Author + information
- Received January 22, 2019
- Revision received February 24, 2019
- Accepted March 3, 2019
- Published online May 27, 2019.
- Geoff Strange, PhDa,∗ (, )
- Simon Stewart, PhDb,
- David S. Celermajer, MD, PhDc,
- David Prior, MBBS, PhDd,
- Gregory M. Scalia, MBBS (Hons), MMedSce,
- Thomas H. Marwick, MBBS, PhD, MPHf,
- Eli Gabbay, MBBS, MDa,
- Marcus Ilton, MDg,
- Majo Joseph, MBBSh,
- Jim Codde, PhDa,
- David Playford, MBBS, PhDa,
- on behalf of the NEDA Contributing Sites
- aUniversity of Notre Dame, Fremantle, Western Australia, Australia
- bUniversity of Cape Town, Cape Town, South Africa
- cFaculty of Medicine and Health, University of Sydney, Sydney, New South Wales, Australia
- dUniversity of Melbourne, St. Vincent’s Hospital, Melbourne, Victoria, Australia
- eUniversity of Queensland, The Prince Charles Hospital, Brisbane, Queensland, Australia
- fBaker Heart and Diabetes Institute, Melbourne, Victoria, Australia
- gMenzies School of Health Research, Royal Darwin Hospital, Tiwi, Northern Territory, Australia
- hFlinders University, Adelaide, South Australia, Australia
- ↵∗Address for correspondence:
Dr. Geoff Strange, University of Notre Dame, 32 Mouat Street, Fremantle, WA 6160, Australia.
Background There is increasing evidence that current thresholds for diagnosing pulmonary hypertension (PHT) underestimate the prognostic impact of PHT.
Objectives The aim of this study was to determine the prognostic impact of increasing pulmonary pressures within the National Echocardiography Database of Australia cohort (n = 313,492).
Methods The distribution of estimated right ventricular systolic pressure (eRVSP) was examined in 157,842 men and women. All had data linkage to long-term survival during median follow-up of 4.2 years (interquartile range: 2.2 to 7.5 years).
Results The cohort comprised 74,405 men and 83,437 women 65.6 ± 17.7 years of age. Overall, 17,955 (11.4%), 7,016 (4.4%), and 4,515 (2.9%) subjects had eRVSP levels indicative of mild (40 to 49 mm Hg), moderate (50 to 59 mm Hg), or severe (≥60 mm Hg) PHT, respectively, assuming a right atrial pressure of 5 mm Hg. These subjects were more likely to die during long-term follow up (for severe PHT, adjusted hazard ratio: 9.73; 95% confidence interval: 8.60 to 11.0; p < 0.001). After adjustment for age, sex, and evidence of left heart disease, those subjects with eRVSP levels within the third (28.05 to 32.0 mm Hg; hazard ratio: 1.410; 95% confidence interval: 1.310 to 1.517) and fourth (32.05 to 38.83 mm Hg; hazard ratio: 1.979; 95% confidence interval: 1.853 to 2.114) quintiles had significantly higher mortality (p < 0.001) than those in the lowest quintile. Accordingly, a clear and consistent threshold of increased mortality (including 1- and 5-year actuarial mortality) around an eRVSP of 30.0 mm Hg was evident.
Conclusions In this large and unique cohort, the prognostic impact of clinically accepted levels of PHT was confirmed. Moreover, a distinctly lower threshold for increased risk for mortality (eRVSP >30.0 mm Hg) indicative of PHT was identified. (A Longitudinal Cohort Study of Echocardiograms From Public and Private Echocardiography Laboratories From Around Australia, Linked With the National Deaths Index; ACTRN12617001387314)
Pulmonary hypertension (PHT) is a hemodynamic state arising from multiple pathogenic etiologies and associated diseases (1). As currently defined (mean pulmonary artery pressure >25 mm Hg at right heart catheterization), PHT is associated with a poor prognosis (2). Beyond the rarer, pre-capillary form of pulmonary arterial hypertension (2–6), treatment options are often limited. It is on this basis that beyond identifying high-risk patients with markedly elevated pulmonary pressures (7,8), proactively identifying those with mild or so-called borderline PHT (mean pulmonary artery pressure 20 to 25 mm Hg) to slow or even prevent disease progression is of therapeutic interest (9,10).
