Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
doi:10.1002/ejhf.664
Biomedical and Surgical Sciences, Cardiology Section, University of Verona, Verona, Italy; 9 Department of Cardiology, Spedali Civili Hospital and University of Brescia, Italy;
10 Division of Cardiology, Fondazione Salvatore Maugeri, IRCCS, Montescano, Italy; 11 Cardiology and Emergency Department, San Antonio Hospital, San Daniele del Friuli, Italy;
and 12 Cardiac, Thoracic and Vascular Department, University of Pisa, Pisa, Italy
Aims To evaluate whether the clinical and echocardiographic correlates and the prognostic significance of right ventricular
(RV) dysfunction are different in heart failure patients with reduced (HFrEF), mid-range (HFmrEF), or preserved
(HFpEF) left ventricular ejection fraction.
.....................................................................................................................................................................
Methods and The study included 1663 patients with heart failure caused by ischaemic or hypertensive heart disease or by idiopathic
results cardiomyopathy. Left ventricular ejection fraction was <40% in 1123 patients (HFrEF), 4049% in 156 patients
(HFmrEF) and 50% in 384 patients (HFpEF). Imaging of the right ventricle was performed by echocardiography;
RV function was defined on the basis of tricuspid annular plane systolic excursion (TAPSE) and its normalization
for pulmonary artery systolic pressure (PASP). All-cause mortality was the endpoint of survival analysis. Non-sinus
rhythm, high heart rate, ischaemic aetiology and E-wave deceleration time <140 ms were associated with a reduced
TAPSE in HFrEF patients, whereas PASP >40 mmHg was by far the strongest correlate of a reduced TAPSE in HFpEF
and HFmrEF patients (interaction analysis, P = 0.0011). TAPSE/PASP proved to be a powerful predictor of prognosis
in all patients.
.....................................................................................................................................................................
Conclusions The correlates of RV dysfunction differ in HFrEF compared with HFpEF and HFmrEF patients. Regardless of the
extent of LV dysfunction, the TAPSE/PASP ratio is a powerful independent predictor of prognosis in all heart failure
patients.
..........................................................................................................
Keywords Heart failure Right ventricle Echocardiography Pulmonary hypertension
*Corresponding author. Tel: +39 0382 503460, Fax: +39 0382 501631, Email: s.ghio@smatteo.pv.it
On behalf of all investigators (see the Supplementary material online, Appendix S1).
........................................................................................................................................................................
life-threatening condition with adverse prognosis other than car-
Right ventricular (RV) systolic dysfunction is an important deter- diovascular disease; and any disease causing precapillary pulmonary
minant of symptoms and a powerful marker of poor prognosis in hypertension. Demographic, clinical, laboratory, and echocardio-
patients with chronic heart failure.1 8 Although data have been graphic data were collected in patients at study entry. The study was
collected primarily in patients with heart failure with reduced ejec- approved by the local institutional review boards and patients gave
written informed consent. The study endpoint was all-cause mortality;
tion fraction (HFrEF), more recently the interest of researchers
survival data were obtained through follow-up visits of patients or, in
has focused on patients heart failure with preserved ejection frac-
the case of missed visits, through telephone contact.
tion (HFpEF).9 12 In contrast, the issue of which factors contribute
to RV dysfunction in heart failure patients is still under scrutiny. It
seems reasonable to hypothesize that there is a spectrum of clin- Echocardiography
ical phenotypes evolving from one to the other, from isolated left A standard transthoracic two-dimensional and Doppler echocardio-
ventricular (LV) dysfunction with normal pulmonary pressures to graphic examination was carried out with commercial equipment.
