Authors
- Nina Jakuš — University of Zagreb School of Medicine, University Hospital Centre Zagreb, Zagreb, Croatia — ORCID: 0000-0001-7304-1127
- Ivo Planinc — University of Zagreb School of Medicine, University Hospital Centre Zagreb, Zagreb, Croatia — ORCID: 0000-0003-0561-6704
- Boško Skorić — University of Zagreb School of Medicine, University Hospital Centre Zagreb, Zagreb, Croatia — ORCID: 0000-0001-5979-2346
- Dora Fabijanović — University of Zagreb School of Medicine, University Hospital Centre Zagreb, Zagreb, Croatia — ORCID: 0000-0003-2633-3439
- Hrvoje Jurin — University of Zagreb School of Medicine, University Hospital Centre Zagreb, Zagreb, Croatia — ORCID: 0000-0002-2599-553X
- Jure Samardžić — University of Zagreb School of Medicine, University Hospital Centre Zagreb, Zagreb, Croatia — ORCID: 0000-0002-9346-6402
- Željko Baričević — University of Zagreb School of Medicine, University Hospital Centre Zagreb, Zagreb, Croatia — ORCID: 0000-0002-5420-2324
- Hrvoje Gašparović — University of Zagreb School of Medicine, University Hospital Centre Zagreb, Zagreb, Croatia — ORCID: 0000-0002-2492-3702
- Bojan Biočina — University of Zagreb School of Medicine, University Hospital Centre Zagreb, Zagreb, Croatia — ORCID: 0000-0003-3362-9596
- Davor Miličić — University of Zagreb School of Medicine, University Hospital Centre Zagreb, Zagreb, Croatia — ORCID: 0000-0001-9101-1570
- Maja Čikeš — University of Zagreb School of Medicine, University Hospital Centre Zagreb, Zagreb, Croatia — ORCID: 0000-0002-4772-5549
Keywords
left ventricular assist device, right ventricular failure, echocardiography
DOI
https://doi.org/10.15836/ccar2016.390Full Text
**Background:** Although right ventricular failure (RVF) is a common complication after left ventricular assist device (LVAD) implantation, its occurrence cannot be reliably predicted by an individual echocardiographic parameter or a simple combination of measurements. (1, 2) We aimed to identify easily obtainable measurements that may successfully predict RVF, combining markers of right atrial (RA) and right ventricular (RV) geometry with hemodynamic measurements. **Patients and Methods:** Preoperative echo examinations and right heart catheterization (RHC) data were studied for 48 consecutive patients (pts) (42 male, mean age 59.1±8.7 years), who underwent continuous flow LVAD implantation (80% HeartMate II/III, 20% HeartWare). A standard preoperative echo examination was performed, assessing RV and RA geometry and function (TAPSE (tricuspid annular plane systolic excursion), FAC (fractional area change)), while haemodynamic parameters were measured by RHC (**Table 1**). The Index of RV asymmetry (RVAi) was used as a novel marker of RV geometry, quantified as a ratio of the difference of the medial and lateral portion of RV area in relation to the total RV area at enddiastole (**Figure 1**). ### Table 1: Comparison of the ability of different measures of right heart morphology and function in predicting RVF after LVAD implantation. | | AUC | Sensitivity | Specificity | p-value | | --- | --- | --- | --- | --- | | RAVol/BSA+RVAi | 0.782 | 0.667 | 0.897 | 0.001 | | RAVol/BSA+RVAi+PVR | 0.722 | 0.444 | 1.000 | 0.001 | | RAVol/BSA+RVAi+TPG | 0.708 | 0.444 | 0.971 | 0.004 | | RAVol/BSA+RVAi+sPAP | 0.708 | 0.444 | 0.710 | 0.004 | | RAVol/BSA+RVAi+C.I. | 0.694 | 0.444 | 0.940 | 0.010 | | RAVol/BSA+RVAi+CVP | 0.637 | 0.333 | 0.941 | 0.054 | | TAPSE | 0.721 | 0.889 | 0.553 | 0.025 | | FAC | 0.384 | 0.444 | 0.324 | 0.258 | | TAPSE+FAC | 0.458 | 0.333 | 0.583 | 0.721 | [†] AUC - area under the curve; RAVol - right atrial volume; BSA - body surface area; RVAi - right volume area index; PVR - pulmonary vascular resistance; TPG - transpulmonary gradient; sPAP - systolic pulmonary artery pressure; C.I. - cardiac index; CVP - central venous pressure; TAPSE- tricuspid annular plane systolic excursion; FAC - fractional area change; P value for chi square test. Cut off values: RVAi– 0.33, RAVol/BSA - 50 mL/m2, TPG 12 mmHg, sPAP 70 mmHg, PVR 5 WU, C.I. 1.8 L/min/m2, CVP 12 mmHg, TAPSE 15 mm, FAC 20%. P value for chi square test. Figure 1. The Index of RV asymmetry (RVAi) is quantified as the ratio of the difference of the medial and lateral portion of RV area in relation to the total RV area at end-diastole. A more asymmetric RV will have a greater asymmetry factor, which is typically seen in preserved RV geometry. **Results:** 19% of pts presented with acute or chronic RVF following LVAD implantation. The RVAi was significantly higher (a more assymetric RV) in nonRVF patients: 0.40±0.13 vs 0.25±0.1, p<0.005. The combination of RAvol/BSA (right atrial volume indexed to BSA) and RVAi was confirmed as a superior predictor of RVF, while the addition of PVR maximized the specificity, yet impaired the sensitivity of the testing. The traditional markers of RV function, FAC and TAPSE, failed to successfully distinguish patients that would develop RVF (**Table 1**). **Conclusion:** Our study demonstrates that echocardiographic parameters of RV and RA geometry combined with RHC parameters of RV afterload may be simple yet reliable predictors of post LVAD RVF; however, limited patient numbers suggest further testing of this finding.
Literature
- Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS). Appendix A: Adverse event definitions: adult and pediatric patients (2013).
- Lampert BC, Teuteberg JJ. Right ventricular failure after left ventricular assist devices. J Heart Lung Transplant. 2015;34(9):1123–30. https://doi.org/10.1016/j.healun.2015.06.015