Initial experience with the Freedom Solo^® stentless aortic valve in a low volume centre

Abstract

Objectives. Freedom Solo is a stentless biological aortic valve which is implanted supra-annularly with a single suture line. An increased risk of postoperative thrombocytopenia in the early postoperative period has been reported in recent studies. In our study we evaluated postoperative haemodynamic performance and thrombocyte-levels. Design. Thirty seven patients who underwent valve implantation of the Sorin Freedom Solo stentless valve were included. The haemodynamic performance of the valve was evaluated by transthoracic echocardiography postoperatively at the fourth day (mean) and after a median of 4.2 months. Results. The mean gradient (mmHg) of Freedom Solo was 7.5 at four days and 8.6 at 4.2 months. Postoperatively no patient had more than grade 1 leakage. Seven percent of the patients had a reduction of thrombocytes to less than 20% of the preoperative level. Seventy six percent had a minimum postoperative thrombocyte level less than 100*10⁹/L. The 30 days mortality in our patient material was zero. Conclusions. Implantation of the Freedom Solo valve was uncomplicated in our experience. Favourable transvalvular gradients and no significant leaks were found. In accordance with the literature, we found a high percentage of patients having a postoperative level of thrombocytes less than 100*10⁹/L after implantation of Freedom Solo.

Key words::

• Cardiac surgery
• valve disease
• aortic valve replacement
• stentless valve
• Freedom Solo
• thrombocytopenia

The Freedom Solo valve is a stentless biological aortic valve which is implanted in a supra-annular position (1). Stentless valves are assumed to result in a better postoperative haemodynamic performance than stented valves. This in turn might be beneficial for regression of left ventricular hypertrophy (2), which is a risk factor for mortality after aortic valve replacement (AVR) (3). Although many cardiac surgical units have experience with implantation of stentless aortic valves, it is obvious that stented valves are still the most commonly used implants. Stentless valves seem to be perceived more difficult and time-consuming to implant. Traditionally stentless valves are implanted using two rows of sutures. A few years ago a new stentless valve, the Freedom solo, was introduced. This valve appears to be associated with shorter cross-clamp times (ACC), probably because only one running suture is required to secure the valve in a supra-annular position (4). Low postoperative thrombocyte levels after implantation of Freedom Solo are reported in two recent studies (5,6). The Freedom Solo has been used in our unit since 2006. The main aim of this retrospective study was to evaluate our initial experience with this valve, including postoperative valve gradients and postoperative thrombocyte-levels.

Material and methods

Patient population

Thirty seven patients (20 males) had a Sorin Freedom Solo stentless valve implanted between December 2006 and June 2009 and were included in this study. During this period a total of 264 patients underwent biological aortic valve implantation at our hospital. Surgical procedures performed were: AVR (11 patients), combined AVR and coronary artery bypass grafting (24 patients), and combined AVR and tricuspid valve reconstruction (two patients).

Surgical technique

The patients followed our standard anaesthetic regimen. All medication was maintained until the day before surgery, beta-adrenergic blockers and nitrates were given also on the day of surgery. Age- and weight-related doses of morphine-scopolamine were given for premedication. Anaesthesia was induced with diazepam, pentothal, fentanyl and pancuronium and maintained with isoflurane and supplements of fentanyl up to a total dose of 10 μg/kg. Propofol was used for sedation during CPB and postoperatively until extubation. Morphine was given for post-operative analgesia.

Cardiopulmonary bypass (CPB) employed standard membrane oxygenator alphastat blood gas control, pump flow of 2.4 L/m², venous temperature 32–34°C and mean arterial blood pressure between 50 mmHg and 70 mmHg. The perfusion circuit was primed with 1 800 ml of Ringer's acetate containing 7 500 U of heparin. Before CPB heparin (300 U/kg) (Leo, Copenhagen, Denmark) was given to achieve a kaolin activated clotting time (ACT) of >480 seconds. Additional heparin was given when needed to keep the ACT above 480 seconds. Cold crystalloid or blood cardioplegia were employed during CPB. Cardiotomy suction was used while the patients were fully anticoagulated. The patients were warmed to a rectal temperature of at least 36°C before termination of CPB. After CPB, protamine sulphate was used as needed to reverse the heparin effect. Blood remaining in the CPB circuit was collected and transfused to the patients.