A recent meta-analysis including 16,482 patients (7,451 with mild PHT and 1,414 diagnosed by echocardiography) derived from 15 single-center and multicenter studies (11) reinforced the findings of the largest study to date examining mortality and “borderline” PHT among 5,030 male veterans (9). Specifically, it demonstrated that during mean follow-up of 5.2 years, even mild PHT was associated with a 1.52-fold (95% confidence interval [CI]: 1.32 to 1.74; p < 0.001) increased risk for mortality with equivalent results on the basis of right heart catheterization versus echocardiographic detection of PHT (12). Given that rapid progression from borderline PHT to clinically defined PHT (mean pulmonary artery pressure >25 mm Hg) has also been reported in 61% of affected patients (10), careful clinical evaluation of patients is required to help identify those at increased risk and who may benefit from disease specific therapy (1). Furthermore, identification of “optimal cutoffs” is needed to identify at-risk populations with PHT at the earliest opportunity in order to inform what further investigation may be warranted (1,11).
Despite the potential effects of increasing age and body mass index (12), echocardiography has become the first-line diagnostic and screening tool for PHT (13). On this basis and given the current limitations of the published research on milder forms of PHT, we applied the unique resources of the National Echo Database Australia (NEDA) (14) to more forensically examine the association between short- and long-term mortality with increasing pulmonary pressures. Critically, this included our ability to identify >45,000 cases that would be below the current threshold for PHT with linkage to actuarial 1- and 5-year mortality.
We hypothesized that independent of age, sex, and evidence of left heart disease (LHD), within the large NEDA cohort (14) we would identify a lower than expected threshold for increased mortality linked to estimated right ventricular systolic pressure (eRVSP) levels measured on echocardiography. To test this hypothesis, we examined: 1) the underlying distribution of eRVSP and indicative prevalence of PHT in adult patients undergoing echocardiography; and 2) the prognostic impact of incremental increases in eRVSP on all-cause and cardiovascular-specific mortality.
Study setting and design
As described previously (14), NEDA is an observational registry that captures individual echocardiographic data (combined with basic demographic profiling) on a retrospective and prospective basis from participating centers throughout Australia with the capacity to link such data to health outcomes. The study conforms to the Declaration of Helsinki (15), and reporting conforms to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines for observational studies (16). NEDA is also registered with the publicly accessible Australian New Zealand Clinical Trials Registry (ACTRN12617001387314). Ethical approval was obtained from all relevant human research ethics committees in each state and territory of Australia. At the time of study census, a total of 12 centers had contributed >500,000 investigations (with close to 20 million individual measurements) from >300,000 subjects undergoing echocardiography.
Once a participating center has received ethical approval, all echocardiographic measurement and report data contained in the echocardiography database are collected (the first through last echocardiograms span the 20-year period from June 26, 1997, to June 13, 2017). Each database is remotely transferred into a central database using a “vendor-agnostic,” automated data extraction process that transfers every measurement for each echocardiogram obtained in an entire echocardiography database into a standard NEDA data format. Precise definitions for each echocardiographic variable are applied. Variables with the same names as the NEDA standard are automatically matched, and those with different names are manually matched with the NEDA standard by a principal investigator (D. Playford), a cardiologist with subspecialist expertise in echocardiography. Duplicate measurements with different naming conventions are combined. Units are transformed to the single NEDA standard, and repeated measures for the same variable are converted to a single variable according to the NEDA study protocol (14). A continuously updated NEDA data dictionary (including all primary data and those derived or calculated from primary data) is maintained through a master NEDA database that forms the basis for all subsequent analyses and outputs. Robust text recognition software was also used to identify any descriptors of right atrial and right ventricular (RV) size and RV function in summary reports of findings.
To address the pre-specified research questions, individual NEDA data are linked to National Death Index data provided by the Australian Institute for Health and Welfare; additional probability matching is used to increase the sensitivity and specificity of mortality data (17). This linkage provided the survival status of everyone (advanced probability matching) for the study census date of October 20, 2017. If a subject had died, the listed primary cause of death was categorized according to International Statistical Classification of Diseases and Related Health Problems-Tenth Revision coding. Subsequently, all chapter codes in the range of I00 to I99 were considered to represent cardiovascular-related deaths; a previous study demonstrated 92.5% sensitivity and 89.6% specificity for this coding (18).
