Table 1:
Summary of the 16 studies included in the full-text review
| Authors | N | LVAD | PAPi associated with RHF? | PAPi cut-off for RHF or death | Summary |
|---|---|---|---|---|---|
| Grandin et al. [12] | 151 | Not reported | No | Not reported | Single-centre study. Low PAC combined with a high CVP: PCWP ratio was the strongest predictor for death at 6 months (HR 8.68, P < 0.001) and RHF (OR 4.74, P = 0.02). PAPI was not significantly associated with RHF (P = 0.10) |
| Morine et al. [13] | 132 | HM2, HVAD | Yes | <1.85 (RHF) | A lower PAPi was significantly associated with RHF P < 0.01. A PAPi < 1.85 provided 94% sensitivity and 81% specificity for predicting RHF and was superior to RAP: PCWP ratio, RVSWI and RAP alone. |
| Kang et al. [14] | 83 | HM2, HVAD | Yes | <2 (RHF) | PAPi was an independent predictor of RHF and RVAD implantation following LVAD therapy. A higher PAPi was associated with reduced risk for RVAD placement (OR 0.31, P < 0.0001). PAPi was more predictive of RVAD placement if inotropes were present at the time of catheterization (OR 0.21 vs 0.49). ROC analysis showed optimal sensitivity and specificity achieved using a PAPi threshold of 2. |
| Nitta et al. [15] | 70 | Nipro-VAD | Yes | <0.88 (RHF) | This study aimed to devise a scoring system for predicting RVAD placement following implantation of the Nipro-VAD paracoporeal device. Patients who required RVAD implantation postoperatively had a significantly lower PAPi (P = 0.001). The authors proposed a combination score using PVR > 4.5WU and RAP: PCWP > 0.8 as a scoring system for predicting RVAD requirement following paracoporeal LVAD therapy. |
| Loforte et al. [16] | 258 | HM2, HM3, HVAD, Jarvik 2000, Berlin Heart | Yes | <2 (RHF) | This study aimed to devise the ALMA risk score for predicting RHF following LVAD implantation. Within the haemodynamic data of this study, a PAPi of <2 was found to be associated with unplanned RVAD support (P = 0.001) and on multivariable logistic regression analysis PAPi < 2 had an OR of 3.3 (CI 1.7–6.1, P = 0.001). The ALMA score employs the following 5 variables: destination therapy intention, PAPI < 2, RVSWi <300 mmHg/ml/m2, RV:LV ratio > 0.75 and MELD-XI score >17. The authors proposed a score of 0–1 implies low risk for RVAD requirement and a score above 4 implies very high risk for requiring RVAD following LVAD implantation. |
| Raymer et al. [17] | 216 | HM2, HVAD | Yes | Not reported | This study reported the combination of TAPSE and HeartMate risk score (HMRS) as a scoring system to predict RHF following LVAD implantation. The RHF group had a lower PAPi (P = 0.001). ROC analysis showed PAPi had an AUC of 0.63 (P < 0.001). When the haemodynamic parameters were analysed combination of TAPSE with the HMRS was the best for predicting RHF compared to HMRS + PAPi and HMRS + sRVCPI. |
| Gudejko et al. [18] | 85 | HM2, HVAD | Yes | Not reported | The data used in this study were intraoperative haemodynamic parameters and also echocardiographic data. Higher CVP, lower pre-CPB and post-chest closure PAPi, post-CPB larger right atrial diameter, larger RVES area, lower FAC and lower TAPSE were all associated with severe RHF. |
| Muslem et al. [19] | 375 | HM2, HM3, HVAD | Yes | Not Reported | PAE was found to be the most robust haemodynamic parameter to predict RHF. PAPi was significantly lower in the Severe RHF group compared to no RHF (1.8 vs 2.2, P = 0.017). |
| Alfirevic et al. [20] | 86 | HM2, HM3, HVAD | No | Not reported | Intraoperative TAPSE measurement was the variable of interest in this study. As a secondary comparison, PAPi and the Michigan risk score were not significantly associated with severe RHF. Intraoperative TAPSE, Michigan risk score and PAPi were poor discriminators of RHF following LVAD therapy. |