A standardized surgical technique was used in all patients. In cases with combined procedures, the distal coronary anastomoses were done before the valve procedure. Coronary surgery was performed with a left internal mammary graft to the left anterior descending coronary artery in all patients where the left anterior descending artery was bypassed. Aortocoronary vein grafts were constructed to all other territories. Distal anastomoses were constructed first and proximal anastomoses were sewn with use of a partial aortic clamp. The decision to implant a Freedom Solo valve was taken following aortotomy and excision of the diseased native aortic valve. A bicuspid native aortic valve and extensive supra-annular calcifications were considered contra-indications to implantation of the Freedom Solo. The size of the annulus and the sino-tubular junction were calibrated carefully using the Freedom Solo valve sizers. A valve labeled with the size that matched the annular size closest was chosen. Polypropylene sutures through the commissures were used to expose the annulus. The valve was implanted using three individual 4/0 polypropylene sutures with a small needle (Ethicon RB-1). The sutures were started in the nadir of each sinus 2–3 mm above the annulus, thereafter passed through the external pericardial sheet midway between each commissure and tied. The sutureline was completed in a running over-and-over fashion up to each commissure taking good bites in the supraannular tissue. The sutures were then tied on the adventitial surface of the aorta over pledgets. The aortotomy was closed in a standard fashion.

All valves were evaluated carefully with perioperative transesophageal echocardiography. A fast track extubation and mobilization policy was followed postoperatively.

Monitoring, measurements and recordings

The haemodynamic performance of the valve was evaluated by transthoracic echocardiography before the patient left the hospital (mean time of evaluation: fourth postoperative day, 95% confidence interval (CI): (3, 5)). In addition, all the patients were routinely called in for a new transthoracic echocardiogram after a median of 4.2 months postoperatively (95% CI: (4, 7)). Mean gradient over the bioprothesis and any leakage were registered. In three patients before leaving the hospital and in five patients at follow up, the mean gradient was not recorded.

Postoperative routine blood samples were taken of all the patients the first days after surgery. The minimum postoperative thrombocyte-level for every patient was registered and evaluated by us. Patients with preoperative thrombocyte levels <100*10⁹/L (two patients), and patients receiving thrombocyte-concentrate (seven patients) during the hospital stay were excluded from the calculations and the statistic of the postoperative thrombocyte levels.

Statistical analyses

All statistical tests were performed in SPSS 16.0. The distribution of any variable was assessed by the q-q plot-function. Variables are presented as mean and 95% confidence interval in brackets. When having a skewed/non-parametric distribution the variables are presented as median and 95% confidence interval in brackets. The p-value of the slope of an estimated regression line where considered when deciding whether a linear relationship between two variables existed.

Results

Patient characteristics

Pre-, peri- and postoperative characteristics are summarized in Tables I and II. Intraoperative transesophageal echocardiography resulted in intraoperative reexploration of the valve in two patients. One patient had a paravalvular leak that was corrected by a single suture upon reexploration. In another patient, one of the suturelines was accidentally damaged on the external surface of the aorta and the valve had to be resuspended. Recovery of both patients was uneventful.

Table I. Preoperative characteristics.

