Application of Outlier Robust Nonlinear Mixed Effect Estimation in Examining the Effect of Phenylephrine in Rat Corpus Cavernosum

Ignoring two main characteristics of the concentration-response data, correlation between observations and presence of outliers, may lead to misleading results. Therefore, the special method should be considered. The present study was designed to apply the outlier robust nonlinear mixed estimation for effects of phenylephrine on rat corpus cavernosum strips. In this study, eight different doses of phenylephrine in eight experimental groups were used. Each group consisted of eight rats. The concentration–response curves to phenylephrine (0.1μM to 300μM) were obtained by the cumulative addition of phenylephrine to the chamber. Because of the existence of an outlier to achieve robust estimations, M-estimation method and Huber function as a dispersion function were used. Cumulative administration of phenylephrine (0.1μM - 300μM) caused concentration-dependent contractions in strips of rat corpus cavernosum (-Log EC50 was 5 ± 0.31, 95% CI= 5.92 to 4.21). The contraction of corpus cavernosum started in the concentration of 0.3 μM and then gradually increased in a dose-dependent manner till it reached a plateau in 100 μM. To consider the clustering feature of concentration-response data, the 4pl regression with a random term has been used. To estimate parameters, because of existence of an outlier in dataset, the robust procedure has been applied. The contraction of corpus cavernosum started in the concentration of 0.3 μM and then gradually increased in a dose-dependent manner till it reached a plateau in 100 μM.