We performed a retrospective, longitudinal and descriptive study. Clinical, demographic, operative and postoperative data were collected through medical records from the database of the Cardiothoracic Surgery Department of Centro Hospitalar São João. The study was approved by the local ethics committee.

Patients who underwent AVR with Freedom Solo or Solo Smart valves at Centro Hospitalar São João between April 2009 and April 2015 were identified from our center's registry, regardless of primary indication for surgery or concomitant procedures.

Preoperative data collected included age, gender, body surface area, body mass index, cardiovascular risk factors (hypertension, diabetes, dyslipidemia, smoking and obesity), creatinine clearance, peripheral arterial disease (defined as carotid occlusion or >50% stenosis, claudication, amputation or previous or planned intervention on the abdominal aorta, limb or carotid arteries), cerebrovascular events (transient ischemic attack or stroke), chronic obstructive pulmonary disease, coronary artery disease, left ventricular dysfunction, previous myocardial infarction, preoperative rhythm, New York Heart Association functional class and urgency of surgery. The European System for Cardiac Operative Risk Evaluation (EuroSCORE) II was calculated for each patient. Information was also collected on both pathology (stenosis, regurgitation or combined) and etiology of aortic valve disease (degenerative calcific, bicuspid, endocarditis [native or prosthetic valve], rheumatic, prolapse or aortic prosthesis dysfunction).

The decision to use a stentless valve was at the discretion of the surgeons, after patient consent was obtained. Implantation of FS/SS valves was mainly performed by two senior surgeons (AFL-M and MJA), who considered all patients undergoing AVR with bioprosthetic valves for FS/SS implantation unless the following exclusion criteria were met: extensive aortic root calcification, severe mismatch between aortic annulus and sinotubular junction, or Sievers type 0 bicuspid aortic valve. Other types of bicuspid aortic valve were not excluded if aortic root symmetry was preserved.

Patients underwent full or partial upper sternotomy and mild hypothermic cardiopulmonary bypass with cold crystalloid anterograde and retrograde cardioplegia. A transverse aortotomy was performed approximately 1 cm above the sinotubular junction. The aortic valve was excised and the annulus was completely decalcified and reinforced with a 5-0 polypropylene suture when necessary. Three 4-0 polypropylene sutures were placed in a supra-annular position at the nadir of each sinus and passed through the Solo valve. Thereafter, the valve was parachuted into the aortic root and tied with sutures running continuously 1 mm above the annulus. The sutures were passed out of the aorta at the level of the commissures and tied with the suture from the adjacent sinus. Immediate outcome was assessed by intraoperative transesophageal echocardiography. All patients underwent our center's standard anesthetic, surgical and postoperative care procedures. Surgical data were gathered regarding valve size and cardiopulmonary bypass and cross-clamp times for both isolated and combined procedures. Postoperative data collected are defined in the follow-up section.

In accordance with local protocol, patients had a postoperative six-month follow-up visit at our center that included transthoracic echocardiographic assessment. Mean gradients and EOA (calculated using the continuity equation) were recorded. PPM was classified by the ratio of prosthesis EOA to patient body surface area as moderate (0.85-0.65 cm2/m2) or severe (<0.65 cm2/m2).13 Thereafter, echocardiographic and clinical follow-up was carried out yearly by the patient's referring cardiologist.

In-hospital mortality (defined as 30-day mortality if the patient was discharged or within any period if the patient was not discharged14) was retrieved from hospital medical records. All-cause mortality (the primary outcome) was obtained from the National Healthcare Registry as of October 1, 2015. For the purposes of analysis, all cases of unknown cause of death were considered cardiovascular deaths.

Secondary endpoints were low cardiac output (need for high-dose inotropic support or intra-aortic balloon pump), stroke, acute renal function impairment (rise of serum creatinine >1.5 times the preoperative value or low urine output [<0.5 ml/kg/h for >6 hours]), atrial fibrillation episodes, permanent pacemaker implantation, severe thrombocytopenia (platelet count <30×109/L), early resternotomy for bleeding, prolonged ventilation (>24 hours mechanical ventilation), length of hospital stay, structural valve deterioration (SVD) or non-structural valve dysfunction, endocarditis or late reintervention for prosthesis-related issues. SVD was defined as changes intrinsic to the valve, such as wear, calcification, leaflet tear or suture line disruption of the valve's components, and non-structural valve dysfunction as any abnormality not intrinsic to the valve itself resulting in dysfunction of the operated valve or hemolysis.14 These data were also retrieved in October 2015, either from clinical records or telephone interview with the referring cardiologist. Due to lack of follow-up information, a post-hoc echocardiogram was performed at our center in 19 patients for the purpose of the current study.

Continuous variables were expressed as mean (standard deviation) or median [interquartile range] (25th-75th percentile), as appropriate. Categorical variables were expressed as frequency and percentage. Comparisons between patients undergoing more than one procedure and those undergoing isolated AVR were performed using the unpaired Student's t test or the Mann-Whitney test for continuous variables. Normality was assessed by the Shapiro-Wilk test and visual inspection of residuals. Kaplan-Meier curves were used to assess time-to-event data. All statistical analyses were performed using IBM SPSS version 21 (IBM Corporation, New York). A p-value less than 0.05 was considered statistically significant.

