Multiple stepwise regression analysis of physiologic tension and secondary messengers in chronic burn trauma

Abstract

In this report, we have used a multiple stepwise regression analysis to determine the best derived multivariate equation between a dependent and multiple independent parameters. All parameters were derived from harvested gastrocnemius muscle subjected to chronic burn trauma. Correlations between seven different parameters were tested. Physiologic tension, calcium, adenosine 3':5' cyclic monophosphate (cAMP), inositol 1,4,5-triphosphate (IP3) and its catabolites comprised one large data matrix. From it, the various parameters were substituted as the dependent variable. The remaining parameters all comprised a multiple independent group. From such a format, the stepwise regression was run. The simultaneous co-governing of independent parameters over one dependent parameter provides a means of mathematically proving multiple correlations. Such findings provide a new perspective in interpreting chemical signalling.