This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
Hau and colleagues have examined an important topic about which much confusing information has been published in recent years. They have chosen 38 carefully matched pairs of patients from a group of 255 patients treated for intra-abdominal infection. In each pair, one patient was treated with planned relaparotomy, and the other underwent reoperation only as clinical circumstances revealed the need. Previous reports examining the utility of planned relaparotomy have compared the authors' experiences with their own prior experiences, with the reported experience in the literature, or with a calculated, expected mortality according to some severity index. All these methods, while useful for generating hypotheses, are too flawed to justify a change in general practice patterns based on this type of information. The present study compared contemporaneous patients treated in the same institutions, using prospective data accumulation but retrospective analysis. Patients were not randomized, so, despite matching, there is a risk
Dellinger EP. Invited Commentary. Arch Surg. 1995;130(11):1197. doi:10.1001/archsurg.1995.01430110055010