If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address reprint requests to Michael P. Fischbein, MD PhD, Department of Cardiothoracic Surgery, Stanford University, 300 Pasteur Dr, Falk CVRB, Stanford, CA 94305.
Limited aortic root repair for acute type A dissection is associated with greater risk of proximal reoperations compared to full aortic root replacement. Surgical outcomes for patients undergoing reoperative root replacement after previous dissection repair are unknown. This study seeks to determine outcomes for these patients to further inform the debate surrounding optimal upfront management of the aortic root in acute dissection. Retrospective record review of all patients who underwent full aortic root replacement after a previous type A dissection repair operation at a tertiary academic referral center from 2004–2020 was performed. Among 57 cases of reoperative root replacement after type A repair, 35 cases included concomitant aortic arch replacements, and 21 cases involved coronary reconstruction (unilateral or bilateral modified Cabrol grafts). There were 3 acute postoperative strokes and 4 operative mortalities (composite 30-day and in-hospital deaths, 7.0%). Mid-term outcomes were equivalent for patients who required arch replacement compared to isolated proximal repairs (81.8% vs 80.6% estimated 5-year survival, median follow-up 5.53 years. Reoperative root replacement after index type A dissection repairs, including those with concomitant aortic arch replacement and/or coronary reconstruction is achievable with acceptable outcomes at an experienced aortic center.
Complex residual and de novo aortic root pathology after index type A dissection repair can be safely addressed during reoperation.
Perspective Statement
Risk-benefit assessments of optimal aortic root management (limited repair vs replace) during acute type A dissection repair are complicated by uncertain risks for reoperations following limited repair. This study reports the largest single-institution series of reoperative root replacements following dissection repair, highlighting the safety of this complex reoperation at an experienced center.
INTRODUCTION
The principal objective when operating on acute type A aortic dissection (ATAAD) patients is to prevent mortality from aortic rupture and/or end organ malperfusion. While surgical replacement of the ascending aorta with a partial arch replacement remains the standard of care, the extent of proximal aortic repair (limited root repair versus full root replacement) remains a topic of debate. Single-center case series have documented similar early outcomes for limited root repairs,
Given the increased complexity of full aortic root replacement, a limited repair strategy may be advisable in the absence of clear indications to replace (eg, root aneurysm, known connective tissue disorder, destroyed valve commissures), particularly in centers without extensive thoracic aortic surgical experience. We have previously advocated referral of dissection patients to specialized, high volume aortic centers (defined as those performing at least 30–40 aortic procedures annually by 2022 ACC/AHA guidelines
2022 ACC/AHA Guideline for the Diagnosis and Management of Aortic Disease: A Report of the American Heart Association/American College of Cardiology Joint Committee on Clinical Practice Guidelines.
In contrast to the nondissected thoracic aortic aneurysm, there is limited data advising when to proceed with redo aortic root repair for progressive aortic root pathology after type A dissection. Proximal aortic root reoperative rates increase with duration and depend on index root repair management strategy
Short- and long-term outcomes of aortic root repair and replacement in patients undergoing acute type A aortic dissection repair: Twenty-year experience.
Moreover, there are limited reports of surgical risk and long-term outcomes for aortic root reintervention after previous initial Type A dissection repair. These reoperations, which frequently encompass complex repair of mixed aortic root pathologies (eg, aneurysmal dilation, aortic valve insufficiency, suture line pseudoaneurysm) and concurrent arch reconstruction, present multiple technical challenges owing to both general risks of sternal reentry and frequent presence of felt material and biologic adhesive glue utilized during the initial root repair. Here, to assess the procedural and long-term risks for reoperative aortic root replacement following prior type A aortic dissection repair, we retrospectively reviewed the contemporary experience of 57 consecutive cases at a single high volume academic center.
PATIENTS AND METHODS
Human Subject Research
All elements of this study were prospectively approved with waiver of individual patient consent by the Institutional Review Board (IRB) at Stanford University (IRB: #50520 Approved 09/14/2021)
Patient Selection
Retrospective medical record review was performed for patients who underwent aortic surgery at Stanford University Medical Center between 2004 and 2020. Patients who had a prior operation for acute type A aortic dissection and subsequently underwent full aortic root replacement (CPT code 33863 or 33864) were included in the study. Limited root repair and isolated root pseudoaneurysm repair operations were excluded. Clinical and operative data were obtained from the electronic medical record and institutional Society of Thoracic Surgeons (STS) database. Postoperative follow-up was determined by electronic health record review and telephone follow-up.
