Hemodynamics in Patients With Transposition of the Great Arteries and Systemic Right Ventricle.
Aug 29, 2024DOI: 10.1161/circheartfailure.124.011882Journal ArticleAHA
Written by: Nael AldweibPayam DeghaniCraig S BrobergAlexandra van DisselAhmad AltibiJoshua WongDavid BakerSalil GindiPaul KhairyAlexander R OpotowskySangeeta ShahAnthony MagalskiJonathan CramerRobert M KaulingMikael DellborgEric V KriegerElizabeth YeungJolien Roos-HesselinkJamil AboulhosnJeremy NicolarsenLuke MashaPastora GallegoDavid S CelermajerJoseph KayIsabelle Vonder MuhllSusan M JamesonClare O'DonnellFlavia FuscoAnitha S JohnConrad MaconPetra AntonovaTimothy CottsBerardo SarubbiFred RodriguezChristopher DeZorziPavithra S JayadevaMarissa KuoShelby KuttyTripti GuptaLuke J BurchillCarla P Rodriguez MonserrateAdam M LubertJasmine GrewalStephen PylypchukMark N BelkinWilliam M Wilson
The systemic ventricle is a morphological right ventricle (RV) in individuals with D-loop transposition of the great arteries (D-TGA) managed with an atrial switch procedure (Senning or Mustard), as well as in those with congenitally corrected transposition of the great arteries (ccTGA).1 Heart failure and RV dysfunction are common in both these groups. Notably, more than half of D-TGA patients exhibit moderate to severe RV systolic dysfunction 25 years after atrial switch procedures and 40% develop heart failure by their fifth decade of life.2–4 Similarly, severe systemic RV dysfunction and heart failure affect 30-50% of ccTGA patients by their sixth decade.5,6Recent research has indicated that the assessment of systemic RV systolic function, the extent of tricuspid valve regurgitation using echocardiogram, and a history of heart failure admissions can be utilized to predict the risk of death, heart transplant and the need for mechanical circulatory support devices in D-TGA and ccTGA patients.4,6 However, the potential for invasive hemodynamics to add to risk stratification in systemic RV patients is not established.Right heart catheterization for invasive hemodynamics, conducted through a pulmonary arterial catheter, has played a vital role in diagnosing and managing heart failure for many years. Routine invasive hemodynamics comprise measurements of pulmonary arterial pressures, filling pressures from both the right- and left-sided chambers of the heart, along with calculation cardiac output and pulmonary vascular resistance.7Calculation of cardiac output, whether using the Fick principle or thermodilution, relies on assumptions that may affect accuracy. Furthermore, these methods have demonstrated only modest and often discrepant correlations with clinical outcomes.8–10 Previous research in hemodynamically stable patients with a systemic left ventricle (LV) admitted with acute decompensated heart failure (ADHF) revealed that elevated central venous pressure (CVP) and pulmonary capillary wedge pressure (PCWP) were better predictors of long-term adverse events than cardiac index.11 Recently, aortic pulsatility index (API), an invasive hemodynamic measurement designed to incorporate cardiac output and filling pressures was found to be independently associated with adverse cardiac outcomes in patients with systemic LV ejection fraction less than 30%, symptoms despite goal-directed medical therapy, and at least one sign or symptom of congestion.12,13This study was conducted with the objective of characterizing a cohort of TGA with systemic RV patients who underwent hemodynamic cardiac catheterization. Our hypothesis was that the API and hemodynamic profile would be associated with adverse clinical outcomes in patients with TGA and systemic RV.
