Pulmonary Artery Pulsatility Index (PAPi)

Clinical background

Right ventricular dysfunction is a decisive prognostic factor in several acute cardiological situations: in inferior STEMI with right ventricular infarction, in cardiogenic shock, and as a complication after mechanical circulatory support or LVAD implantation. The difficulty is that right ventricular function is load-dependent and hard to judge from isolated pressure measurements. The Pulmonary Artery Pulsatility Index (PAPi) was developed precisely to extract, from standard pressures obtained at right heart catheterisation, a measure carried by pulsatility in the pulmonary circulation and which therefore reflects the stroke volume capacity of the right ventricle rather than the filling pressure itself. The decision the tool serves is whether the right ventricle has the reserve to tolerate additional load — for example an LVAD implantation — or whether the patient is at a stage where right ventricular failure is imminent and requires a specific strategy.

Calculating PAPi

PAPi is calculated as the pulmonary artery pulse pressure divided by the right atrial pressure:

$$\text{PAPi} = \frac{\text{PAP}$${\text{sys}} - \text{PAP}{\text{dias}}}{\text{Right atrial pressure}}

where $\text{PAP}${\text{sys}}PAPsys​ is the pulmonary artery systolic pressure, $\text{PAP}${\text{dias}} is the pulmonary artery diastolic pressure and the right atrial pressure corresponds to the central venous pressure. The formula rests on the reasoning that a right ventricle with good stroke volume capacity generates a large difference between systolic and diastolic pulmonary artery pressure, whereas a weak right ventricle produces a low pulse pressure while the right atrial pressure rises. A low value captures both phenomena: low stroke volume and high filling pressure.

The derivation cohort consisted of 20 patients with angiographically confirmed proximal right coronary artery occlusion and clinical suspicion of severe right ventricular dysfunction, together with two control groups of 50 patients without significant coronary artery disease and 14 patients with acute coronary syndrome without right ventricular involvement [1]. Measurements were made at cardiac catheterisation during the same admission. The result was clear-cut: patients with severe right ventricular dysfunction had a mean PAPi of 1.11 compared with 4.32 and 5.52 in the control groups [1]. With an ROC-derived threshold of PAPi ≤ 0.9, a c-statistic of 0.998 was achieved for predicting in-hospital mortality and/or the need for percutaneous right ventricular support [1].

Interpretation in practice

The central message is that PAPi is population-dependent. The threshold used to flag risk must be set according to the clinical context. The original threshold in inferior STEMI, PAPi < 1.85, identified severe right ventricular dysfunction in the derivation cohort [1]. In populations with cardiogenic shock and mechanical circulatory support, lower values, around 1.0, are typically applied.

A patient with a low PAPi in the acute phase should not be left without an active strategy for management of the right ventricle. Before planned LVAD implantation, a low preoperative PAPi is a reason to consider prophylactic temporary right ventricular support.

Validation and performance

PAPi has been externally validated in several populations, but performance varies with the cohort studied and generally concerns outcomes other than those in the derivation study.

A 2026 systematic review identified 12 studies with a total of 3,681 patients and found that a lower PAPi was consistently associated with worse outcomes across different heart failure populations [2]. The reported thresholds varied widely, however: from ≤ 1.9 in cardiogenic shock to ≤ 2.8–2.95 in broader heart failure populations and up to ≤ 3.65 in patients with advanced heart failure [2].

In pre-LVAD optimisation, a retrospective study from the Cleveland Clinic with 315 patients (2008–2017) showed that the optimised PAPi — that is, the best value achieved during intensive care optimisation before implantation — was independently associated with early right ventricular failure, with an odds ratio of 0.64 (95% CI 0.532–0.765, p < 0.0001) [3]. Patients who developed right ventricular failure had a mean optimised PAPi of 3.5 versus 7.5 in those who did not (p < 0.001) [3]. An initial PAPi below 2.0 was regarded as a high-risk marker in this population.

In a French multicentre registry study (ASSIST-ICD) of 117 LVAD patients (2007–2021), the preoperative PAPi predicted 3-month mortality with an ROC-derived threshold of 2.84 and an AUC of 0.68 (95% CI 0.57–0.80) [4]. Patients with a PAPi below 2.84 had 58.1% 3-month survival versus 89.1% for those with a PAPi ≥ 2.84 (HR 0.08, 95% CI 0.02–0.28, p < 0.01) [4]. The predictive value was, however, confined to the first three months; thereafter no significant difference remained.

In heart transplantation, a two-centre study of 173 patients examined PAPi measured immediately on arrival in intensive care and after 6 hours [5]. PAPi at T6 was the best single predictor of the need for mechanical circulatory support for early graft dysfunction, with an AUC of 0.832, and a threshold of 1.22 gave a sensitivity of 81% and a specificity of 65% [5].

Limitations

PAPi requires right heart catheterisation and is therefore not usable without invasive access. The primary problem, however, is that the threshold is population-dependent. A value that is normal in one heart failure cohort may be alarming in another. The original threshold in inferior STEMI cannot be extrapolated to LVAD or transplantation populations, and vice versa.

PAPi is influenced by changes in pulmonary vascular resistance and capacitance. In severe pulmonary hypertension, pulmonary capacitance falls and with it the pulmonary artery pulse pressure, which can lower PAPi independently of right ventricular contractility. This makes the value difficult to interpret in established pulmonary hypertension.

Because PAPi is load-dependent, it is affected by volume status and inotropy. A patient may have a low PAPi on arrival that improves markedly with optimisation, as shown in the LVAD study in which the change in PAPi during optimisation was the strongest predictor [3]. A single static measurement may miss patients with good reserve, and vice versa.

The studies are generally small and retrospective. The derivation cohort included only 20 patients with severe right ventricular dysfunction [1]. External validation in LVAD and transplantation populations is also based on relatively limited cohorts, and AUC values range from 0.68 to 0.83 [4, 5].

References

1. Korabathina R et al. The pulmonary artery pulsatility index identifies severe right ventricular dysfunction in acute inferior myocardial infarction. Catheter Cardiovasc Interv. 2012;80(4):593–600. PMID: 21954053
2. Albdour Z et al. The prognostic utility of the pulmonary artery pulsatility and aortic pulsatility index in patients with heart failure: A systematic review. JRSM Cardiovasc Dis. 2026;15. PMID: 42003950
3. Gonzalez MH et al. Dynamic assessment of pulmonary artery pulsatility index provides incremental risk assessment for early right ventricular failure after left ventricular assist device. J Card Fail. 2021;27(7):777–785. PMID: 33640481
4. Akamkam A et al. Association between pulmonary artery pulsatility and mortality after implantation of left ventricular assist device. ESC Heart Fail. 2024;11(4):2100–2112. PMID: 38581135
5. Yim IHW et al. Pulmonary artery pulsatility index predicts mechanical circulatory support following heart transplantation. JHLT Open. 2024;4:100030. PMID: 40145106