The 9 Stress-free Stitches Technique: Feasibility and Outcomes of a New Technique for Aortic Valve Replacement

Objective: This study aims to evaluate the feasibility and short-term outcomes of a novel surgical technique, named ‘the nine stress-free stitches technique’ for aortic valve replacement (AVR) in patients with aortic valve disease. Methods: From May 2013 to October 2015, 63 consecutive patients underwent aortic valve replacement with the nine stress-free stitches technique, using the Magna Ease bioprosthesis (Edwards Lifesciences, Irvine, CA, USA). Demographics, clinical and echocardiographic data were collected retrospectively. The primary endpoints were paravalvular leak, pacemaker insertion, and mortality. Results: There was one case of moderate paravalvular leak. 2 patients required permanent pacemaker implantation. In-Hospital mortality was 6.3%. All deaths were in redo or combined procedure patients. Conclusion: This preliminary series demonstrates that the nine stress-free stitches technique is an acceptable technique for aortic valve replacement. The potential advantages are reduced ischemic time, especially in combined procedures, with easy replication.


Introduction
Aortic valve replacement is an established technique performed in patients with severe aortic valve disease.
Idiopathic senile degeneration with sclerosis and calcification of the aortic valve has become the most frequent type of valve disease in Europe and North America due to its ageing population (2%-7% of the population is over the age of 65). Greater mechanical stress with age and risk factors such as hypertension, smoking, diabetes, and hypercholesterolaemia contribute to 2%-4% of adults >65 years of age suffering from acquired AS.
Symptomatic patients with severe aortic stenosis can develop major adverse cardiac events that have been reported as 80% at 1 year, 63% at 2 years, and 25% at 5 years. Surgical aortic valve replacement (AVR) is currently the gold-standard treatment for symptomatic aortic stenosis. Excellent short-and long-term outcomes have been reported. (with a mortality rate of 2.6%) Despite these results, advancing age, an increasing presence of comorbidities and increased surgical risk in the patient population has stimulated the development of less invasive procedures that may reduce morbidity and mortality.
Transcatheter aortic valve implantation has provided a valid short-term alternative in high-risk patients, reducing 1-and 3-year mortality by at least one third compared with standard medical treatment.
The 9 stress-free stitches technique for aortic valve replacement is a new method, developed in our Cardio-thoracic surgery department.
In this technique, we used nine interrupted stitches for all valve sizes, owing to the geometry and design of the bioprosthesis valve and the careful placement of each suture; nine stitches were enough to ensure natural sealing of the valve to the aortic annulus.

Methods
This is a retrospective, observational study, with a cohort of consecutive patients.
From Data was retrospectively obtained from our database or the referring cardiologist. There were no differences between the two groups in age, gender, hypertension, COPD, smoking, peripheral vascular disease, pre-existing AF, previous stroke and TIA or conduction disorders. The redo and combined procedure group patients were more likely to have diabetes (27% vs. 57%, P > 0.05), and chronic kidney disease (9% vs. 20%, P < 03), but the isolated procedure group patients were more likely to have a bicuspid valve. Patients in the redo/combined procedure group had a significantly higher incidence of hyperlipidemia and coronary artery disease.
Transthoracic echocardiogram (TEE) was performed before discharge and 1 and 10 months after discharge. The operation was performed using the classic median sternotomy approach (n=30, 47%) for redo and combined procedures, or a less invasive approach (partial sternotomy) (n=33, 52%), for isolated replacements.
In this technique a limited midline skin incision was performed, extending 7cm from the angle of Louis ( fig.1), then the sternum was almost totally divided using an oscillating saw. The interclavicular ligament and the xiphisternum were left intact to give post-operative stability. A left ventricular vent was inserted via the right superior pulmonary vein (Fig.2).
Antegrade cold blood cardioplegia was usually repeated every 20 -30 min and immediately in the event of ECG activity.
An S shape aortotomy was made. Traction sutures were placed through the top of each commissure and clipped to the surgical drapes.

