PURPOSE: Early endoscopy has previously shown a modest improvement in the outcome and the cost of care in lower GI (Gastrointestinal) bleeding. We wanted to confirm, starting with a pilot study, these results in all GI bleeding admissions to the MICU (Medical Intensive Care Unit).
METHODS: 53 patients admitted to the MICU at Downstate Medical Center, for gastrointestinal bleeding, were reviewed retrospectively. Study outcomes were: re-bleeding, length of stay in MICU and in the hospital, and number of packed red blood cells transfused. Predictors of interest were: time to endoscopy, time to GI consult, and time to MICU consult, all categorized as <12, 12-24, or >24 hours from the emergency room presentation. Potential confounders were WBC (White Blood Cells) and admitting diagnosis. Fisher's exact test was used to assess association between re-bleeding and each predictor of interest; the Kruskal-Wallis test was used to assess association between the other outcomes and each predictor of interest. Fisher's exact test was used to assess association between re-bleeding and diagnosis type; Spearman correlation was used to assess association between WBC and outcomes. In a multivariate analysis, exact logistic regression was used to assess association between re-bleeding and each predictor of interest, controlling for WBC, and the admitting diagnosis.
RESULTS: Neither early MICU consult, early admission, early endoscopy, or early GI evaluation had any significant correlation with the re-bleeding rate (P=0.552, P=0.653, and P=0.656, respectively), with a shorter length of hospital stay (P=0.466, P=0.458, P=0.956, respectively), or with a shorter MICU length of stay (P=0.603, P=0.801, P=0.431, respectively). GI bleeding further complicated by other organ dysfunction on admission had a significant correlation with the length of stay in the hospital (P=0.019) and the length of stay in the MICU (P=0.018). This correlation was not found in isolated GI bleeding diagnosis.
CONCLUSIONS: Earlier endoscopy, earlier GI evaluation and earlier MICU evaluation and admission did not improve the outcome of GI bleeding as measured by the length of hospital and MICU stay, as well as by the re-bleeding rate and the total number of PRBC transfused.
CLINICAL IMPLICATIONS: In this pilot study, earlier management of GI bleeding did not change the outcome.
DISCLOSURE: The following authors have nothing to disclose: George Apergis, Francois Abi Fadel, Stephen Kuperberg, Joe Zein, Samir Fahmy
No Product/Research Disclosure Information