Sutureless (SU) technology can be beneficial in surgical cases when stitches are technically problematic, as in minimally invasive surgery,1–3 hypercalcified aortic roots,4,5 or redo surgery. Sutureless technology is also a prerequisite to perform totally endoscopic surgical aortic valve replacement (TEAVR).6
Despite these advantages, SU valves are associated with a greater incidence of paravalvular aortic insufficiency than conventional sutured valves.7,8 In all the three CE-marked SU models (The LivaNova Perceval, LivaNova, Saluggia, Italy; the Medtronic 3f Enable, Medtronic, Minneapolis, MN USA; and the EDWARDS INTUITY, Edwards Lifesciences, Inc, Irvine, CA USA), one of the preeminent causes of this complication seems to be due to an inadequate sizing of the annulus. On the other hand, results of transcatheter therapies are improving, since the rate of grade II aortic insufficiency has decreased form an early incidence, from previous studies, ranging between 10% and 40%, to now less than 3.5%, with the third-generation valves like the SAPIEN 3 (Edwards Lifesciences, Inc, Irvine, CA USA) in the CE mark study.9 In the practice of transcatheter aortic valve implantation (TAVI), sizing of the annulus by transesophageal echocardiography suffers from variability in measurement caused by off-axis cross sections, and multidetector computed tomography is actually the first-line option for choosing the appropriate prosthesis. However, this method is not perfect, and there is still some variability in sizing especially the virtual basal ring leading to an undersized prosthesis, which has been described in several reports.10
Compared to TAVI, image-based sizing is even more inadequate in preoperative determination of SU valve sizes, since it analyzes an anatomical setting which will be modified after removal of the valvular leaflets.
Therefore, surgical sizing needs to be ameliorated to decrease the incidence of paravalvular leakages with SU bioprosthesis; and endoscopic sizers need to be developed, since actual conventional SU sizers cannot be inserted throughout a surgical trocar and cannot be used in the totally endoscopic SU approach.
We present in this article a new concept of sizer, which addresses both these issues.
The Sizer Device
The endoscopic feedback force sizer is a biocompatible polyamide PA2200 printed prototype composed by a holder (in the actual version, the holder is 15 cm long; the shaft diameter is 15 mm) with two extremities. On one extremity, there is a polypropylene expandable tape, and on the other a turning knob (Fig. 1A).
When the operator has introduced the sizer into the thorax via the main working trocar (20 mm; Flexipath, Ethicon, Inc, Somerville, NJ USA) and into the aortic annulus (Fig. 1B), he holds the device with one hand and turns the knob with the other (Fig. 1C). The plastic tape expands (Figs. 1C, D), until full contact with the native aortic annulus is established. The tape is a 17-mm-high soft plastic part that can unroll (from diameter 18 to 39 mm; Figs. 1C, D) and roll from the outside of the thorax, by rotation of a scroll wheel. When the tape comes in contact with the native annulus, the diameter is measured and indicated outside the thorax in a window under the scroll wheel (Fig. 1E). Once the tape is adherent to the aortic annulus (Fig. 1F), the operator checks with the 5-mm fiberoptics that the tape is completely expanded into the aortic native annulus and at the correct level (not in the sinuses of Valsalva or too low into the outflow tract of the left ventricle); the corresponding diameter is reported in the window (Fig. 1E). The evaluation of the resistance corresponding to the further tape expansion (resistance to the oversizing, depending on the native annulus compliance) is physical, although not quantified in this version of the device, and the feedback is just a personal perception of the operator (feedback force). In the following version of the device, the resistance of such further knob rotation is measured by an external torque indicator (Fig. 2) and could be connected to a dynamometric system.
Whatever the diameter of the native aortic annulus is, the sizer has been conceived so the plastic tape can transmit to the surrounding tissues a maximal radial force equivalent to or higher than the radial forces exerted by the actually commercially available SU stent. Radial force of the SU valves has been previously measured in our laboratory at different diameters preceding their full expansion; on this basis, we calibrated the spring inserted into the calibrator so as to absorb the energy of the different degrees of oversizing at different diameters.
Cadavers were prepared at the anatomy laboratory of the St-Etienne University and in the surgical school of the University of Lorraine to perform totally endoscopic SU valve implantations. The endoscopic setting was prepared with the following trocar disposition (Fig. 3A): two working ports (second and third intercostal space), one port in the second space to insert the 30-degree 10-mm optics, a Chitwood clamp in the first intercostal space. In brief, the setting is similar to the setting we adopted previously in the clinical development of TEAVR.6
Once the aorta is transected and the margins exposed via three Prolene stitches extracted transcutaneously, the aortic valve leaflets are removed.
We evaluated the placement and the deployment of the sizer into the aortic root, under camera control (10 mm, 30 degrees). Afterward, SU valve implantation was scheduled.
Two models of nitinol-stented biological SU valves (Medtronic 3f Enable and LivaNova Perceval) have been considered for the experience, since all sizes of both model can be compressed to 19 mm of diameter and be introduced throughout a 20-mm thoracoscopic trocar (Flexipath; Ethicon, Inc) (Figs. 3B, C).
