Remember the use of likelihood ratios in your initial evaluation of a suspected UGIB!
– patient reported melena (LR 5.1 – 5.9)
– melena on exam (LR 25)
– blood/coffee ground on NG lavage (LR 9.6)
– BUN/Cr > 30 (LR 7.5)
Factors associated with severe bleeding:
– red blood on NG lavage (LR 3.1)
– tachycardia (LR 4.9)
– Hgb < 8.0 (LR 4.5 – 6.2)
Signs of hypovolemia – mild to moderate resting tachycardia
15% blood loss – orthostatic hypotension
40% blood loss – supine hypotension
Acid Suppression: when PUD suspected use IV PPI
– decrease hospital stay, rebleeding rate, and need for transfusions
– also shown to promote hemostasis with lesions other than ulcers
Prokinetics: erythromycin/metoclopramide – used 30 minutes prior to EGD for improved visualization
Somatostatin Analogs: octreotide – use in suspected variceal bleeds to lower portal pressure through splenic vasoconstriction
Antibiotics: for cirrhotic patients; studies show prophylactic antibiotics reduce complications and mortality
Rockall Score: requires endoscopy for calculation
Blatchford Score: used at presentation, range from 0-23 (higher score worse). Utilizes BUN, SBP, pulse, presence of melena, syncope, hepatic disease, and/or cardiac failure to generate number.
Testing for H. Pylori:
Endoscopic biopsy: invasive and expensive; requires EGD for diagnosis
Urea Breath Test: non-invasive and inexpensive; sensitivity 88-95%, specificity 95-100%
Serology (IgG): non-invasive; sensitivity 90-100%; specificity 76-96% – not recommended in low prevalence populations (i.e. US resident without travel); conversion of positive serology to negative suggests cure.
Stool Antigen: non-invasive and inexpensive; sensitivity 94%, specificity 86% – good for documenting eradication; false negative if patient on PPI – MUST STOP 2 WEEKS PRIOR TO TESTING!
Treatment for H. Pylori:
Triple Therapy: PPI, amoxicillin (or metronidazole for penicillin allergy), clarithromycin
Quadruple Therapy: PPI/Rantidine, Bismuth Subsalicylate, Metronidazole, Tetracycline (or Doscycline)