METHODS
The data that support the findings of this study are available from the corresponding author upon reasonable request. This retrospective cohort study was conducted by the Alliance for Adult Research in Congenital Cardiology. Approval was obtained from the Institutional Review Board at Oregon Health Science University, as well as the respective research oversight boards at each center. Informed consent requirements were waived due to the retrospective study design. Inclusion criteria encompassed D-TGA with a prior atrial switch procedure (Mustard or Senning) and ccTGA, having been seen at least twice by an adult congenital heart disease specialist over a year or more, and being over 18 years old at the initial assessment. Exclusion criteria included single ventricle anatomy, prior Rastelli procedure or arterial switch operation. Centers were specifically tasked with reviewing older clinical, imaging or catheterization databases for previously seen patients who might have subsequently passed away. The time of cardiac catheterization was identified as the initial visit, with hemodynamic data available from January 1, 1994, onwards. For surviving patients, the most recent follow-up data extended until December 30, 2020. Data in this study were extracted from medical records and entered into the secure online database, Research Electronic Data Capture (REDCap). The initial visit details, including anatomic, interventional and medical history, as well as vital signs and medications, were recorded. Investigations most closely linked to the initial visit, such as laboratory tests, electrocardiography, echocardiography, cardiac magnetic resonance (CMR) and cardiopulmonary exercise data were also documented. Semiquantitative designations (normal, mild, moderate, or severe) for RV systolic dysfunction and tricuspid regurgitation (TR) were assigned. From cardiopulmonary exercise studies (cycle or treadmill), peak heart rate and oxygen consumption (VO2) were gathered. Predicted peak VO2 was derived using published reference equations.14–17
Similar data were collected for the most recent clinic visit, encompassing all the objective findings mentioned earlier. For patients who passed away, underwent mechanical circulatory support (MCS) or had heart transplantation, the most recent clinical information preceding the event was obtained. The date of the last known contact was recorded for each patient. Patients underwent active follow-up and data was retrieved from medical records at each hospital. Patients not seen at least twice a year were considered lost to follow-up and were not included in the study, this was quality control program for data collection to counter a possible bias in measurements of variables. Follow-up data encompassed occurrences of arrhythmias, new pacemaker or implantable cardioverter-defibrillators (ICD) implantation and heart failure admissions (defined as non-elective hospitalization for heart failure symptoms with elevated BNP or the need for diuresis). Atrial arrhythmia was characterized as one necessitating an emergent visit, a new prescription for medication, cardioversion, ablation, or one that was sustained and deemed permanent. Ventricular arrhythmia was defined as at least 3 consecutive ventricular beats >100 beats/min associated with symptoms, lasting >30 seconds with or without symptoms, or any ventricular arrhythmia prompting a therapeutic intervention such as an ICD shock
In this investigation, we examined patients with TGA and systemic RV who underwent cardiac catheterization within the study timeframe (1994-2020). We excluded individuals with pulmonary arterial hypertension and a normal PCWP, which was defined as a mean pulmonary artery pressure (PAP) of ≥ 25 mmHg and a PCWP of ≤ 15 mmHg, given that this subgroup presents a distinct and elevated risk profile for adverse cardiac events (supplementary figure S1).18,19
We identified 242 TGA patients who had undergone invasive hemodynamic evaluation. Information pertaining to medical history, vital signs, and medications was recorded from the visit preceding the cardiac catheterization. Hemodynamic data and vital signs were documented during the cardiac catheterization procedure. The measured values during cardiac catheterization included mean CVP, pulmonary artery systolic pressure (PAS), pulmonary artery diastolic pressure (PAD), mean PAP, mean PCWP and cardiac index. We used the thermodilution method for cardiac index calculation whenever possible, resorting to the Fick method if thermodilution data was unavailable. Patients with D-TGA who have undergone an atrial switch procedure lack a functioning right atrium post-surgery. Consequently, CVP is employed as an alternative metric for right atrial pressure in D-TGA and ccTGA. From the measured hemodynamics, we calculated additional parameters, including the pulmonary artery pulsatility index (PAPI), systemic vascular resistance index (SVRi), left ventricular stroke work index (LVSWI), right ventricular stroke work index (RVSWI), pulmonary vascular resistance index (PVRi), transpulmonary gradient (TPG) and API. We calculated the API using non-invasive brachial blood pressure due to the absence of central aortic blood pressure data. Comprehensive definitions for these estimated parameters have been previously published and are provided in the supplementary material table S1.
