Valve-sparing aortic root replacement in adult patients with congenital heart disease

Abstract

 

OBJECTIVES

Aortic root dilatation is frequently observed in patients with congenital heart defects (CHD), but has received little attention in terms of developing a best practice approach for treatment. In this study, we analysed our experience with aortic valve-sparing root replacement in patients following previous operations to repair CHD.

METHODS

In this study, we included 7 patients with a history of previous surgery for CHD who underwent aortic valve-sparing operations. The underlying initial defects were tetralogy of Fallot (n = 3), transposition of great arteries (n = 2), coarctation of the aorta (n = 1), and pulmonary atresia with ventricle septum defect (n = 1). The patients’ age ranged from 20 to 40 years (mean age 31 ± 6 years).

RESULTS

David reimplantation was performed in 6 patients and a Yacoub remodelling procedure was performed in 1 patient. Four patients underwent simultaneous pulmonary valve replacement. The mean interval between the corrective procedure for CHD and the aortic valve-sparing surgery was 26 ± 3 years. There was no operative or late mortality. The patient with transposition of great arteries following an arterial switch operation was re-operated 25 months after the valve-sparing procedure due to severe aortic regurgitation. In all other patients, the aortic valve regurgitation was mild or negligible at the latest follow-up (mean 8.7 years, range 2.1–15.1 years).

CONCLUSIONS

Valve-sparing aortic root replacement resulted in good aortic valve function during the first decade of observation in 6 of 7 patients. This approach can offer a viable alternative to root replacement with mechanical or biological prostheses in selected patients following CHD repair.

Aortic valve-sparing operations are well-established procedures to treat adult patients with aortic root aneurysms with excellent long-term clinical outcomes in expert centres [1, 2].

INTRODUCTION

Aortic valve-sparing operations are well-established procedures to treat adult patients with aortic root aneurysms with excellent long-term clinical outcomes in expert centres [1, 2]. However, little is known about the outcomes of this technique in patients with previous congenital heart defect (CHD) repair and dilated aortic roots. Studies on this topic in the literature are scarce, and where studies were conducted, the number of included patients tends to be low [3, 4]. There is no clear guidance on indications with only limited information on long-term outcomes for this cohort of patients.

Nevertheless, aortic valve-sparing operations potentially offer considerable advantages such as the preservation of native aortic valve, excellent haemodynamics, the avoidance of long-term anticoagulation, and a low risk of thromboembolism and endocarditis [5]. This is of particular interest for young patients with CHD.

In this study, we evaluate our surgical experience and the respective outcomes with aortic valve-sparing procedures in patients with progressive aortic root dilatation conducted late after initial repair of CHD.

PATIENTS AND METHODS

Study population

All patients provided written informed consent for their inclusion in this retrospective study. Perioperative surgical findings, echocardiographic, Magnetic Resonance Imaging (MRI) or Computed Tomography (CT) scans, and clinical follow-up data including postoperative complications and the indication for reoperation were documented in a database (FileMaker 13, FileMaker Intl., Santa Clara, CA, USA) at Hannover Medical School. The most recent outcome data were collected during a consultation at our outpatient clinic for adults with CHD (GUCH) or by the referring cardiologist.

Between 2006 and 2020, a total of 7 patients with a history of surgery for CHD underwent aortic valve-sparing operations in our unit for congenital heart surgery. Patients with connective tissue disorders (such as Marfan syndrome, Loeys–Dietz syndrome) or other aetiology of aortic aneurysms with aortic regurgitation were excluded.

The underlying initial diseases in the cohort were: tetralogy of Fallot (TOF) (n = 3) and transposition of great arteries (TGA) (n = 2). One TGA patient had a Mustard operation and the other an arterial switch operation (ASO). One patient had coarctation of the aorta (CoA) and 1 patient pulmonary atresia with ventricle septum defect. The patient with CoA had a bicuspid aortic valve (BAV). Two patients had undergone >1 previous operation (patient number 3—2 operations, patient number 5—4 operations).

The indication for an aortic valve-sparing operation in all cases was an increasing aortic root diameter. Four patients had a progressive dilatation of the aortic root and underwent a root procedure with simultaneous repair of right-sided lesions to minimize redo surgery risk. Three patients with an implanted conduit in the right ventricle outflow tract (RVOT) had a compression of the conduit in this position resulting from the dilated aortic root, which contributed to the indication for root surgery (Fig. 1). The type of procedure and timing of the operations were determined in a weekly scheduled heart team meeting between GUCH cardiologists and congenital heart surgeons.

