A practical mortality risk score for emergent colectomy.

BACKGROUND: Critically ill patients requiring emergent colectomy have significant mortality risk. OBJECTIVE: A national administrative database was used to compose a simple scoring scheme for predicting in-hospital mortality risk. DESIGN: The 2007 to 2009 Nationwide Inpatient Sample was queried to identify patients requiring nonelective colectomy. Multivariable binary logistic regression analysis was used to identify predictors that increased mortality. Each predictor was given a point value, based on the corresponding logit, the sum of which constituted a risk score. The scoring system was tested by using k-partitions cross-validation. SETTINGS: This study is based on database analysis. PATIENTS: A total of 338,348 cases were identified. Mean age was 64, and 53% of the patients were women. MAIN OUTCOME MEASURES: The primary outcomes measured were mortality and risk score development. RESULTS: The overall mortality risk was 9%. Regression analysis identified the following risk factors and assigned points: acute renal failure (6), hemodialysis (6), age >65 (4), peripheral vascular disease (4), myocardial infarction (4), chronic obstructive pulmonary disease (2), cardiac arrhythmia (1), and congestive heart failure (1). The maximum score observed was 26 (of a possible 28), which corresponded to 100% mortality. Receiver operator characteristic analysis showed an area under the curve of 0.81. LIMITATIONS: This study was limited because of its retrospective nature, and because it used database data with variability in coding among participating institutions. CONCLUSIONS: With the use of a simple 8-variable scoring system, inpatient mortality estimates can be made for patients requiring emergent colectomy. When used judiciously, it can be used as a tool when counseling patients and family both before and after surgery.

Transfer status is a risk factor for increased in-hospital mortality in patients with diverticular hemorrhage.

PURPOSE: Gastrointestinal tract hemorrhage is a common problem accounting for approximately 1 % of hospital admissions. It is estimated that one third of the episodes of lower gastrointestinal hemorrhage are secondary to diverticular disease. Inter-institutional transfer has been associated with delay in care and increased in-hospital mortality. We hypothesized that patients with diverticular hemorrhage that were transferred from an acute care hospital to tertiary care institutions have increased in-hospital morbidity and mortality when compared to primarily admitted patients. MATERIALS AND METHODS: We performed a retrospective analysis of the Nationwide Inpatient Sample for the year 2008. Patients with a primary discharge diagnosis of diverticular hemorrhage were selected. Multivariate logistic regression was used to identify the relationship between transfer status and in-hospital mortality. RESULTS: A total of 99,415 hospitalizations for diverticular hemorrhage were identified. Transferred patients had higher in-hospital mortality rates compared to primarily admitted patients (3.5 vs. 1.8 %, p < 0.001), as well as increased length of stay (8.4 vs. 5.4 days, p < 0.001) and a higher rate of total abdominal colectomy (1.2 vs. 0.6 %, p < 0.001). Multivariate analysis indicated that transfer status was associated with increased in-hospital mortality [OR 1.8, 95 % CI 1.5-2.8, p < 0.001]. CONCLUSIONS: Inter-institutional transfer for diverticular bleeding is associated with increased in-hospital mortality, increased total abdominal colectomy rate, as well as increased economic burden including mean length of stay and total hospital charges. Further prospective studies are needed to analyze the clinical information of patients requiring transfer to another hospital in order to identify those patients who would truly benefit from inter-institutional transfer.

Hemorrhoids can be a source of obscure gastrointestinal bleeding that requires transfusion: report of five patients.

PURPOSE: Hemorrhoids have been rarely reported to be a source of transfusion-dependent, obscure gastrointestinal bleeding. We report the diagnosis and management of a series of patients in whom hemorrhoids were the source of obscure gastrointestinal bleeding that was severe enough to require transfusion. METHODS: Five patients, who presented with severe hematochezia or obscure gastrointestinal bleeding, during a 24-month period had had an extensive workup for gastrointestinal bleeding. All had required transfusion of 2 units or more of blood. In the absence of other causes of bleeding, the five patients had unprepared flexible sigmoidoscopy on the same day that they reported hematochezia to exclude a proximal source of bleeding. All were found to have internal hemorrhoids as a likely source of bleeding, which was confirmed at exploration under anesthesia, and were treated by surgical hemorrhoidectomy. RESULTS: The five patients underwent surgical, Ferguson-type, hemorrhoidectomy. One patient required 1 unit of blood immediately postoperatively, but none have required further transfusion or reported hematochezia in more than three months of follow-up. CONCLUSIONS: We have shown that hemorrhoids can be a source of transfusion-dependent, obscure gastrointestinal bleeding. This aspect of the common problem of hemorrhoidal bleeding has not been previously reported, which reflects underreporting or missed diagnosis.