progressively more advanced conditions where RV dysfunction is Parameters were categorized as follows a deceleration time of the
the key determinant of prognosis.13 However, it cannot be ruled E-wave (DT) <140 ms was used to define a restrictive LV filling
out that specific mediators may have a different clinical role in pattern in HFrEF patients; an E/e (at septal level) ratio 15 was
the different contexts of HFrEF vs. HFpEF. Recognition of the fac- used to define severe diastolic dysfunction in HFpEF and in HFmrEF
tors associated with RV dysfunction is meaningful to better deter- patients.17 Patients were categorized as having high PASP if the PASP
mine the pathophysiological metrics typical of both conditions, to estimate was >40 mmHg (PASP was obtained as the transtricuspid
improve prognostication, to target the specific mediators of RV pressure gradient + the estimate of right atrial pressure based on
inferior vena cava diameter and collapsibility). The cut-off for RV
dysfunction and, therefore, possibly, to develop more effective ther-
dysfunction was set at a TAPSE value of 14 mm vs. >14 mm as
apeutic strategies in heart failure patients.
this threshold is associated with a strong impact on prognosis in the
In the present multicentre study we aimed at defining the literature.3,8 TAPSE/PASP was categorized as <0.36 mm/mmHg vs.
clinical and echocardiographic correlates of RV dysfunction in a 0.36 mm/mmHg.7 The end-diastolic volume index was categorized
large population of heart failure patients with a broad range of at its median value. Mitral regurgitation was categorized into four
LV systolic function. Right ventricular function was defined by groups based on the evaluation of the jet area within the left atrium.
echocardiography taking tricuspid annular plane systolic excursion For patients in atrial fibrillation (AF), all measurements were repeated
(TAPSE) as the simplest and most robust indicator of RV dysfunc- at least three times and the average value was calculated.
tion. For prognostic purposes we calculated the ratio of TAPSE to
pulmonary artery systolic pressure (PASP), which has been pro- Statistical analysis
posed as a simplified indicator of the right ventricle to pulmonary
Data were described as mean and standard deviation (SD) if continuous
circulation coupling.7 The notion that load-independent indices of and as counts and per cent if categorical. The main echocardiographic
RV function may provide more efficient prognostic information in parameters were categorized as described above. According to sur-
heart failure than the conventional RV echocardiographic param- face electrocardiogram (ECG), patients were classified as sinus rhythm
eters is currently taking hold among researchers also in acute or AF/PM in case of AF/flutter or ventricular stimulation (PM). The
decompensated heart failure.14 remaining continuous variables were dichotomized at their median
value in the whole cohort. Correlation between continuous variables
was computed with the Pearson R coefficient. Univariable and multi-
Methods variable analysis was performed to identify the correlates of RV dys-
function. The variables tested include those hypothesized a priori to
Study patients cause or contribute to RV dysfunction, including age, sex, history of
The present study is a retrospective evaluation of a cohort of 1663 coronary artery disease, rhythm, heart rate, LVEF, DT (and E/e in
patients with chronic heart failure evaluated in 11 Italian Hospitals HFpEF and in HFmrEF patients), PASP. Odds ratios (OR) and their
between January 2004 and December 2014. Inclusion criteria were: 95% confidence intervals (CI) were computed by means of logistic
aetiology caused by coronary artery disease, or hypertensive heart models. This analysis was performed in the whole population and sepa-
disease, or idiopathic cardiomyopathy; stable clinical conditions over rately in the three subgroups with HFrEF, or HFmrEF, or HFpEF. Model
the last 3 months; and age > 18 years. Heart failure was defined discrimination was assessed by the area under the receiver operating
by cardiologist-adjudicated heart failure diagnosis according to the characteristic curve (AUC ROC) of the model together with its 95%
Framingham criteria.15 Patients were defined as having HFrEF if left CI. Centre heterogeneity was considered by computing HuberWhite
ventricular ejection fraction (LVEF) was <40%, heart failure with robust standard errors while clustering for centres.