Statistical Methods
Statistical analysis was processed in RStudio 1.4.1106 (RStudio, Inc., Boston, MA). All categorical variables are presented as a ratio (percentage) and compared using the Pearson's Chi-squared test. Continuous variables are presented as median [interquartile range] and compared with the Wilcoxon rank-sum test. Kaplan-Meier estimates were determined from last known survival date and date of death from the medical record or direct follow up for patients who were missing current follow up. Kaplan-Meier estimates were generated using the survminer R package.
RESULTS
Out of 1272 full aortic root replacement cases, 209 were reoperative (prior sternotomy for any indication) including 57 patients with prior acute type A aortic dissection operations (Fig. 1A). Within this cohort, most cases were referred to our center after index operations at outside centers (84%, n = 49) and underwent elective reoperation (77%, n = 44). Indications for aortic root reintervention (not mutually exclusive) included sinus of Valsalva aneurysm (72%, n = 41), aortic regurgitation (23%, n = 13), suture line pseudoaneurysm (26%, n = 15), endocarditis/graft infection (7%, n = 4), and degeneration of prior aortic root prosthesis (4%, n = 2, Fig. 1B). Supracoronary replacement of the ascending aorta with or without aortic arch replacement was the most common initial operative strategy (86%, n = 49), while 8 patients had composite aortic root replacement during initial repair (14%). The interval from index operation to reoperative aortic root replacement ranged from 2 days to 32 years (median 5 years, IQR 3–10 years). Median age at reoperation was 53 years (IQR 47–64) and 84% of patients were male. Patients with connective tissue disorders comprised 26% of cases (n = 9 Marfan syndrome, n = 5 Loeys-Dietz syndrome, n = 1 undetermined genotype). Table 1 displays preoperative clinical characteristics for the study cohort.
Figure 1Study population (A) Workflow for identification of reoperative redo root after type A dissection cases. (B) Nonexclusive indications for aortic root replacement among cases. SoV, Sinus of Valsalva; PSA, pseudoaneurysm; AR, aortic regurgitation; SVD, structural valve degeneration. (C) Aortic root prosthesis selection at time of reoperation. mCVG, mechanical composite valve-graft; bCVG, bioprosthetic composite valve-graft; VSARR, valve-sparing root replacement. (Color version of figure is available online at http://www.semthorcardiovascsurg.com.)
Aortic root replacement strategies at the time of reoperation primarily consisted of composite valve graft (CVG) in 84.2% of patients (n = 29 with mechanical valve, n = 19 bioprosthetic). Mechanical CVGs were implanted into patients significantly younger than those receiving bioprosthetic conduits (median 49.1 years, [IQR 43.0–53.5] vs 66.3 years [IQR 59.6–72.5], P < 0.001). The remainder of cases had either a valve-sparing reimplantation root replacement (n = 4, 7.0%), cryopreserved homograft (n = 3, 5.3%), or freestyle bioprosthesis (n = 2, 3.5%, Fig. 1C). These less-frequently utilized prostheses reflected a combination of surgical era (homograft and freestyle used only in 2012 and earlier, valve-sparing technique only after 2013) and individual surgeon preference. Concomitant arch operations were performed in 35 cases (61.4%) including peninsula-style partial arch replacement (n = 15), total arch branched graft (n = 10), and total arch with elephant trunk (n = 10). Additional concomitant cardiac operations were infrequent (Table 2). Axillary or innominate chimney grafts were primarily utilized for arterial perfusion (75.4%). Concomitant arch intervention was associated with longer cardiopulmonary bypass and aortic cross-clamp duration (Table 3). Circulatory arrest during arch replacement cases was used with selective antegrade cerebral perfusion (SACP, n = 28 cases), retrograde cerebral perfusion (RCP, n = 2 cases) or combined SACP/RCP (n = 5 cases). There was no significant difference in rate of intraoperative RBC transfusion or other blood product administration for patients undergoing arch reconstruction compared to isolated aortic root operations.
Postoperative morbidity data are presented in Table 4. Nine patients required exploration for postoperative bleeding. Intraoperative intra-aortic balloon pumps were placed in 5 patients for difficulty weaning from cardiopulmonary bypass; these were placed via femoral access in 4 patients without residual dissection in the descending thoracic aorta and antegrade through the aortic graft to ensure true lumen placement in a single case with chronic residual dissection. Mechanical ventilation greater than 24 hours was required in 26 patients (45.6%) and 1 patient required tracheostomy for long term ventilation. Renal failure requiring new hemodialysis occurred in 3 patients. Hemorrhagic strokes occurred in 3 patients and there were 4 operative mortalities (7.0%) including 2 isolated proximal repairs and 2 concomitant root/total arch repairs (details in Table 5). Median ICU length of stay was 4 nights (IQR 3–7) and median total hospital length of stay was 10 days (IQR 7–16).
