The number of elderly patients undergoing surgery has risen in recent years due to the increasing age of the general population, technical improvements operatively and improvements in postoperative care [29]. In most forms of cardiac surgery, octogenarians demonstrate favorable long-term outcomes with low mortality and significantly improved post-operative quality of life [30, 31]. This is especially true in octogenarians undergoing aortic valve surgery [30]. ATAAD on the other hand, is associated with high mortality and morbidity and octogenarian status is a significant risk factor for mortality [4]. Concerns regarding post-operative quality of life and long-term outcomes means that this age demographic is more likely to be managed non-operatively [32, 33]. Aoyama et al. demonstrated that this transition point between patients managed operatively vs non-operatively is approximately 85 years [33]. Short-term mortality is significantly higher in non-operative management of octogenarians as patients pass away rapidly from the sequelae of malperfusion [28, 33]. Once the patient survives the initial event, survival in the long-term (10 years) of the operative cohort may be comparable to the non-operative cohort [34]. Patients and family members should be well informed of the risks of surgery and postoperative recovery.
The surgical approach between both cohorts differs, with a higher incidence of complex surgery in the younger group. Older patients were more likely to undergo isolated ascending aorta replacement rather than a composite graft implantation. Both groups underwent proximal or hemiarch replacement equally often, while younger patients underwent more total aortic arch replacements. Aortic root surgery was almost exclusively performed in the younger cohort, ranging from 1 to 5% in the octogenarian cohort across most studies. An aggressive surgical approach including a full root or arch replacement is associated with an improved freedom from reoperation rate and may not be associated with increased risk in an all-ages cohort, however, this may not be applicable to an elderly cohort [35, 36]. Elderly patients are particularly vulnerable to extended CPB and circulatory arrest times. In addition, the tissue quality may also be more friable making multiple anastomoses technically challenging. Piccardo et al. demonstrated that in an octogenarian cohort, the in-hospital mortality was prohibitively high in patients undergoing total arch replacement [37]. This approach may also seem unnecessary in an elderly cohort where long-term freedom from reoperation is not a pertinent issue [37]. Minimising CPB and circulatory arrest times in an octogenarian cohort is beneficial, by performing a less extensive procedure such as an ascending aorta replacement with an interposition graft [38]. Ghazy et al. termed this a ‘defensive strategy’ and demonstrated improved heath related quality of life (HR-QOL) when elderly patients are managed in this way [39]. Adopting a strategy aimed at minimising operative, CPB, circulatory arrest and rewarming time whilst ameliorating the life-threatening complications of ATAAD is recommended in the octogenarian cohort.
There was a significantly lower number of octogenarians patients undergoing antegrade cerebral perfusion (ACP) during ATAAD repair. Cohort studies demonstrate that ACP can be safely used in an elderly population [40,41,42]. Pacini et al. demonstrated a mortality of 7% and neurological event rate of 5% in 95 elderly patients (age > 75) undergoing aortic arch surgery, with ACP [40]. A limitation is that these outcomes may not extend to emergencies such as an ATAAD. Studies that assess ATAAD demonstrate a mortality benefit utilising the axillary artery as a preferred site of arterial inflow [42]. Future studies assessing an elderly population are warranted.
In the current study, early mortality was twice as high in the octogenarian cohort. This result is consistent with previous published systematic reviews. Biancari et al. (2011) demonstrated a pooled mortality of 36.7% in the octogenarian cohort, with octogenarians twice as likely to die in the short-term. Bruno et al. assessed an elderly population (age > 70), demonstrating an odds ratio of 2.25 and a pooled mortality rate of 19.9%. Since these two systematic reviews, a number of large contemporary retrospective studies have been published comparing octogenarians to non-octogenarians. This study demonstrated that octogenarians were twice as likely to die in the short-term after surgery for ATAAD. There are a number of possible factors that can explain this. Firstly, elderly patients tend to show fewer symptoms such as pain and present later, with more extensive dissections [20]. Secondly, perioperative complications are observed in higher frequency, such as cardiac tamponade, supra-aortic branch vessel involvement and dissections extending to the abdominal aorta [20, 22]. Octogenarians with risk factors such as malperfusion, heart failure or root involvement have a significantly higher mortality [17, 18, 43]. Thirdly, octogenarians do not have the same physiological reserve as younger patients, and do not recover from postoperative sequelae as easily. Our subgroup analysis of study location demonstrated a significantly lower mortality in Japanese studies compared to non-Japanese studies. Octogenarians account for 20% of patients undergoing surgery in Japan and this proportion is increasing [44]. The improved mortality may be a result of the high life expectancy of the Japanese population, lower incidence of atherosclerosis and extensive dissection [45]. Dissections involving this population are more likely to be less complex, Type-B with a retrograde component with fewer entry tears located in ascending aorta [45].