| Sert et al. [21] | 71 | HM2, HM3, HVAD | No | Not reported | This study compared TAPSE, CVP: PCWP ratio, RVSWI, PAPi, Pennsylvania Score, Michigan score, CRITT score, ALMA score and the EUROMACS score. PAPi was not significantly lower in the group with postop RHF (P = 0.304). They concluded that only the EUROMACS and CRITT score had an ROC AUC above 0.7 and that the combination of TAPSE and the Pennsylvania score was found to be the most sensitive (85%) whereas TAPSE + Michigan score + CVP:PCWP ratio was the most specific (97%). |
| Benjamin et al. [22] | 104 | HM2, HVAD | No | Not reported | Primary end point was duration of inotropic support and the association with RVF. They found patients who were on long-term milrinone had a significantly increased risk of developing RHF post-LVAD insertion. PAPi was not significantly different between the RHF and the group without RVF post-LVAD implant (4 ± 3.9 vs 3.2 ± 2.3, P = 0.255) |
| Ruiz-Cano et al. [23] | 80 | HM3, HVAD | No | Not reported | PAPi was not significantly different between the early RHF group versus no early RHF (2.5 vs 3, P = 0.283). This study found that blood urea nitrogen >44.5 mg/dl and CVP/PCWP > 0.55 were the parameters with the strongest association with early RHF. |
| Guglin and Omar [24] | 18608 | Not reported | No data | Not reported | This was a retrospective cohort study looking at data from the INTERMACS database and primarily assessing RAP and its ability to predict death. RAP was the main predictor of mortality in LVAD recipients. PAPi was lower in non-survivors (P < 0.001), but RAP had superior discriminatory value with a difference in AUC of 0.0105 (P = 0.0052). |
| Gonzalez et al. [25] | 315 | HM2, HVAD | Yes | Optimal PAPi <3.3 (RHF); Delta PAPi <2.08 (Death at 6 months) | This study assessed the change in PAPi during preoperative haemodynamic optimization prior to LVAD implantation. The mean optimal PAPi was lower (P < 0.001) in the group that developed early RHF. A delta PAPI of <2.08 during optimization was associated with higher mortality at 180 days (P = 0.003). |
| Cacioli et al. [26] | 75 | HM2, HM3 | Yes | Not reported | A lower PAPi was strongly associated with RHF following LVAD implant. This study also demonstrated in those who did not develop RHF post-LVAD had a significantly higher PAPi following vasodilator challenge at preop RHC (5.3 ± 3.9 vs 2.7 ± 1.3, P = 0.003). Furthermore, PAPi when combined with established risk scores provided incremental risk stratification for post-LVAD RHF. |
| Stricagnoli et al. [27] | 38 | HM3, Jarvik 2000 | Yes | Not reported | PAPi was the most robust haemodynamic parameter which predicted post LVAD RHF (1.52 ± 0.26 vs 3.95 ± 3.39, P = 0.003) with an ROC AUC of 0.85. |
ALMA: Antonio Loforte and Motalto Andrea; CVP: central venous pressure; CRITT: central venous pressure>15, severe right ventricular dysfunction, intubation, tricuspid regurgitation, tachycardia; EUROMACS: European Registry for Patients with Mechanical Circulatory Support; FAC: fractional area change; HR: heart rate; HVAD: HeartWare; INTERMACS: Interagency Registry of Mechanically Assisted Circulatory Support; LVAD: left ventricular assist device; MELD: Model for End-stage Liver Disease; PAC: pulmonary arterial compliance; PAPi: Pulmonary artery pulsatility index; PCWP: pulmonary capillary wedge pressure; RAP: right atrial pressure; RHF: right heart failure; ROC: receiver operating characteristic; RVAD: right ventricular assist device; RVES: right ventricular end systolic; RVSWI: right ventricular stroke work index; sRVCPI: simplified right ventricular contraction pressure index; TAPSE: tricuspid annular plane systolic excursion.