The API is calculated by dividing the pulse pressure (systolic minus diastolic pressure) by PCWP. The numerator, reflecting arterial pulse pressure, has shown correlation with stroke volume under conditions of fixed aortic compliance.20 The denominator, PCWP, drives pulmonary pressures and signals pulmonary congestion. Our decision to use API was based firstly on its similarity to PAPI, which is an effective predictor of RV failure and clinical outcomes in advanced heart failure patients and systemic LV and secondly given previous studies have described the association of API with adverse cardiac outcomes.12,13,21
The preset cutoff values for hemodynamic parameters were established using recognized normal thresholds: mean CVP ≤ 5 mmHg, mean PAP <20 mmHg and mean PCWP < 10 mmHg. However, cardiac pressures are not typically reported in clinical practice as binary values, whereas in reality, they exist along a continuum, therefore, we used the following ranges for analysis: mean CVP ≤ 5 mmHg, 6-10 mmHg, and > 10 mmHg; mean PAP <20 mmHg, 20-40 mmHg and > 40 mmHg; mean PCWP < 10 mmHg, 10-15 mmHg, 15-20 mmHg and > 20 mmHg. For API, an optimal value of < 1.5 has been utilized to predict adverse cardiac outcomes, while an optimal cutoff value of API of 2.9 has been described to predict freedom from adverse outcomes.12,13 Therefore, we chose API <1.5, 1.5-3, and > 3 as our ranges for this variable. Although Receiver Operating Characteristic (ROC) curves could ideally define optimal cutoff values, their use was impractical due to the absence of a gold standard method without error to measure the disease status for each patient.22
We used mean PCWP and cardiac index values from our dataset to establish four distinct hemodynamic profiles for patients with a systemic RV. A PCWP of ≥ 15 mmHg was indicative of congestion, referred to as “wet,” while a cardiac index < 2.2 l/min/m2 signified inadequate perfusion, termed “cold.” Consequently, we defined the four hemodynamic profiles: (1) Wet/Cold: PCWP ≥ 15 mmHg and cardiac index < 2.2 l/min/m2. (2) Wet/Warm: PCWP ≥ 15 mmHg and cardiac index ≥ 2.2 l/min/m2. (3) Dry/Cold: PCWP < 15 mmHg and cardiac index < 2.2 l/min/m2. (4) Dry/Warm: PCWP < 15 mmHg and cardiac index ≥ 2.2 l/min/m2.
The primary endpoint of interest was the first occurrence of major adverse cardiovascular events (MACE) during the follow-up period. MACE encompassed a composite outcome involving all-cause mortality, heart transplantation or the need for MCS devices. The cause of death was reviewed from the medical chart and categorized. Additionally, the secondary outcome was the first episode of ADHF, defined as a non-elective hospital admission requiring diuretic treatment due to symptoms related to heart failure following cardiac catheterization.
Statistical analyses were conducted using STATA 16 (State Corp LLC, College Station, Texas), with significance at a p-value < 0.05. Patients’ baseline characteristics were categorized based on the API index into API < 3.0 and API ≥ 3.0 based on previously published data from The Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness (ESCAPE) study.12 Data underwent internal consistency checks, and discrepancies were addressed by local investigators. Patients with missing data for any measured hemodynamic parameters were excluded from the study, and all outcomes data were available for the included patients. However, there was a significant amount of missing data for confounders and other variables, which was not reported by each participating center. Descriptive statistics outlined the characteristics of each group. Comparative analyses utilized Student’s independent t-test, Chi-square, Mann- Whitney test, or Fischer’s exact test for continuous and categorical variables. Mean ± standard deviation, mediation [first quartile, 3rd quartile] or n (%) were used to summarize data distributions. Univariate Cox proportional hazard regressions assessed associations between predictors and the primary outcome, with hazard ratios (HRs) and 95% confidence intervals (CIs) calculated. Significant variables (P value < 0.05) in univariate analyses were included in the final multivariable Cox proportional hazard models (API, PCWP, mean PAP, and CVP), adjusted for baseline characteristics such as age, gender, and body surface area (BSA). Predictors not significant in the full model (P > 0.05) were eliminated. LVSWI, RVSWI, TPG, SVRi, and PVRi were excluded from the multivariable model to prevent multicollinearity. Kaplan–Meier method, stratified by predefined values for various hemodynamic parameters, utilized the log-rank test to assess equality of survival functions, with the time of the first cardiac catheterization as the origin for survival analyses.
