Transcatheter aortic valve implantation in patients with anomalous origin of a coronary artery
Abstract
Transcatheter aortic valve implantation (TAVI) has become a well-established treatment option for elderly patients with symptomatic severe aortic stenosis. Coronary artery anomalies are an infrequent finding and there have only been few anecdotal reports of patients with coronary anomalies treated with TAVI. We here present a comprehensive overview of existing reports in addition to an own case series to facilitate better understanding of this potentially challenging clinical scenario.
Abbreviations
-
- ACAOS
-
- anomalous origin of a coronary artery from the opposite sinus
-
- BAV
-
- balloon aortic valvuloplasty
-
- CAA
-
- coronary artery anomalies
-
- MDCT
-
- multidetector computed tomography
-
- TAVI
-
- transcatheter aortic valve implantation
-
- THV
-
- transcatheter heart valve
1 INTRODUCTION
Transcatheter aortic valve implantation (TAVI) has become an established treatment option for severe aortic valve stenosis in elderly patients across all surgical risk categories. Preprocedural planning including coronary angiography and multi-detector computed tomography (MDCT) is widely adopted. Within the scope of this diagnostic work-up, coronary artery anomalies (CAA) are in most cases a rare incidental finding. Since CAA are infrequent per se, management of those patients with concomitant aortic stenosis is only derived from a few case reports. However, this clinical entity requires careful attention as it might be associated with an elevated risk of peri-interventional coronary obstruction.
We therefore performed a literature search on existing reports and added a case series from our institution hereby providing the largest available analysis on CAA in patients undergoing TAVI. We sought to address uncertainties concerning preprocedural planning and to highlight potential pitfalls described in the collected cases. A better understanding of this rare scenario may help the operator to decide whether preventive measures must be implemented to optimize procedural safety and success when encountering patients with this unique anatomic feature.
2 METHODS
We performed a systematic literature search in PubMed and Google Scholar for reports published on CAA in patients undergoing TAVI between 2011 and 2022. The following search terms were used: transcatheter aortic valve replacement, TAVI, coronary artery anomaly, coronary obstruction, anomalous coronary artery. After having identified relevant articles, we used those papers as “start set” for further research by using their reference list or the citations to identify more relevant articles. This method is referred to as backward respectively forward snowballing.
In 2011, Weich et al. described for the first time a successful TAVI in a patient with malignant anomalous origin of the right coronary artery.1 Between 2011 and 2022, 17 publications including a total of 20 patients who had CAA and underwent TAVI were identified.1-17 Each publication was analyzed regarding CAA type and course of the anomalous coronary artery, procedural characteristics including the choice of the transcatheter heart valve (THV), whether coronary protection was used or not and if coronary obstruction did ultimately occur after TAVI.
In addition, we screened our institutional database and retrospectively identified 20 consecutive patients with CAA treated by TAVI between 2016 and 2022. Patient and procedural data were prospectively collected and entered into a dedicated institutional database as part of the Leipzig TAVI registry (NCT05015452). Manifestation of CAA by MDCT was documented separately. CAA was classified as previously described by Angelini et al.18 The principal clinical endpoint of interest was the occurrence of coronary obstruction.
3 RESULTS
Patient data from the published experience and our local institution are presented in Tables 1 and 2. There were 20 cases of CAA through literature research and 20 cases in the institutional database identified; 50% of patients were male and the pooled mean age was 80.6 ± 7.3 years. The most common CAA in the total cohort was anomalous origin of a coronary artery from the opposite sinus (ACAOS) (75%, 30/40). Of these 30 patients with ACAOS, 17 had the left circumflex coronary artery, 3 the left anterior descending and 4 the left main coronary artery originating from an opposite sinus. In 6 cases the right coronary artery originated from the left coronary sinus. Seven cases of a single coronary artery were included. A retro-aortic course was present in 22 cases and an interarterial trajectory in 13 cases. CAA detected in the majority of the cases presented here are shown in Figure 1.
