Data from German Registry for Acute Aortic Dissection Type A (GERAADA) [3] showed that 30% were 70 years or older in the population of 2,137 patients. Almost one-third of patients were >70 years of age in International Registry of Acute Aortic Dissection (IRAD) [1] and 19.8% were aged 80 years old or older in Japan registry of aortic dissection (JRAD) [8]. Compared with IRAD, the patients with AAD were significantly younger in Chinese registry (Sino-RAD) [9], of which 6.5% were over 70 years old and 25.4% were over 60 years old. Therefore, it can be seen from these worldwide studies that the proportion of elderly patients (>70 years of age) in China is less than that in western countries. It may be related to the fact that some elderly patients with type A AAD in rural areas die before presentation to hospital, or refuse operation owing to surgical risk and financial problem. In our center, there were only 49 patients aged over 70, accounting for 9.6% in 510 type A AAD patients who underwent surgeries from 2000 to 2015. Because the enrolled sample was too small, we could only expand the age range to ≥ 65 years old in this study.
Whether surgery is justified in elderly type A AAD patients? Evangelista A and his colleagues [1] clearly compared two groups of below and above 70 years and reported that surgical mortality was 21% in patients <70 years of age and 31% in those ≥70 years of age. GERAADA [10] concluded age to be a critical risk factor for morbidity and mortality in 1558 type A AAD patients and clearly confirmed the correlation of older age with early mortality in a large study cohort. The finding that identified age ≥70 years to be a predictor of death could also be verified in IRAD data [11]. All these literatures would suggest that surgery should be considered in all patients with type A AAD regardless of age; however, these studies also revealed that older age was an independent predictor of surgical mortality.
The controversy that still exists regarding the extent of aortic resection for repair for type A AAD in the general population is even more pertinent in elderly patients. As far as we know, Sun’s procedure (TAR+SET) has acquired huge success in China, which not only repairs of extensive parts of the aorta, including the aortic arch and descending aorta, but also has satisfied surgical outcomes and long-term results with decreased rate of re-operation and low risk of adverse aortic events. Sun and his colleagues [4~5, 12~14] reported that total arch replacement could be performed safely without increasing operative mortality and morbidity compared with hemi-arch replacement. Studies from abroad [15~19] also reported no significant differences in terms of hospital mortality between hemi-arch replacement and total arch replacement. Another meta-analysis [20] found that no statistical difference was obtained in aspects of neurological deficit, stroke, intubation >72h, and re-operation for bleeding except for postoperative renal dialysis between total arch replacement and hemi-arch replacement. But, the outcomes from our current study totally differed from the above previous studies. Here the possible reason could be pointed out is that the average age of patients in above studies was much younger than that in the current study. Therefore, we maintain that the conclusion of total arch replacement carrying the same mortality and morbidity compared with hemi-arch replacement cannot be inferred arbitrarily to elderly patients. Actually in the past decade, the indications for LAR (hemi-arch or ascending aortic replacement) or - TAR+SET for elderly were as following: 1. Basically, we resected the tears in ascending and arch as possible as we could. If no tear in arch or no limb malperfusion, we did LAR. 2. Tear in proximal descending aorta was recommended for TAR+SET operation because this kind tear could be covered by stent elephant trunk. 3. Preoperative limb malperfusion was suggested for TAR+SET operation because stent trunk could expand the true lumen and probably restore the flow to lower extremities. The most important thing we should point out is that we do not hesitate to undergo TAR+SET procedure regardless of age if patients’ condition requires. 4. The exceptional condition was that if one patient was dynamic unstable to tolerate an extensive aorta resection, we would only do LAR to rescue his or her life even if he or she needed extensive surgery. In LAR group, 3 patients with tear locating in arch and 3 locating beyond arch were in these conditions.Therefore, the most important principle is one patient one rule, that is to say, which is made upon individual patient.
