A geometric model of the normal human aortic root and design of a fully anatomic aortic root graft.

OBJECTIVE: Available aortic root grafts generally flare outward in the sinus region, and this feature improves procedural ease. However, no current device is based on normal aortic root geometry, and a fully anatomic aortic root graft could further facilitate valve-sparing root operations. METHODS: To develop a model of the normal human aortic root, high-resolution computed tomographic angiogram images from 11 normal human aortas generated high-density x, y, z coordinates of valve and root structures in Mathematica. Three-dimensional least-squares regression analyses assessed geometry of the aortic valve and root. Shapes and dimensions were quantified, and minor variations in geometry were simplified during graft design. RESULTS: Normal aortic valve and root geometry was represented as three leaflet-sinus general ellipsoids nested within a cylindrical aorta. Sinotubular junction diameter was 5 mm larger than the valve base-with a slight funnel-shaped outward commissural flare but cylindrical geometry above the midvalve. The valve base was elliptical, but the midvalve and the sinotubular junction were circular above the midvalve level. Commissural locations on the base circumference were equidistant. On the basis of average three-dimensional geometry, a root graft was designed for root remodeling procedures-to be used with an internal geometric annuloplasty ring of the same design. CONCLUSIONS: An aortic root graft was designed on the basis of mathematical analyses of computed tomographic angiogram images. The design incorporated three anatomic sinuses, commissural symmetry, and compatibility with geometric ring annuloplasty. The anatomic graft may prove useful for restoring aortic root geometry toward normal during aortic valve and root surgery.

Hemodynamic outcomes of geometric ring annuloplasty for aortic valve repair: a 4-center pilot trial.

OBJECTIVES: A geometric annuloplasty ring could improve efficacy and stability of aortic valve repair. Toward this goal, a 1-piece 3-dimensional titanium annuloplasty ring with Dacron covering was developed and tested successfully in animals. The purpose of this study was to define hemodynamic outcomes with this device used as the annuloplasty component of human aortic valve repair. METHODS: In a 4-center pilot trial with informed consent, 16 patients underwent aortic valve repair for aortic insufficiency, with the annuloplasty device sutured into the annulus beneath the leaflets. Preoperative annular diameter averaged 26.5 ± 2.0 (mean ± standard deviation) mm, and average ring size was 22.3 ± 1.2 mm. After annuloplasty, leaflet defects were easy to identify, and 14 of 16 patients (88%) required leaflet plication and/or autologous pericardial reconstruction for leaflet defects. Three patients had ascending aortic replacement, and 2 had remodeling root replacement. One had ultrasonic leaflet decalcification and another tricuspid valve annuloplasty. Follow-up data were from site-specific studies at the 6-month postoperative time point. RESULTS: There were no in-hospital mortalities or major complications. Preoperative aortic insufficiency grade (0-4 scale) was 3.6 ± 1.0 and fell to 1.0 ± 0.8 at 6 months (P < .0001). New York Heart Association class fell from 2.5 ± 0.5 to 1.1 ± 0.3 (P < .0001). Postrepair valve area was 2.7 ± 0.2 cm(2), and 6-month mean systolic gradient was 11.3 ± 3.3 mm Hg. Left ventricular end-diastolic diameter and ejection fraction both normalized (both P < .0001). CONCLUSIONS: Geometric ring annuloplasty facilitated aortic valve repair, allowing more precise reconstruction of leaflet defects. Aortic insufficiency reduction and systolic gradients were excellent, and expansion of valve reconstruction into broader categories of aortic valve disease seems indicated.

Design characteristics of a three-dimensional geometric aortic valve annuloplasty ring.