As shown in Figure 1 (study flowchart), NEDA data as of October 2017 were used to identify the following subjects: 1) men and women ≥18 years of age; 2) with at least 1 echocardiographic investigation (with the last recorded investigation used if multiple investigations documented); and 3) with a recorded peak tricuspid regurgitation velocity (TRV) to derive an eRVSP as a reliable surrogate of pulmonary arterial systolic pressure. A conservative method was used to derive eRVSP by using the Bernoulli equation (eRVSP = 4 × [TRV]2 + 5 mm Hg), noting that previous studies have applied a less conservative right atrial pressure of 10 mm Hg to calculate eRVSP or pulmonary arterial systolic pressure (12). A right atrial pressure of 5 mm Hg approximates the average value recorded overall and removes any variation between laboratories.
All subjects with calculated eRVSPs (n = 157,842) from their last recorded echocardiographic examinations were categorized as follows according to current clinical guidelines (1,3): 1) normal (eRVSP <40.0 mm Hg); 2) mildly elevated (eRVSP 40.0 to 49.9 mm Hg); 3) moderately elevated (eRVSP 50.0 to 59.9 mm Hg); and 4) severely elevated (eRVSP ≥60.0 mm Hg).
These same data (eRVSP) were then characterized according to their quintile distribution (data for men and women were combined given broadly equivalent distributions): first quintile, ≤24.36 mm Hg; second quintile, 24.37 to 28.04 mm Hg; third quintile, 28.05 to 32.04 mm Hg; fourth quintile, 32.05 to 38.82 mm Hg; fifth quintile, ≥38.83 mm Hg. LHD was defined as any of the following: 1) left ventricular ejection fraction <54% (19), measured using the Simpson biplane method if image quality was sufficient and the Teichholz method if image quality was insufficient (20); 2) signs of increased left ventricular filling pressure (manifesting in a ratio of mitral inflow E-wave peak velocity to peak early relaxation tissue Doppler velocity [E/E′] >12) (21); 3) left atrial volume index >34 ml/m2 (19); and/or 4) hemodynamically significant (greater than mild) mitral or aortic valve disease (22).
All fatal events (including cardiovascular-related events) were identified during a median of 4.2 years (interquartile range: 2.2 to 7.5 years) of follow-up. We then explored the relationship between eRVSP and survival with respect to both “current clinical convention” and the “statistical distribution” of this parameter within the study cohort to identify more precisely the level at which the risk for all-cause and cardiovascular-related mortality increased.
No formal calculations of study power were performed given the very large number of cases (n = 157,842), fatal events (n = 37,743), and duration of follow-up (813,000 patient-years). All analyses were conducted according to a pre-specified statistical analysis plan. Discrete variables are summarized by frequencies and percentages (with 95% CIs as appropriate). Continuous variables are summarized using standard measures of central tendency and dispersion. Unless otherwise specified, between-group comparisons were assessed using Student’s t-tests, Mann-Whitney U tests (for continuous data not normally distributed), chi-square tests (with calculation of odds ratios and 95% CIs), or analysis of variance (with post hoc Dunnett t-tests) as appropriate. Actuarial 1- and 5-year survival rates were initially calculated from the 147,874 and 67,072 subjects with complete follow-up for those time points, with construction of 5-year Kaplan-Meier survival curves according to conventional levels of eRVSP indicative of PHT. Multiple logistic regression (entry model) was then used to derive odds ratios and 95% CIs for all-cause mortality according to these same levels of eRVSP and adjusting for age and sex. A series of Cox proportional hazards models (entry model with proportional hazards confirmed by visual inspection of adjusted survival curves) were then used to derive age- and sex-adjusted hazard ratios (HRs) and 95% CIs according to both conventional levels of eRVSP and their quintile distribution (equivalent for men and women). On visual inspection of actuarial survival rates per 2 mm Hg increase in eRVSP, a clear and consistent cut point of increased risk for all-cause mortality equal to eRVSP >30 mm Hg was evident. This was further explored using post hoc Cox proportional hazards models adjusting for age (as a continuous variable and then a cut point of 65 years), sex, and the presence or absence of LHD from the long-term survival data of those subjects with eRVSPs between 20.00 and 39.99 mm Hg (figure legends). All analyses were performed using SPSS version 24.0 (SPSS, Chicago, Illinois), and statistical significance was accepted at a 2-sided p value of <0.05.