Figure 1:

Preoperative images. (A) CT imaging from patient number 5 showing a dilated ascending aorta. (B) MRI imaging from patient number 4 showing a dilated ascending aorta with compression of pulmonary artery. Ao: aorta; PA: pulmonary artery.

Operative technique

All patients underwent elective surgery. The operations were performed under combined intravenous general anaesthesia via median sternotomy. Aortic and right atrial cannulation (with two-stage venous cannula or bicaval where a simultaneous RVOT procedure was conducted) was used as a standard cardiopulmonary bypass (CPB) in all cases. A left ventricular vent was placed via the right upper pulmonary vein. For myocardial protection, antegrade ‘Buckberg’ cold blood cardioplegia was administrated and repeated every 30 min. After aortic cross-clamping and cardioplegic cardiac arrest, the proximal portion of the ascending aorta was transected at the sinotubular junction. In cases of simultaneous RVOT replacement, the right-sided conduit was first completely resected to achieve better exposure of the aortic root (Fig. 2). The aortic root was dissected as deeply as possible to facilitate the placement of the subannular polyester pledged sutures. The valve-sparing reimplantation was performed using a straight tube graft in which the native aortic valve was resuspended (David I). It was not necessary to reduce the aortic annulus in addition to the valve-sparing reimplantation. Plasty or plication of the valve leaflets was similarly not required. The coronary arteries were mobilized and reimplanted into the graft. Transoesophageal echocardiography was performed intraoperatively to evaluate the function of the preserved aortic valve.

Figure 2:

Intraoperative photos from patient number 4. (A) Operative situs from the surgeon’s view: the degenerated Contegra^® graft was completely removed from the right ventricle outflow tract and the aortic cuff, including the commissures, was reimplanted into the Hemashield^® vascular graft. (B) The decellularized pulmonary homograft is implanted with continuous suture technique. (C) Final view of aortic root replacement and implantation of the decellularized pulmonary homograft. AV: aortic valve; DPH: decellularized pulmonary homograft; PV: pulmonary valve; RVOT: right ventricle outflow tract.

RESULTS

The age of the patients at the time of aortic root surgery ranged from 20 to 40 years (mean age 31 ± 6 years), and all were male. The mean interval between the corrective procedure for CHD and aortic valve-sparing surgery was 26 ± 3 years. The mean diameter of aortic root at the time of operation was 51.3 ± 3.6 mm. Aortic valve regurgitation was mild in most cases and was not the primary indication for surgery. Patient history data are summarized in Table 1.

Table 1:

Patient characteristics

Patients	Gender	Age	Primary diagnosis	Type of corrective surgery	Years since repair	Ao. root diameter (aneurysm) (mm)	Ventriculo- arterial junction (mm)	Sinuses of Valsalva (mm)	Sinotubular junction (mm)	Ao. Ascendens (mm)	Degree of AR
1	M	30	TGA	Mustard operation	28	50	32	50	47	41	Mild
2	M	20	TGA	ASO	20	52	31	52	30	26	Moderate
3	M	32	TOF	TOF repair	29	55	32	55	50	43	Mild
4	M	28	TOF	TOF repair	27	56	35	56	41	34	Moderate
5	M	34	PA/VSD	Rastelli operation	24	45	31	45	43	38	Mild
6	M	33	TOF	TOF repair	30	51	30	51	47	32	Mild
7	M	40	CoA, BAV	CoA resection, EEA	23	50	31	50	45	42	Mild
Mean		31 ± 6			26 ± 3	51.3 ± 3.6

AR: aortic regurgitation; ASO: arterial switch operation; BAV: bicuspid aortic valve; CoA: coarctation of the aorta; EEA: end-to-end anastomosis; PA: pulmonary atresia; TGA: transposition of great arteries; TOF: tetralogy of Fallot; VSD: ventricle septum defect.

Surgical procedures

David aortic valve reimplantation was performed in 6 patients and Yacoub aortic root remodelling was carried out in 1 patient. A Hemashield^® platinum-woven double velour prosthetic graft (Maquet, Getinge Group, Germany) was used in 4 patients and the Gelweave^® graft (Vascutek Ltd., a Terumo Company, Scotland, UK) was implanted in 3 patients. The size of the vascular graft was chosen to fit the diameter of the aorto-ventricular junction. Four patients underwent simultaneous pulmonary valve replacement (PVR) or RVOT conduit exchange (Contegra^® graft, n = 1; Hancock^® conduit, n = 1; and ESPOIR PV^® decellularized pulmonary homograft, n = 2). The mean CPB time was 281 ± 83 min; the mean aortic cross-clamping time was 142 ± 49 min. Intraoperative data are shown in Table 2.

