Esophageal perforation is uncommon and traditionally has a high rate of morbidity and mortality. Our aim was to perform a 13-year retrospective review of the cases managed in our district general hospital. Thirty-four cases of esophageal perforation diagnosed between 1995 and 2008 were retrospectively analyzed. There were 20 males and 14 females with a median age of 64 (range 23-86) years. The etiology of the perforations were Boerhaave's syndrome (n= 19), therapeutic endoscopy (n= 9), diagnostic endoscopy (n= 2), gastric lavage injury (n= 1), foreign body (n= 1), blunt chest trauma (n= 1), and spontaneous tumor perforation (n= 1). Only 11 cases (32%) had evidence of surgical emphysema upon examination. In 50% of cases, another clinical diagnosis was initially suspected. Twenty-four were treated surgically and 10 cases managed non-operatively. Surgical treatment included thoracotomy with primary repair (n= 9), T-tube drainage (n= 7), emergency esophagectomy (n= 1), or intra-operative stent insertion (n= 1). Four cases had primary repair and fundal wrap via abdominal approach without thoracotomy. Two patients were treated with washout and drainage only. Eight patients died overall (in-hospital mortality 23.5%). Esophageal perforations are often initially misdiagnosed and the majority do not have surgical emphysema. There are a wide variety of methods to manage esophageal perforation. Management tailored to the location and size of perforation, degree of contamination, and underlying cause appears to result in a reasonable prognosis.

The information requirements of nurse managers have been uncreasing rapidly in recent years. The demand for access to relevant and accurate information has led to the development of computerized information systems for use a ward level. However, once implemented, few information systems have been evaluated to identify the benefits that will be realized. This paper presents the results of a study carried our in Oxford to evaluate one nurse management information system using a multi-method approach to identify ways in which returns may be manifested. It suggests that evaluation should be an active process planned at the same time is defining information requirements.