OBJECTIVE: A full geometric annuloplasty ring could facilitate aortic valve repair. The purpose of this report was to document the design of such a ring using mathematical analyses of normal human aortic valve computed tomographic angiograms. METHODS: One-millimeter axial slices of high-resolution computed tomographic angiograms from 11 normal aortic roots were used to generate high-density x, y, and z coordinates of valve structures in Mathematica. Three-dimensional least squares regression analyses of leaflet-sinus coordinates were used to assess geometry of aortic valve and root structures. RESULTS: Normal valve geometry could be represented as three leaflet-sinus general ellipsoids nested within an elliptical aortic root. Minor-major diameter ratio of the valve base was 0.60 ± 0.07, and elliptical geometry extended vertically up the commissures. By contrast, leaflet-sinus horizontal circumferences were fairly circular (diameter ratios, 0.82-0.87), and the left coronary/noncoronary commissural post was located at the posterior base minor diameter-circumference junction, with the center of the right coronary leaflet opposite. Post location on the circumference was symmetrical, with a deviation of only ±2% to ±3% from 33.3% symmetry. Commissural posts flared outward by 5 to 10 degrees, and leaflet areas were statistically equivalent (P > 0.10). From end diastole to midsystole, the aortic root became less elliptical (diameter ratio increased by 0.15), but root area expanded minimally (less than +5%). A one-piece rigid annuloplasty ring was designed with 2:3 base ellipticality, three 10-degree outwardly flaring symmetrical posts, and post height = base circumference/2π. CONCLUSIONS: A three-dimensional aortic annuloplasty ring was designed that could prove useful for enhancing applicability and stability of aortic valve repair.

In vivo testing of an intra-annular aortic valve annuloplasty ring in a chronic calf model.

OBJECTIVE: To increase applicability and stability of aortic valve repair, a three-dimensional aortic annuloplasty ring has been developed for intra-annular placement. The goal of this study was to test the safety of this device with in vivo implantation in the calf model. METHODS: In 10 chronic calves, the HAART annuloplasty ring was sutured to the aortic valve annulus using cardiopulmonary bypass. The animals were recovered and followed for 1-2 months. Serial echocardiography was used to evaluate valve competence, and contrast aortograms and CT angiograms were obtained in selected animals. After completion of follow-up, each animal was euthanized, and aortic endoscopy was performed under water distension in five. Full autopsies with histologic examinations were performed. RESULTS: All animals survived surgery. Two were euthanized in the first week for complications, and the remaining eight calves were followed uneventfully for the 1-2 months. Serial echocardiography showed completely competent valves in all but one animal, in which the ring was intentionally up-sized to test the sizing strategy. Contrast aortographic and CT angiographic findings were similar to the echocardiograms. Postmortem examination showed proper seating of all rings with endothelialization at 1-2 months. All valves demonstrated good leaflet coaptation and no abnormalities. CONCLUSIONS: In vivo testing of a three-dimensional aortic annuloplasty ring in a chronic calf model proved to be very successful and safe. Using the sizing and implant strategies developed, human trials seem indicated.

Technique for aortic valve annuloplasty using an intra-annular "hemispherical" frame.

OBJECTIVE: A need exists for a stable annuloplasty method for aortic valve repair. On the basis of a "hemispherical" model of aortic valve geometry, a prototype annuloplasty frame was constructed and initially tested. Specific goals were to develop a clinically applicable transaortic implant technique and to evaluate the system in isolated and intact animal preparations. METHODS: Eight isolated porcine aortic roots were perfused from a water reservoir at a constant pressure of 100 mm Hg, and valve leak was measured by timed collection in a beaker. Baseline leak was negligible, and the 2 commissures adjacent to the right coronary leaflet were incised vertically to create severe valve insufficiency. Through a transverse aortotomy, a "hemispherical" annuloplasty frame was sutured to the aortic valve annulus with horizontal mattress sutures. The 3 posts of the frame were sutured first to the 3 subcommissural regions to align the device horizontally and vertically. The 3 frame curvatures then were sutured into the leaflet annuli using horizontal mattress "looping" sutures, supported with supra-annular pledgets. Post-repair valve leak was measured, and significance of the change was assessed with a 2-tailed paired t test. One survival implant was evaluated in an intact calf. RESULTS: Subcommissural incision disrupted annular geometry and created a valve leak of 1400 ± 847 mL/min (mean ± standard deviation). Suturing the 3-dimensional annuloplasty frame into the aortic valve annulus reestablished appropriate geometry of leaflet coaptation and restored valve competence, with a reduction in valve leak to 102 ± 86 mL/min (P = .004). After 6 weeks of chronic calf implantation, the frame was well healed and the native valve functioned normally. CONCLUSIONS: Transaortic insertion of a "hemispherical" annuloplasty frame into severely disrupted and insufficient porcine aortic valves routinely and effectively restored valve competence. These data support the continued development and testing of this device as a stable method of annuloplasty during aortic valve repair.