mid-range ejection fraction (HFmrEF) if LVEF was 4049%, and Cumulative survival was calculated on the basis of KaplanMeier
HFpEF if LVEF was 50%, according to the 2016 European heart estimates; the end-point of survival analysis was all-cause death. Cox
failure guidelines.16 Coronary artery disease was diagnosed on the univariable and multivariable analysis was performed to identify the
basis of documented previous myocardial infarction or significant independent predictors of survival. Two analysis were performed: the
disease on coronary arteriography. Exclusion criteria were: myocar- first included the following variables: age, sex, New York Heart Associ-
dial infarction, or coronary artery bypass graft, or percutaneous ation (NYHA) class, coronary artery disease, systolic blood pressure,
coronary angioplasty in the previous 3 months; implantation of a rhythm, heart rate, mitral regurgitation, furosemide dose, therapy
cardiac resynchronization device in the previous 6 months; organic with beta-blockers, therapy with angiotensin-converting enzyme
HFrEF, heart failure with reduced ejection fraction; HFmrEF, heart failure with mid-range ejection fraction; HFpEF, heart failure with preserved ejection fraction; CAD, coronary
artery disease; SBP, systolic blood pressure; DBP, diastolic blood pressure; AF, atrial fibrillation; NYHA, New York Heart Association; BNP, brain natriuretic peptide; ACE,
angiotensin converting enzyme; ARB, angiotensin receptor blockers; ICD, implantable cardioverter defibrillator; LV EDVi, left ventricular end-diastolic volume index; DT,
E-wave deceleration time; TAPSE, tricuspid annular plane systolic excursion; PASP, pulmonary artery systolic pressure.
(ACE) inhibitors/angiotensin receptor antagonists, end-diastolic vol- and according to LVEF. About two-thirds of the population had
..........................................................
ume index, DT, TAPSE, and PASP. In the second analysis, TAPSE and HFrEF (n = 1123) and about one-third had either HFpEF or HFm-
PASP were replaced by the TAPSE/PASP ratio. The relative risk of rEF (n = 384 and n = 156, respectively). The echocardiographic
dying and its 95% CI were computed. As the number of patients in the estimate of PASP was feasible in 1411 (85%) patients and 17.2%
HFpEF and HFmrEF groups was substantially smaller than in the HFrEF of the patients had a PASP >40 mmHg. The TAPSE was measured
group, and the results of the above analysis showed that the correlates
in all patients, and 12.6% of the patients had a TAPSE 14 mm.
of RV dysfunction were similar in patients in the HFpEF and HFmrEF
In addition, DT was measured in 98% of patients and E/e was
groups, these two groups were pooled for survival analysis. Model
discrimination was assessed by the Harrells C-statistics. For all models,
measured in 89% of the HFpEF and HFmrEF patients.
the interaction with LVEF groups was tested with the likelihood ratio
test (the model with the interaction was compared with the model Correlates of right ventricular
without). Stata 14.1 (Stata Corporation, College Station, TX, USA)
was used for computation. A two-sided P-value <0.05 was considered
dysfunction
statistically significant. In the entire population LVEF <40% (OR 3.14, 95% CI 2.324.26,
P < 0.001), coronary artery disease (OR 1.69, 95% CI 1.182.44,
P = 0.005), AF/PM (OR 1.76, 95% CI 1.302.39, P < 0.001), heart
Results rate >70 b.p.m. (OR 1.52, 95% CI 1.152.00, P = 0.003), a restric-
tive mitral inflow pattern (OR 2.38, 95% CI 1.334.35, P = 0.003),
Clinical characteristics and PASP >40 mmHg (OR 2.26, 95% CI 1.144.46, P = 0.019)
Table 1 shows the main demographic, clinical and echocardio- showed an independent association with RV dysfunction. The
graphic characteristics of the entire population (1663 patients) independent predictors of TAPSE 14 mm in the HFrEF, in the
........................................................................................................................................................................