A total of 157,842 (50.3%) men and women with documented peak TRVs to generate eRVSP values within the broader NEDA cohort were studied. Figure 2 shows the normally distributed frequency distribution of eRVSP levels (median 25.00 mm Hg; interquartile range: 20.33 to 31.36 mm Hg). Both men (65.6 ± 17.2 years vs. 57.9 ± 16.5 years) and women (65.7 ± 18.1 years vs. 56.7 ± 17.9 years) with calculable eRVSPs were significantly older than those without these data recorded (p < 0.001 for both comparisons). On an age- and sex-adjusted basis, those individuals without eRVSPs were less likely to die in the longer term (adjusted HR: 0.861; 95% CI: 0.847 to 0.876).
Table 1 summarizes the demographic and echocardiographic characteristics of the study cohort. As expected, eRVSP levels rose steadily (in both men and women) with increasing age, rising from a mean of 25.9 ± 8.3 mm Hg among those 20 to 25 years of age (n = 2,981) to 39.3 ± 12.6 mm Hg among those ≥85 years of age. Overall, 128,356 subjects (81.3%) had eRVSP levels <40.0 mm Hg. An additional 45,104 subjects (28.6%) had eRVSPs between 30.0 and 39.9 mm Hg. Of the remainder, 17,955 (11.4%), 7,016 (4.4%), and 4,515 (2.9%) subjects, respectively, were classified as having mild, moderate, and severe PHT (18.7% combined for any form of PHT). If it were conservatively assumed that all subjects without measurable TRVs for eRVSP had values <40 mm Hg (i.e., a “normal value”), the prevalence of PHT within the entire NEDA cohort would be 29,486 of 313,492 (9.4%; 95% CI: 9.3% to 9.5%). The proportion of men and women in each category of incremental eRVSP broadly reflected the overall proportion of men (47.1%) and women (52.9%) in the study cohort. Consistent with an age gradient in eRVSP levels overall, among both men and women, those with elevated eRVSPs were markedly older than the rest (p < 0.001 for both comparisons). A dilated right ventricle and impaired RV function were reported in 12,409 (7.8%) and 6,045 (3.8%) subjects. Evidence of LHD was also present in 40.4% and 34.7% of men and women, respectively, with increasing levels according to increasing age and eRVSP levels (>50% in those with severe PHT).
Table 2 summarizes the survival profile of the cohort according to severity of echocardiographically determined PHT at 1 and 5 years (actuarial survival), as well as the impact of PHT on longer-term survival on an age- and sex-adjusted basis. As expected, the risk for mortality rose markedly with eRVSP (p < 0.001 for all comparisons between those with eRVSPs <40.0 mm Hg and the 3 categories of elevated eRVSP on an unadjusted and an adjusted basis). This was evidenced by a range in 1- and 5-year actuarial mortality from a low of 6.8% and 20.3% to a high of 44.2% and 78.0% in those with normal to severely elevated eRVSPs. On an unadjusted basis, median survival among those with full 5-year follow-up ranged from 14 to 47 months according to the severity of PHT on the basis of eRVSP (Figure 3); those with severely elevated eRVSP were almost 10-fold more likely to die within 5 years when adjusting for age and sex. Within all models, advancing age (HR: 1.058; 95% CI: 1.057 to 1.059 per year; p < 0.001) and being male (HR: 1.403; 95% CI: 1.375 to 1.432) were also correlated with mortality.
