Aims:
Upper gastrointestinal bleeding (UGIB) is a common condition in the emergency room (ER), with high morbimortality. Nevertheless, there is no evidence about the impact of the implementation of an urgent intervention protocol (UGIB-CODE) for its management.
Our aims are: 1) Evaluate the impact of the UGIB-CODE implementation in patients with non-variceal upper bleeding (NV-UGIB) at the ER. 2) Identify the variables that are independently associated with the final success in patients with NV-UGIB. 3) Costs analysis.
Methods:
Observational cohort study including a retrospective cohort (RC) and a prospective cohort (PC), before and after the implementation of UGIB-CODE. We recruited adult patients attended at the ER with NV-UGIB during 2014 (RC) and 2016 (PC). Univariate and multivariate analysis were done to determine the impact of UGIB-CODE implementation and the variables associated with the final success (no mortality, no re-bleeding or re-bleeding controlled by endoscopy).
Results:
We included 66 patients in the RC (age 68 ± 1.87;30.3% women) and 89 patients in the PC (age 69 ± 1.65;42.7% women). The multivariate analysis showed a reduction of red blood cell concentrate (RBCC) administration (OR:1.840;IC95% 1.066 – 3.175;p = 0.028) and a decrease in hospital admission (OR:4.729;IC95% 1.306 – 17.114;p = 0.018). Regarding the final success no differences were found between the two cohorts (93.9% vs. 87.6%;p = 0.190). Blatchford Score < 12 (OR:4.460;IC95% 1.366 – 14.568;p = 0.013) and non-emergency endoscopy (> 6 hours) (OR:5.449;IC95% 1.133 – 26.209;p = 0.034) were independently associated with the final success. The implementation of UGIB-CODE saved 24,780 € per 100 patients related to RBCC administration and hospital admission.
Conclusions:
The implementation of the UGIB-CODE is a cost-effective strategy, decreasing RBCC administration and hospital admission. We reaffirm that Blatchford Score is a useful tool for predicting the evolution of NV-UGIB. The finding that an emergency endoscopy was associated with a worse outcome should be confirmed in a prospective study that particularly assess this issue.