RESULTS
Out of the initial 1,721 patients in the registry, 410 individuals (23.8%) underwent cardiac catheterization during the study period. Among this group, 261 patients had complete data for all measured hemodynamic parameters. Nineteen patients with pulmonary arterial hypertension were excluded from the analysis. Therefore, the final cohort consisted of 242 patients (supplementary figure S1). These individuals were followed for a median duration of 11.4 years (IQR: 7.5-15.9) from their initial cardiac catheterization to the occurrence of their first major adverse cardiac event. Cardiac catheterization was performed more frequently in patients with cc-TGA than those with D-TGA (19.5% vs. 15.2%, p < 0.001). D-TGA patients were younger, with ages of 37.4± 8.3 years, compared to 40.4± 14.0 years for cc-TGA patients (p = 0.031). Of note, six out of the 19 patients excluded due to pulmonary arterial hypertension experienced the primary composite outcome (31.5%), (table 1).
The mean age at the time of cardiac catheterization was 38.5±10.8 years. Notably, patients who underwent cardiac catheterization were more likely to reside in the USA than those who did not (79.3% vs. 59.6%, p<0.001).
All adverse cardiac events, including major cardiac surgical interventions, ADHF, heart transplants, and all-cause mortality, were more frequent among patients who underwent cardiac catheterization (supplementary table S2).
The mean CVP and PCWP were 10±5 mmHg and 17±8 mmHg, respectively. The mean cardiac index was 2.7±0.9 l/min/m2, the mean SVRi was 31.2±10.2 and the PVRi was 5.6±5.7 Wu*m2. The mean API was 3.8±2.7.
In comparison to patients with an API ≥ 3, those with an API < 3 had a higher rate of: diuretic prescriptions, less favorable hemodynamic profiles, severe RV systolic dysfunction, and moderate to severe tricuspid valve regurgitation, as determined through transthoracic echocardiography assessment (table 1).
In the univariate regression analysis, mean CVP, mean PAP, PCWP, API, RVSWI, LVSWI, PVRi and TPG were associated with higher risk of primary and secondary outcomes. Cardiac index, PAPI and SVRi were not associated with higher risk of primary or secondary outcomes. The API was associated with the lowest risk for the primary composite outcome among hemodynamic parameters; the risk was decreased by 41% per one unit increase in API (HR: 0.59 (95% CI: 0.47-0.74), p<0.001) (table 2).
Significant variables (P value <0.05) from the univariate analyses (mean CVP, mPAP, PCWP, API) were included in the final multivariable Cox proportional hazard model. Results of the Cox proportional hazard model showed that PCWP, API and mean PAP were significant predictors of primary and secondary outcomes, while mean CVP was not associated with the primary or secondary outcomes.
Further details regarding the variables identified in the multivariable COX proportional hazard model include:
An elevated mean PCWP was linked to an increased risk of primary and secondary adverse outcomes. A mean PCWP > 20 mmHg was associated with a threefold higher risk of death/transplant/MCS and twice the risk of ADHF (adjusted HR of 3.98 (95% CI: 1.82-8.70, p<0.001) and 2.72 (95% CI: 1.46-5.08, p<0.001), respectively) (supplementary table S3). The Kaplan-Meier curve for the primary and secondary outcomes, categorized by PCWP, is presented in Figure 1 and supplementary figure S2, respectively. The adjusted HR for PCWP > 15mmHg was 17.4 (95% CI: 2.12-142.02, p<0.001) if RV systolic function was normal versus 2.34 (95% CI: 0.91-6.01, p=0.06) in the presence of moderate-severe RV systolic dysfunction. Patients with a PCWP < 15mmHg and normal RV systolic function had a very low mortality rate (supplementary figure S3, S4).