Figures 4A (all-cause mortality) and 4B (cardiovascular-related mortality) show survival curves derived from age- and sex-adjusted Cox proportional hazards models on the basis of the pre-specified clinical levels of eRVSP in comparison with the same models but using the quintile distribution of eRVSP. The latter survival curves suggested a threshold (or “pivot point”) of increased risk for all-cause mortality in the range of borderline PHT (30.0 mm Hg) relative to the lowest quintile of eRVSP (≤24.36 mm Hg). There was no increased risk for mortality in those within the second quintile of eRVSP (24.37 to 28.04 mm Hg; adjusted HR: 1.002; 95% CI: 0.932 to 1.078; p = 0.955) but increasing risk for those with values in the third (28.05 to 32.0 mm Hg; HR: 1.410; 95% CI: 1.310 to 1.517; p < 0.001), fourth (32.05 to 38.83 mm Hg; HR: 1.979; 95% CI: 1.853 to 2.114; p < 0.001), and fifth (>38.84 mm Hg; HR: 4.404; 95% CI: 4.132 to 4.694; p < 0.001) quintiles. This observation was also confirmed with respect to cardiovascular-related mortality (HR: 1.354 [95% CI: 1.289 to 1.423] to 3.491 [95% CI: 3.342 to 3.464] for the third to fifth quintiles of eRVSP; p < 0.001 for all comparisons above the third quintile). Figure 5 plots the adjusted HRs for longer-term, all-cause mortality (absolute 1- and 5-year mortality rates also shown for those with follow-up data) above and below the statistically identified threshold of an eRVSP of 30.0 mm Hg. Post hoc analyses confirmed evidence of a lower threshold of risk persisted when examining cardiovascular- and respiratory-related mortality (Online Figures 1 and 2). On the basis of a strong and consistent cut point of increased 5-year actuarial mortality around an eRVSP of 30.0 mm Hg in younger age groups (Online Figure 3), we then further examined the potential confounding effects of age (including 66,648 and 54,026 subjects >65 and <65 years of age, respectively). This showed the same phenomenon with respect to all-cause mortality but stronger in younger subjects (Figure 6). This was also observed in the presence and absence of LHD (Figure 7) and adjusting for body mass index. As expected, RV dilatation and RV dysfunction (mild to severe) were both associated with increased risk for 5-year mortality in those with eRVSPs of 20.0 to 40.0 mm Hg (adjusted odds ratios: 2.51 [95% CI: 2.25 to 2.81; p < 0.001] and 1.81 [95% CI: 1.58 to 2.06; p < 0.001], respectively). However, the same cut point at an eRVSP level of 30.0 mm Hg persisted when adjusting for these and other potential confounders.
Among the 37,743 subjects who died overall, ischemic heart disease (16.3%), malignancy (21.3%), and respiratory disease (28.9%) were common. In keeping with a lack of association between PHT and malignancy and a competing risk for contributory respiratory and heart disease (above a threshold of 30.0 mm Hg), the former and latter were a decreasingly more minor and an increasingly more major contributor to death, respectively, in those with increasingly elevated pulmonary pressures (Online Figure 4). Hence, the adjusted risk for mortality was markedly elevated among the 45,104 subjects (28.6% of the study cohort) with eRVSP levels indicative of what has been previously referred to as borderline PHT (30.0 to 39.9 mm Hg), irrespective of the primary listed cause of death, age, or evidence of LHD (Central Illustration).
Consistent with the design and rationale of NEDA (14), we present, to the best of our knowledge, the largest ever study of mortality (including 5-year actuarial survival rates) across the full spectrum of PHT. Accordingly, we confirm the deadly nature of currently defined PHT, the 9.4% of patients with eRVSPs >40 mm Hg having an almost 3-fold increased risk for 5-year mortality compared with the rest. We also confirm the additional increased risk for mortality imposed by mild or borderline PHT, as revealed by relatively large cohort studies (9) and a recently published meta-analysis of available data (11). Moreover, using the granularity of detail afforded to us by NEDA, we identified a clear and consistent “cut point” (eRVSP 30.0 mm Hg) at which short-term and, particularly, longer-term mortality increased. Specifically, within the additional 14.4% of the entire NEDA cohort with milder elevations of eRVSP, those with eRVSP levels between 30.0 and 39.9 mm Hg were, on an adjusted basis, 1.4- to 1.9-fold more likely to die compared with the rest. This phenomenon was observed in all age groups, did not appear to be confounded by LHD or other comorbidity, and was consistent with increasing levels of RV dilatation and impaired function in those with eRVSPs >30.0 mm Hg. Critically, our more detailed calculations of mortality according to increasing pulmonary pressures are entirely consistent with the broader findings of the meta-analysis published by Kolte et al. (11).
