Risk assessment in patients with gastrointestinal bleeding

Upper gastrointestinal (GI) bleeding is one of the most common medical problems needing hospital admission. To provide adequate medical care and appropriate medical resource utilization, several clinical and endoscopic scoring systems have been developed, implemented, and internally and externally validated. The initial evidence suggests that these scoring systems can improve clinical decision-making.

Hospital admissions for GI bleeding in the United States account for up to 10% of all hospitalizations and cost 3,180-8,990 US dollars per admission. 1,2The severity of GI bleeding affects mortality and hospital cost. 3Furthermore, elderly patients will have worse outcomes and higher mortality rates since they often have multiple comorbidities. 4To provide the best medical care and to use resources optimally in some hospital settings, several researchers developed tools for risk stratification by identifying and distinguishing between high-risk patients and low risk patients.The goal is to anticipate the risk of rebleeding within 30 days, hospital mortality, urgent endoscopic treatment, and the need for ICU admission.Also, these tools could provide prognostic information regarding the length of hospital stay. 5The literature suggests that the mortality of patients with non-variceal GI bleeding admitted to a GI service is lower than those admitted to an internal medicine service. 6herefore, the evaluation of severity of GI bleeding is important to identify who will receive the most benefit from hospitalization for close monitoring and aggressive resuscitation.

Gi bleeDinG risk assessment tools
Clinical evaluation, including the presence or absence of postural symptoms and the assessment of hemodynamic status by monitoring vital signs, urine output, level of consciousness, and estimated blood loss, is very important when managing patients with upper GI bleeding.In addition, clinical presentation, age, and comorbidities are important factors.As a standard of care for the treatment of GI bleeding, patients with hemodynamic compromise need immediate resuscitation and endoscopy within 24 hours after admission. 1e best-established GI bleeding scoring sys-C Mankongpaisarnrung, et al.Risk assessment in patients with GI bleeding tems include the Rockall score 7 and the Glasgow-Blatchford score (GBS). 8Recently, the AIM65 was also found to be useful. 9me studies have demonstrated that clinical decisions made by certified emergency physicians are also important and perhaps more useful than using a GI bleeding score tool in triage decisions either to admit the patient or to discharge the patient to outpatient evaluation. 10,11However, there has only been one clinical study in the US regarding the use of a GI bleeding scoring system to triage into an ICU or non-ICU bed. 12per Gi bleeDinG sCorinG systems 1. Pre-endoscopic Rockall score and complete Rockall score 2. Glasgow-Blatchford score (GBS)

pre-enDosCopiC roCkall sCore anD Complete roCkall sCore 5
This scoring tool may help predict the risk on in-hospital bleeding and/or mortality.At the initial evaluation before endoscopy, patients can be evaluated using the pre-endoscopic Rockall score.Patients are assigned point values for each of the clinical variables (age, shock, and comorbidity) and endoscopic variables (diagnosis and stigmata of recent hemorrhage).The Rockall score is equal to the sum of the points assigned.Scores can range from 0-7 for the preendoscopic score and 0-11 points for the complete score.Patients with complete Rockall scores of 2 or less are considered low risk for developing adverse outcomes (rebleeding 4%, mortality <0.1%).Patients with a pre-endoscopic Rockall score of 0 are considered low risk.This scoring system may help anticipate the risk of in-hospital bleeding and mortality.

GlasGow-blatChforD sCore 10
Patients with upper GI bleeding who have a total GBS of 0 are considered low risk.These patients can be discharged from the emergency room without admission and/or in-hospital intervention.Instead, early out-patient endoscopy would be a more effective option.This scoring system has also been shown to anticipate the risk of intervention (blood transfusion, endoscopic or surgical therapy) and death.Many studies have found it to be a simple and useful scoring system when applied to triage patients in emergency rooms awaiting endoscopic results.

aims65 sCore
Patients with upper GI bleeding who have the total AIMS65 of 1 or less are considered at low risk.

Clinical parameters Score value
Albumin level <3.0 g/dL

Case review usinG sCorinG tools
Our patient presents with symptomatic anemia, most likely from GI bleeding given the melenic stools.She has a low BP and tachycardia likely secondary to blood loss.Her estimated blood loss would be 15-30% since she is tachycardic but there is no other information regarding orthostasis.She needs immediate fluid resuscitation and hospitalization for evaluation.
Using the upper GI bleeding scoring systems discussed, her calculated pre-endoscopic Rockall score would be 3 from tachycardia (pulse > 100 bpm) and low blood pressure (< 100 mmHg).Her GBS would be 9 from her BUN level, hemoglobin level, heart rate, systolic blood pressure, and presentation of melena.Therefore, this patient would be stratified as high risk by both scoring systems, requiring immediate resuscitation and urgent endoscopic evaluation.

Comparison of Gi bleeDinG sCorinG tools
Gralnek, et al. concluded that the complete Rockall score helps identify more low risk patients (complete Rockall score less than 2) than the pre-endoscopic Rockall score or GBS. 14However, another study demonstrated that the GBS is more effective than the pre-endoscopic Rockall score. 11Since the complete Rockall score requires endoscopy results for its scoring, it cannot be utilized in the clinical settings where endoscopy cannot be performed soon after evaluation in the emergency room.Clinical evaluation and the use of a clinical scoring system such as the pre-endoscopic Rockall score would help triage this group of patients. 5The GBS can better predict clinical outcomes of death, the need for blood transfusion, endoscopic therapy, or surgery than the complete Rockall score. 15,16Another study reported that the GBS identified high risk patients with variceal hemorrhage needing intervention; however, it was not useful in predicting mortality. 17No known cardiac/hepatic disease = 0 Total score 3 9

reCommenDations
Upper GI bleeding scoring systems are useful instruments to help evaluate patients with upper GI bleeding.However, clinical acumen is also important and necessary to provide the best level of medical care to patients with upper GI bleeding.With the use of a scoring system, low risk patients can be identified more effectively and triaged to less costly care.key points 1. Detailed clinical evaluation and hemodynamic assessment can risk stratify patients with upper GI bleeding.
2. Upper GI bleeding scoring tools can assist physicians in the triage of patients.However, the scoring system cannot substitute for clinical acumen.In order to provide the most appropriate level of medical care, incorporating an upper GI bleeding scoring system into a clinical practice should allow internists to evaluate patients with GI bleeding requiring hospitalization more effectively.
3. Low risk criteria in the different upper GI bleeding scoring system are listed below.

•
Pre-endoscopic Rockall score of 0 and complete Rockall score of 2 or less -Predict the risk of further bleeding and death • GBS of 0 -Predicts the risk of intervention and death • AIMS65 of 1 or less 4. Patients with upper GI bleeding who have the total GBS of 0 are considered low risk, and they can be discharged from emergency department safely with < 1% chance of requiring an immediate intervention after discharge.

13
disease = known history or clinical/laboratory evidence of chronic or acute liver disease ** Cardiac failure = known history of clinical/echocardiographic evidence of cardiac failure key worDs-Upper gastrointestinal bleeding, upper GI bleeding scoring system Rockall score, Glasgow Blatchford score C Mankongpaisarnrung, et al.Risk assessment in patients with GI bleeding