Sarah R Lieber1, Michael L Volk. 1. University of Michigan Medical School, M4101 Medical Sciences I, SPC 5624, 1301 Catherine Road, Ann Arbor, MI 48109, USA. srlieber@med.umich.edu
Abstract
BACKGROUND: Non-adherence to medical therapy after liver transplantation is confounded by different methods of measurement. AIMS: (1) To compare the performance of three different methods of measuring non-adherence: (a) biochemical (standard deviation [SD] tacrolimus levels), (b) clinician report, (c) self-report. (2) To identify pre-transplant predictors of post-transplant non-adherence. (3) To evaluate whether SD tacrolimus is an accurate predictor of graft outcomes. METHODS: In this retrospective cohort study, charts of adult recipients of a liver transplant 2003-2009 (sample A, n = 444) were reviewed to determine pre-transplant predictors of non-adherence and clinician report of non-adherence. SD tacrolimus levels were measured between 6 and 18 months post-transplant. A subset of sample A (n = 122) completed a survey on non-adherence. The three methods were compared using linear and logistic regression. Multivariable analysis was used to investigate pre-transplant predictors of non-adherence. In sample B (transplant recipients 1995-2003, n = 544) Cox regression was used to determine the relationship between SD immunosuppressant level and graft failure. RESULTS: Non-adherence was found in 22-62 % of subjects, with the highest rates indicated by self-report. Clinician report of non-adherence was associated with both self-report and SD tacrolimus. On multivariable analysis, unemployment at time of listing and chart evidence of pre-transplant non-adherence were significant predictors of higher SD of tacrolimus. History of substance abuse and pre-transplant chart evidence of non-adherence were also significant independent predictors of post-transplant chart evidence of non-adherence. Drug variability in the immediate post-transplant setting was independently associated with graft failure over time (hazard ratio 1.005 per unit increase in standard deviation, p = 0.04). CONCLUSIONS: Non-adherence among liver transplant recipients is a common problem associated with increased risk of graft failure. SD tacrolimus can be used to measure non-adherent behavior and perhaps target patients for behavioral interventions.
BACKGROUND: Non-adherence to medical therapy after liver transplantation is confounded by different methods of measurement. AIMS: (1) To compare the performance of three different methods of measuring non-adherence: (a) biochemical (standard deviation [SD] tacrolimus levels), (b) clinician report, (c) self-report. (2) To identify pre-transplant predictors of post-transplant non-adherence. (3) To evaluate whether SD tacrolimus is an accurate predictor of graft outcomes. METHODS: In this retrospective cohort study, charts of adult recipients of a liver transplant 2003-2009 (sample A, n = 444) were reviewed to determine pre-transplant predictors of non-adherence and clinician report of non-adherence. SD tacrolimus levels were measured between 6 and 18 months post-transplant. A subset of sample A (n = 122) completed a survey on non-adherence. The three methods were compared using linear and logistic regression. Multivariable analysis was used to investigate pre-transplant predictors of non-adherence. In sample B (transplant recipients 1995-2003, n = 544) Cox regression was used to determine the relationship between SD immunosuppressant level and graft failure. RESULTS: Non-adherence was found in 22-62 % of subjects, with the highest rates indicated by self-report. Clinician report of non-adherence was associated with both self-report and SD tacrolimus. On multivariable analysis, unemployment at time of listing and chart evidence of pre-transplant non-adherence were significant predictors of higher SD of tacrolimus. History of substance abuse and pre-transplant chart evidence of non-adherence were also significant independent predictors of post-transplant chart evidence of non-adherence. Drug variability in the immediate post-transplant setting was independently associated with graft failure over time (hazard ratio 1.005 per unit increase in standard deviation, p = 0.04). CONCLUSIONS: Non-adherence among liver transplant recipients is a common problem associated with increased risk of graft failure. SD tacrolimus can be used to measure non-adherent behavior and perhaps target patients for behavioral interventions.
Authors: Gerda Drent; Elizabeth B Haagsma; Sabina De Geest; Aad P van den Berg; Els M Ten Vergert; Hillegonda J van den Bosch; Maarten J H Slooff; Jan H Kleibeuker Journal: Transpl Int Date: 2005-08 Impact factor: 3.782
Authors: Ronan E O'Carroll; Lesley M McGregor; Vivien Swanson; George Masterton; Peter C Hayes Journal: Liver Transpl Date: 2006-12 Impact factor: 5.799
Authors: Marina Serper; Rachel E Patzer; Peter P Reese; Kamila Przytula; Rachel Koval; Daniela P Ladner; Josh Levitsky; Michael M Abecassis; Michael S Wolf Journal: Liver Transpl Date: 2015-01 Impact factor: 5.799
Authors: E Shemesh; J C Bucuvalas; R Anand; G V Mazariegos; E M Alonso; R S Venick; M Reyes-Mugica; R A Annunziato; B L Shneider Journal: Am J Transplant Date: 2017-04-22 Impact factor: 8.086
Authors: Emily A Leven; Rachel Annunziato; Jacqueline Helcer; Sarah R Lieber; Christopher S Knight; Catherine Wlodarkiewicz; Rainier P Soriano; Sander S Florman; Thomas D Schiano; Eyal Shemesh Journal: Clin Transplant Date: 2017-05-03 Impact factor: 2.863
Authors: Anthony W Castleberry; Muath Bishawi; Mathias Worni; Loretta Erhunmwunsee; Paul J Speicher; Asishana A Osho; Laurie D Snyder; Matthew G Hartwig Journal: Ann Thorac Surg Date: 2016-09-10 Impact factor: 4.330
Authors: Rachel E Patzer; Marina Serper; Peter P Reese; Kamila Przytula; Rachel Koval; Daniela P Ladner; Josh M Levitsky; Michael M Abecassis; Michael S Wolf Journal: Clin Transplant Date: 2016-08-29 Impact factor: 2.863