Reply to "association of Blood Transfusion During Resection for Hepatocellular Carcinoma with Postoperative Recurrence and Overall Survival: A Cautionary Comment".
Tian Yang,Feng Shen,Myron Schwartz
DOI: https://doi.org/10.1016/j.jhep.2016.04.002
IF: 25.7
2016-01-01
Journal of Hepatology
Abstract:Perioperative blood transfusion does not influence recurrence-free and overall survivals after curative resection for hepatocellular carcinoma: A Propensity Score Matching AnalysisJournal of HepatologyVol. 64Issue 3PreviewWhether perioperative blood transfusions (PBTs) negatively impact oncologic outcomes after curative resection for HCC remains controversial. We aimed to identify the independent predictive factors of PBT for curative resection of hepatocellular carcinoma (HCC), and to investigate the impact of PBT on long-term recurrence and survivals after resection. Full-Text PDF Association of blood transfusion during resection for hepatocellular carcinoma with postoperative recurrence and overall survival: A cautionary commentJournal of HepatologyVol. 65Issue 1PreviewAs researchers at a large tertiary liver care center with a longstanding interest in peri- and post-resection prognosis of patients with hepatocellular carcinoma (HCC) [1], we were pleased that Yang and coworkers [2] took on the controversial question of whether perioperative blood transfusion (PBT) during curative liver resection for HCC affects risk of recurrence and overall survival. Their results prior to propensity score matching indicated that, indeed, PBT was associated with significantly higher risk of recurrence-free survival and significantly lower overall survival. Full-Text PDF We thank Zhong et al. for their interest in our manuscript recently published in the Journal of Hepatology [[1]Yang T. Lu J.H. Lau W.Y. Zhang T.Y. Zhang H. Shen Y.N. et al.Perioperative blood transfusion does not influence recurrence-free and overall survivals after curative resection for hepatocellular carcinoma: a propensity score matching analysis.J Hepatol. 2016; 64: 583-593Abstract Full Text Full Text PDF PubMed Scopus (84) Google Scholar]. We assumed from the outset that the results of our propensity-matched analysis of the impact of intraoperative blood transfusion on hepatocellular carcinoma (HCC) recurrence after resection would engender controversy, as they challenge a long-standing preconception that a cause-and-effect relationship exists between the two. The comments of Zhong et al. focused primarily on the fact that the p values for our univariate analyses of morbidity, overall survival (OS), and recurrence-free survival (RFS) in our propensity-matched cohort were relatively low (even though greater than the 0.05 threshold), leaving room for continued speculation that a cause-and-effect relationship between transfusion and cancer recurrence indeed exists. Zhong et al. have perhaps not adequately considered that in our subsequent multivariate Cox regression analyses, the p values and hazard ratios (95% confidence interval) for transfusion as an independent predictor were 0.225 and 1.250 (0.871–1.794) for OS, and 0.396 and 1.204 (0.809–1.770) for RFS, respectively, as shown in Tables 4 and 5 of the publication [[1]Yang T. Lu J.H. Lau W.Y. Zhang T.Y. Zhang H. Shen Y.N. et al.Perioperative blood transfusion does not influence recurrence-free and overall survivals after curative resection for hepatocellular carcinoma: a propensity score matching analysis.J Hepatol. 2016; 64: 583-593Abstract Full Text Full Text PDF PubMed Scopus (84) Google Scholar]. As skepticism concerning our findings was anticipated, we preemptively addressed the issues raised Zhong et al. in the discussion of our manuscript, third paragraph from the end. We would point out to Zhong et al. that a number of recently published studies employing propensity-score matching in other cancers including rectal [[2]Warschkow R. Guller U. Koberle D. Muller S.A. Steffen T. Thurnheer M. et al.Perioperative blood transfusions do not impact overall and disease-free survival after curative rectal cancer resection: a propensity score analysis.Ann Surg. 2014; 259: 131-138Crossref PubMed Scopus (41) Google Scholar], prostate [[3]Boehm K. Beyer B. Tennstedt P. Schiffmann J. Budaeus L. Haese A. et al.No impact of blood transfusion on oncological outcome after radical prostatectomy in patients with prostate cancer.World J Urol. 2015; 33: 801-806Crossref PubMed Scopus (35) Google Scholar], and cholangiocarcinoma [[4]Muller S.A. Mehrabi A. Rahbari N.N. Warschkow R. Elbers H. Leowardi C. et al.Allogeneic blood transfusion does not affect outcome after curative resection for advanced cholangiocarcinoma.Ann Surg Oncol. 2014; 21: 155-164Crossref PubMed Scopus (27) Google Scholar], have failed to demonstrate any independent impact of transfusion on long-term survival, similarly to our study. It has been well demonstrated that allogeneic blood transfusion induces changes in the immune system that are associated with immunosuppression, and a link between transfusion and cancer recurrence is thus easy to rationalize. The complex relationship between immunity and cancer is slowly being unraveled. On the other hand, from a surgeon’s perspective, it is also easy to rationalize that transfusion is associated with worse outcomes because cases that require transfusion are worse in a variety of ways that can be difficult to distill into discrete variables. As by its nature this is not a question that can be addressed by a prospective randomized trial. Propensity score matching is a way that we can, to the best of our abilities, minimize the effect of such confounding factors [[5]Austin P.C. The performance of different propensity score methods for estimating marginal hazard ratios.Stat Med. 2013; 32: 2837-2849Crossref PubMed Scopus (554) Google Scholar]. Statistical analyses such as those performed in our study do not offer proof; they deal with probability. In this sort of analysis we begin with a null hypothesis, in this case, we attempt to disprove this null hypothesis, that transfusion has no independent influence on the likelihood of cancer recurrence. We accept by convention p = 0.05, in other words, a less than 1 in 20 chance that the null hypothesis is true, as a basis for rejecting it. This criterion was not met in our study, so the null hypothesis that there is no relationship between transfusion and cancer recurrence stands intact. Zhong et al. appear to have taken the opposite approach, assuming that a relationship exists between transfusion and cancer recurrence and requiring our study to prove that the null hypothesis is true. Their approach, unfortunately all too common in clinical research, is contrary to the scientific method; we all have our biases, but it is up to us to prove, rather than to disprove them. There remains a small likelihood that there is a small effect of transfusion on HCC recurrence, but our study nicely demonstrates that there probably isn’t. This study was supported by the National Natural Science Foundation of China (Nos. 81472284, 81172020 and 81372262) and State Key Project on Infectious Diseases of China (2012ZX10002-016). The authors declared that they do not have anything to disclose regarding funding or conflict of interest with respect to this manuscript.