Table 2:

Intraoperative data

Patients	Type of procedure	Vascular graft (mm)	Concomitant procedures	HLM time (min)	X-Clamp time (min)
1	David	Hemashield®, 28		356	123
2	David	Gelweave®, 28		312	170
3	David	Hemashield®, 30	PVR; Contegra®, 22 mm	223	120
4	David	Hemashield®, 28	PVR; DPH, 22 mm	326	180
5	David	Gelweave®, 30	PVR; Hancock®, 26 mm	380	218
6	Yacoub	Gelweave®, 30	PVR; DPH, 29 mm	217	77
7	David	Hemashield®, 28		155	107
Mean				281 ± 83	142 ± 49

DPH: decellularized pulmonary homograft; PVR: pulmonary valve replacement.

One patient (patient number 5) required coronary artery bypass grafting as a result of poor right ventricular function after coronary reimplantation. In this high-risk patient (4 previous cardiac operations—procedure 1: modified Blalock–Taussig shunt; procedure 2: Rastelli correction with implantation of pulmonary valved conduit; procedure 3: re-PVR; procedure 4: re-re-PVR). Thereafter, the right ventricular function improved rapidly, and CPB weaning and the subsequent postoperative recovery of this patient was uneventful. Patient number 2 required resternotomy due to a haemothorax on postoperative day 2. Aside from these 2 events (representing 28.5% perioperative morbidity), no other complications have been documented.

Postoperative course

There was no in-hospital or late mortality. Transthoracic echocardiography at hospital discharge showed mild-to-no aortic valve regurgitation in all patients. The mean follow-up was 8.7 years (range 2.1–15.1 years) and complete. Standard echocardiography and/or MRI were performed in all patients to assess the competence of the preserved aortic valve (Fig. 3). At the latest follow-up, the aortic valve regurgitation was mild or negligible in 6 patients. One patient, however, developed initially moderate and later severe aortic valve regurgitation. He was re-operated after 2.1 years and underwent aortic valve replacement with a 21-mm mechanical valve. Freedom from aortic valve dysfunction at 15 years was 85.7 ± 13.2%. Clinically, the status of patients remained consistently in the New York Heart Association (NYHA) class I or I–II (Table 3).

Figure 3:

Cardiac magnetic resonance imaging of patient number 4. (A) Preoperative imaging showing aortic root dilatation (arrow). (B) Postoperative imaging demonstrating the aortic root after aortic valve-sparing replacement (arrow). (C) Postoperative MRI 3D reconstruction. Ao: aorta; PA: pulmonary artery.

Table 3:

Postoperative data and outcome

Patients	Mortality	Post OP AR	Post OP NYHA	Reoperation	Follow-up (years)
1	None	Mild	I	None	13.1
2	None	Severe	I	AVR	2.1
3	None	Mild	I	None	15.1
4	None	None	I	None	2.6
5	None	Mild	I–II	None	2.3
6	None	None	I	None	5
7	None	Mild	I	None	0.2

AR: aortic valve regurgitation; AVR: aortic valve replacement; NYHA: New York Heart Association.

DISCUSSION

Valve-sparing aortic root replacement procedures for aortic root aneurysms have proven to be a good alternative to mechanical-valved conduits, and gained popularity in the 1990s. The preserved native aortic valve had physiological haemodynamics and did not require permanent anticoagulation. The 20-year results of aortic valve-sparing operations in expert centres have been excellent [1, 2].

Aortic valve-sparing operations in children with connective tissue disorders have provided a safe and effective option for these patients [6]. Some centres avail significant experience with these procedures in up to 100 paediatric patients [7]. The use of this approach to treat dilated aortic root in patients with congenital heart disorders is, however, limited and the available evidence in the literature is sparse. We have summarized the data of some recent publications on valve-sparing procedures in patients with CHD conducted late after corrective surgery in Table 4. Most published studies are case reports or small series. The underlying congenital diseases of patents included in these studies are heterogeneous, including TOF, truncus arteriosus, transposition of the great arteries, CoA or palliated complex heart defects with univentricular physiology after a Fontan procedure. It appears likely that the long-term results following aortic-sparing procedures in patients with various original forms of CHD and resulting residual anatomical defects and specific physiology after correction of these defects will vary. At present, the follow-up in most publications is very limited and no midterm results are available.