Figure.2. CPB established by aortic and atrial cannulation and a left ventricular vent via the right superior pulmonary vein
Three mattress stitches with pledgets were placed at the level of the three commissures with the pledgets in the sub-annular position. The stitches were placed sub-annularly, to create a neo-annulus where the prosthesis could sit and eliminate the dead space between the prosthetic valve and the patient's aorta to reduce leakage. Then 6 large everting U stitches were placed supra-annularly; 2 for each leaflet. These stitches were intended to strengthen and marginalize the annulus below the prosthesis.
The sutures were then passed through the valve sewing ring as close as possible to the valve stent (Fig.3).

348
The 9 Stress-free Stitches Technique: Feasibility and Outcomes of a New Technique for Aortic Valve Replacement The valve was located above the aortic annulus, and the sutures were tied very gently to obtain a "stress-free fixation." The sutures at each of the three commissures were tied first: the stitch at the commissure between the right and left coronary ostia was tied first, then the one at the commissure between the left ostium and the non-coronary leaflets, and finally the suture at the commissure between the right ostium and non-coronary cusp.
The knots at the commissures were not tied tight; they just needed to ensure that the valve sat appropriately on the neo-annulus.
The remaining stitches were tied beginning on either side of the commissure between the left and the right coronary ostia.
This resulted in a true "supra-annular" position of the bioprosthesis using the sewing ring for a pressure dependent sealing.

Statistical Analysis
Continuous values are expressed as mean +/-standard deviation (SD) and categorical variables are displayed as percentages.
For comparison of qualitative variables, the Chi-square test was used.
A p-value less than 0.05 was considered significant.

Results
Operative data was collected during the operation in a common database.
In the redo and combined group procedure, the operative approach used the classic median sternotomy, while in the isolated procedure group employed the limited skin incision and partial sternotomy.
There was no conversion of a small incision into a full sternotomy.
There was no significant difference in valve size implanted. In the redo and combined group, a bigger size prosthesis tended to be used.
Coronary artery bypass grafting (CABG) was the most common other procedure performed at the time of aortic valve replacement in the redo/combined procedure group. The remaining concomitant procedures included simplified MAZE, ascending aorta replacement, mitral valve replacement, and mitral valve repair.
CPB times and ACC times were longer in the redo/combined procedure group. The mean duration of ACC and CPB times was 57 +/-8.9 min and 81 +/-14.7 min respectively for single procedures; 100 +/-26.7 min and 129.5 +/-35 min respectively for combined and redo procedures. (Table 2) Intraoperative TEE revealed satisfactory hemodynamic performance of the valve prostheses, without any significant paravalvular regurgitation, in all cases except in one patient.
There were 4 deaths (6.3%), all due to multiple organ failure, all in the combined and redo procedure group.
One patient without preoperative conduction disorders, operated for acute aortic dissection, developed per-operative atrioventricular (AV) block. He died on the 4 th postoperative day. 2 patients showed third-degree AV block, one had a pacemaker implantation, but the pacemaker check after 2 months showed a normal conduction; the other one resolved after stopping beta blockers.
4 patients presented new minor conduction disorders: 6 patients displayed a first degree AV block, 6 patients had a left bundle branch block (LBBB) (4 complete and 2 partial), and 2 patients had a right bundle branch block (RBBB) (1complete and 1 partial). (Table 3) TEE was routinely performed 8.9 +/-7.5 days post-operatively in all patients. (Table 4) The mean trans-aortic valve gradient was 12.6 +/-4 mmHg for the patients in the isolated AVR group and 14.4 +/-3.4 mmHg in the redo and combined procedure group. The short and middle term result of post-operative echocardiogram revealed no significant difference about the gradients or paravalvular leak between the two groups of patients confirming the success of the operation in both groups.
There was one moderate paravalvular leak without clinical consequences; this complication occurred in the first patient who had a bicuspid aortic valve.
Unfortunately, only 17 patients in the combined and redo procedure group and 24 patients in the isolated AVR group attended the TEE at the last follow up. (Table 4)  There was no migration or structural damage to the prosthetic valve.
None of the patients required a conversion of the 9 stress-free stitches technique into a conventional replacement procedure during the operation, and none of the patients required re-operation for valve failure or paravalvular leak.
One-year mortality was 6.7%, un-related to the AVR.

