Aortic valve replacement using autologous pericardium : single centre experience with the Ozaki technique

METHOD: Between September 2015 and May 2017 30 patients (20 males, mean ± standard deviation age 66.83 ± 10.55 years) suffering from aortic stenosis (AS, n = 7), aortic regurgitation (AR, n = 12), or a combination of both (AS/AR, n = 11) were assigned for an Ozaki procedure. The glutaraldehyde-treated autologous pericardium was intraoperatively customised and tailored according to individual sinus measurements and appropriate Ozaki templates (CE marked).


Introduction
Heart valve disease represents a serious and growing public health problem, with aortic valve stenosis being the most common entity [1].Heart valve disease is frequently caused by degenerative atherosclerotic processes, congenital abnormalities, or rheumatic disease [2,3].Open aortic valve replacement remains the gold standard in the treatment of patients with severe aortic stenosis (and aortic regurgitation) [4].Mechanical valves are preferred in younger patients (<60 years) because of their longer lifetime, whereas biological valves are used for elderly patients in order to avoid the need for oral anticoagulation [4].Transcatheter aortic valve implantation (TAVI) has become increasingly popular in recent years.To date, this technique has mostly been used in patients for whom conventional aortic valve replacement was considered too risky.More recently, new surgical approaches, which aim at reconstructing the aortic valve rather than replacing it with a prostheses, have been developed.In 2011, Ozaki et al. published their technique, which involves the use of autologous pericardium and was studied in a large patient cohort with excellent long-term follow-up [5,6].Potential benefits of the so-called Ozaki procedure include avoidance of oral anticoagulation, avoidance of foreign material, and suitability for patients with small aortic annuli and in infectious endocarditis.However, excellent long-term outcomes have not yet resulted in widespread use of the technique [7].Further evidence from short-and long-term studies is needed to prove the efficacy of this technique in the treatment of aortic valve stenosis and regurgitation.As the first centre for cardiac surgery in Europe, we started performing aortic valve replacement with autologous pericardium in September 2015 after CE certification.The aim of this study was to analyse short-term results from the first 30 patients undergoing aortic valvuloplasty by means of the Ozaki procedure at the University Hospital of Basel.

Material and methods
In September 2015, the first aortic valve replacement using autologous pericardium was performed in our clinic.The indication for operation was based on published guidelines [4].Data were collected between September 2015 and May 2017.The decision whether or not to use the Ozaki procedure was extensively discussed with the patient preoperatively.In our opinion, every patient who is eligible for a biological valve replacement or has a reason for avoiding oral anticoagulation is suitable for an Ozaki procedure.Contraindications were previous thoracic irradiation and re-operation, because of potential damage to the pericardium.All operations were performed by two experienced surgeons.Primary endpoints were in-hospital mortality and thromboembolic events.Intra-and perioperative data, as well as 3-month follow-up data (including echocardiographic results), were also collected.

Surgical technique
All patients were under general anaesthesia and were continuously monitored with two-lead electrocardiography (II, V5), pulse-oximetry, and invasive measurement of arterial blood and central venous pressure by a board-certified cardiac anaesthesiologist, according to our routine protocol for patients undergoing open heart surgery.Transoesophageal echocardiography (TEE) was performed throughout the procedure.The operative technique has already been described by Ozaki et al. [5].In brief, after median sternotomy, the pericardium was dissected and treated for 10 minutes with a 0.6% glutaraldehyde solution, before it was rinsed three times for 6 minutes with sterile saline.In the meantime, the ascending aorta and the right atrium were cannulated and cardiopulmonary bypass was established.All operations were performed with use of extracorporeal circulation (ECC).Cardioplegic arrest was accomplished with blood cardioplegia in all cases.After aortotomy, the diseased cusps were resected.The distance between each commissure was measured, using a special measuring device.The pericardium was subsequently cut according to a template that corresponded to the measured size.At this point, the varying thickness of the pericardium had to be considered.The pericardium is usually thicker in vicinity of the diaphragm than in the area of the aortic root.Thus, the larger cusps were usually taken from the diaphragmatic area.Moreover, the inner side of the pericardium (serous lamina) is smoother and therefore this side faces the ventricle.The cusps were then sewn along the annuli with a 4-0 running suture, and commissural coaptation was secured with additional 4-0 sutures.All sutures were knotted and reinforced with an autologous patch outside the aorta.The functionality of the substituted valve was reassessed intraoperatively with water testing, and postoperatively by TEE.A pericardial patch was used to replace the removed pericardium.Postoperatively, the patients were routinely monitored on the intensive care unit (ICU) and lifelong antiplatelet therapy was prescribed (100 mg aspirin per day).