Table 4:

Current literature on valve-sparing aortic root operations in congenital heart defects

Author, year	Number of patients	Primary diagnosis	Diameter of aortic root (mm)	Age (years)	Type of procedure	Follow-up	Reoperation
Moreau de Bellaing et al., 2020 [8]	2	Not specified	a	a	David, Yacoub	a	AVR; n = 1
Liebrich et al., 2019 [9]	1	TGA	53	31	David	Discharge	None
Kluin et al., 2016 [6]	1	CoA+BAV	68	8.7	David	a	None
Kanzaki et al/, 2015 [10]	1	UVH	39	11	David	Discharge	None
Koolbergen et al., 2015 [11]	8	TOF, CoA + BAV	a	a	David	a	None
Baliulis et al., 2015 [3]	11	TOF, CoA, TGA, UVH, TAC, VSD	a	a	David	Median 32 months	None
Kuwauchi et al., 2014 [12]	1	TAC	46	24	David	6 months	None
Liebrich et al., 2014 [13]	1	TGA	49	15	David	Discharge	None
Ootaki et al., 2013 [14]	1	TOF	40	16	David	1 month	None
Lange et al., 2013 [15]	1	CoA + BAV	60	18	David	a	None
Erez et al., 2012 [16]	4	UVH	a	a	David	Mean 2.6 years	AVR; n = 2
Rakhra et al., 2012 [4]	4	TOF, TB + CoA, TGA, UVH	a	a	David	a	None
Pizarro et al., 2011 [17]	1	UVH	54	10	David	1 year	None
Chen et al., 2010 [18]	1	TB + CoA	51	8	David	Discharge	None

AVR: aortic valve replacement; BAV: bicuspid aortic valve; CoA: coarctation of the aorta; TAC: truncus arteriosus communis; TB: Taussig–Bing complex; TGA: transposition of great arteries; TOF: tetralogy of Fallot; UVH: univentricular heart after Fontan operation; VSD: ventricle septum defect.

^aInformation was not available for all parameters within cited references.

Despite the variety of initial diagnoses in this cohort, all patients with CHD can, in our opinion, be divided into the following groups: (i) TOF/TOF-pulmonary atresia, (ii) TGA, (iii) CoA + BAV and (iv) patients with univentricular heart physiology.

Patients with TOF were among the first patients with CHD in whom dilatation of the ascending aorta due to Tunica media degeneration induced by intrauterine volume overload (right-to-left shunt through ventricle septum defect), was studied in detail [19, 20]. In our opinion, these patients are the most suitable cohort for the strategy of aortic root surgery, due to the advantage of a surgically unchanged native aortic root. Our hypothesis, backed by our currently available data, is that these patients will have more stable long-term results after successful aortic valve-sparing surgery.

Dilatation of the aortic root with or without neoaortic valve regurgitation is one of the most frequent late complications in patients with TGA after ASO [21, 22]. A major risk is posed by changes in the aortic root geometry after coronary artery reimplantation leading to the enlargement of the neoaortic root. Associated pulmonary artery-aorta size discrepancy and poor coaptation of the valve leaflets are other important factors. The valve-sparing operations in these patients are more complex due to the surgically modified aortic root. After dissection of the pulmonary artery and preparation of the aortic root, the placement of subannular sutures should be done as deeply as possible to correctly position the tube graft, thus eliminating the potential for recurrent root dilatation and neoaortic valve regurgitation. Care should be taken to avoid the distortion of both semilunar valves, which are in close proximity after an arterial switch procedure.

CoA in 85% of cases is associated with BAV, which is the most common congenital heart valve anomaly. It has been reported that the simultaneous presence of BAV and CoA is more often associated with aortic dilatation compared with the isolated pathology [23]. This group should be considered separately from conotruncal CHD. The management of aortic dilatation in these patients is largely based on experience with aneurysmal disease associated with connective tissue disorders and has seen major advances over the last decade [24, 25].