API was associated with both the primary and secondary outcomes. Patients with an API <1.5 had a higher incidence of death/transplant/MCS (adjusted HR of 5.90 (95% CI: 3.01-11.62, p<0.001)) as well as a higher risk of ADHF (HR of 3.40 (95% CI: 1.74-6.47, p<0.001)); (supplementary table S4). The Kaplan-Meier curve for the primary and secondary outcomes, stratified by API, is depicted in figure 2 and supplementary figure S5, respectively.
Using multivariable Cox regression analysis, a mean PAP ≥ 20 mmHg is associated with a higher risk of primary and secondary outcomes, with adjusted HR 2.24 (95% CI: 1.01-4.95), P=0.048 and 2.22 (95% CI:1.18-4.17), P=0.013, respectively (supplementary table S5). The Kaplan-Meier curve for the primary and secondary outcomes, stratified by mean PAP is shown in Figure 3 and supplementary figure S6, respectively.
Approximately 74% of the patients exhibited a normal cardiac index (classified as ‘warm,’ with values ≥ 2.2 l/min/m2) and 47.5% had normal pulmonary capillary wedge pressure (referred to as ‘dry,’ with values < 15 mmHg). The most common hemodynamic profile was ‘wet/warm’ (n=94, 38.8%), followed by ‘dry/warm’ (n=85, 35.1%), ‘wet/cold’ (n=33, 13.6%) and ‘dry/cold’ (n=30, 12.4%) profiles.
The highest incidence rates of death/transplant/MCS and ADHF were observed among patients with a ‘wet/cold’ hemodynamic profile (39.4% and 51.5%), followed by those with a ‘wet/warm’ profile (38.3% and 46.8%). In contrast, patients with ‘dry/cold’ hemodynamic profiles had lower incidence rates of death/transplant/MCS and ADHF (12.3% and 23.3%) (figure 4).
Multivariable regression, based on clinical profiles, was employed to evaluate the risk of primary and secondary outcomes. Patients with a ‘wet/cold’ hemodynamic profile, characterized by a PCWP ≥ 15 mmHg and a cardiac index < 2.2 L/min/m2, had the highest risk of death/transplant/MCS, and ADHF with adjusted HR of 3.83 (95% CI: 1.63-9.02, p<0.001) and 3.50 (95% CI: 1.73-7.07, p<0.001), respectively (supplementary table S6). The Kaplan-Meier curve for the primary and secondary outcomes, stratified by hemodynamic profiles (figure 5 and supplementary figure S7).
Patients with D-TGA and ccTGA exhibited similar clinical and hemodynamic characteristics, with the exception of diuretic use and the degree of tricuspid valve regurgitation, which was higher among ccTGA patients (supplementary table S7).
DISCUSSION
Our study investigates the prognostic significance of invasive hemodynamic parameters in a relatively large cohort of patients with a systemic RV undergoing cardiac catheterization.
The primary findings of our study can be summarized as follows: API emerged as the most robust prognostic factor for long-term adverse cardiac outcomes among the various hemodynamic parameters studied. The invasive hemodynamic assessment facilitated the identification of distinct profiles utilizing PCWP and cardiac index among patients with a systemic RV. These profiles were found to be significant determinants of outcomes and hold promise as guiding tools for therapeutic decision-making.
In our multivariable analysis, the only hemodynamic parameters demonstrating statistical significance were PCWP, API and mean PAP;
The prognostic value of systemic ventricle filling pressures has been established among patients with a systemic LV, where elevated PCWP was associated with poor 1-year survivals of < 40%.23–26 We demonstrated similar findings among patients with systemic RV; a PCWP > 20 mmHg was associated with a 3-fold increase in the primary composite outcome (death/transplant/MCS) and a 2-fold increase in the secondary outcome (ADHF).