Data collection
Perioperative data were exported from our quality management database.These data were regularly checked for completeness and correctness.An echocardiographic examination was routinely performed prior to the patients' discharge, and resulting data were collected for each patient.Data for the 3 month follow-up analyses were obtained from rehabilitation centres and ambulant cardiologists.

Ethical considerations
The study was approved by the local ethics committee in accordance with the principles of the declaration of Helsinki.All authors had unlimited access to the complete data set and had taken responsibility for its integrity.All authors have read and agreed to the manuscript.

Preoperative demographic data
Between September 2015 and May 2017, a total of 30 patients were operated on by means of the Ozaki technique.Demographic data for these patients are summarised in table 1.The patients' mean ± standard deviation (SD) age was 66.83 ± 10.55 years, and 66.67% (n = 20) of the patients were male.Mean EuroSCORE II score was 2.04 ± 1.39.A combined aortic pathology of stenosis and regurgitation occurred in 11 patients (36.67%).Seven patients (23.33%) had pure aortic stenosis and 12 patients (40%) regurgitation only.One patient (3.33%) had active endocarditis.Eleven patients (36.67%) had concomitant coronary heart disease.Angina pectoris class II was present in 9 patients (30%) and class III in 1 patient (3.33%).Twelve patients (40%) had dyspnoea of New York heart Association (NYHA) class III, and 1 patient (3.33%) suffered from dyspnoea of NYHA class IV.NYHA classes II and I occurred in 12 (40%) and 5 (16.67%) patients, respectively.

Intraoperative data
Intraoperative data are presented in table 3. Overall, mean ± SD operation time was 230.53 ± 43.78 minutes, aortic cross clamp time was 95.43 ± 22.96 minutes, and extracorporeal perfusion time was 118.63 ± 40.65 minutes.In isolated Ozaki procedures, the mean operation time was 209.41 ± 35.30 minutes, aortic cross clamp time 85.18 ± 18.10 minutes, and extracorporeal perfusion time was 104.76 ± 38.52 minutes.Mean operation time in combined procedures was 258.15 ± 38.9 minutes, and aortic cross clamp time and extracorporeal perfusion time were 108.85 ± 22.22 and 136.77 ± 36.99 minutes, respectively.Implant sizes for the left and right coronary cusp varied from 23 to 35 mm.For the left coronary cusp, the most frequent sizes were 25 mm (n = 7, 24.14%) and 29 mm (n = 6; 20.69%).For the right coronary cusp, 25 mm (n = 7; 23.33%) and 31 mm (n = 7; 23.33%) were the most frequent sizes.In one patient, the native left coronary cusp could be retained.Sizes for the non-coronary cusp varied from 25 to 35 mm; 8 patients (26.67%) received a 27-mm cusp.In 17 patients (56.67%), an isolated aortic valve replacement was performed, and 13 patients (43.33%) had an additional procedure.These included coronary artery bypass grafting, mi-  resection was accomplished.There was no pacemaker implantation after an isolated Ozaki procedure (n = 0; 0%).Postoperative atrial fibrillation occurred in seven patients (23.33%), of whom five had an isolated Ozaki procedure.The postoperative echocardiographic follow-up before discharge (n = 29) showed a mean LVEF of 56.3 ± 10.67% (n = 27).None of the patients had postoperative aortic stenosis.Twenty-three patients (79.31%) had no or only trace aortic regurgitation before discharge.Mild aortic regurgitation was seen in six patients (20.69%), moderate or severe aortic regurgitation was not seen before discharge.
In patients with preoperative aortic stenosis, the postoperative mean pressure gradient before discharge was 8.00 ± 3.55 mm Hg, and the peak gradient was 14.8 ± 6.21 mm Hg (both n = 15).Postoperative peak velocity was 2.55 ± 2.36 m/s (n = 15), and mean velocity was 1.67 ± 1.33 m/s (n = 13).Postoperative stroke occurred in two patients (6.67%), one of whom had pronounced stenosis of the intracranial blood vessels on postoperative computed tomographic (CT) angiography.The other patient suffered from generalised arteriosclerosis and had had previous thromboendarterectomies on both carotid bifurcations.
The 30-day mortality rate was 3.33% (n = 1).This patient died acutely on the first postoperative day from a transfusion-triggered shock reaction that led to multiorgan failure.