Aortic dilatation has been reported in approximately half of the patients with univentricular physiology after a Fontan procedure [26]. In some HLHS patients, this may be linked to the exposure of the native pulmonary root to systemic pressures, similar to TGA patients following arterial switch procedures. Neoaortic valvular dysfunction in HLHS and in Fontan aortopathy may be an indication for reoperation, but both the decision for reoperation only for aortic root dilatation and the choice of procedure remains controversial. In addition, such procedures, due to the extensive nature of this operation, can pose a significant risk of morbidity and mortality.

The timing of these operations with regard to the diameter of the aortic root is challenging. In general, surgical management is recommended in adult patients with an aortic root aneurysm larger than 55 mm in diameter to avoid aortic dissection or rupture [27]. Different underlying mechanisms, however, make it difficult to make a general decision for or against an aortic root operation based only on the diameter of the aorta. Compression of surrounding tissues such as the pulmonary artery, superior vena cava, or a bronchus caused by a dilated aortic root is also a determining factor when considering reintervention.

Moreover, indication for surgery on aortic roots, which are smaller than 55 mm, is also based on additional right-sided cardiac lesions requiring reoperations [3]. Our cohort included patients with implanted pulmonary grafts during initial repair of TOF, pulmonary atresia and marked pulmonary graft compression from a dilated aortic root. In 2 of these patients, we opted for simultaneous aortic root procedures despite moderate aortic dimensions (aortic root 45 and 51 mm) in view of the risky re-do surgery, as described in the literature [10]. In addition, both patients were not viewed as good candidates for percutaneous pulmonary valve implantations by the heart team due to known risk factors for percutaneous pulmonary valve implantations [28].

Our operative morbidity was low despite the complexity of the procedures, which may be attributed to the long-term experience of our centre with this type of root repair procedure [2]. One patient needed intraoperative bypass grafting to the RCA. This was the fifth complex heart procedure for this patient and intraoperative anatomical dissection was very cumbersome, as indicated by the long procedure times. It is our understanding that this was more related to the multiple previous procedures than to the aortic valve-sparing root procedure in itself.

There was no operative or late mortality and the results compare well with the excellent results obtained in primary aortic valve-sparing procedures in adults [1, 2]. There was one failure in a TGA patient following ASO. These patients have specific anatomical features, which render them at increased risk for early failures [29]. In general, we would recommend aortic valve-sparing procedures in patients following ASOs only after thorough discussion.

Limitations

The main limitation of the present study is the small number of patients and the heterogeneity of the underlying CHD. Due to the small size of the patient cohort, advanced statistical analyses were not performed. Moreover, the follow-up period for the most of patients in our study did not exceed mid-term. However, as the patient follow-up continues, more data on the long-term results will become available in the future. Data consolidation from multiple centres is also warranted to assess the durability of valve-sparing techniques in this cohort of patients.

CONCLUSION

Valve-sparing aortic root replacement following CHD repair in early childhood can be performed with excellent initial results and good aortic valve function for the first decade of follow-up. This approach may be a viable alternative to aortic root replacement with mechanical or biological prostheses in young adult patients. Longer follow-up and a larger series are needed before standard recommendations for such procedures can be considered.

ABBREVIATIONS

ASO

Arterial switch operation

BAV

Bicuspid aortic valve

CPB

Cardiopulmonary bypass

CoA

Coarctation of the aorta

CHD

Congenital heart defects

PVR

Pulmonary valve replacement

RVOT

Right ventricle outflow tract

TOF

Tetralogy of Fallot

TGA

Transposition of great arteries

 

Author contributions

Dmitry Bobylev: Conceptualization; Data curation; Methodology; Project administration; Writing—original draft; Writing—review & editing. Murat Avsar: Data curation; Investigation; Methodology. Samir Sarikouch: Conceptualization; Data curation; Project administration; Supervision; Writing—review & editing. Tomislav Cvitkovic: Data curation; Formal analysis; Visualization. Dietmar Boethig: Data curation; Formal analysis; Software; Supervision. Mechthild Westhoff-Bleck: Conceptualization; Data curation; Project administration; Supervision. Harald Bertram: Data curation; Formal analysis; Methodology; Visualization. Philipp Beerbaum: Conceptualization; Formal analysis; Methodology; Supervision; Validation. Axel Haverich: Conceptualization; Data curation; Formal analysis; Project administration; Supervision. Alexander Horke: Conceptualization; Data curation; Formal analysis; Supervision; Writing—review & editing.

Reviewer information

Interactive CardioVascular and Thoracic Surgery thanks Emile Bacha, Antonios Kallikourdis, Guido Michielon and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.