PURPOSE: GI bleeding(GIB) is a major cause of morbidity and deaths in ICUs. However there is scant literature related to this condition among Hispanics and African Americans. In addition, illicit drug use, smoking and alcoholism are common social factors in this population that are suspected to affect outcomes with GIB. Previously published studies have shown that APACHE-II scores overestimate mortality among GIB patients. We proposed to study the predictors of survival among patients with GIB in our MICU.
METHODS: This is prospective observational study of patients admitted to MICU at LMMHC with a diagnosis of GIB from Jan 2006 to Dec 2007. Data regarding clinical history, laboratory and interventions were collected by an independent trained investigator. Predictors such as demographics, APACE II, etiology of GIB, coagulopathy and other factors were analyzed. Using stepwise logistic regression, odds ratios were calculated and 95% confidence intervals are reported. A p-value < 0.05 was considered significant.
RESULTS: Of the 173 patients included, 69% of patients were Hispanics. Baseline characteristics are described in Table1. There were 27(15.6%) deaths (all due to upper GI source) in our study which is comparable to published data. Subjects who died from GIB were noted to have significantly higher median APACHE-II score compared to survivors (19 versus 12, OR=1.18, 95%CI 1.08–1.29;p value < 0.0001). None of other characteristics predicted mortality associated to GI bleed. (Table2) Interestingly, those who died had a marginally decreased odds of prophylactic aspirin use than survivors. (OR=0.31, 95%CI 0.1–0.99;p = 0.049).
CONCLUSION: Despite observed mortality being lower than expected mortality, higher APACHE-II scores at admission may still offer a good estimate to predict death among inner city minority patients with GIB. It is however less clear why prophylactic aspirin use had a marginally protective effect and this needs to be validated in larger studies.
CLINICAL IMPLICATIONS: This is probably the largest reported series of patients with GIB reported among inner-city minority MICU subjects and offers insight into understanding characteristics and predictors of mortality in this population.
DISCLOSURE: Raghu Loganathan, No Financial Disclosure Information; No Product/Research Disclosure Information