According to the latest clinical guidelines for the detection and management of PHT, further follow-up of those patients in the so-called intermediate-probability range on echocardiography is recommended only if other signs or risk factors are present (1). Indeed, the probability indicator for the presence of PHT on the basis of echocardiographic assessment is at best qualitative and at worst may perpetuate clinical indecision in the intermediate range. When combined with the existing evidence (11), our findings now suggest that those presenting with eRVSPs >30.0 mm Hg, no matter the eventual disease, should be carefully evaluated clinically, and where appropriate, further investigation should be undertaken to unmask disease that may benefit from disease-specific therapy. Whether this approach will change the trajectory toward increased mortality has not yet been investigated.
Because these data are based on indirect estimates of PAP derived from echocardiography rather than from right heart catheterization, it is first important to comment on the overall validity of our approach to determining the indicative prevalence of PHT and prognostic impact using NEDA data. Although it would be ideal to rely on right heart catheterization alone to detect PHT, echocardiography will continue to be the first detection point for PHT in most cases. As such, many groups have correlated eRVSP with invasive pulmonary artery systolic pressure, producing areas under the receiver-operating characteristic curve ranging from 0.74 to 0.97 (7,23,24), and equivalent findings of mortality risk associated with so-called borderline PHT have been derived from studies using right heart catheterization and echocardiography (11).
As already noted, the novel collection of such large data on the full spectrum of PHT and long-term follow-up enabled us to identify (if it indeed existed) with more certainty a threshold for increased mortality risk among those with so-called borderline or mild PHT. Consistent with our own findings, a small population (n = 179) from the Olmsted County data also had a clear survival inflection point at an equivalent level of 32 mm Hg (7). From a clinical perspective, these data reaffirm current recommendations (1) that patients should be carefully clinically evaluated for symptoms that may be consistent with PHT (usually exertional dyspnea). Symptomatic patients should undergo echocardiography, and those with TRV levels at either end of the spectrum should be considered either at low (TRV <2.8 m/s) or high (TRV >3.4 m/s) risk for PHT and/or a poorer prognostic outcome. However, there is still an area of uncertainty around the clinical management of those patients currently classified at intermediate risk for PHT (TRV 2.9 to 3.4 m/s, correlating to a range of eRVSP of 33 to 46 mm Hg). A recent study focusing on the value of initially screening for PHT on the basis of echocardiography, in this case associated with interstitial lung disease (25), found that current recommendations were of limited value in excluding PHT (with many “low-risk patients” on the basis of TRV misclassified). Whether efforts to target subjects with PHT associated with an eRVSP >30.0 mm Hg (or equivalent) for more intense investigation and treatment will alter their outcomes requires further investigation.
Beyond our reliance on echocardiography, there are important limitations to be considered when interpreting these data. First, the NEDA cohort typically comprises subjects being investigated for possible or pre-existing cardiovascular disease. Although NEDA has the capacity to capture detailed echocardiographic data on a substantial number of subjects with reliable individual linkage to long-term mortality, it does not (yet) capture some important clinical details pivotal to health outcomes, including a patient’s pharmacological and surgical management. Although we can assume that many of those subjects with calculable eRVSP values were symptomatic, we cannot confirm this important component of suspicion for invasively investigating PHT on the basis of initial echocardiography (1). The presence of PHT does not inform the cause of the PHT, and as such we also have no means (other than identifying echocardiographic evidence of LHD) of determining the specific type of PHT (including pulmonary arterial hypertension) present. Given that the absence of a tricuspid regurgitation jet does not exclude PHT (26), our estimates of the prevalence and prognostic impact of PHT should be interpreted as the minimum indicative prevalence from an epidemiological perspective. Although we have plans to extend the number of contributing NEDA centers to improve the applicability of the data to the entire population, these data were largely derived from specialist centers and clinics in Australia, but noting at least the similarity in pulmonary artery pressure (using similar methods) derived from the Olmstead County population cohort (median pulmonary artery systolic pressure 25 mm Hg; interquartile range: 24 to 30 mm Hg) (7). Moreover, although Australia is a truly multicultural society, unlike reports derived from North America (8), it is predominantly of European descent, with a minimum of subjects of African descent. Last, although our data suggest that even mild PHT may affect right heart structure and/or function, without quantitative RV assessment, we are unable to definitively explore this important observation, nor can we definitively determine the excess mortality attributable to right heart abnormalities in that group. This represents an important focus for future research.
