COMMENT
Before advent of the ET procedure, the earlier techniques (clamping the descending aorta or aortic arch for Stage II, or performing Stage II under DHCA through a left thoracotomy) carried a significant risk of hemorrhage. Hemorrhage could easily occur during dissection of the enlarged, fragile upper descending aorta, especially after inflammation from a preceding aortic arch replacement. The back wall of the aorta or the pulmonary artery could easily be injured in surrounding the upper descending aorta.[11]. Multiple ET modifications have been suggested and tried, and reports on the mid and long-term post-operative mortality are becoming available.
One-step repair of the entire aorta through a clamshell [12], trans-mediastinal [13], or left posterolateral thoracotomy [14] have been attempted in order to avoid the need for two separate operations. However these challenging techniques have achieved only limited clinical application due to multiple factors, including long operating time, higher pulmonary complications rates (ranging from 15 to 50%), and near inability to repair segments below the diaphragm [15].
The staged ET procedure is thought to decrease the pulmonary complications and negates the need for proximal descending aorta dissection, hence avoiding injury to adjacent anatomic structures such as the pulmonary artery, esophagus, and phrenic nerve [4].
Frozen elephant trunk is another technique introduced in 2003 [6] utilizing a custom-made prosthesis. The proximal portion consists of a Dacron sleeve for traditional surgical anastomosis; and the distal part consists of a stent graft [16]. Experience with the frozen elephant trunk is rapidly accumulating. It can be said, in summary, that the initial operation is generally well tolerated, but a risk of paraplegia is incurred, especially with all but the shortest stent portions [17]. In contrast, paraplegia is quite unlikely with a non-frozen Stage I elephant trunk (except with very long elephant trunks) . In support of the frozen elephant trunk option is the issue that the conventional elephant trunk (used in our experience) can be difficult to retrieve and utilize as a proximal landing zone for a later endovascular extension.
Indeed, subjecting patients to two major surgeries carries real risks of morbidity and mortality. Several retrospective reports have discussed the mortality rates for each stage. In our cohort, the operative and 30-day mortality rate after Stage I in 152 patients was 3.3%, with reported rates ranging from 2% to 14% [7, 11, 18, 19]. Our inter-stage mortality rate in patients awaiting Stage II was 8.3%, with reports ranging between 2% and 11% [7, 20, 21]. This comes in a patient population with 31.6 % having undergone a previous cardiac surgery before Stage I. Fortunately, most of our patients who were planned for Stage II eventually underwent repair. We had only 1 patient who was lost to follow up and 5 patients who expired, not from rupture, advocating for earlier rather than later second stage [7]. It is notable that in the inter-stage period – beyond the first 30-days – there were no cases of aortic rupture. Our mortality rate in the second stage performed in 58 patients was 10.3%, which again is consistent with previous reports [7, 11, 18]. The continued mortality of 6 patients within the first year after completion of Stage II shows how serious is the disease of mega-aorta, often with substantial non-aortic comorbidities. In our patients, prophylactic placement of an elephant trunk was performed in 87 patients in whom the descending aorta was mildly dilated. To date, four patients eventually required repair, and a secure upper landing zone was already present for replacement of their descending aorta[21].
Aortic diameter remains a significant predictor of rupture and dissection in TAA [22, 23] however, it is difficult to predict the operative mortality and morbidity following repair. The lack of statistical significance in our variable analysis is understandable due to low number of events, which is a problem encountered in previous studies and metanalyses [24]. Our results showed increased composite adverse events in patients with prior neurologic abnormalities including prior strokes. Not surprisingly, smoking, coronary artery disease and prolonged clamp time were also factors associated with increased complications after Stage I, although this failed to reach statistical significance.
With advanced techniques of neuronal protection, the risk of spinal cord injury (SCI) and strokes has decreased. During Stage I, we used straight DHCA for cerebral protection, which we have shown to be safe and effective for a duration of 50 minutes or less [15]. We do not usually use antegrade cerebral perfusion [25]. In our study, the rate of strokes after Stage I was 4.6%.
Our paraplegia rates were low (3 patients). Routine visualization of the anterior spinal artery prior to repair may have contributed to these low paraplegia rates [26]. Utilization of a spinal drain to keep the lumbar pressure at 5 to 12 mm Hg postoperatively in descending, thoracoabdominal, and thoracic endovascular aortic repair (TEVAR) has been shown effective in reducing the risk of spinal cord injury [27].
An important point to consider has to do with choosing an appropriate length of the elephant trunk (10-15 cm) to avoid thrombus around the graft yet permit easy accessibility at Stage II. It is important to avoid entrapment of the graft in the false lumen in patients with chronic dissection [20].
Some centers apply a stent graft in the descending aorta prior to discharge instead of a later Stage II open repair [28]. Although this hybrid approach has gained popularity, there remain serious questions about the durability of these stents and the unique associated complications—especially endoleaks– often mandating later re-operation.
Enthusiasm regarding ET staged repair may have waned over the years due to complexity and fear of the inter-stage mortality; however, we feel that the inter-stage mortality rates and overall outcomes are acceptable for a disease of this severity. Long-term survival finds our results durable.