Back

Risk of incident cardiovascular disease among patients with gastrointestinal disorder: prospective cohort study of 340,862 individuals

Chen, J.; Sun, Y.; Fu, T.; Lu, S.; Shi, W.; Zhao, J.; Li, S.; Li, X.; Yuan, S.; Larsson, S. C.

2023-04-09 gastroenterology
10.1101/2023.04.09.23288329 medRxiv
Show abstract

Background and AimsThe associations between gastrointestinal diseases (GIs) and cardiovascular disease (CVD) were unclear. We conducted a prospective cohort study to explore their associations. MethodsThis study included 340,862 individuals without baseline CVD from the UK Biobank cohort. Individuals with and without GIs were followed up until the ascertainment of incident CVDs, including coronary heart disease (CHD), cerebrovascular disease (CeVD), and peripheral artery disease (PAD). The diagnosis of diseases was confirmed with combination of the nationwide inpatient data, primary care data, and cancer registries. A multivariable Cox proportional hazard regression model was used to estimate the associations between GIs and the risk of incident CVD. ResultsDuring a median follow-up of 12.4 years, 28,787 incident CVD cases were diagnosed. Individuals with GIs had an elevated risk of CVD (hazard ratio 1.38; 95% confidence interval 1.35-1.42, P<0.001). Eleven out of fifteen GIs were associated with an increased risk of CVD after Bonferroni-correction, including cirrhosis, non-alcoholic fatty liver disease, gastritis and duodenitis, irritable bowel syndrome, gastroesophageal reflux disease, peptic ulcer, celiac disease, pancreatitis, diverticular disease, biliary disease, and appendicitis. The associations were stronger among women, individuals aged [&le;] 60 years, and those with body mass index [&ge;] 25 kg/m2. ConclusionsThis large-scale prospective cohort study revealed the associations of GIs with an increased risk of incident CVD, in particular CHD and PAD. These findings support the reinforced secondary CVD prevention among patients with gastrointestinal disorders.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.