Clinical background
Pulmonary arterial hypertension (PAH) is a progressive disease in which treatment intensity should be guided by the patient's risk profile. The higher the risk, the more aggressive the therapy, and the goal is to reach or maintain low-risk status. The decision is difficult without a structured instrument because PAH is heterogeneous and individual variables, such as the 6-minute walk distance or NT-proBNP, do not capture the whole picture on their own. Moreover, studies show that clinicians' assessment of risk level differs from objective estimates in more than half of cases, with both overestimation and underestimation of risk [3].
The REVEAL Registry Risk Score 2.0 was developed to replace the original REVEAL calculator with improved thresholds and an expanded set of variables. The score is a weighted multiparameter tool intended to predict 1-year survival and thereby guide decisions about treatment escalation.
Calculating the REVEAL Registry Risk Score 2.0
The score is a weighted sum of ten physiological, functional and haemodynamic variables in this implementation. The full tool comprises twelve variables; those omitted here are PAH aetiology, right atrial pressure and hospitalisation within the past 6 months.
The variables and their points:
| Variable | Points |
|---|---|
| WHO/NYHA functional class I | |
| WHO/NYHA functional class II | |
| WHO/NYHA functional class III | |
| WHO/NYHA functional class IV | |
| Male sex AND age >60 years | (otherwise 0) |
| Systolic blood pressure <110 mmHg | (otherwise 0) |
| Heart rate >96 beats/min | (otherwise 0) |
| 6-minute walk distance | divided into bands, a shorter distance scoring more points |
| NT-proBNP or BNP | divided into bands, a higher value scoring more points |
| eGFR <60 mL/min/1.73 m² (or clinical renal insufficiency) | (otherwise 0) |
| DLCO <40% of predicted | (otherwise 0) |
| Pericardial effusion (echocardiography) | (otherwise 0) |
| PVR <5 Wood units (right heart catheterisation) | (otherwise 0) |
Note that a low PVR lowers the score, reflecting the fact that a low pulmonary vascular resistance is a favourable sign. In the same way, functional class I makes a negative contribution. NT-proBNP and BNP must not be entered at the same time; if both are available, NT-proBNP should be used.
The derivation cohort consisted of a subpopulation from the US REVEAL registry (Registry to Evaluate Early and Long-Term PAH Disease Management) who had survived at least 1 year after enrolment [1]. This subpopulation forms the baseline for the REVEAL 2.0 calculation, that is, the score is intended for patients already established in care rather than for newly diagnosed patients at their first right heart catheterisation. The outcome modelled was 12-month survival.
Interpretation in practice
The calculator grades the patient into three risk categories:
| Score | Risk category | Clinical action |
|---|---|---|
| Low risk | Maintain current treatment. Reassess regularly, but there is no indication for escalation. | |
| 7–8 | Intermediate risk | Consider treatment escalation. Assess whether the patient achieves low-risk targets with add-on therapy. |
| High risk | Treatment escalation indicated. Consider referral for transplant assessment if not already initiated. |
The aim of modern PAH treatment is to reach low-risk status, and the score should therefore be used iteratively at follow-up, not only on isolated occasions. A patient who remains at intermediate or high risk despite optimal oral dual or triple therapy should be considered for advanced treatment options including parenteral prostacyclin and transplant assessment.
Validation and performance
In the derivation cohort, REVEAL 2.0 showed good discrimination and separated the risk categories well. A systematic review of risk stratification models in PAH reported a c-statistic for REVEAL 2.0 in the range 0.65 to 0.74 across several studies [2]. By comparison, the c-statistic was 0.62 to 0.77 for COMPERA and 0.39 to 0.69 for the FPHR (non-invasive) in the same review [2].
The REVEAL calculators are the only risk stratification models in PAH that use weighted variables, in which variables with at least a doubled hazard ratio are assigned two points and the rest one point [2]. This may explain their generally higher discrimination compared with models based on the mean of unclassified variables, such as COMPERA, or on the number of low-risk criteria met, such as the FPHR.
External validation has been performed in the Australian and New Zealand PAH registry cohort (PHSANZ, n=1,011), in which REVEAL 2.0 showed robust separation of 12- and 60-month survival across the risk categories [4]. The simplified three-category model is preferred for clinical use over the full eight-level model, since survival estimates overlapped within some intermediate and high-risk groups in the full model [4].
In a comparative study of patients with functional class II PAH (n=119 with complete data for REVEAL 2.0), REVEAL 2.0 classified 58% as low risk, 27% as intermediate and 15% as high risk [3]. Clinicians' own assessment agreed with REVEAL 2.0 in only 43 to 54% of cases, and both overestimation and underestimation of risk occurred [3]. The commonest reason for discordance was that clinicians relied on the patient's symptomatic stability over time while objective parameters showed a different picture [3].
Limitations
This implementation covers ten of the twelve variables in the full REVEAL 2.0 tool. The missing variables are PAH aetiology, right atrial pressure and hospitalisation within the past 6 months. REVEAL 2.0 permits scoring with at least 7 of the 12 variables provided that functional class, natriuretic peptide and the 6-minute walk distance are included, but for definitive grading the official calculator should be used.
The score was derived from a US registry of patients who had survived at least 1 year after enrolment. It therefore reflects a prevalent cohort rather than newly diagnosed patients at baseline. Validation in incident patients has shown a somewhat lower c-statistic than in the prevalent cohort [4].
REVEAL 2.0 is not recommended in the 2022 ESC/ERS guidelines, which instead favour the four-level COMPERA 2.0 model for risk stratification at follow-up [2]. This does not reflect poorer performance by REVEAL 2.0, but rather that the ESC/ERS prefer a simpler, non-invasive model with fewer variables. REVEAL 2.0 requires both haemodynamic data from right heart catheterisation (PVR) and lung function (DLCO), which limits its usefulness outside specialist centres.
The score applies only to group 1 PAH. It is not validated for pulmonary hypertension due to left heart disease, lung disease or chronic thromboembolic pulmonary hypertension. Nor is it validated for paediatric PAH, where only two small studies have applied REVEAL 2.0, with limited discrimination [2].
References
- Benza RL et al. Predicting Survival in Patients With Pulmonary Arterial Hypertension: The REVEAL Risk Score Calculator 2.0 and Comparison With ESC/ERS-Based Risk Assessment Strategies. Chest 2019. PMID: 30772387
- Lokhorst C et al. Risk stratification in adult and pediatric pulmonary arterial hypertension: A systematic review. Front Cardiovasc Med 2022. PMID: 36440049
- Sahay S et al. Risk assessment in patients with functional class II pulmonary arterial hypertension: Comparison of physician gestalt with ESC/ERS and the REVEAL 2.0 risk score. PLoS One 2020. PMID: 33175857
- Anderson JJ et al. Retrospective Validation of the REVEAL 2.0 Risk Score With the Australian and New Zealand Pulmonary Hypertension Registry Cohort. Chest 2020. PMID: 31563497