Preoperative characteristics of the study population are described in Table 1. Patients’ mean age was 72±8 years and 52% were female. The most prevalent cardiovascular risk factors were hypertension (79%) and dyslipidemia (68%); 35% of patients had diabetes, 25% were obese and 15% were current or former smokers. Degenerative calcific disease was the most common etiology for aortic valve disease (76%), 8% of patients had bicuspid and 8% rheumatic valves, and 5% had endocarditis (4% native valve, 1% prosthetic). The median EuroSCORE II was 2.7 [1.5-4.7]: 1.8 [1.1-3.1] for isolated AVR and 3.3 [2.2-6.4] for combined procedures. Out of 345 patients, 318 received FS and 27 SS. Medium-term clinical and echocardiographic follow-up (5±3 months) was retrieved from 98% of patients and was 100% (complete) for all-cause mortality. Mean follow-up was 39±22 months and maximum follow-up was 78 months.

The most frequently implanted valve size was no. 23 (39%), followed by sizes 25 (25%) and 21 (23%). One or more concomitant procedures were performed in 52% of patients (summarized in Table 2). Coronary artery bypass grafting (CABG) was the most frequent combined procedure (29%); mitral and tricuspid valves, as well as the ascending aorta, were also commonly treated. Compared with isolated AVR, combined procedures had longer median cardiopulmonary bypass and cross-clamp times, 95 [83-118] vs. 152 min [120-201] and 67 [59-85] vs. 110 min [85-140] (p<0.001), respectively. There was no intraoperative mortality.

Overall in-hospital mortality was 2.6% (n=9), 1.8% (n=3) for isolated AVR and 3.4% (n=6) for combined procedures. Causes of death were prosthetic endocarditis (n=1), cardiogenic shock (n=2), septic shock (n=3), multiorgan failure (n=2) and iatrogenic complication of intensive care unit procedure (n=1). EuroSCORE II was lower in surviving patients (2.6 [1.5-4.6] vs. 10.0 [4.7-11.9], p=0.001).

Length of hospital stay was 7 [6-11] days and was significantly longer in patients who underwent combined procedures (8 [6-12] vs. 7 [6-9] days, p<0.001). Concerning in-hospital morbidity (Table 3), 30% of patients showed acute renal function impairment, 21% presented low cardiac output (requiring a high-dose single inotropic agent [9%], two or more inotropic agents [12%] or intra-aortic balloon pump [0.3%]), 7% had severe thrombocytopenia, 3% underwent pacemaker implantation due to atrioventricular (AV) conduction disturbances, and 2% suffered stroke. Only four patients (1%) underwent resternotomy for bleeding and none developed hemorrhagic stroke. In accordance with local protocol, patients without contraindication were discharged on vitamin K antagonists (changed to an antiplatelet agent three months after surgery); alternatively, patients were medicated with an antiplatelet agent.

Mean transvalvular gradient was 11.9±4.5 mmHg and EOA was 1.9±0.5 cm2 (Figure 1). PPM occurred in 38 patients (13.7%) and was severe in only one case. This patient had a BSA of 1.86 m2, a number 23 valve was implanted and the mean transprosthetic gradient on follow-up echocardiographic assessment was 13 mmHg. Mean transprosthetic gradient in patients with moderate or severe PPM was 16.3±5.6 mmHg, EOA was 1.33±0.18, BSA was 1.78±0.17 m2 and the most frequent prosthesis size was 21. Three cases of SVD were identified during follow-up (two of these patients were reoperated).

EOA: effective orifice area.'> EOA: effective orifice area.' title='Mean gradient and effective orifice area according to prosthesis size. EOA: effective orifice area.'/>

Figure 1.

Mean gradient and effective orifice area according to prosthesis size. EOA: effective orifice area.

(0.08MB).

Forty-seven patients (14%) died after discharge. The underlying cause was non-cardiovascular in 25 patients and cardiovascular in 22. Two deaths related to SVD were identified: one patient died after reoperation and the other was considered unsuitable for reintervention. The one-, three-, and six-year cumulative survival rate was 94%, 87% and 72%, respectively. Patients who underwent isolated AVR showed better survival than those who underwent combined procedures (p=0.005, log-rank test) (Figure 2).

AVR: aortic valve replacement.'> AVR: aortic valve replacement.' title='Six-year cumulative survival of all-cohort, isolated and combined procedures. AVR: aortic valve replacement.'/>

Figure 2.

Six-year cumulative survival of all-cohort, isolated and combined procedures. AVR: aortic valve replacement.

(0.14MB).

Six patients were reoperated: three due to endocarditis (two, 15 and 19 months after surgery), two for SVD (41 and 67 months after surgery) and one for periprosthetic fistula (two months after surgery). Two cases of prosthetic endocarditis were attributed to Staphylococcus aureus and one to Enterococcus faecalis infections; extensive root and subvalvular abscesses were found, with small or no vegetations on the cusps (two of these patients required root replacement due to extensive tissue infiltration). In the two cases of SVD (a 59-year-old male and a 67-year-old female), the surgeon found immobile and severely calcified cusps, but with only mild to moderate thickening. A visible delamination plane between the prosthesis pericardium and the native aortic root enabled easy en bloc explantation of the valves (Figure 3). The periprosthetic fistula was due to fracture of the 4-0 polypropylene suture in the right coronary sinus, but there was no valve deterioration or root dilation, and so the leak was closed with a continuous 4-0 polypropylene suture.

Figure 3.

Solo bioprosthetic valve with structural valve deterioration (left); en bloc explantation of Solo valve (right).

(0.17MB).

Throughout follow-up, five strokes and three cases of endocarditis, treated medically, were recorded. Loss of follow-up occurred in eight cases (2.8%) (Table 4).