EUROMACS-RHF Score for right heart failure after LVAD

Clinical background

Right heart failure is one of the commonest and most serious complications after implantation of a left ventricular assist device (LVAD). In the derivation cohort, 21.7% of patients developed early right heart failure within 30 days, and these patients had markedly worse survival: 1-year survival was 53% compared with 71% in those who did not develop right heart failure [1]. The decision to implant an isolated LVAD rests on a judgement that the right ventricle will cope with the increased preload that arises when the left ventricle is unloaded. If the right ventricle fails, right-sided mechanical support, prolonged inotropic treatment or pulmonary vasodilators are required, which dramatically increases morbidity and mortality. Without a structured risk assessment the decision becomes a subjective clinical guess, particularly in patients with borderline right ventricular function.

The EUROMACS-RHF score was developed to give a quantitative estimate of the risk of precisely this complication, and to outperform the earlier, more weakly performing models in the field [1].

Calculating the EUROMACS-RHF score

The score is calculated as the sum of five preoperative variables:

$$\text{EUROMACS-RHF} = 2 \times \mathbb{1}$${\text{RA/PCWP} > 0{.}54} + 1 \times \mathbb{1}{\text{Hb} \leq 10} + 2{.}5 \times \mathbb{1}{\text{multiple IV inotropes}} + 2 \times \mathbb{1}{\text{INTERMACS 1--3}} + 2 \times \mathbb{1}_{\text{severe RV dysfunction}}

where $\mathbb{1}$ is an indicator function taking the value 1 if the variable is present and 0 if absent. The score ranges from 0 to 9.5.

The variables are:

• RA/PCWP ratio > 0.54: the ratio of right atrial pressure to pulmonary capillary wedge pressure, a measure of right ventricular load relative to the left.
• Haemoglobin ≤ 10 g/dL: a marker of chronic heart failure and cardiac anaemia.
• Multiple IV inotropes: a need for more than one intravenous inotrope preoperatively, reflecting the severity of the heart failure.
• INTERMACS class 1–3: patients in critically to moderately decompensated condition according to the Interagency Registry for Mechanically Assisted Circulatory Support.
• Severe right ventricular dysfunction on echocardiography: a semiquantitative assessment of right ventricular function.

The derivation cohort consisted of 2,988 adult patients from the EUROMACS registry (European Registry for Patients with Mechanical Circulatory Support) who underwent implantation of a continuous-flow LVAD of an established model. The cohort was randomly split into a derivation group (n = 2,000) and an internal validation group (n = 988). The primary outcome was early (<30 days) severe postoperative right heart failure, defined as a need for right-sided mechanical circulatory support, continuous inotropic treatment for ≥14 days or inhaled nitric oxide for ≥48 hours [1].

Interpretation in practice

The score is divided into three risk bands by the calculator:

The risk percentages are taken from the derivation cohort and should be read as approximate [1]. In practice the risk is also influenced by intraoperative factors such as right ventricular ischaemic time, the duration of cardiopulmonary bypass and fluid management, which cannot be captured preoperatively [2].

A high score correlates not only with right heart failure but also with mortality. In the derivation cohort, patients with right heart failure had a 1-year survival of 53% compared with 71% in those without, and the median stay in the intensive care unit was 24 days compared with 7 days [1].

Validation and performance

In the derivation cohort the score achieved a c-statistic of 0.70, and in the internal validation cohort 0.67. It thus performed better than previously published models and than individual echocardiographic and haemodynamic markers [1].

The external validation is, however, less encouraging. In an analysis from the University of Minnesota of 254 patients with a continuous-flow LVAD (2007–2017), the AUC was only 58% (95% CI 52–66%) for the score as a whole. When the outcome was restricted to the more rigorous criteria of a need for an RVAD or prolonged inotropic treatment, the AUC rose to 67% (95% CI 54–79%) [2]. Two of the five variables in the score, a haemoglobin ≤10 g/dL and an RA/PCWP ratio >0.54, were not predictive of right heart failure in this cohort. Instead, markers of long-standing heart failure such as a higher creatinine, a lower albumin and a higher total bilirubin were more strongly associated with the outcome [2].

The differences between the cohorts are worth noting. In the Minnesota cohort, 49% were destination therapy compared with 14% in the EUROMACS cohort, and 65% of the patients classified as having right heart failure met the criterion solely through prolonged use of pulmonary vasodilators, compared with 1% in EUROMACS. This reflects large differences in clinical practice between centres and makes direct comparisons of incidence figures difficult [2].

A recent review from 2025 notes that the EUROMACS-RHF score is one of several published models, and that all of them have variable to weak performance on external validation. The review points out that the pulmonary artery pulsatility index (PAPi) and right ventricular longitudinal strain could potentially add predictive value, and that machine learning methods may shift the field [3].

Limitations

The score was derived exclusively from patients who received a continuous-flow LVAD of an established model. It has not been validated for pulsatile pumps or, to a sufficient extent, for newer centrifugal pumps, and caution is required when extrapolating to such populations.

The definition of right heart failure varies between centres, which directly affects how well the score performs. In the EUROMACS study, prolonged use of pulmonary vasodilators was included in the definition, but the proportion of patients meeting the criterion solely through such treatment differed greatly in external cohorts [2]. The incidence figures in the risk bands should therefore be interpreted with caution.

Two of the variables, haemoglobin and the RA/PCWP ratio, lost their predictive power in the largest external validation [2]. This may be explained in part by the fact that haemodynamic parameters change rapidly during preoperative optimisation with diuretics and inotropes, and that a single measurement at one point in time does not necessarily reflect the actual risk profile at implantation.

The score does not capture intraoperative events. Right ventricular ischaemia during hypotension, a prolonged bypass time, the surgical placement of the inflow cannula and the volume load of transfusion can all "unmask" a right ventricular function that was not predictable from preoperative variables [2].

Finally, the derivation cohort was predominantly European with a lower proportion of destination therapy patients than in many North American cohorts. Patients with more comorbidities, who are commoner in destination therapy populations, may have a different risk profile that the score does not fully capture [2].

References

1. Soliman OII, Akin S, Muslem R, et al. Derivation and Validation of a Novel Right-Sided Heart Failure Model After Implantation of Continuous Flow Left Ventricular Assist Devices: The EUROMACS Right-Sided Heart Failure Risk Score. Circulation. 2018;137(9):891–906. PMID: 28847897
2. Shah H, Murray T, Schultz J, et al. External assessment of the EUROMACS right-sided heart failure risk score. Scientific Reports. 2021;11:16064. PMID: 34373475
3. Vogel F, Sollie ZW, Kilic A, Kung E. Prediction of Right Heart Failure in LVAD Candidates: Current Approaches and Future Directions. Journal of Cardiovascular Development and Disease. 2025;12(7):240