BACKGROUND: Right ventricular (RV) failure is a significant complication after implantation of a left ventricular assist device (LVAD). It is therefore important to identify patients at risk a priori. However, prognostic models derived from multivariate analyses have had limited predictive power. METHODS: This study retrospectively analyzed the records of 183 LVAD recipients between May 1996 and October 2009; of these, 27 later required a RVAD (RVAD(+)) and 156 remained on LVAD only (RVAD(-)) until transplant or death. A decision tree model was constructed to represent combinatorial non-linear relationships of the pre-operative data that are predictive of the need for RVAD support. RESULTS: An optimal set of 8 pre-operative variables were identified: transpulmonary gradient, age, right atrial pressure, international normalized ratio, heart rate, white blood cell count, alanine aminotransferase, and the number of inotropic agents. The resultant decision tree, which consisted of 28 branches and 15 leaves, identified RVAD(+) patients with 85% sensitivity, RVAD(-) patients with 83% specificity, and exhibited an area under the receiver operating characteristic curve of 0.87. CONCLUSIONS: The decision tree model developed in this study exhibited several advantages compared with existing risk scores. Quantitatively, it provided improved prognosis of RV support by encoding the non-linear, synergic interactions among pre-operative variables. Because of its intuitive structure, it more closely mimics clinical reasoning and therefore can be more readily interpreted. Further development with additional multicenter, longitudinal data may provide a valuable prognostic tool for triage of LVAD therapy and, potentially, improve outcomes.
BACKGROUND: Right ventricular (RV) failure is a significant complication after implantation of a left ventricular assist device (LVAD). It is therefore important to identify patients at risk a priori. However, prognostic models derived from multivariate analyses have had limited predictive power. METHODS: This study retrospectively analyzed the records of 183 LVAD recipients between May 1996 and October 2009; of these, 27 later required a RVAD (RVAD(+)) and 156 remained on LVAD only (RVAD(-)) until transplant or death. A decision tree model was constructed to represent combinatorial non-linear relationships of the pre-operative data that are predictive of the need for RVAD support. RESULTS: An optimal set of 8 pre-operative variables were identified: transpulmonary gradient, age, right atrial pressure, international normalized ratio, heart rate, white blood cell count, alanine aminotransferase, and the number of inotropic agents. The resultant decision tree, which consisted of 28 branches and 15 leaves, identified RVAD(+) patients with 85% sensitivity, RVAD(-) patients with 83% specificity, and exhibited an area under the receiver operating characteristic curve of 0.87. CONCLUSIONS: The decision tree model developed in this study exhibited several advantages compared with existing risk scores. Quantitatively, it provided improved prognosis of RV support by encoding the non-linear, synergic interactions among pre-operative variables. Because of its intuitive structure, it more closely mimics clinical reasoning and therefore can be more readily interpreted. Further development with additional multicenter, longitudinal data may provide a valuable prognostic tool for triage of LVAD therapy and, potentially, improve outcomes.
Authors: Marion Verduijn; Peter M J Rosseel; Niels Peek; Evert de Jonge; Bas A J M de Mol Journal: J Biomed Inform Date: 2007-07-26 Impact factor: 6.317
Authors: D J Farrar; J D Hill; D G Pennington; L R McBride; W L Holman; R L Kormos; D Esmore; L A Gray; P E Seifert; G P Schoettle; C H Moore; P J Hendry; J N Bhayana Journal: J Thorac Cardiovasc Surg Date: 1997-01 Impact factor: 5.209
Authors: J Raymond Fitzpatrick; John R Frederick; Vivian M Hsu; Elliott D Kozin; Mary Lou O'Hara; Elan Howell; Deborah Dougherty; Ryan C McCormick; Carine A Laporte; Jeffrey E Cohen; Kevin W Southerland; Jessica L Howard; Mariell L Jessup; Rohinton J Morris; Michael A Acker; Y Joseph Woo Journal: J Heart Lung Transplant Date: 2008-12 Impact factor: 10.247
Authors: R L Kormos; T A Gasior; A Kawai; S M Pham; S Murali; B G Hattler; B P Griffith Journal: J Thorac Cardiovasc Surg Date: 1996-04 Impact factor: 5.209
Authors: Sanjiv J Shah; Daniel H Katz; Senthil Selvaraj; Michael A Burke; Clyde W Yancy; Mihai Gheorghiade; Robert O Bonow; Chiang-Ching Huang; Rahul C Deo Journal: Circulation Date: 2014-11-14 Impact factor: 29.690
Authors: Anthony E Peters; LaVone A Smith; Priscilla Ababio; Khadijah Breathett; Timothy L McMurry; Jamie L W Kennedy; Mohammad Abuannadi; James Bergin; Sula Mazimba Journal: J Card Fail Date: 2019-02-18 Impact factor: 5.712
Authors: Natasha A Loghmanpour; Manreet K Kanwar; Marek J Druzdzel; Raymond L Benza; Srinivas Murali; James F Antaki Journal: ASAIO J Date: 2015 May-Jun Impact factor: 2.872
Authors: Natasha A Loghmanpour; Robert L Kormos; Manreet K Kanwar; Jeffrey J Teuteberg; Srinivas Murali; James F Antaki Journal: JACC Heart Fail Date: 2016-06-08 Impact factor: 12.035
Authors: Michael S Kiernan; E Wilson Grandin; Marshall Brinkley; Navin K Kapur; Duc Thinh Pham; Robin Ruthazer; J Eduardo Rame; Pavan Atluri; Edo Y Birati; Guilherme H Oliveira; Francis D Pagani; James K Kirklin; David Naftel; Robert L Kormos; Jeffrey J Teuteberg; David DeNofrio Journal: Circ Heart Fail Date: 2017-10 Impact factor: 8.790
Authors: Salim Hayek; Daniel B Sims; David W Markham; Javed Butler; Andreas P Kalogeropoulos Journal: Circ Cardiovasc Imaging Date: 2014-03 Impact factor: 7.792