© 2013 by the American College of Gastroenterology The American Journal of GASTROENTEROLOGY Letters to the Editor 1533 in mortality were signiicant according to comorbidity score in the three classes of age ( P < 0.001, P < 0.001, P < 0.007 by chi-square test for < 50, 50–75, and >75 years), but not according to age in the three classes of comorbidities ( P >0.20)). In logistic regression, the two independent predictors of death were the occurrence of a non-bleeding complication and Rockall’ s score (which includes comorbidity in its deinition). Death was related to a bleeding cause in 28.3% of cases and to a non-bleeding cause in 60% of cases (cardiopulmonary causes ( n = 11), cerebrovascular disease ( n = 3), multiorgan failure ( n = 1), terminal malig- nancy ( n = 2), uncontrolled sepsis ( n = 8), cirrhosis ( n = 9)), with the cause remaining undetermined in 10.7% of cases. herefore we conirm the conclusions of Leontiadis et al. Moreover, in our patients, as in the cohort of Sung et al., mortality was unrelated to peptic ulcer bleeding in the majority of cases (3). ACKNOWLEDGMENTS his study was supported by Sanoi- Aventis and the French Society of Gastroenterology (Société Nationale Française de Gastroentérologie). CONFLICT OF INTEREST he authors declare no conlict of interest. REFERENCES 1. Leontiadis GI, Molloy-Bland M, Moayyedi P et al. Efect of comorbidity on mortality in patients with peptic ulcer bleeding: systematic review and meta analysis. Am J Gastroenterol 2013;108:331–45; quiz 346. 2. Nahon S, Hagège H, Latrive JP et al. Pariente A; Groupe des Hémorragies Digestives Hautes de l’ANGH. Epidemiological and prognostic factors involved in upper gastrointestinal bleed- ing: results of a French prospective multicenter study. Endoscopy 2012;44:998–1008. 3. Sung JJ, Tsoi KK, Ma TK et al. Causes of mortality in patients with peptic ulcer bleeding: a prospective cohort study of 10,428 cases. Am J Gastroenterol 2010;105:84–9. 1 Service d’hépatogastroentérologie, Centre Hospitalier Intercommunal Le Raincy-Montfermeil, Montfermeil, France; 2 Centre Hospitalier de Pau, Pau, France. Correspondence: Stéphane Nahon, MD, Service d’hépatogastroentérologie, Centre Hospitalier Intercommunal Le Raincy-Montfermeil, 10 avenue du Général Leclerc, 93370 Montfermeil, France. E-mail: snahon@ch-montfermeil.fr or snahon@club-internet.fr Treat the Patient, not Just the Source of Bleeding Riccardo Marmo, MD 1 , Cristina Bucci, MD 1 , Matilde Rea, MD 1 and Gianluca Rotondano, MD, FACG 1,2 doi:10.1038/ajg.2013.190 To the Editor: We have read with great interest the paper by Leontiadis et al. (1) on the efect of comorbidity on mortal- ity in patients with peptic ulcer bleed- ing. Longer life expectancy, together with efective treatment of concurrent illnesses worldwide, have contributed to the selection of an aged population with progressively multiple comorbidi- ties. he occurrence of an episode of gastrointestinal bleeding, related to the use of nonsteroidal anti-inlammatory drugs or antiplatelet medications in 60% of the cases (2,3), induces the breakage of an unstable balance of the comorbidi- ties afecting the patient. Such a derange- ment occurs not only for major peptic ulcer bleeding, but also for bleeding from lesions so far considered “minor”, i.e., vascular lesions, Mallory–Weiss tears, or gastro-duodenal erosions (4). he criti- cal impact of comorbid illnesses on the risk of mortality from upper nonvariceal bleeding is well known, as witnessed by the recent publication of several risk scores addressing the pre-eminent role of comorbidity (2,5,6). Early endoscopy, aggressive endoscopic and pharmacologic treatment of high risk lesions, and careful assessment of the patient’s overall health status have deinitely contributed to the improved outcomes of patients with upper GI nonvariceal bleeding (7). We now have an estimate of the mean survival gain in these patients, but research should focus on those subgroups of patients at a higher risk of death. In patients with ASA score 1–2, the risk of death from nonvariceal bleeding is signiicantly lower than those with ASA scores 3–4 (4). he most rele- vant comorbidities already identiied and conirmed by the Leontiadis paper (1) are essentially three: disseminated malig- nancy, liver cirrhosis, and renal failure; the others have a less relevant role and can be globally summarized with the clinical assessment of the ASA score 3–4. Patients with in-hospital bleeding have a signii- cantly higher risk of death not because they are older, but because they are sicker than outpatients admitted for non-variceal bleeding (8,9). he risk of death of in- hospital bleeders increases for the pres- ence of severe comorbidity, categorized by an ASA score of 3 or 4 (odds ratio 2.52 (95% conidence interval 1.98–3.20)). he most relevant clinical risk factors among comorbidities are the presence of chronic renal failure (odds ratio 2.93 (1.28–6.68)) and the presence of neoplasia (odds ratio 1.9 (0.73–4.94)) (9). Moreover, the risk of mortality in patients with cirrhosis is two- fold greater than noncirrhotic patients sufering an episode of nonvariceal bleed- ing (7.8 vs. 4.1%, odds ratio 1.99 (1.23– 3.20), P = 0.004) (10). In patients with cirrhosis, the main causes of nonvariceal bleeding are peptic ulcer and gastro-duo- denal erosions. he presence of chronic renal failure and neoplasia (hepatocel- lular carcinoma) have been identiied as independent predictors of death. Moreo- ver, cirrhotic patients bleeding from a duodenal ulcer have a 13% risk of death, three times higher than noncirrhotic patients (10). he frequency of death for nonvariceal bleeding is similar to that of variceal hemorrhage, where again the stage of liver insuiciency and concomi- tant presence of comorbidities increase the risk of death (11). All these elements highlight that the role of comorbidities is the true clinical nodal point afect- ing the outcome of nonvariceal bleeders. A true multidisciplinary approach, i.e., the early involvement of other specialty professionals to share critical decisions about comorbidity management in these patients, is the key to further reduce the risk of death from nonvariceal bleeding. he recent demonstration that overuse of transfusions increases the risk of death (12) and that continuation of aspirin in patients with nonvariceal bleeding improves overall survival (13) clearly go in this direction.