Motohiko Goda [21] found that risk factors for hospital mortality in patients with type A AAD were cardiopulmonary resuscitation, renal dysfunction, and lower-extremity ischemia. Another study [22] indicated that pre-existing cardiac disease (RR=3.7, 95% CI=1.8-7.4) and cardiopulmonary resuscitation (RR=6.8, 95% CI=2.3-20.2) were independent predictors of in-hospital death for 487 type A AAD. Therefore, variables associated with patients’ characteristics and the surgical procedures, such as surgical technique, cannulation site, CPB time, cross-clamp time, and circulatory arrest time with a potential influence on mortality have been investigated a lot in type A AAD. Although most previous studies revealed that hemi-arch and total arch replacement had no significant difference on early death [15~19], it was quite different regarding hospital mortality, which was much higher (19.4%) in TAR+SET group than in LAR group (4.9%) in the current study. This was most likely due to the small number of sample in each operative method.
There is always a fear that patients would die of rupture of the residual false lumen after LAR, although the present study confirmed that elderly patients, who underwent LAR had lower surgical mortality and morbidity than those who underwent TAR+SET in the setting of type A AAD. Residual patent false lumen is a well-known risk factor for progressive aortic dilatation and poor long-term outcomes following type A AAD surgery [16, 23, 24]. Fichadiya A and his colleagues [25] analyzed that false lumen thrombosis was achieved in 57% and 9% of patients undergoing extended-arch (total arch) and hemi-arch repair, respectively. Rate of growth in the proximal descending aorta was 0.7 ± 2.3 mm/year in the extended-arch group versus 2.7 ± 3.9 mm/year in the hemi-arch group. In our present study, expansion of the true lumen and regression of the false lumen was observed in almost all the patients except 3 patients in TAR+SET group. In contrary, for LAR patients, false lumen in aortic arch or proximal descending aorta was observed in 35 discharged patients (35/39, 89.7%) with aortic aneurysm progression during follow-up. The increased diameter of proximal descending aorta was (19.5±9.3) mm, with expansion occurring at (3.1±1.6) mm/y for these 35 patients. The expansion speed is higher than that in Fichadiya A and his colleagues’ study, probably due to poor control of blood pressure in elderly patients in China. Aizawa K and his colleagues [26] compared 225 patients who underwent ascending or hemi-arch replacement and 42 underwent total arch replacement for type A AAD patients and found that the actuarial survival rates were 80.7% for ascending/hemi- arch group versus 84.3 % for total arch group after 5 years, and 66.4% for ascending/ hemi-arch group versus 74.6 % for total arch group after 10 years (p=0.94). For ascending/hemi-arch and total arch groups, reoperation-free survival rates were 72.1% versus 77.1 % after 5 years, and 62.0 versus 67.1 %, respectively, after 10 years (p=0.85). For our study, the 5-year survival rate was 82.5±6.0% in LAR group versus 75.2±5.6% in TAR+SET group (p=0.151). But, the freedom from adverse aortic events at 5-year follow-up was (84.3±6.5) % in LAR group versus (97.9±2.1) % in TAR+SET group (p=0.03), which was different from the previous study. We understand the fact that LAR did fail to achieve the objective of massive resection of dissection and realized that average age nearly 70 years old in this study could mainly highlight the shortness of LAR. Despite adverse aortic events risk in LAR group was increased; however, there was no statistically significant difference in mid-term survival rate between the two groups.
There are certain limitations of this study as follows. Firstly, the current study is a retrospective analysis with all known limitations of such a study design. Secondly, the
data come from a single center with a small sample; however, it might also decrease the potential bias because surgical, anesthetic and intensive care techniques even the data collection methods are consistent. Thirdly, the choice of procedure is not random, but determined upon single patient’s condition. The difference in aspects of limb malperfusion and tear location exists, which may lead to different procedures. Last one, it is difficult for us to set up a completely consistent control group in clinical. So, we calculated the Euro-score II for the two groups (17.1±5.5 vs. 16.7±6.3, p=0.733), which emphasizes the weight of the patients’ preoperative status.