LAUSR

We read with great interest the article from Winzer et al. [1] entitled ‘‘Local Intra-arterial Vasodilator Infusion in Non-Occlusive Mesenteric Ischemia Significantly Increases Survival Rate.’’ First of all, we commend the authors for their management of these very severe and challenging patients and thank them for sharing their experience. On a general matter, the pathophysiology of non-occlusive mesenteric ischemia (NOMI) questions the use of local intra-arterial bowel treatments. In occlusive mesenteric ischemia, gut ischemia drives the evolution toward bowel necrosis, organ failure, and death when left untreated. Restoring gut perfusion is, therefore, of utmost importance because it targets the cause of the disease. In NOMI, it is the severe underlying condition that causes gut ischemia and not the other way around. The gut is only part of the multi-organ failure that also affects the liver and kidneys. Therefore, if attempting to improve patients’ prognosis with local treatments is a valid therapeutic option, but it only deals with one part of the problem. Although the authors report interesting results, significant methodological concerns question their interpretation. First, and as pointed out by the authors, groups that received different treatments were hardly comparable. Except for the level of lactates, other liver and kidney laboratory tests associated with NOMI prognosis were not provided [2]. Retrospective assessment of treatment efficacy is challenging, of course, because many factors may have influenced the decision to treat or not. As a result, many of these factors would need to be controlled in the analysis (e.g., need for surgery, diagnostic criteria of NOMI, fluid resuscitation protocols, presence and extent of necrosis). Second, this study failed to control immortal time bias [3]. As patients had to be alive to receive vasodilators, the assessment of their survival is likely biased and overestimated. Using the treatment variable as a time-dependent variable in the Cox regression model may control this bias. In conclusion, while local intra-arterial vasodilator infusion may improve NOMI patients, we feel this study fails to close the debate. We would be grateful if the authors could provide a further adjusted/matched analysis (e.g., propensity score matching analysis) for controlling immortal time bias.