The LivaNova Perceval Model
The endoscopic implantation technique of the Perceval model follows the same technique established for implantation via full sternotomy or minimally invasive open surgery. Three guiding stitches are passed into the three nadirs of the aortic valve commissures. They are extracted, in the thoracoscopic setting, outside the main (second space) working trocar before insertion of the Perceval flange, taking care to manage them to avoid any rotation of the device during the descent of the valve (Fig. 3C). Balloon expansion is routinely performed (Figs. 3D, E).
The Medtronic 3f Enable Model
The endoscopic implantation technique has been recently detailed6,11: in brief, after a manual collapse of the leaflet destined to the noncoronary aortic annulus, the 3f Enable valve is kept collapsed using two stitches of prolene 4/0, one passed into the flange and one at the level of the upper part of the nitinol frame. The prosthesis can then be introduced into the thorax and scrolled to the level of the aortic valve annulus using an endoscopic grasper. Once the guiding stitch is knotted, the valve position is adjusted and the prolene stitches cut to decollapse the valve. Further adjustments of the level of sitting of the valvular flange can be performed.
Five implantations of each model were scheduled (total of 10 implantations). Once the sizing of the prosthesis is achieved, the correct deployment of the valve (without migration toward the aortic root or the ventricle) is performed with the camera. Afterward, further control of the sealing is performed with a nerve hook around the flange of the prosthesis to exclude a passage toward the left ventricle outflow tract (Fig. 2F).
Sizing of the Annulus and Choice of the Appropriate Valve Size
Once the operator has sized the annulus, exerting an expansive force similar to that of the nitinol stent of the valve, the choice of the corresponding size of bioprostheses to implant has been assessed as indicated in Table 1.
After removal of the native valve leaflets, introduction of the sizer through the main working trocar into the aortic root, positioning, and expansion under camera control were technically successful in the 10 scheduled cases. In the five cases of implantation of the LivaNova Perceval, sizes of the implanted prostheses were one small, three large, and one medium, respectively. In the five cases of 3f Enable Medtronic implantation, selected sizes were 27, 27, 23, 21, and 19 mm, respectively. Correspondence between the effective native annular diameters seized by the calibrators and the sizes of implanted bioprostheses is reported in Table 2. After expansion of the Nitinol stents of the bioprosthesis, the sealing was checked with a nerve hook around the flange of the valves, without evident passage into the outflow tract of the left ventricle (Fig. 3F).
Enthusiasm regarding SU new technology is mitigated by the issue of the selection of the valve size, since inappropriate size of the prosthesis seems to be the preeminent etiology of paravalvular leakages or device migration. The development of an appropriate sizer aims both to address an ideal sizing, taking into account the dynamic stent expansion that anchors the SU valves to the annulus, and to facilitate its use in the minimally invasive approaches.
The concept of our prototype was originally designed to enable a physical sizing of the aortic annulus throughout a thoracoscopic trocar (15–20 mm of diameter), aiming to increase TEAVR reproducibility.
The current version of the sizer presents some limitations linked to the diameter of the holder (15 mm of transverse diameter), still partially obstructive in small aortic roots. Although during all the cases of this cadaver experience the visualization of the deployment of the plastic tape into the aortic annulus has been possible using fiberoptics (10 mm, 30 degrees), a thinner holder, in clinical practice, would be even less obstructive in the aortic root. Moreover, the actual holder is still rigid, requiring a perfect alignment of the working trocar point of entry into the thorax and the long axis of the aortic root. That alignment can condition the correct positioning of the sizer into the annulus. Although guiding stitches can be helpful to correct and adjust the aortic annular plane, an evolution of the actual prototype, with a steerable holder, is under development.
Beyond the thoracoscopic approach, the calibrator addresses a general issue linked to the actual system of calibration of the aortic annulus before implantation of a SU valve. Anchoring of SU valves into the aortic annulus depends mainly from the ratio between the radial force of the nitinol frame on one hand and the opposite resistance to the expansion that the native aortic annulus can oppose on the other hand. Moreover, it is known that depending on the degree of surgical decalcification, aortic annuli present a variable compliance. An aortic annulus deeply decalcified, especially in some anatomical configurations (bicuspid type 0 of Sievers) can approach the compliance of annuli with a purely regurgitant aortic valve. That annular compliance is the main trigger of delayed paravalvular leakage, since delayed annular dilatation can follow the valve implantation. Although sizing during our TEAVR initial experience was only done by multidetector computed tomography and transesophageal echocardiography image6 without postoperative paravalvular leakages, the image-based sizing before SU valve implantation cannot be comparable to sizing before TAVI procedures owing to the decalcification of the annulus, which alters the compliance of the aorta.