It is important to note that PCWP does not always correlate with ventricular end diastolic pressure (EDP) and PCWP, as an intrathoracic measurement, often underestimates ventricular EDP. In patients with TGA post atrial switch repair, intra-atrial baffle issues are not uncommon (among 151 D-TGA patients in this study, 34 (22.5%) had history of prior baffle interventions, and 27 (17.8%) required baffle interventions). The presence of pulmonary baffle stenosis, in particular, will mean that use of PCWP as a surrogate for EDP will lead to overestimation of EDP, therefore it is worth considering measurement of both values during catheterization in this population.
We found that PCWP was particularly predictive of adverse outcomes in those with preserved RV systolic function, which may suggest alternative mechanisms for PCWP elevation driving clinical outcomes in this group (eg tricuspid regurgitation, diastolic dysfunction or baffle obstruction in those with an atrial switch procedure).
API is an innovative biomarker that integrates pulse pressure and filling pressures. The potential advantage of API over PCWP is API was designed to simultaneously represent cardiac function and filling pressures, which is representative of the relationship described in a pressure volume loop. A low API, characterized by a low pulse pressure and high PCWP, is suggestive of decreased myocardial efficiency27 (usually reflective of high energy expenditure and oxygen consumption).28
Previous studies have demonstrated the utility of API in predicting adverse cardiac outcomes among patients with systemic LV failure. A study by Belkin et al. of 224 patients with systemic LV systolic dysfunction found that individuals with an API < 1.45 had higher rate of the need for advanced therapies at 30 days including inotropes, orthotopic heart transplant, or left ventricle assist device placement (52% versus 21% if API ≥ 1.45).13 An analysis of the ESCAPE trial demonstrated that an API ≥ 2.9 was the optimal cutoff value for predicting freedom from the primary outcome (freedom from primary outcome was 58.4% among patients with an API < 2.9 in contrast to 83.5% in patients with an API ≥ 2.9).12
Our study revealed that individuals with a systemic RV and higher API experienced a low rate of adverse cardiovascular events (with each unit increase in API associated with a 41% relative decrease in the risk of death, transplant, or mechanical circulatory support) as seen in table 2. An API less than 1.5 is linked to a five-fold increase in death, transplant, or mechanical circulatory support and a three-fold increase in ADHF (supplementary table S4).
It is worth acknowledging that API does not necessarily factor in aortic bioelasticity. This may be of relevance when interpreting API, given decreased aortic distensibility and increased pulse wave velocity have been found in this group (ie. the API may be normal or elevated despite poor cardiac output owing to a stiff aorta).29
In our study, where those with ‘precapillary’ pulmonary hypertension (ie pulmonary HT with PCWP<15mmHg) were excluded, individuals with a mean pulmonary artery pressure (PAP) of ≥20 mmHg face a two-fold increased risk of both primary and secondary outcomes. If the mean PAP exceeds 40 mmHg, the risk of death, transplant, or MCS escalates six-fold, and the risk of ADHF quadruples. Our findings align with previous research, confirming a linear relationship between mean PAP and all-cause mortality.30,31
In a study by Chaix et al of patients with D-TGA who had undergone Mustard or Senning procedures, pulmonary hypertension (mean PAP > 25mmHg) was linked to higher cardiovascular mortality compared to patients without pulmonary hypertension.32 In that study, pulmonary hypertension was largely ‘post capillary’ in nature (elevated PCWP seen in all 18 patients with pulmonary HT (mixed picture in two patients)), implying that pulmonary hypertension relates to systemic RV dysfunction and abnormal atrial anatomy with subsequent pulmonary vascular remodelling (in the TGA/atrial switch population).
Our findings indicate that an elevated CVP (> 10 mmHg) is not associated with adverse cardiac outcomes. This observation contrasts with data from the ESCAPE trial, where baseline and persistently elevated mean CVP (>10 mmHg) following treatment were indicative of an increased risk of combined adverse outcomes, including death, cardiovascular hospitalization, and heart transplantation. Similarly, Damman et al. reported that mean right atrial pressure >6 mmHg was independently associated with mortality.33,34 A possible explanation for our discrepant finding is that CVP is less reflective of systemic ventricular failure in patients with a subpulmonic LV (compared to those with a subpulmonic RV), as the former may be better equipped to adapt to increased afterload ( eg. due to elevated pulmonary pressures in heart failure).