Clinical endpoints
Between postoperative day 30 and the 3-month followup, one more patient died from aspiration pneumonia.This patient had undergone a combined procedure consisting of Ozaki aortic valve replacement, coronary artery bypass grafting, mitral valve reconstruction and septal resection, and suffered a postoperative stroke due to pronounced stenosis of the intracranial blood vessels.Thus, 28 patients completed the 3 months of follow up (table 5).No patient required reoperation or experienced an additional thromboembolic event within the first 3 months.One patient developed moderate aortic valve regurgitation postoperatively.This was found to be due to valvular endocarditis; the patient underwent reoperation 5 months after the initial operation and a biological valve was implanted.

Echocardiographic characteristics after 3 months
The echocardiographic 3-month follow-up was performed in 28 patients (93.33%), a mean of 106.68 ± 63.66 days after the initial operation (table 6).Mean LVEF was 58.89 ± 11.29% (n = 27).None of the patients had evidence of aortic valve stenosis.Moderate aortic valve regurgitation was seen in one patient (3.57%).Mild aortic regurgitation was seen in three patients (10.71%), whereas no or only trace aortic regurgitation was seen in the majority of patients (n = 24; 85.71%).The mean transvalvular pressure gradient was 6.57 ± 3.53 mm Hg (n = 22); peak gradient was 13.51 ± 8.88 mm Hg.