Valve replacement during same hospitalization	9 (24%)
Elective surgery	28 (76%)
Standard EuroSCORE	8 (7, 9)
Age (years)	76 (75, 78)
Operative procedure
AVR	11 (30%)
AVR + CABG	24 (65%)
AVR + tricuspid valve repair	2 (5%)
Male/female	20/17 (54% v. 46%)
Previous open-heart surgery	4 (10.8%)
ASA
3	14 (38%)
4	23 (62%)
NYHA
1	3 (8%)
2	9 (24%)
3	24 (65%)
4	1 (3%)
Diabetes mellitus	2 (5%)
BMI (kg/m^2)	26.4 (24.6, 28.2)
Weight (kg)	76 (71, 81)
Hypertension	22 (60%)
Hypertrophy left ventricle	31 (84%)
AMI within last 4 weeks prior to surgery	3 (8%)
AMI within last 3 months prior to surgery	4 (11%)
Preoperative CKMB (μg/l)	4.10 (†) (3.91, 5.16)
Preoperative creatinine (μmol/l)	83.08 (75.61, 90.55)
Renal disease (creatinine > 140)	1 (3%)
Preoperative Hbg (g/dL)	13.4 (12.9, 13.8)
Preoperative mean transvalvular aortic gradient (mmHg)	52 (46, 59)
Pre-operative CRP (mg/l)	5 (†) (4, 8)
Pre-operative CRP-value > 5	6 (17%)

AMI: Acute myocardial infarct, ASA: anaesthesiologists physical status, AVR: aortic valve replacement, BMI: body mass index, CABG: coronary artery bypass grafting, CKMB: creatinine kinase-MB, CRP: C-reactive protein, Hbg: haemoglobin, NYHA: New York Heart Association.

Values are mean or median (†), and 95% confidence intervals in brackets.

Table II. Per- and postoperative characteristics.

Cardioplegia
Cold crystalloid	2 (5%)
Cold blood	35 (95%)
CPB (min)	113 (99, 127)
ACC (min)	79 (71, 86)
Size implanted valve
21 mm	7 (19%)
23 mm	11 (30%)
25 mm	9 (24%)
27 mm	10 (27%)
Heart insufficiency
postoperatively	3 (8%)
Inotropic agent(s)	2 (5%)
IABP
Hgb when leaving hospital (g/dl)	9.8 (9.6, 10.1)
CKMB 1^st postoperative day (μg/l)	15.60 (†) (15.55, 24.94)
CKMB 2^nd postoperative day (μg/l)	7.70 (†) (7.43, 12.62)
SvO2 arrival ICU (%)	61.6 (58.0, 65.1)
SvO2 6 hours after arrival ICU (%)	63.1 (59.0, 67.12)
SVO2 1st post-operative morning (%)	62.2 (59.6, 64.8)
Postoperative bleeding (ml)	630 (†) (647, 1035)
CRP 2^nd postoperative day (mg/l)	142 (123, 161)
Maximum postoperative creatinine (μmol/l)	88 (†) (86, 132)
Postoperative renal failure creatinine > 140	5 (14%)
Prisma/dialysis	1 (3%)
ICU stay (days)	1 (†) (1, 3)
Postoperative intubation (min)	270 (†) (184, 892)
Reoperation for bleeding during hospital stay	4 (11%)
MOF	2 (5%)
Post-operative myocardial infarction	0
Re-intubation	2 (5%)

ACC: Aortic cross-clamping, CPB: cardio pulmonary bypass, CKMB: creatinine kinase-MB, CRP: C-reactive protein, Hgb: haemoglobin, IABP: intra-aortic balloon pump, MOF: multiple organ failure (failure of three organs or more), Postoperative bleeding: total postoperative bleeding from surgery until the first postoperative morning, SvO₂: systemic venous oxygen saturation.

Values are mean or median (†), and 95% confidence intervals in brackets.

Haemodynamics

The mean gradient (mmHg) of Freedom Solo after implantation is not reduced throughout the 4.2 first postoperative months, and was 7.5 (95% CI: (6.3, 8.8)) at four days and 8.6 (95% CI: (7.0, 10.2)) at 4.2 months respectively. In Figure 1 each valve size is considered separately with regard to mean pressure gradients at 4.2 months. The dataset suggests a negative linear relationship between the size of the valve and the mean gradient of the valve at 4.2 months. The estimated regression line was estimated by SPSS to be Mean gradient = 32.778–0.993*(valve size), (p=0.005).

