Prediction of right ventricular failure after ventricular assist device implant: systematic review and meta-analysis of observational studies

Diego Bellavia, Attilio Iacovoni, Cesare Scardulla, Lorenzo Moja, Michele Pilato, Sudhir S Kushwaha, Michele Senni, Francesco Clemenza, Valentina Agnese, Calogero Falletta, Giuseppe Romano, Joseph Maalouf, Michael Dandel

Research output: Contribution to journalReview article

33 Citations (Scopus)

Abstract

Right ventricular failure (RVF) after left ventricular assist device (LVAD) implantation is associated with increased morbidity and mortality, but the identification of LVAD candidates at risk for RVF remains challenging. We undertook a systematic review and meta-analysis of observational studies of risk factors associated with RVF after LVAD implant. Thirty-six studies published between 1 January 1995 and 30 April 2015, comprising 995 RVF patients out of a pooled final population of 4428 patients, were identified. Meta-analysed prevalence of post-LVAD RVF was 35%. A need for mechanical ventilation [odds ratio (OR) 2.99], or continuous renal replacement therapy (CRRT; OR 4.61, area under the curve 0.78, specificity 0.91) were the clinical variables with the highest effect size (ES) in predicting RVF. International normalized ratio [INR; standardized mean difference (SMD) 0.49] and N-terminal pro-brain natriuretic peptide (NT-proBNP) (SMD 0.52) were the biochemical markers that best discriminated between RVF and No-RVF populations, though NT-proBNP was highly heterogeneous. Right ventricular stroke work index (RVSWI) and central venous pressure (CVP) (SMD -0.58 and 0.47, respectively) were the haemodynamic measures with the highest ES in identifying patients at risk of post-LVAD RVF; CVP was particularly useful in risk stratifying patients undergoing continuous-flow LVAD implant (SMD 0.59, P < 0.001, I2 = 20.9%). Finally, pre-implant moderate to severe right ventricular (RV) dysfunction, as assessed qualitatively (OR 2.82), or a greater RV/LV diameter ratio (SMD 0.51) were the standard echocardiographic measurements with the highest ES in comparing RVF with No-RVF patients. Longitudinal systolic strain of the RV free wall had the highest ES (SMD 0.73) but also the greatest heterogeneity (I2 = 74%) and was thus only marginally significant (P = 0.05). Patients on ventilatory support or CRRT are at high risk for post-LVAD RVF, similarly to patients with slightly increased INR, high NT-proBNP or leukocytosis. High CVP, low RVSWI, an enlarged right ventricle with concomitant low RV strain also identify patients at higher risk.

Original languageEnglish
Pages (from-to)926-946
Number of pages21
JournalEuropean Journal of Heart Failure
Volume19
Issue number7
DOIs
Publication statusPublished - Jul 2017

Fingerprint

Heart-Assist Devices
Observational Studies
Meta-Analysis
Central Venous Pressure
International Normalized Ratio
Brain Natriuretic Peptide
Odds Ratio
Stroke
Right Ventricular Dysfunction
Renal Replacement Therapy
Leukocytosis
Artificial Respiration
Population
Area Under Curve
Heart Ventricles
Biomarkers
Hemodynamics
Morbidity
Mortality

Keywords

  • Journal Article
  • Review

Cite this

Prediction of right ventricular failure after ventricular assist device implant : systematic review and meta-analysis of observational studies. / Bellavia, Diego; Iacovoni, Attilio; Scardulla, Cesare; Moja, Lorenzo; Pilato, Michele; Kushwaha, Sudhir S; Senni, Michele; Clemenza, Francesco; Agnese, Valentina; Falletta, Calogero; Romano, Giuseppe; Maalouf, Joseph; Dandel, Michael.

In: European Journal of Heart Failure, Vol. 19, No. 7, 07.2017, p. 926-946.