Patient [Ref] | Age | Sex | Coronary anomaly | Arterial course | Valve type | Size | Protection strategy | Coronary obstruction | comments |
---|---|---|---|---|---|---|---|---|---|
15 | 86 | F | ACAOS LCx from the RCS | Retro-aortic | Sapien XT | 26 | Guiding catheter + wire LCx | Yes | BMS-Implantation, valve oversizing |
27 | 81 | F | ACAOS LMT from the RCS | Retro-aortic | Lotus | 23 | Wire + provisional balloon | Yes | ECMO, Stent-Implantation |
39 | 86 | F | ACAOS LCx from the RCS | Interarterial | BAV | 23 | Wire in LCx | Yes | Refrained from TAVI |
413 | 76 | M | ACAOS RCA from the LCS | Interarterial | Sapien 3 | 26 | None | Yes | Emergency coronary bypass graft |
515 | 89 | M | LMT from the non-coronary sinus | Retro-aortic | Evolut Pro+ | 29 | 2 wires in LMT | Yes | DES-Implantation, Bicuspid valve |
61 | 79 | M | ACAOS RCA from LCS | Interarterial | Sapien | 26 | None | No | First description |
72 | 84 | M | ACAOS LCx from the RCA | Retro-aortic | CoreValve | 26 | None | No | |
83 | 87 | F | SCA from the RCS | LCA retro-aortic | Sapien XT | 23 | None | No | |
93 | 89 | F | SCA from the RCS | LCA retro-aortic | CoreValve | 26 | None | No | |
104 | 88 | F | SCA with LMT from RCC | Interarterial | Sapien | ? | Wire in LAD | No | |
116 | 81 | F | SCA from the RCS | CoreValve | 26 | None | No | ||
128 | 77 | F | ACAOS LAD from the RCS | Retro-aortic | Sapien 3 | 26 | None | No | |
1310 | 84 | M | ACAOS LCx from the RCS | Retro-aortic | Acurate Neo | M | None | No | |
1411 | 80 | M | ACAOS LCx from the RCS | Retro-aortic | Sapien 3 | 29 | Wire in LCx + RCA, provisional balloon LCx | No | |
1511 | 87 | M | ACAOS LCx from the RCS | Retro-aortic | Sapien 3 | ? | None | No | |
1611 | 80 | M | ACAOS LCx from the RCS | Retro-aortic | Sapien 3 | ? | None | No | |
1712 | 72 | M | ACAOS LCx from the RCS | Retro-aortic | Sapien 3 | 26 | Wire + guideliner + provisional stent LCx | No | Bicuspid valve |
1814 | 65 | F | SCA | Evolut Pro | 26 | None | No | Bicuspid valve | |
1916 | 82 | F | ACAOS LCx from the RCS | Retro-aortic | Sapien 3 | ? | Wire in LCx | No | |
2017 | 82 | F | ACAOS LCA from the non-coronary sinus | Retro-aortic | Acurate Neo2 | S | Wire + provisional stent LCA | No | Intentional misalignment |
21 | 66 | M | ACAOS LCx from the RCS | Retro-aortic | Sapien 3 | 29 | None | No | |
22 | 83 | F | ACAOS RCA from the LCS | Interarterial | Portico | 25 | None | No | |
23 | 83 | F | ACAOS LCx from the RCS | Retro-aortic | Evolut R | 29 | None | No | Pericardial effusion |
24 | 55 | M | ACAOS LCx from the RCS | Retro-aortic | Sapien 3 | 29 | Wire in LCx | No | |
25 | 82 | M | ACAOS RCA from the LCS | Interarterial | Symetis | L | None | No | |
26 | 86 | M | Common ostium RCA/LAD | Interarterial | Sapien 3 | 26 | None | No | |
27 | 74 | M | ACAOS RCA from the LCS | Interarterial | Sapien 3 | 26 | None | No | |
28 | 81 | M | ACAOS LCx from the RCA | Retro-aortic | Sapien 3 | 29 | None | No | |
29 | 84 | F | ACAOS RCA from the LCS | Interarterial | Evolut Pro | 26 | None | No | |
30 | 78 | F | ACAOS LMT from the RCS | Interarterial | Evolut R | 29 | None | No | |
