We read the recent report entitled Utility of the Penn Classification in Predicting Outcomes of Surgery for Acute Type A Aortic Dissection that was published in The American Journal of Cardiology , in which Kimura et al investigated whether Penn classes predict outcomes after surgery for acute type A aortic dissection (AAAD) by comparing the hospital outcomes of 351 patients divided into 4 groups based on Penn classification. The investigators proposed that morbidity and mortality were high in patients with generalized ischemia (Penn Ac and Penn Abc groups). In addition, they recommended that changing the surgical strategies based on the Penn classification is important. A more aggressive surgical approach, such as total arch replacement, should be performed in Penn class Aa cases, especially if patients are young. In contrast, we agree with the comment by Kimura et al that prolonged operation time might increase the already high operative risk through infection (e.g., mesenteric ischemia) or multiorgan failure (e.g., Penn Ac and Penn Abc groups) as our recent reports pointed out. Therefore, the hospital outcomes of surgery in patients in Penn Ac and Penn Abc groups might be improved through less invasive hemiarch replacement based on patient’s age and preoperative co-morbidity. Kimura et al have also pointed out that the mortality rate of Penn Aa group was 2.7% in their study cohort; in addition, 3.1% in the Penn group and 13.7% in the Karolinska group. This is in line with the observation that we have (1.6%, 1 of 63) in patients who presented with stable hemodynamics associated with no significant electrocardiographic abnormalities. Furthermore, the mortality rate of Penn Abc group (branch vessel malperfusion and circulatory collapse) was 21.6% in the study by Kimura et al, 40.0% in the Penn group, and 44.4% in the Karolinska group, which was very similar to the mortality rate (41.2%) of our patient group who had preoperative shock combined with myocardial ischemia on electrocardiogram. Thus, the validity of the Penn classification is further supported by our findings. However, this well-organized study could still benefit from further discussion on the patients with branch vessel malperfusion with ischemia, so-called Penn class Ab. First, Kimura et al showed excellent surgical outcomes for AAAD patients with mesenteric ischemia (in-hospital mortality: 6%, 1 of 17). AAAD complicated with mesenteric malperfusion is an uncommon but ominous complication carrying a high risk of in-hospital mortality. The recent reports demonstrated that cases with mesenteric malperfusion were associated with dismal mortality rates (75% [6 of 8] from a German group and 63.3% [43 of 68] from the International Registry of Acute Aortic Dissection, respectively). We advocate Kimura et al that individualized and detailed surgical strategies should be carried out on AAAD patients with mesenteric malperfusion based on their excellent results in this special patient group. Second, regarding the brain malperfusion, it has been acknowledged that a wide spectrum of symptoms is associated with brain malperfusion. Conscious patients with a transient or even persistent neurological deficit preoperatively usually have a similar operative and longer term survival as patients without neurological deficit. In contrast, patients with dense deficits or coma have a poor postoperative prognosis, and resuscitative surgery in such patients might be considered futile. Nonetheless, neurological recovery has been reported in cases in which rapid reperfusion on comatous patients within 10 or 12 hours of presentation is achieved. The most comprehensive study of patients with AAAD presenting with major brain injury was based on the International Registry of Acute Aortic Dissection database, which included 87 patients with cerebrovascular accident and 54 patients with coma. The overall hospital mortality was 22.7% without brain injury, 40.2% with cerebrovascular accident, and 63.0% with coma. Thus, brain injury at presentation significantly (p <0.001) affects hospital survival of patients with AAAD. Third, coronary circulation is not an uncommon site of malperfusion reported in the literature with an approximate incidence of 6% to 19%. Kimura et al shared a similar experience that we have in regard to the incidence of coronary malperfusion (6.3%, 22 of 351 vs 9.9%, 14 of 142, p = 0.17) and in-hospital mortality (27.3%, 6 of 22 vs 21.4%, 3 of 14, p = 0.69). In the presence of coronary malperfusion, in-hospital mortality doubles, and some patients will die intraoperatively from low output syndrome. In a series from the Leipzig group, in-hospital mortality was significantly higher in patients with coronary malperfusion than in patients without it (39.4% vs 17.1%, p = 0.004). It is important to note that Kimura et al addressed that the in-hospital mortality was significantly higher in patients with coronary malperfusion. They emphasized that such organ-specific ischemia (i.e., coronary malperfusion) remains a surgical challenge and is associated with worse unfavorable outcomes.

The influence of ischemia presentation on outcomes after surgical intervention for AAAD has been studied by a number of investigators. Patients with AAAD presenting with organ-specific ischemia (including mesenteric ischemia, sustained cerebral ischemia, and coronary malperfusion) are known to carry extremely high surgical mortality rate. Finally, the clinical validity of Penn classification has been proposed by several reports. Nevertheless, we do concur with the comments by Kimura et al that Penn classification might underestimate the surgical risk of patients with these critical organ-specific ischemias. Thus, a reasonable modification for the Penn class Ab would be considered. We suggest that the Penn class Ab be subclassified as follows.

Ab-1 denotes branch vessel malperfusion with ischemia but not involving mesenteric, cerebral, or coronary.

Ab-2 denotes branch vessel malperfusion with ischemia but involving mesenteric, cerebral, or coronary.

After this modification, we think the validity of Penn classification could be easier to apply and more adequately to point out patients with higher risk of in-hospital morbidity and in-hospital mortality. For patients and their families, we can provide better understanding of the inherent preoperative risk and unmodifiable and potentially modifiable factors.