Figure 1. Transvalvular mean gradients according to valve size. The figure presents transvalvular mean gradients according to valve size at 4.2 months postoperatively. Each box represents the distribution of an individual valve size falling between the 25^th and the 75^th percentile. The horizontal line within the box represents the median, while the vertical lines outside includes all the registered values except outliers and extreme cases. *Extreme case.

Both on the fourth postoperative day and at 4.2 months three patients had a leakage of grade 1 (Table III). The other patients had a minimal or no leakage at all.

Table III. Type and degree of leakage.

	Valvular, grade 1	Paravalvular, grade 1	Unknown location, grade 1
At the 4^th postoperative day	1 (3%)	1 (3%)	1 (3%)
After 4.2 months	2 (6%)	1 (3%)	0

Thrombocyte-levels

The thrombocyte level was at its minimum at a mean of two days postoperatively. Two patients (7%) had a reduction of their thrombocytes to less than 20% of the preoperative level. Twenty two patients (76%) had a minimum postoperative thrombocyte level less than 100*10⁹/L.

Thirty days mortality

None of the patients in this study died within 30 days postoperatively.

Discussion

This retrospective analysis summarizes the initial experience with the Freedom Solo stentless aortic prosthesis in a low volume cardiac surgical unit. The decision to implant a Freedom Solo valve was primarily dependent on the operating surgeon’s preference. In a number of cases a Freedom Solo valve was not considered, because a bicuspid valve or extensive calcifications were found. The final decision to implant this stentless valve was taken following excision of the native valve. The technique of valve implantation is different from the usual technique used for stented biological valves. However this was found to be neither a problem nor an advantage with respect to ACC in our study. In our series ACC (minutes) all procedures were higher compared to previously published data of Freedom Solo implantation: 79 in our study vs. 56 (4). The ACC for isolated valve replacement was 64 in our study vs. 39 reported in a previous study (1). It seems that our careful and unhurried implantation policy resulted in that no implant time advantage was achieved. This may reflect a learning curve. Technical problems during implantation were not encountered. Moreover, in patients with extensive calcification at an annular or subannular level, implantation of the Freedom Solo prosthesis appeared to be the easier than implantation of a standard stented valve. Also, tension free closure of the aortotomy is easier compared to situations where a stented prosthesis is used. Another potential advantage for stentless valves is that it may be easier to perform a percutaneous valve in valve implantation at a later point of time, in cases where the implant fails. However, we did not find a reference in the literature to support this concept.

The Freedom Solo is implanted in a supraannular position with a continuous sutureline. The suture does not cross the annulus at all and it can be speculated that the risk for damage of conductive tissue with subsequent AV-block should be lower compared to standard aortic valve implants. None of our patients had a temporary or permanent AV-block. Naturally, our material is not large enough to draw firm conclusions. This issue has to our knowledge not been discussed in the literature.

In our patient material the biological aortic valve Freedom Solo gives haemodynamic advantages such as a low mean gradient and no significant leakage. Stentless valves are assumed to give a larger effective aortic orifice area than stented valves, and therefore better postoperative valve haemodynamics. But within the groups of stented and stentless valves, different types and differences in size make comparisons complicated. Measured gradients in our material (mean gradient 8.6 at 4.2 months) were significantly lower than published data on stented prosthesis (7) (Edwards Perimount Magna (EPM) and Medtronic Mosaic (MM): mean gradients 10 and 17 respectively after one year) when comparing the overall gradient in the population, irrespective of implanted valve size. Also when considering the valve size, the mean gradients of the individual Freedom Solo valve sizes in our material were lower than the gradients to corresponding sizes of stented valves in the EPM and the MM (7). Furthermore, when we compared the individual stentless valve sizes within our data-material we observed differences in gradients between them. This dataset suggests a negative linear relationship between the size of the valve and the mean gradient of the valve at late follow-up. When comparing the performance of stented and stentless valves per size it has to be taken into account that stentless valves are usually oversized to allow optimal function. According to the manufacturer, the Freedom Solo is already oversized. This means that a valve labeled 21 really is a 23 valve. Nevertheless the median size implanted (according to the label) was larger compared to our own experience with stented valves (data not shown).

