Response

Abstract

We thank Dr Chandra for the letter and interest in the AIMS65 score. Dr Chandra reports that the AIMS65 score accurately predicted 30- and 90-day all-cause mortality among patients with upper GI hemorrhage (UGIH). Mortality at 30 and 90 days is an important metric, because it captures patients who die shortly after hospitalization. The AIMS65 score has now been shown to reliably predict in-hospital 30- and 90-day mortality among patients with UGIH.1Hyett B.H. Abougergi M.S. Charpentier J.P. et al.The AIMS65 score compared with the Glasgow-Blatchford score in predicting outcomes in upper GI bleeding.Gastrointest Endosc. 2013; 77: 551-557Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar, 2Saltzman J.R. Tabak Y.K. Hyett B.H. et al.A simple risk score predicts in-hospital mortality, length of stay and cost in acute upper GI bleeding.Gastrointest Endosc. 2011; 74: 1215-1224Abstract Full Text Full Text PDF PubMed Scopus (263) Google Scholar Dr Chandra also assessed the AIMS65 score's performance in predicting a composite outcome of endoscopic treatment, surgery, and packed red blood cell (PRBC) transfusion. It is critical to recognize that various studies of UGIH have defined “need-for-intervention” differently. To objectively evaluate the need for intervention, we applied the definition used for the derivation of the Glasgow-Blatchford risk score, but without PRBC transfusion.1Hyett B.H. Abougergi M.S. Charpentier J.P. et al.The AIMS65 score compared with the Glasgow-Blatchford score in predicting outcomes in upper GI bleeding.Gastrointest Endosc. 2013; 77: 551-557Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar We did not include PRBC transfusion because both the American College of Gastroenterology guidelines3Laine L. Jensen D.M. Management of patients with ulcer bleeding.Am J Gastroenterol. 2012; 107: 345-360Crossref PubMed Scopus (494) Google Scholar and the International Consensus Conference recommendations on the management of patients with nonvariceal UGIH4Barkun A.N. Bardou M. Kuipers E.J. et al.International Consensus recommendations on the management of patients with nonvariceal upper gastrointestinal bleeding.Ann Intern Med. 2010; 152: 101-113Crossref PubMed Scopus (879) Google Scholar classify PRBC transfusion as part of the initial resuscitation and not as an outcome. Furthermore, the paradigm for PRBC transfusion in patients with UGIH is changing after Villanueva et al5Villanueva C. Colomo A. Bosch A. et al.Transfusion strategies for acute upper gastrointestinal bleeding.N Engl J Med. 2013; 368: 11-21Crossref PubMed Scopus (888) Google Scholar showed that conservative hematocrit transfusion thresholds are associated with improved survival compared with traditional thresholds. A similar transfusion threshold was recommended in 2012 by the American College of Gastroenterology guidelines.3Laine L. Jensen D.M. Management of patients with ulcer bleeding.Am J Gastroenterol. 2012; 107: 345-360Crossref PubMed Scopus (494) Google Scholar Dr Chandra's letter further adds to our understanding of the AIMS65 score's performance at predicting mortality in patients with UGIH. We believe the need-for-intervention composite outcome in patients with UGIH should not include PRBC transfusion, and the AIMS65 score is a reliable predictor of the need for intervention when properly defined. More studies are needed to fully compare the different prognostic scores for patients with UGIH. AIMS65 score predicts short-term mortality but not the need for intervention in acute upper GI bleedingGastrointestinal EndoscopyVol. 78Issue 2PreviewI read the article by Hyett et al1 with great interest. The investigators have validated the predictive performance of the AIMS65 score for inpatient mortality and compared its predictive accuracy with the Glasgow-Blatchford score for need for blood transfusion in acute upper GI bleeding. However, the AIMS65 score was not predictive for the need for blood transfusion. Full-Text PDF