Introduction
Upper gastrointestinal bleeding (UGIB) is a common medical emergency, with peptic ulcer bleeding (PUB) accounting for up to 59% of cases. Effective risk assessment is essential for patient stratification, guiding clinical decisions on endoscopy timing, level of care, and discharge planning. Although several scoring systems exist for UGIB, most are designed for general use rather than specifically for PUB.
Aims & Methods
This study aims to evaluate the accuracy of six pre-endoscopic scoring systems—Glasgow-Blatchford Score (GBS), Modified GBS (MGBS), Rockall Score (RS), AIMS65, H3B2 and MAP score—in predicting intervention needs, clinical outcomes, and level of care in patients admitted with PUB. A retrospective cohort study was conducted between 2020 and 2023, including patients admitted with PUB at a tertiary center. Pre-endoscopic scores were assessed, and data on the hemostatic treatment, level of care, mortality, and rebleeding rates were collected and analyzed.
Results
A total of 364 patients (66.8% male; mean age 71.7±17 years; median Charlson Comorbidity Index (CCI): 5.2) were admitted. Most were hospitalized at level 1 (68.1%) and level 2 (30.2%) care units, with 75.3% requiring emergency endoscopic hemostasis. The best scores for predicting endoscopic treatment were GBS, MGBS, and H3B2 (AUROC 0.93, 0.91, and 0.87, respectively; p<0.001). For rebleeding risk, GBS and MGBS performed best (AUROC 0.70 and 0.68). However, overall predictive value for 1-year mortality was low, with RS performing best (AUROC 0.67), closely followed by the CCI (AUROC 0.66). GBS, H3B2, and MAP were the most effective for differentiating levels of care (p<0.001)
| Score | Need for endoscopic treatment*
| Rebleeding Risk*
| 1-Year Mortality
| Differentiation of Care LevelΔ
|
|---|
Glasgow-Blatchford Score (GBS)
| 0.930 (0.899 – 0.961) | 0.704 (0.596 – 0.812)
| 0.518 (0.446 – 0.590)
| p < 0.001 |
Modified GBS (MGBS)
| 0.907 (0.876 – 0.942)
| 0.682 (0.563 – 0.801)
| 0.557 (0.485 – 0.629)
| p < 0.001
|
Rockall Score (RS)
| 0.609 (0.546 – 0.673)
| 0.642 (0.539 – 0.744)
| 0.670 (0.607 – 0.733)
| p = 0.410
|
AIMS65
| 0.630 (0.569 – 0.692)
| 0.643 (0.533 – 0.744)
| 0.626 (0.561 – 0.690)
| p = 0.051 |
H3B2
| 0.872 (0.832 – 0.913)
| 0.615 (0.515 – 0.715)
| 0.517 (0.446 – 0.588) | p < 0.001
|
MAP score
| 0.721 (0.656 – 0.785)
| 0.557 (0.444 – 0.669)
| 0.587 (0.516 – 0.659)
| p < 0.001 |
Table 1 – Performance of Pre-Endoscopic scores in Peptic Ulcer Bleeding
*Area under ROC curve - 95% confidence interval; Δ Chi-square test value
Conclusion
GBS and H3B2 demonstrated the highest accuracy in predicting the need for endoscopic treatment and rebleeding risk in PUB. However, while these scores are valuable for acute management, long-term outcomes may require additional prognostic factors in these patients.
References
1. Sasaki Y, Abe T, Kawamura N, et al. Prediction of the need for emergency endoscopic treatment for upper gastrointestinal bleeding and new score model: a retrospective study. BMC Gastroenterol. 2022;22(1):337. 2022 doi:10.1186/s12876-022-02413-8
2. Akhila Arya PV, Thulaseedharan NK, Raj R, et al. AIMS65, Glasgow-Blatchford bleeding score and modified Glasgow-Blatchford bleeding score in predicting outcomes of upper gastrointestinal bleeding: An accuracy and calibration study. Indian J Gastroenterol. 2023;42(4):496-504. doi:10.1007/s12664-023-01387-z
3. Tham J, Stanley A. Clinical utility of pre-endoscopy risk scores in upper gastrointestinal bleeding. Expert Rev Gastroenterol Hepatol. 2019;13(12):1161-1167. doi:10.1080/17474124.2019.1698292