31 | 81 | M | ACAOS LCx from the RCA | Retro-aortic | Evolut R | 34 | None | No | |
32 | 80 | M | ACAOS LAD from the RCS | Interarterial | Sapien 3 | 26 | None | No | |
33 | 76 | M | ACAOS LMT from the RCS | Interarterial | Sapien 3 | 29 | None | No | |
34 | 83 | F | High ostium RCA | Evolut Pro | 26 | None | No | ||
35 | 88 | F | SCA | Evolut R | 26 | None | No | ||
36 | 86 | F | ACAOS LAD from the RCS | Interarterial | Acurate Neo2 | S | None | No | |
37 | 80 | F | ACAOS LCx from the RCA | Retro-aortic | Evolut R | 29 | None | No | |
38 | 91 | F | ACAOS LCx from the RCS | Retro-aortic | Evolut Pro+ | 26 | None | No | |
39 | 68 | M | SCA | Sapien 3 | 29 | None | No | ||
40 | 84 | M | ACAOS LCx from the RCA | Retro-aortic | Sapien 3 | 26 | Wire+provisional stent LCx | No |
- Note: Patient characteristics from case reports 1−20 and from own institution 21−40.
- Abbreviations: ACAOS, anomalous origination of a coronary artery from the opposite sinus; BAV, balloon aortic valvuloplasty; LAD, left anterior descending; LCS, left coronary sinus; LCx, left circumflex artery; LMT, left main trunk; RCA, right coronary artery; RCS, right coronary sinus; SCA, single coronary artery.
Overall (n = 40) | Published experience (n = 20) | Local institution (n = 20) | |
---|---|---|---|
Age [years] | 80.6 ± 7.3 | 81.8 ± 6.0 | 79.4 ± 8.3 |
Male sex | 20 (50%) | 9 (45%) | 11 (55%) |
ACAOS | 30 (75%) | 14 (70%) | 16 (80%) |
Anonmalous coronary artery | |||
|
17 (56.7%) 3 (10%) 4 (13.3%) 6 (20%) |
9 (64.3%) 1 (7.3%) 2 (14.3%) 2 (14.3%) |
8 (50%) 2 (12.5%) 2 (12.5%) 4 (25%) |
Trajectory | |||
|
13 (32.5%) 22 (55%) |
4 (20%) 14 (70%) |
9 (45%) 8 (40%) |
THV type | |||
|
18 (46.2%) 20 (51.3%) 1 (2.6%) |
7 (36.8%) 11 (57.9%) 1 (5.3%) |
11 (55%) 9 (45%) 0 (0%) |
Coronary protection | 11 (27.5%) | 9 (45%) | 2 (10%) |
Coronary obstruction | 5 (12.5%) | 5 (25%) | 0 (0%) |
Coronary protection by vessel wiring was performed in 11 cases. Periinterventional coronary obstruction occurred in 5 (12.5%) patients. In those cases, the trajectory of the anomalous artery was retro-aortic in three and interarterial in two patients. Protective measures had been taken in four of these patients. Coronary obstruction occurred after the use of balloon-expandable THV in two cases, a mechanically-expanding and a self-expanding THV in one case respectively and after balloon aortic valvuloplasty (BAV). Management of coronary obstruction included coronary stent placement in three cases, transient extracorporeal life support in one case and emergency coronary artery bypass grafting in another case. TAVI was aborted in one case. There was no procedural mortality. In consideration of the small case number, we did not perform a formal statistical analysis. However, the patient and procedural characteristics are grossly comparable between the published and the local institutional series. The main difference between the two cohorts was the frequency of coronary protection and obstruction. Protective measures were taken in 45% of the cases found in the published experience versus 10% in our local cohort. All cases of coronary obstruction occurred in the published cohort.