Following reoperative aortic root replacement, there was no significant difference in 5-year actuarial survival for patients who underwent concomitant arch repairs (81.8%) compared to isolated root operations (80.6%) with median follow up of 5.53 [IQR: 3.52–9.97] years, (Fig. 2). Seven patients underwent subsequent endovascular or open thoracoabdominal repairs for enlarging distal aneurysm at Stanford during the follow up period, including 4 patients with Marfan syndrome (Table 6).
Figure 2Long term survival for root after prior type A dissection repair cohort. (A) Survival for total cohort. (B) Survival data split by concomitant arch replacement status. P value indicates result for logrank test comparing patients undergoing concomitant arch replacements vs isolated proximal repairs. Shading indicates 95% confidence interval. (Color version of figure is available online at http://www.semthorcardiovascsurg.com.)
Coronary artery reconstruction was performed during reoperative root replacement in 21 cases (36.8%): bilateral woven Dacron Cabrol extension grafts were used in 11 cases, unilateral Cabrol graft was used for the left main coronary artery reimplantation in 5 cases and the right coronary in 2 cases, and a short saphenous vein interposition graft to the left main was utilized in 3 cases. There were no acute coronary graft complications in-hospital. One patient developed complete occlusion of the right coronary artery Cabrol limb 9 years after repair and all other grafts remained patent on surveillance imaging throughout the follow up period.
DISCUSSION
While surgical management of the aortic root in acute type A dissection has been widely debated, outcomes data for reoperative aortic root replacement remains limited. Similar cases within broader reports of single-institution thoracic aortic surgical experience demonstrated in-hospital mortality ranging from 6.4 to 23.5%, though these series were limited to small numbers of patients
In the 63 proximal aortic repairs reviewed by Wang et al. (encompassing aortic valve, root, and ascending operations), they reported similar stroke and operative mortality rates to our experience (6.4% and 7.0%, respectively). Reporting the largest series of full aortic root replacement after prior ATAAD repair, we describe a 7.0% operative mortality and reasonable mid-term survival outcomes (81.3% estimated survival at 5 years). These results suggest reoperative root replacement after prior ATAAD repair can be performed safely at experienced aortic centers within the limitations of a small total sample size (57 patients). These outcomes are in line with previous, larger single-center series of broader reoperative root replacement following prior cardiac surgery
despite the high frequency of concurrent coronary artery and aortic arch reconstruction in these chronically dissected roots. We highlight several key points regarding the surgical management of these patients.
Coronary Button Reimplantation
Dissecting out the coronary buttons remains the critical part of the operation, often made difficult due to the presence of felt and biologic glue. Our primary strategy is to mobilize and directly reimplant the coronary buttons, which may require dissection from inside the root. When button mobilization is not possible, we utilize modified Cabrol extension grafts of woven dacron—a reconstructive strategy employed in 36.8% cases. This high rate of indirect coronary reimplantation reflects the challenges of reoperation on the dissected root and frequently the presence of dense nonbiologic material (eg, felt, bioglue). When required, the left main button can often be reimplanted with a short length (few corrugations) of Dacron, and we usually redirect the right coronary to the neo-noncoronary sinus during reconstruction. It is critical to fully dissect out the small, scarred root to prevent extrinsic compression of Cabrol grafts when the heart is filled and pulsatile. Coronary reconstruction rates vary among clinical series by institution, ranging from 3.2 to 23% for all reoperative root replacements
with higher rates in ‘true’ redo root operations (prior full root replacement). In the Toronto experience of 473 reoperative root replacements, 23% required indirect coronary reimplantation, which was an independent predictor of early morbidity and mortality in their series. They report avoidance of full Cabrol reconstruction (single graft to both coronaries), instead utilizing separate interposition grafts with approximately equal utilization of saphenous vein and synthetic grafts. In our cohort, we observed no acute coronary complications and a single late graft occlusion in extended follow up after Cabrol reconstructions. These outcomes are consistent with long-term patency data for modified Cabrol extension coronary reconstruction grafts from the Yale
Here, we found similar operative and long-term outcomes for patients undergoing isolated aortic root reintervention compared to combined root and arch replacement (hemi- and total arch operations). Although most cases in this series had an unclamped distal anastomosis of some extent during index repair, in many cases arch aneurysm or residual intimal tear was addressed during the reoperation. Our approach to these cases typically encompasses hypothermic circulatory arrest and unilateral SACP via the right axillary artery with a low threshold to add a left carotid perfusion catheter in the setting of reduced left cerebral oxygen saturations monitored using near-infrared spectroscopy. Patients in this cohort undergoing redo arch replacements constituted a heterogeneous population with surgical indications including suture line pseudoaneurysm, arch aneurysm, and residual arch primary intimal tears. In some cases, total arch replacement and distal elephant trunk deployment was utilized to address expanding residual dissection in the distal aorta, either to cover a large fenestration or set up further endovascular treatment of the descending aorta. Though limited to a small heterogeneous cohort, this experience suggests that single-stage total mediastinal aortic replacement is a feasible operative strategy for complex reoperative aortic dissection cases.
