The relationship between serum uric acid level and gastrointestinal bleeding frequency in patients using new generation oral anticoagulants

Introduction and Aim: The primary goal of using non-vitamin K antagonist oral anticoagulants is to prevent the development of stroke. The most common complication of this treatment is bleeding. Gastrointestinal system bleeding is the most common type of bleeding and bleeding predictive scoring systems have been developed. However, none of these scoring systems are specific and sensitive in detecting bleeding risk. Uric acid is the end product of purine metabolism and it has been shown in many studies that elevated uric acid levels impair endothelial functions. The effect of uric acid level on bleeding has not been investigated in this patient group. We investigated whether the uric acid levels (measured at baseline and at follow-up) have an effect on predicting gastrointestinal system bleeding. Material and Method: The study group consisted of 501 patients aged 18-80 years who were using non-vitamin K antagonist oral anticoagulants, who applied to the Cardiology Polyclinic of Balıkesir University Medical Faculty Hospital. The patients were screened retrospectively according to the history of previous gastrointestinal bleeding. The serum uric acid levels of the patients which were studied during their previous outpatient clinic admission were evaluated. In addition urea, creatinine, electrolytes, albumin, hemogram and lipid profile values were also examined if it was studied in the previous admissions of the patients. In the control group, patients who were treated with direct oral anticoagulants and who did not have previous gastrointestinal system bleeding were evaluated. The data were evaluated with the Statistical Package for Social Science (SPSS) 21.0 package program. The study was carried out between 2022-2023. Results: The study group consists of 501 people aged 18-80. Of the participants, 284 (56.7%) were female and 217 (43.3%) were male. The youngest age is 36, the oldest age is 79, and the mean age is 69.59±8.04. There were 68 (13.6%) patients who had gastrointestinal system bleeding. 132 patients (26.3%) use rivaroxaban, 186 patients (37.1%) use edoxaban, 143 patients (28.5%) use apixaban and 40 patients use (8.0%) dabigatran. The lowest uric acid value detected in the participants was 2.60, the highest uric acid value was 15.40, with a mean of 6.51 ± 2.31, and a median of 6.51. The uric acid reference interval normal value is <5.6, and 271 participants (54.1%) have high uric acid levels. The mean of uric acid value was 7.87 mg/dl in gastrointestinal bleeding group and 6.30 mg/dl in control group. The difference between the mean uric acid levels is statistically significant (p<0.001). The Uric acid/Albumin ratio was 2.27 in patients with GIS bleeding and 1.67 in those without GIS bleeding, and the difference between UA/Alb rates of 0.6 (CI: -0.77 and -0.41) was statistically significant (p<0.001) Conclusions: In our study, we found that high uric acid levels were predictive of gastrointestinal system bleeding in patients under non-vitamin K antagonist oral anticoagulant therapy. Considering the uric acid levels in these patients, we believe that closer follow-up of the patients who are thought to be at high risk of bleeding is appropriate.