4 DISCUSSION
CAA are a rare congenital condition comprising an ample number of anatomic variations affecting the origin and course of the coronary arteries. An extensive classification has been proposed elsewhere.18 The prevalence of CAA was historically studied in coronary angiography populations and autopsied patients, but MDCT has proven to be a valid tool to visualize these anomalies as well.19 Although the exact prevalence remains unclear, it is estimated to range between 0.2 and 2.3% in the general population.20
Some CAA may cause life-threatening symptoms and severe myocardial ischemia leading to sudden cardiac death.21 Patients with ACAOS with certain high-risk anatomical features such as an interarterial course (between the pulmonary artery and the aortic root), slit-like coronary ostium and intramural course are especially prone to myocardial ischemia.20 Higher rates of sudden cardiac death are found for anomalous left coronary arteries compared to anomalous right coronary arteries.22, 23 A rare entity associated with higher risk of sudden cardiac death are single coronary artery origins supplying the right coronary artery, the left anterior descending and circumflex artery.20
In contrast to the interarterial trajectory, the retro-aortic course is considered to be less malignant and often found when the left circumflex artery originates from the right coronary artery or from the right sinus of Valsalva.24
4.1 Coronary obstruction
In general, several factors have been identified to increase the risk of coronary obstruction during TAVI. These include compression of the coronary ostium by bulky calcified leaflets, low coronary take-off height (<12 mm), shallow sinus of Valsalva (width < 30 mm), a small and oval annulus and untreated underlying coronary artery disease.7, 25, 26 Interestingly, coronary calcification was regarded by some authors as a protective factor counteracting vessel compression.1, 11
Importantly, apart from ostial obstruction, compression may occur in other vessel segments. This appears to be especially relevant for patients with CAA, since four of the five reported cases with coronary obstruction were not located at the ostium but along the course of the vessel, with obstruction occurring in the proximity of the aortic annulus. In the remaining case, the exact location of obstruction was not specified. Due to the small number of reported cases, it remains unclear whether interarterial or retro-aortic trajectory increases the risk of coronary obstruction.
Higher prevalence of CAA was also described in patients with bicuspid compared to tricuspid aortic valves.27 Bicuspid aortic valves have several anatomic features like longer leaflets and higher calcification burden that may additionally increase the risk of coronary obstruction.
4.2 Preprocedural planning
Besides measurements of the aortic annular dimensions, the exact course of the coronary arteries can be mapped on MDCT. Possible implications of interarterial or retro-aortic vessel trajectory have been stated above. However, not only the course of the artery, but also the spatial relationship to the annulus may play a role. Proximity to the annulus may be hazardous for coronary compression compared to higher courses of the anomalous artery at leaflet level.12
Furthermore, anatomic features increasing risk for coronary obstruction (low coronary ostium height, shallow sinus of Valsalva, small and oval annulus) are assessed by MDCT. In the presence of these risk factors for coronary obstruction, the operator must assess whether measures for coronary protection must be implemented or not.
4.3 Choice of THV
There is no clear recommendation for the choice of a specific THV in patients with anomalous coronary arteries. Selection criteria should primarily focus on the valve anatomy, taking annular dimensions, degree of calcification, access route and operator experience with the device into account. In the cases presented, the use of balloon- and self-expanding THVs was balanced (20 balloon-expandable THVs vs. 18 self-expanding THVs vs. 1 mechanically-expanding THV).
There are potential advantages and disadvantages of different THV types in the setting of CAA. Some authors emphasize the benefit of self-expanding THVs leading to reduced aortic wall tension and stretching of the annulus, whereas others fear continued pressure after deployment due to radial forces.1, 2, 12 The possibility of re-sheathing after initial deployment is regarded as beneficial by some authors. On the other hand, coronary reaccess may be facilitated by implantation of a shorter balloon-expandable THV.