NEDA provides a unique opportunity to examine the prognostic implications of varying levels of PHT derived from echocardiography in a very large cohort of adults over a prolonged period. Our unique findings suggest that those currently considered to be at intermediate risk for underlying PHT have a significant risk for mortality once they reach the threshold of eRVSP >30.0 mm Hg. These subjects should be thoroughly clinically evaluated, and where appropriate, further investigation should be undertaken with the goal of unmasking underlying disease that may be more pro-actively managed than suggested by current expert guidelines (1).
COMPETENCY IN MEDICAL KNOWLEDGE: Short- and longer-term all-cause and cardiovascular mortality rates in patients with PHT are markedly elevated above an echocardiographical eRVSP of >30 mm Hg.
TRANSLATIONAL OUTLOOK: Future studies should characterize patients with borderline PHT on the basis of eRVSP to determine the mechanisms of excess mortality and evaluate the efficacy of therapeutic interventions to prolong survival.
The National Echo Database Australia was initially supported (database engineering and infrastructure costs) through unrestricted research grants from Actelion, Bayer, and GlaxoSmithKline. It is also supported by National Health and Medical Research of Australia funding (grant 1055214). Dr. Celermajer works at an institution that has received clinical trials funding and educational grants from Actelion. Dr. Prior has received payments for talks from Actelion. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Bradley A. Maron, MD, served as Guest Associate Editor for this paper.
Listen to this manuscript's audio summary by Editor-in-Chief Dr. Valentin Fuster on JACC.org.
- Abbreviations and Acronyms
- confidence interval
- estimated right ventricular systolic pressure
- hazard ratio
- left heart disease
- National Echo Database Australia
- pulmonary hypertension
- right ventricular
- tricuspid regurgitation velocity
- Received January 22, 2019.
- Revision received February 24, 2019.
- Accepted March 3, 2019.
- 2019 American College of Cardiology Foundation
- Galie N.,
- Humbert M.,
- Vachiery J.L.,
- et al.
- Strange G.,
- Playford D.,
- Stewart S.,
- et al.
- Strange G.,
- Williams T.,
- Kermeen F.,
- Whyte K.,
- Keogh A.
- Peacock A.J.,
- Murphy N.F.,
- McMurray J.J.,
- Caballero L.,
- Stewart S.
- Lam C.S.,
- Borlaug B.A.,
- Kane G.C.,
- Enders F.T.,
- Rodeheffer R.J.,
- Redfield M.M.
- Choudhary G.,
- Jankowich M.,
- Wu W.C.
- Maron B.A.,
- Hess E.,
- Maddox T.M.,
- et al.
- Assad T.R.,
- Maron B.A.,
- Robbins I.M.,
- et al.
- Kolte D.,
- Lakshmanan S.,
- Jankowich M.D.,
- Brittain E.L.,
- Maron B.A.,
- Choudhary G.
- McQuillan B.M.,
- Picard M.H.,
- Leavitt M.,
- Weyman A.E.
- Parasuraman S.,
- Walker S.,
- Loudon B.L.,
- et al.
- Strange G.,
- Celermajer D.S.,
- Marwick T.,
- et al.
- ↵WMA Declaration of Helsinki—ethical principles for medical research involving human subjects. Available at:. https://www.wma.net/policies-post/wma-declaration-of-helsinki-ethical-principles-for-medical-research-involving-human-subjects/. 1442019.
- Magliano D.,
- Liew D.,
- Pater H.,
- et al.
- Lang R.M.,
- Badano L.P.,
- Mor-Avi V.,
- et al.
- Ponikowski P.,
- Voors A.A.,
- Anker S.D.,
- et al.
- Nagueh S.F.,
- Smiseth O.A.,
- Appleton C.P.,
- et al.
- Currie P.J.,
- Seward J.B.,
- Chan K.L.,
- et al.
- Keir G.J.,
- Wort S.J.,
- Kokosi M.,
- et al.
- O’Leary J.M.,
- Assad T.R.,
- Xu M.,
- et al.