Improved analysis of the vascular response to arterial ligation using a multivariate approach.

Ligation of the murine common carotid artery induces a reproducible remodeling response. The contribution of individual genes can be determined by comparison of the phenotypes of genetically modified mice. Although studies have shown the response to carotid artery ligation is influenced by many factors, individual analyses typically only consider a single factor, the presence of a gene of interest. Because of this limitation, measurements of the response to ligation show large variation, making the determination of significant difference between test groups difficult. In this study, we examine the hypothesis that the variation in the response to ligation is due to non-genetic factors in addition to genetic factors. Distance from the ligature, a variable common to all arterial ligation experiments, is an important source of variation and a significant predictor of the remodeling response. We find that the use of statistical regression is an improved analysis technique, as it allows the simultaneous consideration of multiple variables. We demonstrate this by showing improved sensitivity and novel findings in the analysis of the remodeling response in mice genetically mutant for the osteopontin gene. We conclude regression analysis provides a simple way to improve both comparative power and description of vascular remodeling.