Research output: Contribution to journalReview article

@article{6e95c111b18c4ffa8824bb2c0b725213,
title = "Prediction of right ventricular failure after ventricular assist device implant: systematic review and meta-analysis of observational studies",
abstract = "Right ventricular failure (RVF) after left ventricular assist device (LVAD) implantation is associated with increased morbidity and mortality, but the identification of LVAD candidates at risk for RVF remains challenging. We undertook a systematic review and meta-analysis of observational studies of risk factors associated with RVF after LVAD implant. Thirty-six studies published between 1 January 1995 and 30 April 2015, comprising 995 RVF patients out of a pooled final population of 4428 patients, were identified. Meta-analysed prevalence of post-LVAD RVF was 35{\%}. A need for mechanical ventilation [odds ratio (OR) 2.99], or continuous renal replacement therapy (CRRT; OR 4.61, area under the curve 0.78, specificity 0.91) were the clinical variables with the highest effect size (ES) in predicting RVF. International normalized ratio [INR; standardized mean difference (SMD) 0.49] and N-terminal pro-brain natriuretic peptide (NT-proBNP) (SMD 0.52) were the biochemical markers that best discriminated between RVF and No-RVF populations, though NT-proBNP was highly heterogeneous. Right ventricular stroke work index (RVSWI) and central venous pressure (CVP) (SMD -0.58 and 0.47, respectively) were the haemodynamic measures with the highest ES in identifying patients at risk of post-LVAD RVF; CVP was particularly useful in risk stratifying patients undergoing continuous-flow LVAD implant (SMD 0.59, P < 0.001, I2 = 20.9{\%}). Finally, pre-implant moderate to severe right ventricular (RV) dysfunction, as assessed qualitatively (OR 2.82), or a greater RV/LV diameter ratio (SMD 0.51) were the standard echocardiographic measurements with the highest ES in comparing RVF with No-RVF patients. Longitudinal systolic strain of the RV free wall had the highest ES (SMD 0.73) but also the greatest heterogeneity (I2 = 74{\%}) and was thus only marginally significant (P = 0.05). Patients on ventilatory support or CRRT are at high risk for post-LVAD RVF, similarly to patients with slightly increased INR, high NT-proBNP or leukocytosis. High CVP, low RVSWI, an enlarged right ventricle with concomitant low RV strain also identify patients at higher risk.",
keywords = "Journal Article, Review",
author = "Diego Bellavia and Attilio Iacovoni and Cesare Scardulla and Lorenzo Moja and Michele Pilato and Kushwaha, {Sudhir S} and Michele Senni and Francesco Clemenza and Valentina Agnese and Calogero Falletta and Giuseppe Romano and Joseph Maalouf and Michael Dandel",
note = "{\circledC} 2017 The Authors. European Journal of Heart Failure {\circledC} 2017 European Society of Cardiology.",
year = "2017",
month = "7",
doi = "10.1002/ejhf.733",
language = "English",
volume = "19",
pages = "926--946",
journal = "European Journal of Heart Failure",
issn = "1388-9842",
publisher = "John Wiley & Sons, Ltd",
number = "7",

}

TY - JOUR

T1 - Prediction of right ventricular failure after ventricular assist device implant

T2 - systematic review and meta-analysis of observational studies

AU - Bellavia, Diego

AU - Iacovoni, Attilio

AU - Scardulla, Cesare

AU - Moja, Lorenzo

AU - Pilato, Michele

AU - Kushwaha, Sudhir S

AU - Senni, Michele

AU - Clemenza, Francesco

AU - Agnese, Valentina

AU - Falletta, Calogero

AU - Romano, Giuseppe

AU - Maalouf, Joseph

AU - Dandel, Michael

N1 - © 2017 The Authors. European Journal of Heart Failure © 2017 European Society of Cardiology.

