According to the power analysis, G-POWER 3.0 (Heinrich-
Heine University, Du¨ sseldorf, Germany) computed a required
sample size of 159 using F-test analysis of variance (ANOVA)
method under the following assumptions: the absolute
difference between the pain scores on the operation day
(POD-0), the 1st postoperative day (POD-1) and the 2nd
postoperative day (POD-2) being equal to a medium effect
size of 0Æ25, a power of 0Æ80 and a 2-tailed alpha of 0Æ05
(Erdfelder et al. 1996, Polit & Beck 2004). A total of 174
children were enrolled over a six-month period of which two
patients died shortly after the surgery and two patients
received ECMO, leaving 170 participants for final data
analysis. The Ethical Research Committee of the hospital
approved this study. Parents were informed about the study
and signed the informed consents before the study commenced.