BACKGROUND: After left ventricular assist device (LVAD) implantation, post-operative renal failure is a fairly high risk. While comparing the differences in indicators of renal function associated with the use of three different types of LVAD, this retrospective study was designed to evaluate predictors and outcomes of acute renal failure after LVAD insertion. MATERIAL/ METHODS: Data for 28 patients who had received left ventricular assist device (LVAD) implantation at Osaka University Hospital from January 2002 through February 2007 were retrospectively analyzed. We correlated preoperative creatinine levels with the number of days needed for the creatinine level to drop below 1.5 mg/dL. Using the preoperative serum creatinine level as a univariate risk predictor, receiver operating characteristic (ROC) curves were analyzed to determine the best cutoff points for predicting ICU stays more than 30 days. Patients were characterized according to the type of LVAD pump implanted, their hemodynamic data, and other laboratory test data that were collected and analyzed. RESULTS: In our population, the type of LVAD was not related to the number of days needed for creatinine to drop to 1.5 mg/dL. Preoperative serum creatinine, however, did correlate with prolonged high postoperative creatinine levels: ROC analysis shows that a preoperative creatinine level of more than 1.95 mg/dL predicts an ICU stay more than 30 days. Ten patients implanted with Toyobo LVADs had preoperative creatinine levels above 1.5 mg/dL. The Pearson's correlation of their body weight and their final creatinine level when transferred from the ICU was 0.758 (P=0.011). CONCLUSIONS: Preoperative renal function predicts post-implantation renal function and length of ICU stay. Especially when implanted in heavier patients, the Toyobo LVAD tends to provide poorer renal perfusion.
BACKGROUND: After left ventricular assist device (LVAD) implantation, post-operative renal failure is a fairly high risk. While comparing the differences in indicators of renal function associated with the use of three different types of LVAD, this retrospective study was designed to evaluate predictors and outcomes of acute renal failure after LVAD insertion. MATERIAL/ METHODS: Data for 28 patients who had received left ventricular assist device (LVAD) implantation at Osaka University Hospital from January 2002 through February 2007 were retrospectively analyzed. We correlated preoperative creatinine levels with the number of days needed for the creatinine level to drop below 1.5 mg/dL. Using the preoperative serum creatinine level as a univariate risk predictor, receiver operating characteristic (ROC) curves were analyzed to determine the best cutoff points for predicting ICU stays more than 30 days. Patients were characterized according to the type of LVAD pump implanted, their hemodynamic data, and other laboratory test data that were collected and analyzed. RESULTS: In our population, the type of LVAD was not related to the number of days needed for creatinine to drop to 1.5 mg/dL. Preoperative serum creatinine, however, did correlate with prolonged high postoperative creatinine levels: ROC analysis shows that a preoperative creatinine level of more than 1.95 mg/dL predicts an ICU stay more than 30 days. Ten patients implanted with Toyobo LVADs had preoperative creatinine levels above 1.5 mg/dL. The Pearson's correlation of their body weight and their final creatinine level when transferred from the ICU was 0.758 (P=0.011). CONCLUSIONS: Preoperative renal function predicts post-implantation renal function and length of ICU stay. Especially when implanted in heavier patients, the Toyobo LVAD tends to provide poorer renal perfusion.
Authors: Brittney H Davis; Amelia K Boehme; Salpy V Pamboukian; Michael Allon; James F George; Chrisly Dillon; James K Kirklin; Jose Tallaj; Emily B Levitan; Russell Griffin; Gerald McGwin; T Mark Beasley; Nita A Limdi Journal: ASAIO J Date: 2020-03 Impact factor: 3.826