Discussion
The aim of this study was to evaluate the feasibility, transferability and short-term performance of the Ozaki procedure using data from the first 30 patients operated on with this method at the University Hospital of Basel.Despite very promising results published by the inventor of the procedure, to date only limited and unpublished data are available from other centres [5,6].To the best of our knowledge, these are the first data from a cohort of patients undergoing the Ozaki procedure from a centre other than Toho University Ohashi Medical Centre, Tokyo, Japan.After its CE certification in September 2015, we started using the Ozaki technique, one of the first cardiosurgical centres in Europe to do so.Obvious advantages of this new technique, beside the excellent postoperative echocardiographic outcomes that were published earlier, include avoidance of oral anticoagulation, avoidance of foreign material, lower gradients, a larger orifice area and the feasibility of anatomical reconstruction [5,6].Aortic valve replacement with a mechanical or biological prosthesis is a safe and established procedure, and considered the gold standard in the treatment of aortic valve stenosis or regurgitation [4].Therefore, we strived to maintain high safety standards when we introduced the Ozaki technique in our institution.The 30-day mortality rate was 3.33% (n = 1), and the rate of reoperation due to structural valve failure within 3 months was 0%, which underlines the safety of this procedure.There were two thromboembolic events (6.67%); this rate was lower in the study by Ozaki et al. [6].However, one of our patients underwent a combined procedure consisting of aortic valve replacement, myocardial revascularisation, mitral valve reconstruction, and septal resection, and the postoperative CT angiography showed pronounced stenosis of the intracranial blood vessels.[6].In their series, 309 (75.7%) of 416 patients undergoing the Ozaki procedure had a small aortic annulus, which is suggestive of the importance of this issue [6].The literature suggests favouring stentless biological aortic prostheses over stented biological prostheses in patients with small aortic annuli, which results in improved left ventricular function and functional class [7].Similar benefits can be expected from the Ozaki procedure as a result of the anatomical replacement (rather than prosthetic replacement) of the aortic valve and the larger resultant orifice area.However, to date no mid-and long-term echocardiographic data are available to prove these assumptions.Shultz et al. recently published the, to date, largest propensity-matched study comparing stented and stent-less biological prostheses [8].Compared to these results, 30-day mortality was lower in patients treated with the Ozaki procedure in the original publication (2.7% stented, 2.9% stentless, 1.9% Ozaki) [6,8].However, in our study cohort, the 30-day mortality was slightly higher (3.3%); this might be the result of the limited statistical power of our study, with only 30 patients, and because we performed other procedures concomitantly in some cases.As for long-term outcomes, the follow-up survival rate was 83.3% at 73 months after the Ozaki procedure in the original publication (215 concomitant procedures, 201 isolated Ozaki aortic valve replacements), which is slightly higher than the follow-up survival rates of 80.5 and 80.7% at 5 years after implantation of stent-less prostheses and stented prostheses, respectively, published by Shultz et al. [6,8] Unfortunately, Shultz et al. provided no long-term echocardiographic data or information about reoperation [8].In 2016, Repossini et al. published data from a cohort of 565 patients undergoing isolated (n = 350) or combined (n = 215) aortic valve replacement with the stentless Freedom Solo (FS) bovine pericardial valve (Sorin Group, Milan, Italy) [9].After a mean ± SD follow-up time of 6.9 ± 3.7 years, 28 patients (5.2%) underwent reoperation, for several reasons: endocarditis in 9 patients, blunt trauma in 1, and structural valve deterioration in 18 [9].No reoperation due to structural valve deterioration was required within 73 months of follow-up after Ozaki aortic valve replacement in the cohort of Ozaki et al. [6].Sponga et al. recently published their long term echocardiographic data on the FS valve in 109 patients [10].They observed a deterioration in echocardiographic findings in the long term: at discharge, 3-5, and 7-9 years postoperatively, the mean ± SD gradients were 8 ± 4, 12 ± 11, and 19 ±19 mm Hg, respectively [10].Mean ± SD peak pressure gradient 5.5 years after Ozaki aortic valve replacement was 14.3 ± 5.0 mm Hg and echocardiographic follow-up showed no signs of any structural valve deterioration [6].On this basis and the combination of our results with the excellent mid-term results Ozaki et al. [6] published, we consider the Ozaki procedure a potentially effective alternative for younger patients requiring aortic valve surgery for whom a mechanical valve is not an option.Also, the avoidance of artificial material makes the technique favourable in infective endocarditis.

Study limitations
This was a nonrandomised, retrospective single centre study.Because the study protocol was finalised at the end of the observation period, 3-month echocardiographic follow-up data were not complete in all patients and echocardiography was performed by various examiners.Moreover, we present only short-term data.Therefore, a standardised 12-month echocardiographic and clinical followup study to confirm our preliminary results is currently being implemented.Due to the fact that the glutaraldehyde substance is not available on the market in a pharmacopeia quality, the preparation of the glutaraldehyde soaking solution is a high-risk procedure and can be restricted by health authorities.

Conclusion
With our short-term results on aortic valve replacement with use of the Ozaki technique, we were able to confirm the excellent results that Ozaki et al. had published before [5,6].Further evidence from long-term studies and randomized clinical trials are required to guide the treatment and to assess the degree of degeneration of the treated pericardium.
Swiss Medical Weekly • PDF of the online version • www.smw.chPublished under the copyright license "Attribution -Non-Commercial -No Derivatives 4.0".No commercial reuse without permission.See http://emh.ch/en/services/permissions.html.

Table 5 :
Clinical endpoints after 3 months (after in-hospital period).
Swiss Medical Weekly • PDF of the online version • www.smw.chPublished under the copyright license "Attribution -Non-Commercial -No Derivatives 4.0".No commercial reuse without permission.See http://emh.ch/en/services/permissions.html.