A-072 The Impact of Short and Prolonged Fasting on Routine Lab Tests

Abstract Background Fasting is commonly recommended before blood tests. While the effects of fasting on blood glucose and lipid profiles have been extensively studied, its impact on other laboratory tests remains unclear. Some analytes may be influenced by short fasting intervals, whereas others change only after prolonged fasting (beyond 8 or 12 hours). This study aimed to investigate the effect of fasting duration on various clinical laboratory test results. Methods This retrospective study analyzed 33,414,109 test results from 181,919 outpatients between February 2021 and December 2023. Starting in February 2021, outpatients were routinely asked about the time of their last meal before blood collection, with fasting intervals electronically recorded in 30-minute increments up to 7.5 hours. For fasting durations exceeding 8 hours, data were categorized as "more than 8 hours" or "more than 12 hours" based on patient reports. Of the 360 test items with available fasting records, 74 were included after excluding those with fewer than 1,000 results. Pearson correlation and adjusted linear regression assessed associations between fasting time and test results, calculating p-values and regression coefficients. A generalized linear model (GLM) and estimated marginal means (EMMEAN) were used to compare results between three fasting groups: non-fasted, >8 hours, and >12 hours. Age and department were included as covariates. Statistical significance was set at p < 0.001. Changes exceeding the total allowable error (TAE) were considered clinically meaningful. Results Regression analysis identified 15 analytes with significant p-values. However, only glucose demonstrated a ß1 coefficient exceeding the TAE, indicating a clinically meaningful change with fasting duration. In the GLM and EMMEAN analyses comparing the three fasting groups, 37 analytes showed statistically significant differences. Of these, six—alanine aminotransferase (ALT), glucose, triglycerides, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, and gamma-glutamyl transferase (GGT)—exceeded TAE thresholds. Notably, while lipid profile components (cholesterol, HDL, LDL, triglycerides) were unaffected by short fasting intervals in the regression analysis, they displayed meaningful differences with prolonged fasting in the GLM analysis, suggesting that extended fasting influences these parameters. Conclusion Fasting duration significantly impacts the results of specific clinical laboratory tests. Glucose levels were uniquely sensitive to both short and extended fasting periods. Lipid profile components and select liver enzymes (ALT, GGT) exhibited clinically meaningful differences only after prolonged fasting. These findings highlight the importance of standardized fasting protocols for accurate interpretation of certain laboratory tests.