Table 2 Independent predictors of reduced tricuspid
the correlates of TAPSE 14 mm were the same as in the entire
annular plane systolic excursion (TAPSE) by left
population, with the exception of heart rate and PASP which
ventricular ejection fraction group
did not reach statistical significance (for heart rate >70 b.p.m.:
OR 1.30, 95% CI 0.921.85, P = 0.142; for PASP >40 mmHg: OR
OR 95% CI P-value
1.78, 95% CI 0.963.31, P = 0.067). In the HFpEF and HFmrEF ................................................................
groups, non-sinus rhythm, heart rate >70 b.p.m., end-diastolic vol- HFrEF (n = 1123)
ume greater than median value and PASP >40 mmHg were found Aetiology
to be independently associated with a TAPSE 14 mm; notably, in CAD vs. no CAD 2.29 1.533.45 <0.001
both groups PASP >40 mmHg carried an extremely high risk for Rhythm
RV dysfunction. A formal interaction analysis was performed and AF vs. SR 4.96 2.639.34 <0.001
PM vs. SR 2,79 1.904.09 <0.001
the results of this analysis confirmed that the phenotype of LV
DT (ms)
dysfunction (HFrEF, or HFpEF, or HFmrEF) modifies significantly
140 vs. >140 2.38 1.234.55 <0.001
(P = 0.0011) the relationship between TAPSE and PASP. Coronary
HFmrEF (n = 156)
artery disease was associated with a higher risk of RV dysfunction Aetiology
in HFrEF but not in HFpEF patients and was associated with a sig- CAD vs. no CAD 0.69 0.130.62 0.017
nificantly lower risk of RV dysfunction in HFmrEF patients (Table 2); Rhythm
interaction analysis was statistically significant (P = 0.0010). A DT AF vs. SR 4.96 2.639.34 <0.001
140 ms was associated with a higher risk of RV dysfunction in PASP (mmHg)
HFrEF but not in HFpEF or HFmrF patients (Table 2); interaction >40 vs. 40 32.9 14.375.7 <0.001
analysis was not statistically significant (P = 0.6126). As shown in HFpEF (n = 384)
Figure 1, the proportion of HFpEF and HFmrEF patients with a Rhythm
reduced TAPSE was negligible in the absence of high PASP, whereas AF vs. SR 3.20 2.514.07 <0.001
PM vs. SR 5.34 4.226.75 <0.001
a substantial proportion of HFrEF patients had a reduced TAPSE in
Heart rate (b.p.m.)
the presence of normal PASP.
>70 vs. 70 1.77 1.182.67 0.006
EDVi (mL/m2 )
Prognostic significance of right >113 vs. 113 1.37 1.061.79 0.017
PASP (mmHg)
ventricular dysfunction >40 vs. 40 5.50 1.0728.5 <0.001
Two hundred and fifty-eight patients died during a median follow-up
OR, odds ratio; CI, confidence interval; HFrEF, heart failure with reduced ejection
period of 56 months. The independent predictors of survival at fraction; CAD, coronary artery disease; AF, atrial fibrillation; SR, sinus rhythm;
multivariable analysis for the HFrEF group and for the HFpEF PM, pacemaker rhythm; DT, E-wave deceleration time; HFmrEF, heart failure with
and HFmrEF groups are shown in Table 3 using TAPSE and PASP mid-range ejection fraction; PASP, pulmonary artery systolic pressure; HFpEF,
heart failure with preserved ejection fraction; EDVi, end-diastolic volume index.
separately and in Table 4 using the TAPSE/PASP ratio. Owing to HFrEF: Discrimination [area under the curvereceiver operating characteristic
the strict relationship between TAPSE and PASP in patients with (AUC ROC)]: 0.76, 95% CI 0.720.79.
LVEF 40%, TAPSE was not an independent predictor of prognosis HFmrEF: Discrimination (AUC ROC): 0.94, 95% CI 0.830.98.
