UPPER GASTROINTESTINAL BLEEDING: COMPARISON BETWEEN THE ROCKALL SCORE AND THE GLASGOW-BLATCHFORD SCORE

Aims Upper gastrointestinal bleeding (UGIB) is a common cause of hospital admission .Multiple scoring systems have been developed to stratify patients : including Rockall and Glasgow-blatchford.Many clinical studies have validated their predictive values ; yet, their clinical effectiveness and their cut-offs are subjects to debates

Methods This is a prospective,descriptive and analytical study conducted between April 2020 and November 2021.We included patients over 18 years of age who presented with upper gastrointestinal bleeding

Results 100 patients were included, mean age 60[40.5-72] .The sex ratio: 1.2 .56 . The prevalence of rebleeding was 37.6% of patients.The mean of the two scores was significantly higher in patients who rebleeded compared to those who did not : GBS (8.32±2.6 Vs 7,17±3,5 , p 0.005: ) ; RS (5.39±1.64 Vs3,48±1.23 p<0,001). In multivariate analysis, the two scores were predictive of re-bleeding : Glasgow score (OR=1.13 , p<0.001), RS (OR=1.7, p=0.003). The GBS – using a cutt off of 4 has an excellent sensitivity and NPV: (Sensitivity: 100% – Specificity: 9.38% – PPV:29.3% -NPV:100% ) ; In comparison with The RS : using a cut-off of 3 Sensitivity:83.33% – Specificity: 50% – PPV: 38.5% – VPN:88.9%

Conclusions Our study showed that , the two scores are predictive factors of re bleeding , despite a rather average diagnostic ability . We also underline that the GBS has an excellent and better NPV than the Rockall score. It would make outpatie nt care possible to those with a GB score of 0.

Publication History

Article published online:
14 April 2022