Of the three cases with coronary obstruction in which a THV was deployed, the use of a second and third generation (Sapien XT/Sapien 3) balloon-expandable THV led to obstruction in two cases and a self-expanding THV (Evolut Pro+) in another case. Considering only those cases in which a current generation THV was used, 15 cases with Sapien 3, 9 cases with either Evolut R, Evolut Pro or Evolut Pro+ and three cases with Acurate Neo were unproblematic. In summary, the use of balloon-expandable as well as self-expanding THVs may be considered in this particular scenario.
4.4 Management of TAVI in CAA patients
Early detection of coronary obstruction during the intervention is of great importance. In addition to continuous hemodynamic and electrocardiographic monitoring, the authors of some cases described selective angiography or repeated aortic root injections during implantation. The use of intravascular ultrasound may confirm vessel compression.15
A strategy to identify possible coronary artery occlusion before TAVI is the utilization of BAV as described by Tabachnick et al.9 In this case, BAV was done in a patient with a retro-aortic course of an anomalous LCx, which led to compression of the vessel during balloon inflation. Subsequently, the authors refrained from performing TAVI. Such an approach may be a proof of concept but leaves the patient with an untreated condition and most patients in this situation are not eligible for surgical valve replacement.
In cases of elevated risk of coronary obstruction some authors opted to implement precautions to increase peri-interventional safety. The most common measure taken was to protect the coronary artery at risk with a 0.014 inch wire for rapid re-access after THV implantation.4, 5, 7, 9, 11, 12, 15-17 A guide catheter extension catheter was placed over the wire to disengage the larger coronary guide catheter during valve deployment in one case.12 Provisional balloon or stent placement was performed in five cases.7, 11, 12, 17 Coronary stent implantation was necessary in three of those cases and was performed without difficulty.5, 7, 15 In all other cases when post implantation angiography showed sufficient coronary flow, the wire and/or balloon and stents were successfully removed from the coronary artery. Nevertheless, cannulation of the anomalous ostium can be challenging, and emergency bypass graft surgery may be a bail out strategy to restore coronary flow.13
Recently, increasing interest has been developed concerning the possible benefits of commissural alignment of transcatheter aortic valves.28 However, in patients with CAA, commissural alignment might even be counterproductive. Costa et al. described a case in which they intentionally misaligned the neo-commissures to prevent potential impairment of coronary reaccess.17
Finally, careful attention should be given to bioprosthetic valve oversizing and high-pressure balloon inflation because it intuitively increases the risk for coronary compression.5 An algorithm for the management of patients with CAA undergoing TAVI is proposed in (Figure 2).
5 LIMITATIONS
We did not perform formal statistical analysis in view of the small number of reported patients with CAA treated with TAVI. Hence, the proposed mechanistic explanations and highlighted risk factors for coronary obstruction in this particular scenario are merely hypothesis-generating. Although we present a consecutive series of patients from our institution, the findings from the published experience may underlie a certain selection bias. Underreporting of procedural complications and unsuccessful interventions may contribute to this fact and ultimately obscure the overall incidence of coronary obstruction in patients with CAA. The proposed algorithm for the management of patients with CAA has been derived from the published reports and our own experience. External validation in a broader cohort is needed.
6 CONCLUSION
With a total of 40 cases, this is the largest body of evidence of patients diagnosed with aortic stenosis and CAA undergoing TAVI. Little is known about feasibility of TAVI in patients with CAA. In the absence of evidence-based clinical practice guidelines, a patient-tailored and risk-adapted approach is recommended. Awareness of this clinical scenario and meticulous preprocedural assessment together with protective measures could help minimize procedural complications.
ACKNOWLEDGMENTS
Ahmed Abdelhafez acknowledges funding received from the European Society of Cardiology in form of an ESC Training Grant (ESC Training Grant App000056788). Open Access funding enabled and organized by Projekt DEAL.
CONFLICTS OF INTEREST STATEMENT
Steffen Desch reports that his hospital received speakers' honoraria and/or consulting fees on his behalf from Medtronic. Mohamed Abdel-Wahab reports that his hospital received speakers' honoraria and/or consulting fees on his behalf from Medtronic and Boston Scientific. The remaining authors declare no conflict of interest.