DOI: 10.1097/SLA.0000000000000455
,
,
PMID: 24424148
Issn Print: 0003-4932
Publication Date: 2015/07/01
Facebook
Twitter
LinkedIn
Email
 |
Checking for direct PDF access through Ovid | |
Excerpt
We enjoyed reading the article by Turrini et al1 on portal vein resection during pancreaticoduodenectomy but feel that there are a number of points relating to the article that would benefit from clarification.
It would be useful to know how many pancreaticoduodenectomies were performed in total over the 10-year period in the authors' institution. Although 78 pancreaticoduodenectomies with mesenteric vein resection is commendable, the final number of 34 for the purposes of analysis seems small by comparison. It is not clear how many of the excluded patients (ie, those with a wedge venous resection) were R0.
All pancreatic surgeons face the difficult intraoperative decision of whether there is venous involvement or not. It is known that intraoperative decision-making is unreliable (in this article, a little worse than 50:50), and indeed there are very little conclusive data about the benefit of preoperative imaging in determining histologically involved margins. Was the decision to proceed with venous resection in this cohort a purely intraoperative decision or was there some preoperative imaging (computed tomography or endoscopic ultrasound) available to support the decision?
The survival curves appear to show an excess of early deaths in the group with histologically positive portal vein or superior mesenteric vein (+PV/SMV); however, the survival curves are parallel thereafter. The analysis uses the Wilcoxon test, which is biased toward early differences, compared with the more common log-rank test. Was the latter analysis performed and was it significant?
The conclusions stated by the authors are somewhat confusing. Their inclusion of K-ras evaluation, perhaps as a prelude to further research, is not clearly related to the aims or outcome of this article.
Notwithstanding the points mentioned earlier in relation to the article itself, we feel that it is important to recognize that these graphs yet again remain almost static in relation to other survival curves from numerous studies on pancreatic adenocarcinoma over the last 25 years or more2–5—in that there is early progression and death in a percentage of the patients. These are patients who have been deemed suitable for surgical cure on sound radiological and clinical characteristics, yet who drop off the curve early and all too rapidly. Irrespective of venous resection or not, perhaps surgery was not the correct initial treatment option for this group. The same argument the authors make in favor of vein resection can equally be made against it, given that uncinate margin involvement by definition will include neural and connective tissues as well and not just vascular structures exclusively, this of course being the argument for neoadjuvant therapy. We note that in both groups, almost equal numbers had neoadjuvant chemoradiation. It would be interesting if the authors could provide a subgroup analysis of this particular cohort, including what regimen exactly they received.