Implications for Index Repair and Follow Up
Robust outcomes data for reoperative aortic root replacement after type A dissection repair provides a useful adjunct to the broader discussion of operative decision-making during index repairs. While retrospective analyses comparing root replacement to limited repair may be confounded by selection bias, multiple studies have identified heightened risk of reoperation for limited repairs.
In our institutional experience, 11.8% of patients who underwent limited root repair required proximal reoperations at 10 years with a median interval to reoperation of 2.3 years.
While certain intraoperative findings (root aneurysm, young patients with suspected connective tissue disorder) should prompt the surgeon to strongly consider full root replacement, others (eg, root rupture, destroyed sinus tissues, coronary artery dissection) may mandate root replacement in order to complete repair. Our general philosophy is to aggressively perform root replacement (including valve-sparing approaches) during index repair in these cases. We note that 26% of the reoperative cases in this series had connective tissue disorders, consistent with the recognized high rate of root reoperation for these patients after supracoronary repairs.
Though these diagnoses are not always known at the time of aortic dissection, this finding further underscores the importance of committing to full aortic root replacement when there is clinical suspicion for Marfan or Loeys Dietz syndrome. In cases without these clear indications to replace the root, we maintain that limited root repair and referral to an aortic center for follow up is a reasonable approach, particularly in emergency settings performed at centers with limited experience and comfort with elective root replacement. While minimizing the need for reoperation is important, the ultimate goal of index dissection repair must always be a surviving patient. We posit that the excellent outcomes for the small subset of patients that ultimately do require root reoperations further support this premise.
Surveillance imaging is a critical component of long-term management after aortic dissection repair.
While the operative indications and intervals from acute dissection in this series vary, we note that a quarter of cases presented over a decade after index repair. These prolonged intervals underscore the need for lifelong surveillance imaging in patients with residual chronic dissection or aortopathy after initial operative management of type A dissection. Ultimately, the decision to reoperate on the aortic root is multifactorial—in some cases prevention of sudden death from aneurysm or pseudoaneurysm rupture is the primary motivation whereas the insidious onset of heart failure from chronic aortic regurgitation may prompt surgery in others. In the absence of abundant clinical experience from which to generate guidelines, sound clinical judgement must also be deployed in decisions to not engage in a reoperative root on a case-by-case basis (eg, on slowly enlarging dissected roots or stable pseudoaneurysms in elderly or frail patients).
Limitations
This study is inherently limited by its retrospective nature and the narrow focus of patients included for analysis (only full aortic root replacement after prior type A dissection). This patient population is also certainly subject to substantial selection bias as patients who underwent surgery represent those deemed healthy enough to survive long enough to require a reoperation (up to decades after index surgery) and fit enough to be offered surgery for root pathology. Given that most of these patients are referred to our center from outside health systems, it is impossible to know what fraction of patients undergoing follow up after type A dissection repairs this cohort represents. Finally, the small sample size and heterogeneous operative extent within this series limits the utility of statistical methods to determine risk factors for adverse outcomes during these infrequently encountered reoperations.
Conclusion
In summary, we report the largest single institution experience with reoperative aortic root replacement in patients with prior type A dissection repair. Despite technically complex operations frequently involving coronary ostial reconstruction and concomitant arch repairs, we report excellent operative outcomes and long-term survival (Fig. 3).
Figure 3Graphical abstract. This study reviews single-center experience for patients undergoing reoperative aortic root replacement after a previous type A dissection repair, encompassing 57 cases. Overall estimated survival at 5 years was 81.3% with median follow-up of 5.53 years. (Color version of figure is available online at http://www.semthorcardiovascsurg.com.)
2022 ACC/AHA Guideline for the Diagnosis and Management of Aortic Disease: A Report of the American Heart Association/American College of Cardiology Joint Committee on Clinical Practice Guidelines.
Short- and long-term outcomes of aortic root repair and replacement in patients undergoing acute type A aortic dissection repair: Twenty-year experience.
Funding: The authors have no funding sources relevant to this study to disclose.
Authors have no relevant commercial support to disclose.
IRB/Informed Consent: All elements of this study were prospectively approved with waiver of individual patient consent by the Institutional Review Board (IRB) at Stanford University (IRB: #50520 Approved 09/14/2021)
Median (IQR); n (%)
00016-3&id=ga1.jpg){kind=link}
00016-3&id=gr4.jpg){kind=link}
00016-3&id=gr1.jpg){kind=link}
00016-3&id=gr2.jpg){kind=link}
00016-3&id=gr3.jpg){kind=link}