Our study did not find any association between cardiac index alone and primary or secondary outcomes. This is consistent with prior research on pulmonary arterial catheter use in hemodynamically stable patients admitted with ADHF, which demonstrated that PCWP was associated with long-term adverse events, whereas a decreased cardiac index was not.11
It has been well-established that clinical and invasive hemodynamic profiles can predict short-term survival in heart failure patients. These profiles indicate increased mortality when congestion is present, and outcomes become even worse when both congestion and hypoperfusion coexist.35–37 However, there has been a lack of evidence regarding whether these profiles are associated with adverse cardiac effects in adults with systemic RV.
Our study describes four distinct hemodynamic profiles based on median PCWP and cardiac index, which carry independent, long-term prognostic information. These profiles mirror those seen in patients with systemic LV in the context of acute myocardial infarction and chronic cardiomyopathy.35–38 Among these profiles, the ‘wet/cold’ hemodynamic profile, characterized by a PCWP ≥ 15 mmHg and a cardiac index < 2.2 L/min/m2, is linked to the highest risk of adverse outcomes (death, heart transplant or need for MCS), as well as risk of ADHF.
Two key observations regarding patients with ‘wet/warm’ and ‘wet/cold’ profiles are noteworthy. First, regardless of perfusion status, these patients have a worse prognosis compared to those with ‘dry’ hemodynamic profiles, consistent with our finding that PCWP is a better predictor of outcomes than cardiac index. Second, long-term survival is notably poor among these patients, underscoring the importance of considering heart transplantation as a potential treatment option earlier in patients with a wet/cold hemodynamic profile, in particular. Future investigations should aim to establish whether the clinical assessment of hemodynamic profiles can offer prognostic value comparable to invasive assessments in adults with systemic RVs.
LIMITATIONS
This is a retrospective study, thus sampling bias is expected and findings therefore are not necessarily applicable to all patients with a systemic RV. For example, in our cohort, patients referred for cardiac catheterization were older, more often had prior arrhythmias, were more commonly prescribed cardiac medications, and had a higher likelihood of death/transplant/MCS as well as admission for ADHF.
The cutoff values for hemodynamic parameters were established using recognized normal thresholds and we used the mean PCWP and cardiac index values to establish four distinct hemodynamic profiles; future research could validate these cut-off points in another cohort.
Our analysis combines both D-TGA and ccTGA patients into a single group. Despite their similar characteristics and comparable mean PCWP levels, it is worth acknowledging that PCWP elevation, in particular, may relate to different mechanisms other than RV dysfunction in the two groups (eg. TR (common in ccTGA) and pulmonary venous baffle stenosis in TGA/atrial switch).
We used non-invasive brachial blood pressure to calculate API rather than invasive central aortic pressure measurements (which were not universally recorded), which is in keeping with prior studies;12,13 non-invasive BP may vary from central aortic pressures thus may affect the API validity.39,40 Nonetheless, a computed API using non-invasive brachial blood pressure was associated with clinical outcomes in patients with systemic LV and heart failure.12,13
It is recognized that there is only modest agreement between thermodilution and Fick cardiac output calculations (for example, 38% of patients had a difference of more than 20% between Fick and thermodilution CO calculations in a large cohort study).9 Cardiac output calculation in this study was based upon thermodilution or the indirect Fick method if the thermodilution value is not reported.
Patients who undergo sedation during cardiac catheterization may be have variations in their hemodynamics; we do not have data regarding anesthesia, sedation, or the use of vasopressors during cardiac catheterization.
CONCLUSION
Calculated API, and hemodynamic profiles provide long-term prognostic information in individuals with a systemic right ventricle. API < 1.5 appears to be the strongest prognostic factor of all-cause death/transplant/MCS among hemodynamic parameters. A hemodynamically directed assessment based on the cardiac index and PCWP relationship is also prognostic.
MB
Dr. Mark Belkin
Cardiology · Advanced Heart Failure & TransplantationUniversity of Chicagomarkb