| Objectives Upper gastrointestinal bleeding(UGIB)is one of the most common clinical emergencies.Some studies have shown that diabetes is a risk factor for the development of UGIB,which may be related to the fact that prolonged elevated blood-glucose can lead to neurovascular disease.Studies on the morbidity of UGIB in China are rare,while studies on the association between fasting blood glucose(FBG)levels and the risk of UGIB has not been reported.Therefore,this study used a prospective cohort study to analyze the morbidity of first-onset UGIB and to evaluate the association between FBG levels with the risk of developing first-onset UGIB in a Chinese population,and finally to further explore the effect of FBG on the first-onset UGIB by gender and age groups.Methods The research subjects consisted of 100003 in-service and retired employees of Kailuan Group who received a physical examination in 2006 or 2007,had no missing FBG data and no history of UGIB.Examinees were divided into three exposure groups based on FBG levels:’normoglycemia group’(non-exposeure group,FBG<6.1mmol/L),’prediabetes group’(low-exposure group,6.1mmol/L≤FBG<7.0mmol/L),and’diabetes group’(high-exposure group,FBG≥7.0mmol/L),the firs-onset of UGIB was followed up and recorded year by year,with final follow-up until December 31,2018.Analysis of the incidence density of first-onset UGIB in these three exposure groups,the Kaplan-Meier method was applied to analysis of survival curves and tested by the log-rank method.The Cox proportional risk regression model was used to analyze the effect of FBG level on the first-onset UGIB event.The study further subgroup analysis by gender and age to evaluate the effect of FBG levels on first-onset UGIB in specific populations.Finally,sensitivity analysis was performed.Results The mean age of 100003 subjects was(51.86±12.66)years,and after a mean follow-up of(10.57±1.71)years,we documented 942 first-onset UGIB cases(an incidence rate of 8.91 per 100000 person-years),each exposure group results showed 711 cases for’normoglycemia group’(8.35 per 100000 person-years),84 cases for’prediabetes group’(10.45 per 100000 person-years)and 147 cases for’diabetes group’(11.67 per100000 person-years).UGIB incidence curves among exposure groups were statistically significant(χ~2=14.84,P<0.01).Cox regression model analysis showed that the HR for the first-onset UGIB was 1.30(95%CI:1.02-1.66,P<0.05)and 1.35(95%CI:1.09-1.67,P<0.01)in the’prediabetes group’and’diabetes group’,compared with the’Normal glucose group’.Further subgroup analyses by sex and age showed that elevated FBG levels increased the risk of first-onset UGIB in the male population,and the risk of first-onset UGIB is also increased in people with diabetes aged≥60 years,while FBG levels were not significantly associated with the risk of first-onset UGIB in the female population and those aged<60years.Conclusions In this study,we found that not only the incidence of UGIB increased with the increase of FBG level,but also the risk of first UGIB showed an increasing trend.This study suggests that only prediabetes(FBG≥6.1 mmol/L)can lead to an increased risk of UGIB,providing a theoretical basis for the prevention and control of UGIB in Chinese residents.Figure 2;Table 9;Reference 121... |
PDF Full Text Request