Estimating the Timeframe of Lactate Reduction in Acute Cardiovascular Disease Using Functional Data Analysis Based on Nonlinear Mixed Effects Models

Abstract

In patients with acute cardiovascular disease, treatment aimed at reducing lactate levels is crucial for im-proving prognosis. Trends in blood lactate levels for each specific cardiovascular disease can provide an accu-rate evaluation of the patient’s condition. We used functional data analysis with nonlinear mixed effects models to estimate the timeframe of lactate reduction in 3 cardiovascular diseases (acute heart failure, aortic dissection, and ischemic heart disease) by analyzing lactate trends. Among 1,816 patients admitted to the intensive care unit (ICU) or intensive cardiovascular care unit (ICCU) of St. Luke’s International Hospital for cardiology or cardiovascular surgery from December 31, 2010, to June 31, 2020, 1,249 adults with a diagnosis of acute heart failure (39%), aortic dissection (24%), or ischemic heart disease (37%) were included in the present study. Using functional data analysis with nonlinear mixed effects models, our study estimated the timeframe of lactate reduction based on blood lactate level trends. Lactate reduction took 30 hours (95% CI, 23-37 hours) in patients with acute heart failure, 40 hours (95% CI, 33-47 hours) in patients with aortic dissection, and 95 hours (95% CI, 49-failed estimate) in patients with ischemic heart disease. We were able to estimate the timeframe of lactate reduction with different cardiovascular diseases. Recog-nizing the differences in lactate reduction may be useful in developing treatment plans tailored to each disease.