Use of devices measuring “dynamically” the real diameter of the aortic annulus, taking into consideration the elastic compliance of tissues, needs tables of correspondence between those effective native annular diameters and the prosthetic sizes. Some variables, like the thickness and the elasticity of the flange of each prosthesis, may condition the choice of the correct size for each aortic annulus diameter. The table that we provided, in reference to the LivaNova Perceval and the Medtronic 3f Enable valves, is based on an initial experience of use of our calibrator, and thus it is perfectible.
Some anatomical settings are critical both in sizing and in implanting of a SU valve. When the annulus is not properly decalcified or keeps some deep major calcifications, which make the annulus remain elliptic or asymmetrically compliant, the plastic tape expands inside and adapts to those irregularities, providing the mean diameter corresponding to that area. However, the radial force of the SU valve may not be sufficient to circularize the annulus, with a suboptimal sealing.
Sizing with an underlying bioprosthesis may be critical, as with a conventional sizer, and implantation of a SU prosthesis may be contraindicated when the mitroaortic distance is not sufficient. We estimate that a minimal mitroaortic distance of 5 mm is necessary; otherwise, the sizer will not expand correctly (the tape may pop up from the native annulus) because of an insufficient surface of contact in the noncoronary annulus portion.
We could not test the effectiveness of the device in partially decalcified annuli (condition that can happen in some bailout implantation of SU prosthesis), since we had to adapt to the noncalcified valves that were found in cadavers. This issue can limit also the usefulness of future animal essays. We could test the feasibility of a rapid sizing through a trocar before implantation of a SU valve: absence of paravalvular leakage was presumed when the hook did not pass between the prosthetic flange and the native annulus into the left ventricle outflow tract. Nevertheless, since we worked in flaccid hearts, this study did not explore a preclinical hemodynamic performance.
After the study, in August 2015, the 3f Enable bioprosthesis was withdrawn from the market. The interest of the study is now restricted to the Perceval valve, since we have not yet performed this endoscopic cadaver testing with the INTUITY prosthesis (Edwards Lifesciences), which needs an open access (ministernotomy or minithoracotomy) for implantation.
Further laboratory work is scheduled to test the appropriateness of the selected size of the implanted bioprostheses (and the incidence of paravalvular leakage) by a controlled pressurization of the aortic root. Afterward, the prototype will be used in a pilot study for clinical testing.
The proof of concept performed in our anatomy laboratory seems encouraging concerning the technical feasibility of a dynamic endoscopic measure of the aortic annulus. Further in vitro and in vivo testing is needed.
1. Martens S, Zierer A, Ploss A, et al. Sutureless aortic valve replacement
via partial sternotomy. Innovations
2. Glauber M, Miceli A, Gilmanov D, et al. Right anterior minithoracotomy versus conventional aortic valve replacement
: a propensity score matched study. J Thorac Cardiovasc Surg
3. Santarpino G, Pfeiffer S, Concistrè G, Fischlein T. Perceval S aortic valve implantation in mini-invasive surgery: the simple sutureless solution. Interact Cardiovasc Thorac Surg
4. Concistrè G, Farneti P, Miceli A, Glauber M. Sutureless aortic bioprosthesis
in severe aortic root calcification: an innovative approach. Interact Cardiovasc Thorac Surg
5. Gatti G, Benussi B, Camerini F, Pappalardo A. Aortic valve replacement
within an unexpected porcelain aorta: the sutureless valve option. Interact Cardiovasc Thorac Surg
6. Vola M, Fuzellier JF, Chavent B, Duprey A. First human totally endoscopic aortic valve replacement
: an early report. J Thorac Cardiovasc Surg
7. Eichstaedt HC, Easo J, Härle T, Dapunt OE. Early single-center experience in sutureless aortic valve implantation in 120 patients. J Thorac Cardiovasc Surg
8. Folliguet TA, Laborde F, Zannis K, Ghorayeb G, Haverich A, Shrestha M. Sutureless Perceval aortic valve replacement
: results of two European centers. Ann Thorac Surg
9. Webb J, Gerosa G, Lefèvre T, et al. Multicenter evaluation of a next-generation balloon-expandable transcatheter aortic valve. J Am Coll Cardiol
. 2014;64: 2235–2243.
10. Dementhon J, Rioufol G, Obadia JF, et al. A novel contribution towards coherent and reproducible intravalvular measurement of the aortic annulus by multidetector computed tomography ahead of transcatheter aortic valve implantation. Arch Cardiovasc Dis
11. Vola M, Campisi S, Anselmi A, Faure M, Fuzellier JF, Gerbay A. Video-assisted minithoracotomy approach: technical developments towards totally endoscopic sutureless aortic valve replacement
. J Card Surg
This proof of concept study describes an endoscopic expandable valve sizer to facilitate thoracoscopic aortic valve replacement with a sutureless prosthesis. The authors examined this device in 10 aortic torsos. In these experiments, endoscopic sizing was technically feasible and the scheduled aortic sutureless valve implantation was successfully performed. These data support the feasibility of endoscopic expandable sizers in a cadaveric model. Further experimental evaluation will be needed in more clinically relevant animal models to prove the effectiveness of this device. However, their preliminary results are encouraging, and support continued investigation of this technology.