We thank Dr Shao and colleagues for their comments on our study.
1
First, we agree that the association between the baseline characteristics and patient outcomes may cause the chance bias on a statistical test of outcome,2
and covariate adaptive allocation may reduce the imbalance of baseline covariates.3
In this study, we used blocked randomization (restricted randomization), and imbalance of baseline patient ages between the self-expandable metal stent (SEMS) group and the plastic stent (PS) group had occurred. However, we performed covariate adjusted analysis to correct for the imbalance in baseline characteristics or, more specifically, age. We used multiple logistic regression and Cox proportion hazards regression for successful drainage rate and patient survival, respectively. In both regression analyses, age was not the statistically significant explanatory variable. Besides, the presence or absence of age in the models did not significantly affect the magnitude and direction of the coefficient of stent type in the models; therefore, the imbalance in mean ages of the patients in the SEMS and PS groups should not alter the conclusions of the study.Second, concerning the deviations from random allocation, Figure 1 in the study showed that only 1 patient in the PS group received SEMS after the first PS occlusion.
1
According to the protocol, some of patients with unsuccessful stent insertion and unsuccessful biliary stent drainage in both groups received PTBD as a rescue therapy, and the rest received palliative medication to control the symptoms of hilar cholangiocarcinoma. Regarding a missing response, there was no missing response on the part of survival analysis. However, in the comparison of successful drainage rate between the PS and SEMS groups, missing response occurred in patients who died before the first appointment, and they were assigned to unsuccessful drainage in the analysis.Finally, in this study, we focused on the efficacy of SEMS and PS in unresectable hilar cholangiocarcinoma; thus, the patients in both groups did not receive any treatment modalities that affected patient survival such as chemotherapy, brachytherapy, and external beam radiation.
References
- Efficacy of metal and plastic stents in unresectable complex hilar cholangiocarcinoma: a randomized controlled trial.Gastrointest Endosc. 2012; 76: 93-99
- Understanding controlled trials: baseline imbalance in randomised controlled trials.BMJ. 1999; 319: 185
- Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial.Biometrics. 1975; 31: 103-115
Article info
Identification
Copyright
© 2013 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.
ScienceDirect
Access this article on ScienceDirectLinked Article
- Is a self-expandable metal stent better than a plastic stent in unresectable complex hilar cholangiocarcinoma with regard to the adequacy of drainage and survival of the patients?Gastrointestinal EndoscopyVol. 77Issue 1
- PreviewWe read with great interest the recent article by Sangchan et al.1 The authors concluded that endoscopic biliary drainage with a self-expandable metal stent (SEMS) provides better adequacy of drainage and longer survival compared with the plastic stent (PS) in patients with unresectable complex hilar cholangiocarcinoma. It is an interesting study. Nevertheless, several issues seem worthy of comment.
- Full-Text
- Preview
