Between 2004 and 2018, 200 consecutive patients (mean age: 43.2 ± 17.5 years; 148 males: 74%) underwent BAV repair at the same surgical setting (MJ), the results of which have already been published elsewhere [9,10]. Of these, 24 patients with BAV who developed aortic regurgitation requiring valve repair with annuloplasty without concomitant aortic root surgery were prospectively referred for CMR and transthoracic echocardiography (TTE) one year after the operation. Generally accepted contraindications to magnetic resonance imaging were used. Subsequently, 21 patients have had CMR performed. The size of the study and length of the follow-up were determined on the basis of a statistical power calculation of similar methods that were previously co-published by the main author [11]. This study was approved by the Local Ethics Committee. Informed consent was obtained from each subject.

All CMR examinations were performed on a 1.5 T clinical scanner (Signa HDxt, GE Healthcare, Milwaukee, WI). Images were obtained during suspended respiration at end-expiration. For cine imaging balanced steady-state free precession sequence was used. Patients underwent a left ventricular (LV) function study as previously described [9]. Cine images were acquired in three LV long-axis views, LV outflow tract view and a set of multiple contiguous short-axis slices from atrioventricular ring to the apex. Cine imaging of a BAV orifice was performed. The acquisition was repeated several times a few millimetres further and closer from valve orifice to ensure optimal direct planimetry of valve area. Then double oblique cine images were acquired as multiple cross-sections and long-axis views in order to visualize aortic root, sinotubular junction (STJ), and ascending aorta. After choosing a plane few millimetres above the tips of aortic cusps (usually at the level of STJ) 2D through-plane phase-contrast sequence was used for quantification of forward and backward flow. Then additional phase-contrast imaging was performed at the level of valve orifice and also few millimetres distal from tips of aortic cusps for maximal velocity assessment. Meticulous care was taken to orient imaging planes perpendicular to the blood flow. Encoding velocity was carefully adjusted to avoid aliasing. All flow sequences were acquired with the region of interest located at the isocenter of the magnet.

Images were analysed with commercial software (Qmass MR, Medis Medical Imaging Systems, Leiden, the Netherlands) by an experienced, observer-blinded to the patients’ profile (KMJ). LV volumes, ejection fraction, and compacted mass were calculated as previously described [12]. Published sex-specific normal values were used to assess the normalization of LV parameters [13]. Three-chamber and LV outflow tract cine images were used for aortic annulus assessment. The average diameter of the annulus was calculated from minimal and maximal annulus diameters. Anatomical aortic valve area (AVA) planimetry was performed at the systolic frame (largest aortic valve opening) after careful confirmation of the correct imaging plane. Maximum cross-sectional measurements of the aortic root, STJ, and ascending aorta were performed. From phase-contrast images, forward and backward flow through the aortic valve was measured. Regurgitant volume and regurgitant fraction were calculated. In order to minimize phase offset errors background correction was applied with the region of interest in the stationary tissue (pectoralis muscle).

All transthoracic echocardiograms were obtained one year after the surgery by an experienced, the same every time, observer-blinded to the patients’ profile. Standard echocardiographic parameters of the LV and aortic valve were assessed including the specific bicuspid anatomy, magnitude, and character of the aortic regurgitation, as well as transvalvular aortic gradients. Recommendations for non-invasive evaluation of valvular regurgitation developed by the American Society of Echocardiography in collaboration with the Society for Cardiovascular Magnetic Resonance has been used [14].

All operations were performed through a median sternotomy, with the use of standard cardiopulmonary bypass. In each case, the myocardium was protected with blood cardioplegia. Techniques of aortic valve repair have been standardized and described previously by the author [10,15]. Briefly, the first step has been an evaluation of the effective height of leaflets [9,16], as well as the central leaflet coaptation. Secondly, the relative lengths of the leaflet free margins were assessed by suturing together and identifying prolapse [10,15,17‐19]. The mechanism of AR was related to both, prolapse of conjoined fused leaflet and annulus enlargement. Prolapsing leaflet management consisted of plication with or without triangular excision followed by direct suturing. Annulus dilatation has been addressed either by external annuloplasty band or ring, or by subcommissural annuloplasty with regard to the height of the leaflets [17,20]. The postoperative assessment of successful repair was based on perioperative transoesophageal echocardiography (TOE), described elsewhere [10]. Briefly, postoperative valve evaluation included measurement of the aortic annulus, the presence of coaptation and assessment of effective coaptation height, as well as analysis of residual regurgitation, including quantification of regurgitation jet and its direction.

Twenty-four consecutive patients fulfilling preselection criteria operated on between 2013 and 2014 by a single surgeon (MJ), were prospectively allocated to one of two groups based on the aortic annulus stabilization technique. Three patients have been excluded due to CMR contraindications. The first group comprised patients who had an external annuloplasty (EA) in a circumferential fashion with the use of a Dacron strip made of aortic graft chosen for STJ remodelling and aorta replacement. The external annuloplasty consisted of the placement of a circular, transverse line of 6–8 interrupted pledgeted 2–0 braided sutures at the level of the aortic ring below leaflets nadirs, from inside to outside where were supported by a circular band from Dacron strip. The second group had subcommissural annuloplasty (SCA) performed with two braided 2–0 sutures enhanced with pledgets to narrow two subcommissural triangles. Additionally, all patients from the EA and SCA group underwent STJ remodelling and replacement of the ascending aorta.

The preselection criteria included: BAV type I (17 valves with left–right coronary cusp fusion and 4 valves with right-noncoronary cusp fusion, similarly distributed in both groups), moderate to severe aortic regurgitation, an enlarged aortic annulus (median: 26 mm (first quartile: 24; third quartile: 28)) with no aortic root dilatation (aortic root measurements below 40 mm), thus not requiring root replacement and valve reimplantation in accordance with the recommendations [21]. Both groups were comparable in terms of the baseline characteristics including patient age, dimensions of the left ventricle, and aortic annulus diameter. All patients presented with normal left ventricular contractility at baseline.

Data are presented as counts (percentages) for categorical variables, and as medians with first and third quartiles of the distributions for continuous variables. Nonparametric tests were used due to the small sample size. For continuous variables, the Mann–Whitney test was performed for unpaired samples. Proportions comparison between groups were analysed with Fisher’s exact test. Correlations between continuous variables were evaluated with the use of Spearman’s rank correlation coefficient. All statistical tests were two-sided. A p-value was considered to indicate statistical significance at the nominal 0.05 level. Statistical analyses were performed using SPSS version 20.0 (SPSS Inc., Chicago, IL).