The literature on the haemodynamic superiority of stentless valves is inconclusive. A study by Sensky et al. (8) reports haemodynamic advantages of stentless valves: Mean thansprosthetic gradients of the stentless porcine valves are smaller than the gradients of stented. Furthermore Walther et al. (2) concludes that stentless valves results in a higher degree of left ventricular hypertrophy-regression, probably because of better transvalvular dynamics. In contrast a study by Risteski et al. (9) reports no significant haemodynamic differences between patients having implanted stented or stentless bioprotheses (mean gradient of 9.9+/−4.8 mmHg vs. 10.2+/−4.2 mmHg). Furthermore Payne et al. (10) have performed a meta-analysis comparing stented and stentless valves, and are concluding that there is no difference in postoperative mean gradient between the valves (but there is a lower peak gradient among the patients having implanted stentless valves). A weakness of this meta-analysis is that only three studies are included to evaluate mean transvalvular pressure early outcome and only two studies are included to evaluate mean transvalvular pressure late outcome. One (the study by Sensky et al. discussed above (8)) of the three (8,11,12) studies included actually disagrees with the conclusion of the meta-analysis saying that there is a lower mean thansprosthetic gradient with the stentless porcine valves compared to the stented. A study by Doss et al. (11) is also included and evaluates early and late mean pressure gradient outcomes. This publication includes the same patient material as the study of Risteski discussed above. The main drawback of this study is that only 20 patients are included in each group. Finally the study by Cohen et al. (12) also included in the meta-analysis reports internal valve diameters similar for stented and stentless valves although manufacturer`s labeled valve size was much larger for stentless valves. This is surprising and may explain why the stentless valve did not perform better in this study.

In view of the fact that stentless valves are assumed to give a larger aortic valve orifice than corresponding stented valves, it is reasonable to think that especially patients measured to have a small aortic bioprothesis implanted would gain a haemodynamic advantage by a stentless valve. When making subgroups of patients in regard to the size of the aortic annulus (implanted valve <19 mm), Narang et al. (13) found a difference in both mean gradient and left ventricular mass between patients with stented vs. stentless valves. The difference was not statistical significant, probably because of few patients in this category of small annulus. In the same study, among a subgroup of patients with low left ventricular ejection fraction (<50%) patients having implanted a stentless valve had a better improvement in left ventricular mass regression than patients having implanted a stented valve.

In the literature, low thrombocyte levels in Freedom Solo patients are reported. In a study comparing thrombocytes levels, Hilker et al. (5) found a significant higher number of patients with postoperative thrombocyte levels <100 between the second and the fifth postoperative day among patients having received stentless valves (Freedom Solo) compared to patients having received stented valves (71.9% vs. 36.6%). This corresponds to our result (76% of the patients in our study had postoperative thrombocyte levels <100*10⁹/L). In contrast our number of patients having a reduction of thrombocytes to less than 20% of its preoperative level is much lower than reported in a study (6) by Yerebakan et al. (2/29 versus 8/20). The study by Yerabakan et al. also compared thrombocyte levels between patients having implanted a stentless valve (Freedom Solo) vs. a stented valve, observing differences between them already from day one.

In our experience implantation of the Freedom Solo valve was uncomplicated. We measured low transvalvular gradients postoperatively and no significant leak was found. A high percentage of patients had a low postoperative level of thrombocytes.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.