The indication for emergency surgery in a type A dissection, according to the Stanford classification, is clearly established. Likewise, it is considered that the cannulation of the right axillary artery (RAA) is the first option to initiate antegrade perfusion with cardiopulmonary bypass (CPB), being an indication IIa according to current clinical guidelines. Several studies emphasize the benefits of this type of cannulation, including that of Rosinski et al., carried out at the Cleveland Clinic, where it was observed that this cannulation route was safe and offered good results for reestablishing adequate systemic perfusion. However, although the indication is clear, it is not well established what to do in the presence of a dissected axillary artery. This scenario could pose aborting the use of this route for the initiation of systemic perfusion. This study seeks to clarify whether the use of a dissected right axillary artery (DRAA) is safe and if there are repercussions on the evolution of these patients, with the primary objective being hospital mortality and stroke with sequelae.

The paper we will review is an observational retrospective cohort study. The records of all patients who underwent type A aortic dissection surgery from January 2016 to November 2020 were reviewed. The strategy for perfusion in this center was normally direct right axillary artery cannulation unless anatomical problems (depth for the approach, small diameters, arterial calcification) or technical problems made it impossible, in which case alternative femoral cannulation was performed (in some cases double axillary and femoral cannulation), over the brachiocephalic trunk or central cannulation by the Seldinger technique. The primary endpoints were hospital mortality and established stroke. Secondary objectives included complications related to RAA cannulation.

From a purely statistical point of view, an initial univariate analysis of all variables involved in the study was carried out, followed by a bivariate analysis of each predictor variable with the response, after which the most important variables were selected to perform a multivariate analysis using a propensity score adjustment. Finally, a multivariable logistic regression was performed to define the specific contribution of each predictor.

A total of 931 type A aortic dissections were urgently intervened at the Guandong Provincial Hospital, of which only 835 patients would have an adequate study with CT. Of the 835 patients, 124 (14.9%) presented right axillary artery dissection; the rest, 711 (85.1%) did not. The majority of the patients were male in both branches. There was a greater presence of moderate-severe aortic insufficiency and poor cerebral perfusion in those patients with a DRAA.

The rate of failed cannulation was higher in patients with DRAA, but not statistically significant (2.4% vs. 0.7%, p = 0.102). Five patients presented vascular complications related to cannulation, however, all of them belonged to the group of non-dissected right axillary artery. No case of upper limb ischemia was observed after cannulation.

With a propensity analysis, the percentages of in-hospital mortality (13.4% vs. 12.5%, p = 0.842) and stroke (9.8% vs. 7.1%, p = 0.472) did not show statistically significant differences. Through a multivariable logistic regression analysis, it was concluded that age (p = 0.045), circulatory collapse (p = 0.010), coronary disease (p = 0.046), the need for coronary bypass (p = 0.012) and time on CPB (p = 0.001) were independent predictive factors for in-hospital mortality. With the same analysis, the presence of DRAA was not considered a predictive factor for hospital mortality (p = 0.431) or stroke (p= 0.276).

The conclusion reached after the analyses performed was that cannulation over a dissected right axillary artery is possible and safe at least in experienced centers.

COMMENTARY:

Despite the accumulated experience on the surgical approach in type A aortic dissections, there may be some controversy about the arterial cannulation route to reestablish systemic perfusion through true lumen, especially given the heterogeneity of cases and the variability of findings that we can encounter in imaging tests. This article by Tong et al. addresses a topic of great clinical impact, as it is part of our attitude in managing a type A dissection and specifically in how to act in the face of a dissection that progresses towards the right axillary artery, considering that it is the most commonly used arterial cannulation route.

Several experts advise not to perform a cannulation over a dissected artery due to its fragility and susceptibility to new ruptures which could produce a new reentry and an alteration of the flow with the risk of expanding the false lumen, so they would opt for an approach of a femoral artery even when it does not follow a physiological flow as it is retrograde flow. Given the absence of a consensus in this problem, there are those who mention that cannulation over a dissected right axillary artery is possible, without presenting vascular or systemic perfusion complications in CPB, which represents the hypothesis that is studied in this article.

Although the conclusions and results observed in this article, in terms of hospital mortality, established stroke or vascular complications, encourage cannulation over a dissected right axillary artery, it would be correct to clarify certain points:

First, it is a single-center study which clearly limits its external validity. Second, the statistical techniques selected for this study are correct, although a sufficient number of events is not available so that the estimates of the coefficients of each predictor variable are sufficiently precise. We assume that the absence of collinearity or autocorrelation between an excessive number of predictor variables has been tested, although the article does not comment on it, since this could be a cause of unstable estimates (wide confidence intervals of the odds ratio) of the coefficients of the predictor variables. It is noteworthy within the analysis of Tong et al., that once the adjustment by propensity analysis was made, the most appropriate thing to know the causal effect of the dissection of the right axillary artery against the non-dissection on the primary and secondary objectives, would have been necessary to carry out directly a bivariate analysis between the matched variable (DRAA vs. RAA) and the result, since the two branches (DRAA vs. RAA) are already comparable in terms of causality. Certain events such as cannulation failure and vascular complications probably should not be used as secondary outcomes due to their low incidence. In summary, a greater number of events is needed to reach statistical power and precision in the estimates. Third, the extent of the dissection on the right axillary artery is not clearly defined, so it could be that the cannulation was performed only in cases of proximal dissection from the subclavian origin. Considering that a dissected artery is very fragile and susceptible to intimal ruptures, it would draw attention to the realization of a direct cannulation and the almost null vascular complications. The work also does not specify the technique used, whether semi-Seldinger with open vessel exposure, pure percutaneous Seldinger or with the interposition of an 8 mm Dacron conduit anastomosed end-to-side. In case of extensive dissection of the axillary artery, it could be suggested the opening and intimal inspection of the artery and subsequent performance of an 8 mm Dacron conduit anastomosis, which would offer greater safety of remaining in true lumen and of avoiding intimal ruptures posterior to the cannulation point produced by a direct cannulation by Seldinger techniques. In conclusion, this is a very interesting article that encourages cannulation over a dissected right axillary artery, but it must be interpreted with caution given the limitations of the study. Finally, the technique used for the initiation of perfusion in CPB will continue to depend on the individual characteristics of each patient and even on the way of working of each center.

REFERENCE:

Tong G, Zhao S, Wu J, Sun Z, Zhuang D, Chen Z, et al. Right axillary artery cannulation in acute type A aortic dissection with involvement of the right axillary artery. J Thorac Cardiovasc Surg. 2024 Jul;168(1):50-59.e6.  https://doi.org/10.1016/j.jtcvs.2022.09.058