There were no significant differences in three out of the four of the secondary endpoints. The rate of stroke and ARF in both cohorts were comparable, with low heterogeneity between the studies. This result is consistent with previously published literature [8]. Though, the presence of postoperative ARF or stroke is associated with higher mortality in the elderly cohorts [46]. The results of ICU LOS were varied across the included studies and there was no significant difference between the two cohorts on pooled analysis. It should be noted that two high quality studies reported a longer ICU LOS in the elderly cohort [17, 20]. Elderly patients may require more time to wean from the ventilator and demonstrate protracted time to full neurological recovery postoperatively and therefore spend more time in ICU. None of the included studies demonstrated a significantly higher stroke rate in either population group. This result is in line with published data from large registry databases whereby the incidence of postoperative stroke was unrelated to age [47]. Causes of re-exploration include uncontrolled bleeding, tamponade and infection. This study did not demonstrate a significant difference in rates of re-exploration between either cohort. An element of selection bias may account for this, as there may have been higher threshold to manage unwell octogenarians (more prone to coagulopathy) operatively.
Actuarial survival was significantly lower in the octogenarian cohort, with a five-year survival in the octogenarian cohort of 54% compared to 76% in the non-octogenarian cohort. There was also significantly less attrition in the younger cohort over time. Studies report that the majority of deaths in the octogenarian cohort were caused by non-aortic events such as pneumonia and frailty, which may be more common postoperatively [18]. Two studies compared octogenarians post-surgical repair of ATAAD to an age-matched sample, demonstrating a significantly lower actuarial survival revealing the impact of surgery in this demographic [17, 26]. This current study demonstrated that octogenarians have a significantly lower mid-term survival after surgery for ATAAD, compared to non-octogenarians. Few studies compare non-operative to operative management of ATAAD in the elderly. In the short-term, survival of those surgically treated was superior [34, 43, 46]. There is a paucity of evidence assessing long-term outcomes of surgically treated ATAAD in the elderly cohort. The limited data available demonstrates that in the mid-term, outcomes are significantly worse in the non-operative arm, however, overall survival in the long-term (10 years) becomes comparable between both arms [34, 48, 49].
Two studies included in this review assessed HR-QOL outcomes in the elderly cohort [14, 26]. This is a useful indicator as it captures information on the physical and mental health status of a patient. Tang et al. compared octogenarians to non-octogenarians, demonstrating significantly lower physical functioning scores in the octogenarian cohort [26]. Bojko et al. demonstrated the same result, however this did not reach significance [14]. Studies that assess postoperative HR-QOL in the elderly following ATAAD repair also demonstrated significant attrition of physical health [50, 51]. Elderly patients have a greater degree of frailty and are also more vulnerable to the cerebral insult from deep hypothermic circulatory arrest which is required for some ATAAD repairs. As a result, the elderly cohort are more prone to lasting physical limitations of surgery compared to younger patients. There is a paucity of evidence assessing long term HR-QOL in octogenarians post-ATAAD repair. The availability of this data would be useful in making decision whether operative management is beneficial to octogenarians.
Limitations
The main limitation of this study is the degree of study heterogeneity. All studies were retrospective in nature with inherent biases in design. A number of studies did not report whether octogenarians were recruited consecutively with resultant selection bias as only ‘well’ octogenarians may have been selected for surgery. Nine studies reported consecutive recruitment of octogenarians for surgery, with the exception of patients who were moribund or refused surgery [18]. This bias may have also extended to the recruitment of stable and robust octogenarians only. We aimed to reduce this bias by conducting a subgroup analysis of studies which stated consecutive recruitment of patients in the methodology. Furthermore, there was insufficient data reporting the use of open and closed distal anastomoses, DHCA times, FET in octogenarian populations, limiting analysis in this area. There was also variable reporting of cerebral perfusion strategies, which is an important measure as it has an impact on outcome [52]. Studies that ensure that relevant datapoints are collected pre- and post-operatively will minimize bias. Such studies are of course, difficult to facilitate in an emergent setting such as an ATAAD.
Population samples also demonstrate significant heterogeneity. All studies are from developed countries with significantly longer life expectancies, and therefore results of this study cannot be extrapolated to developing countries. The majority of studies come from Japan, which has a high life expectancy amongst developed countries [53]. Population samples also varied with time; six studies recruited patients from the 1990’s. Evidence suggests that operative mortality has improved significantly since these recruitment periods [6]. This is reflected in the mortality rate reported in these earlier studies, with studies recruiting from this period reporting a significantly higher mortality rate [15, 19]. We aimed to minimize this bias by including studies from 2000 onwards. Operative technique also varied significantly between the studies, where some studies adopted a defensive strategy and others had a high proportion of total arch repair and aortic root surgery in the elderly population. This would therefore have a considerable impact on primary and secondary endpoints.