[Skip to Content]
[Skip to Content Landing]
Views 192
Citations 0
Feb 2012

Surveillance Bias and Postoperative Complication Rates—Reply

Author Affiliations

Author Affiliations: Department of Surgery, University of California Irvine, Orange.

Arch Surg. 2012;147(2):199-200. doi:10.1001/archsurg.2011.1492

In reply

We appreciate the comments of Ms Rosenberg and Dr Haut regarding possible selection bias. We acknowledge that this is an inherent limitation of an analysis of any population-based database. We do not have specific data related to the number of duplex ultrasonography tests or computed tomographic scans performed for the 2 groups because this level of information is lacking in the Nationwide Inpatient Sample database; however, we believe that the discovery of deep vein thrombosis and/or a pulmonary embolism, as an incidental finding, as Ms Rosenberg and Dr Haut hypothesize, is not common enough to bias the overall findings to a significant degree. In terms of their comment that “the mean length of hospital stay was significantly longer after open surgery than after laparoscopic surgery (9.5 vs 6.5 days; P < .001), these patients may have been at higher risk of having additional diagnostic testing performed,” although we agree that this is likely a minor variance biasing our findings, it is more likely that this longer length of stay and the decreased ambulation inherent in the recovery after open surgery is in fact a major factor influencing the increased risk of venous thromboembolism.

First Page Preview View Large
First page PDF preview
First page PDF preview