PY - 2017/7

Y1 - 2017/7

N2 - Right ventricular failure (RVF) after left ventricular assist device (LVAD) implantation is associated with increased morbidity and mortality, but the identification of LVAD candidates at risk for RVF remains challenging. We undertook a systematic review and meta-analysis of observational studies of risk factors associated with RVF after LVAD implant. Thirty-six studies published between 1 January 1995 and 30 April 2015, comprising 995 RVF patients out of a pooled final population of 4428 patients, were identified. Meta-analysed prevalence of post-LVAD RVF was 35%. A need for mechanical ventilation [odds ratio (OR) 2.99], or continuous renal replacement therapy (CRRT; OR 4.61, area under the curve 0.78, specificity 0.91) were the clinical variables with the highest effect size (ES) in predicting RVF. International normalized ratio [INR; standardized mean difference (SMD) 0.49] and N-terminal pro-brain natriuretic peptide (NT-proBNP) (SMD 0.52) were the biochemical markers that best discriminated between RVF and No-RVF populations, though NT-proBNP was highly heterogeneous. Right ventricular stroke work index (RVSWI) and central venous pressure (CVP) (SMD -0.58 and 0.47, respectively) were the haemodynamic measures with the highest ES in identifying patients at risk of post-LVAD RVF; CVP was particularly useful in risk stratifying patients undergoing continuous-flow LVAD implant (SMD 0.59, P < 0.001, I2 = 20.9%). Finally, pre-implant moderate to severe right ventricular (RV) dysfunction, as assessed qualitatively (OR 2.82), or a greater RV/LV diameter ratio (SMD 0.51) were the standard echocardiographic measurements with the highest ES in comparing RVF with No-RVF patients. Longitudinal systolic strain of the RV free wall had the highest ES (SMD 0.73) but also the greatest heterogeneity (I2 = 74%) and was thus only marginally significant (P = 0.05). Patients on ventilatory support or CRRT are at high risk for post-LVAD RVF, similarly to patients with slightly increased INR, high NT-proBNP or leukocytosis. High CVP, low RVSWI, an enlarged right ventricle with concomitant low RV strain also identify patients at higher risk.

AB - Right ventricular failure (RVF) after left ventricular assist device (LVAD) implantation is associated with increased morbidity and mortality, but the identification of LVAD candidates at risk for RVF remains challenging. We undertook a systematic review and meta-analysis of observational studies of risk factors associated with RVF after LVAD implant. Thirty-six studies published between 1 January 1995 and 30 April 2015, comprising 995 RVF patients out of a pooled final population of 4428 patients, were identified. Meta-analysed prevalence of post-LVAD RVF was 35%. A need for mechanical ventilation [odds ratio (OR) 2.99], or continuous renal replacement therapy (CRRT; OR 4.61, area under the curve 0.78, specificity 0.91) were the clinical variables with the highest effect size (ES) in predicting RVF. International normalized ratio [INR; standardized mean difference (SMD) 0.49] and N-terminal pro-brain natriuretic peptide (NT-proBNP) (SMD 0.52) were the biochemical markers that best discriminated between RVF and No-RVF populations, though NT-proBNP was highly heterogeneous. Right ventricular stroke work index (RVSWI) and central venous pressure (CVP) (SMD -0.58 and 0.47, respectively) were the haemodynamic measures with the highest ES in identifying patients at risk of post-LVAD RVF; CVP was particularly useful in risk stratifying patients undergoing continuous-flow LVAD implant (SMD 0.59, P < 0.001, I2 = 20.9%). Finally, pre-implant moderate to severe right ventricular (RV) dysfunction, as assessed qualitatively (OR 2.82), or a greater RV/LV diameter ratio (SMD 0.51) were the standard echocardiographic measurements with the highest ES in comparing RVF with No-RVF patients. Longitudinal systolic strain of the RV free wall had the highest ES (SMD 0.73) but also the greatest heterogeneity (I2 = 74%) and was thus only marginally significant (P = 0.05). Patients on ventilatory support or CRRT are at high risk for post-LVAD RVF, similarly to patients with slightly increased INR, high NT-proBNP or leukocytosis. High CVP, low RVSWI, an enlarged right ventricle with concomitant low RV strain also identify patients at higher risk.

KW - Journal Article

KW - Review

U2 - 10.1002/ejhf.733

DO - 10.1002/ejhf.733

M3 - Review article

C2 - 28371221

VL - 19

SP - 926

EP - 946

JO - European Journal of Heart Failure

JF - European Journal of Heart Failure

SN - 1388-9842

IS - 7

ER -