Author Affiliations: Departments of Visceral and Transplantation Surgery (Drs Andres, Majno, Morel, Toso, and Mentha) and Clinical Epidemiology (Dr Gerstel), Geneva University Hospitals, Geneva University Hospitals, Switzerland.
We thank Drs Tamhane and McGwin for their critical comments on our article entitled “Complications of Elective Liver Resections in a Center with Low Mortality.”1 Their first comment was linked to the use of a linear regression model to select variables associated with the occurrence of perioperative complications after elective liver resections. Although the outcome variable was binary (complication vs no complication), a logistic regression would have been more appropriate. This said, all 4 selected variables (type of liver resection, indication, American Society of Anesthesiologists [ASA] score and, perioperative blood transfusion) also remained the only significant variables in the multivariate logistic regression model. The proposed score would not have been modified by the use of the logistic model (Table 1). In our original score, the areas under the receiver operating characteristic curve were 0.66 and of 0.59 for the derivation and the validation cohorts, respectively.
Andres A, Gerstel E, Majno P, Morel P, Toso C, Mentha G. Analytical Concerns Regarding Complications of Elective Liver Resections in a Center With Low Mortality—Reply. Arch Surg. 2011;146(12):1455. doi:10.1001/archsurg.146.12.slt110001