HFpEF: Discrimination (AUC ROC): 0.79, 95% CI 0.690.89.
in such patients when PASP was entered first in the multivariable
model; the opposite was also true, as PASP was not an independent
predictor of prognosis in such patients when TAPSE was entered
first in the multivariable model. The TAPSE/PASP ratio allowed
Correlates of right ventricular
us obtain good values of Harrell C-coefficient in both groups. dysfunction
Figure 2 shows survival for HFrEF patients and for HFmrEF and The presence of an elevated pressure level in the pulmonary cir-
HFpEF patients according to the TAPSE/PASP ratio. There was no culation has traditionally been considered a plausible cause for RV
interaction between LVEF groups and results of survival analysis dysfunction in heart failure patients. This hypothesis has a sound
(P = 0.8818). pathophysiological background.18,19 However, in a substantial pro-
portion of HFrEF patients, RV function may be reduced despite nor-
mal pulmonary artery pressures. Information on the determinants
Discussion of RV dysfunction is poorer in HFpEF.12 To our knowledge, the
The main finding of the present study is the demonstration that the present study is the first specifically planned to evaluate the clini-
clinical and echocardiographic correlates of a dysfunctioning right cal and echocardiographic correlates of RV dysfunction in patients
ventricle differ in heart failure patients according to the extent of having HFrEF, or HFmrEF, or HFpEF. The first parameter corre-
LV systolic dysfunction. In particular, pulmonary hypertension is by lated with RV dysfunction is LV systolic function itself: having a LVEF
far the strongest correlate in HFpEF and HFmrEF patients whereas <40% was associated with a more than threefold increased risk of
it is not statistically significant in HFrEF. Dysfunction of the right having a reduced TAPSE. Pure physiological considerations could
ventricle is a powerful predictor of poor prognosis regardless of explain this result; the right and the left ventricles share the same
the extent of LV systolic dysfunction. visceral cavity (the pericardium), have common myofibres, and
........................................................................................................................................................................
Table 4 Independent predictors of survival using the
tricuspid annular plane systolic excursion pulmonary
artery systolic pressure (TAPSE/PASP) ratio
LVEF, left ventricular ejection fraction; HR, hazard ratio; CI, confidence interval; NYHA, New
York Heart Association; SBP, systolic blood pressure.
LVEF <40%, Harrells C 0.71.
LVEF 40%, Harrells C 0.75.
for the necessity of discovering effective treatments for pulmonary is a simple non-invasive estimate that allows the application of
........................................................................................................................................................................
hypertension in such patients.23 this concept in clinical practice. Distinguishing between isolated
The observation that AF and permanent RV pacing are strongly postcapillary and combined precapillary and postcapillary pul-
related to RV dysfunction in HFrEF, HFmrEF, and HFpEF patients monary hypertension could be important as it could be related
is in agreement with previous suggestions that the mechanical to the development of RV dysfunction and failure; however, such
function of the atria and ventricles are closely coupled on the right distinction is not feasibly accurate using echocardiography. Finally,
side of the heart and highlights the relevance of RV dyssynchrony, the duration of the disease and the duration of pulmonary hyper-
which is an issue that is largely underestimated.9,24 tension could both be important determinants of RV dysfunction
but both are extremely difficult to assess in clinical practice.
9. Mohammed SF, Hussain I, AbouEzzeddine OF, Takahama H, Kwon SH, Forfia 17. Nagueh SF, Appleton CP, Gillebert TC, Marino PN, Oh JK, Smiseth OA,
.........................................................................
P, Roger VL, Redfield MM. Right ventricular function in heart failure with Waggoner AD, Flachskampf FA, Pellikka PA, Evangelisa A. Recommendations for
preserved ejection fraction: a community-based study. Circulation 2014;130: the evaluation of left ventricular diastolic function by echocardiography. Eur J
23102320. Echocardiogr 2009;10:165193.
10. Burke MA, Katz DH, Beussink L, Selvaraj S, Gupta DK, Fox J, Chakrabarti S, 18. Haddad F, Doyle R, Murphy DJ, Hunt SA. Right ventricular function in cardiovas-
Sauer AJ, Rich JD, Freed BH, Shah SJ. Prognostic importance of pathophysiologic cular disease, part II: pathophysiology, clinical importance, and management of
markers in patients with heart failure and preserved ejection fraction. Circ Heart right ventricular failure. Circulation 2008;117:17171731.