Discussion
Elderly patients referred for AVR often have multiple comorbidities and new implantation techniques, like sutureless or rapid deployment valves, have been developed to minimize the surgical risk [15][16][17] .
We wanted to create a standardized and simplified procedure to reduce the ACC and CBP time while maintaining the same efficacy, quality, and safety of a conventional approach.
We report a series of AVR cases using the 9 stress-free stitches technique in patients with severe aortic stenosis.
The peculiarity of our method is that suture knots at two of the three commissures are tied very gently, they just need to ensure that the valve will sit appropriately; in fact, one is passed in the muscular septum and the other one near the His bundle so they cannot be tied too tight. They just need to create a new annulus where the prosthetic valve sits naturally.
The 6 U-sutures placed supra-annularly, are intended to give a better support on to a fragile annulus. They help avoid protrusion of the residual annulus we see with sub-annular U-stitches.
With this technique, it is important to pass the stitches near the metallic ring at the base of the valve, to avoid deforming the sewing ring; this decreases the risk of paravalvular leak.
Considerable care needs to be taken to ensure that the biological prosthesis is implanted with perfect sealing at the natural annulus, using the pressure environment of the valve and the design of the prosthetic ring to avoid a para-prosthetic leak.
Compared to what has been reported in the literature, we don't have an increased incidence of leaks despite the limited number of stitches used to implant the valve.
In our series there was one moderate paravalvular leak without clinical consequences. This complication occurred in the first patient who had a bicuspid aortic valve.
According to the literature the incidence of paravalvular leak in patients undergoing aortic valve replacement is 2-10% and 1 to 5% of patients have hemodynamic repercussions and become symptomatic.
The risk of developing an early paravalvular leak is increased by technical difficulties such as annular calcification, suturing technique, prosthetic size, and shape [18][19] .
To decrease these risks we created a new annulus where the prosthesis can sit, and we used the soft sewing ring of the bioprosthesis to model against the patient aorta and ensure the sealing; choosing the right sizing of the prosthesis is a fundamental step of our technique; in the case of the bicuspid valve we center the geometry of the bioprosthesis to the fusion commissure.
In our series, 33 patients underwent an isolated AVR surgery through a small skin incision without conversion.
Compared to a full sternotomy or a J-ministernotomy, our small incision gives a better operative position and an improved cosmetic appearance.
Compared to the literature, our ACC time is equal if not shorter than conventional replacement through full sternotomy or ministernotomy 7,9,12,13 (Table 5 -6). This is due to the association of a small skin incision giving a good exposure to the aortic valve with a fast implanting technique.    (Table 6).
When compared to sutureless and rapid deployment valve implantation, our ACC times are significantly longer, but our rate of pacemaker implantation and paravalvular leak is lower. Out of a total of 14 patients with bicuspid aortic valve, we had just one moderate paravalvular leak in the very first patient of the series.
In the articles by Nguyen et al. published in 2015 that studied the feasibility of sutureless technique in patients with bicuspid aortic valves, the results describe a 20% rate of permanent pacemaker implantation and 12% of paravalvular leak at the pre-discharge TTE 20 (Table 8).

352
The 9 Stress-free Stitches Technique: Feasibility and Outcomes of a New Technique for Aortic Valve Replacement Our results suggest that our procedure is technically feasible and safe and it is associated with satisfactory hemodynamic and clinical results.
However, follow-up is limited, and no conclusion can be drawn regarding the long-term results.

Conclusions
This study demonstrates that the 9 stitches technique is a safe and acceptable treatment for critical aortic valve disease The potential advantages of this procedure can be a shorter aortic cross-clamp (ACC) and cardiopulmonary bypass (CPB) time; especially in combined procedures. It can be used without increasing the risk of paravalvular leak.
Further investigation and data collection will allow the assessment of valve performance over a more extended period