Abstract
BACKGROUND AND AIMS: The aim of this study is to investigate the association between visit-to-visit variability in fasting plasma glucose (FPG) and the risk of digestive cancers among individuals with and without diabetes. METHODS: Using data from Kailuan cohort, a prospective population-based study, individuals who had at least two measurements of FPG between 2006 and 2012 without prior cancer were included in this study. Four indexes of variability were used, including standard deviation (SD), coefficient of variation (CV), variability independent of the mean (VIM), and average successive variability (ARV). Cox regression was used to evaluate the relationship between the quartiles of FPG variability and digestive cancers. RESULTS: A total of 98,161 individuals were studied. Over a mean follow-up of 9.32 ± 0.81 years, 1103 individuals developed incident digestive cancer (1.21 per 1000 person-years). Compared to the individuals in the lowest quartile, those in the highest quartile of FPG variability by SD had 38.7% higher risk of developing overall digestive cancers after adjusting for the significant confounders (hazard ratio, 1.387; 95% confidence interval, 1.160-1.659; P = 0.0003). Higher FPG variability was associated with significantly higher risks of colorectal cancer (fully adjusted HR 1.432, 95% CI [1.073-1.912], P = 0.015) and pancreatic cancer (fully adjusted HR 2.105, 95% CI [1.024-4.329], P = 0.043), but not liver cancer (fully adjusted HR 1.427, 95% CI [0.973-2.092], P = 0.069) or esophageal and gastric cancer (fully adjusted HR 1.139, 95% CI [0.776-1.670], P = 0.506). Subgroup analyses showed that individuals who were younger (<65 years), male, and those without diabetes experienced a predominantly higher risk of developing digestive cancers. Similar results were observed when using CV, VIM, and ARV. CONCLUSIONS: FPG variability was significantly associated with increasing risk of digestive cancers, especially for pancreatic and colorectal cancer. Our study suggested a potential role of FPG variability in risk stratification of digestive cancers. Approaches that reduce FPG variability may lower the risks of incident digestive cancers among the general population. This trial is registered with ChiCTR-TNRC-11001489.