Fail 2014;7:288299. 19. Vatta M, Stetson SJ, Jimenez S, Entman ML, Noon GP, Bowles NE, Towbin JA,
11. Melenovsky V, Hwang SJ, Lin G, Redfield MM, Borlaug BA. Right heart dysfunction Torre-Amione G. Molecular normalization of dystrophin in the failing left and
in heart failure with preserved ejection fraction. Eur Heart J 2014;35:34523462. right ventricle of patients treated with either pulsatile or continuous flow-type
12. Aschauer S, Kammerlander AA, Zotter-Tufaro C, Ristl R, Pfaffenberger S, ventricular assist devices. J Am Coll Cardiol 2004;43:811817.
Bachmann A, Duca F, Marzluf BA, Bonderman D, Mascherbauer J. The right 20. Damiano RJ Jr, La Follette P Jr, Cox JL, Lowe JE, Santamore WP. Significant
heart in heart failure with preserved ejection fraction: insights from cardiac left ventricular contribution to right ventricular systolic function. Am J Physiol
magnetic resonance imaging and invasive haemodynamics. Eur J Heart Fail 2016;18: 1991;261:H1514H1524.
7180. 21. Hoffman D, Sisto D, Frater RW, Nikolic SD. Left-to-right ventricular interac-
13. Vachiry JL, Adir Y, Barber JA, Champion H, Coghlan JG, Cottin V, De Marco tion with a noncontracting right ventricle. J Thorac Cardiovasc Surg 1994;107:
T, Gali N, Ghio S, Gibbs JS, Martinez F, Semigran M, Simonneau G, Wells A, 14961502.
Seeger W. Pulmonary hypertension due to left heart diseases. J Am Coll Cardiol 22. Lam CS, Roger VL, Rodeheffer RJ, Borlaug BA, Enders FT, Redfield MM.
2013; 62(25 Suppl):D100D108. Pulmonary hypertension in heart failure with preserved ejection fraction: a
14. Frea S, Pidello S, Bovolo V, Iacovino C, Franco E, Pinneri F, Galluzzo A, Volpe A, community-based study. J Am Coll Cardiol 2009;53:11191126.
Visconti M, Peirone A, Morello M, Bergerone S, Gaita F. Prognostic incremental 23. Guazzi M. Pulmonary hypertension in heart failure preserved ejection fraction:
role of right ventricular function in acute decompensation of advanced chronic prevalence, pathophysiology, and clinical perspectives. Circ Heart Fail 2014;7:
heart failure. Eur J Heart Fail 2016;18:564572. 367377.
15. McKee PA, Castelli WP, McNamara PM, Kannel WB. The natural history of con- 24. Bazaz R, Edelman K, Gulyasy B, Lopez-Candales A. Evidence of robust coupling
gestive heart failure: the Framingham study. N Engl J Med 1971;285:14411446. of atrioventricular mechanical function of the right side of the heart: insights from
16. Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JG, Coats AJ, Falk V, M-mode analysis of annular motion. Echocardiography 2008;25:557561.
Gonzlez-Juanatey JR, Harjola VP, Jankowska EA, Jessup M, Linde C, Nihoy- 25. Ghio S, Perlini S, Palladini G, Marsan NA, Faggiano G, Vezzoli M, Klersy C, Cam-
annopoulos P, Parissis JT, Pieske B, Riley JP, Rosano GM, Ruilope LM, Rus- pana C, Merlini G, Tavazzi L. Importance of the echocardiographic evaluation of
chitzka F, Rutten FH, van der Meer P. 2016 ESC Guidelines for the diagnosis right ventricular function in patients with AL amyloidosis. Eur J Heart Fail 2007;9:
and treatment of acute and chronic heart failure: the Task Force for the Diagno- 808813.
sis and Treatment of Acute and Chronic Heart Failure of the European Society of 26. Sunagawa K, Maughan WL, Sagawa K. Optimal arterial resistance for the max-
Cardiology (ESC). Developed with the special contribution of the Heart Failure imal stroke work studied in isolated canine left ventricle. Circ Res 1985;56:
Association (HFA) of the ESC. Eur J Heart Fail 2016;18:891975. 586595.