[Percutaneous cardiac valve replacement: a review]. / Remplacement percutané des valves sigmoïdiennes. Revue de la littérature.

Surgery is no longer the only technique to replace a cardiac valve. New percutaneous procedures allow aortic or pulmonary valve implantation. Even if the feasibility of these procedures has been proved, cases reported are very rare and selected. This emergent technology is still at an early stage of development and new prospective studies will be necessary to evaluate these procedures correctly before concluding their clinical benefit. At this time surgery remains the gold standard in terms of cardiac valve replacement.

The aortomitral angle is suspended by the anterior mitral basal "stay" chords.

OBJECTIVES: The function of the anterior mitral basal "stay" chords (SC) is not yet known. Collagen fiber orientation of the anterior mitral leaflet (AML) suggests that local stress is directed from papillary muscles (PM) over SC and AML to fibrous trigones (FT), maintaining the aortomitral angle (AMA). It has been shown that narrowing of AMA increases risk of systolic anterior movement (SAM). METHODS: Sonomicrometry crystals were implanted in six sheep at the left ventricular (LV) apex, PM tips (M1, M2), FT (T1, T2), posterior mitral annulus (PMA), and base of aortic right coronary sinus (RCS). The retracting force of ascending aorta was measured. RESULTS: Transection of SC resulted in an increase of distance M1-T1 and M2-T2. Consequently, the AMA narrowed at end systole by -3.26+/-0.85 degrees (p<0.05) and at end diastole by -4.16+/-1.28 degrees (p<0.05). A force of 1.8+/-0.2 N was needed to pull the recoiling ascending aorta back to its original position. CONCLUSIONS: The elastic recoil of ascending aorta is balanced by SC, which connect PM to FT and constitute the center of the LV base. Transection of SC narrows AMA and increases the risk of SAM.

A four-dimensional study of the aortic root dynamics.

OBJECTIVE: Although aortic root expansion has been well studied, its deformation and physiologic relevance remain controversial. Three-dimensional (3-D) sonomicrometry (200Hz) has made time-related 4-D study possible. METHODS: Fifteen sonomicrometric crystals were implanted into the aortic root of eight sheep at each base (three), commissures (three), sinuses of Valsalva (three), sinotubular junction (three), and ascending aorta (three). In this acute, open-chest model, the aortic root geometric deformations were time related to left ventricular and aortic pressures. RESULTS: During the cardiac cycle, aortic root volume increased by mean+/-1 standard error of the mean (SEM) 33.7+/-2.7%, with 36.7+/-3.3% occurring prior to ejection. Expansion started during isovolumic contraction at the base and commissures followed (after a delay) by the sinotubular junction. At the same time, ascending aorta area decreased (-2.6+/-0.4%). During the first third of ejection, the aortic root reached maximal expansion followed by a slow, then late rapid decrease in volume until mid-diastole. During end-diastole, the aortic root volume re-expanded by 11.3+/-2.4%, but with different dynamics at each area level. Although the base and commissural areas re-expanded, the sinotubular junction and ascending aorta areas kept decreasing. At end-diastole, the aortic root had a truncated cone shape (base area>commissures area by 51.6+/-2.0%). During systole, the root became more cylindrical (base area>commissures area by 39.2+/-2.5%) because most of the significant changes occurred at commissural level (63.7+/-3.6%). CONCLUSION: Aortic root expansion follows a precise chronology during systole and becomes more cylindrical - probably to maximize ejection. These findings might stimulate a more physiologic approach to aortic valve and aortic root surgical procedures.

Aortic and pulmonary root: are their dynamics similar?

OBJECTIVES: The long-term behavior of the pulmonary autograft in the aortic position (Ross procedure) remains uncertain. Using three-dimensional (3D) sonomicrometry (200 Hz) we compared the dynamics of the aortic and pulmonary roots. METHODS: Twenty-four crystals were implanted in each aortic (eight sheep) and pulmonary roots (six sheep) at: base (3 x 2), commissures (3 x 2), sinotubular junction (3 x 2), ascending aorta (3) and pulmonary trunk (3). Under stable hemodynamic conditions, geometric changes were time-related to left ventricular pressure (LV) and aortic pressure. RESULTS: The expansion of the aortic root is twice that of the pulmonary root. During the cardiac cycle, the aortic root volume increased by 37.7 +/- 2.7% (mean +/- SEM) versus 20.9 +/- 1.0% for the pulmonary root. Both were cone-shaped at end diastole. Because expansion at commissures was twice that of the base, both roots became more cylindrical during ejection. Although both roots started to expand prior to ejection and reached maximal expansion during the first third of ejection, the commissural and sinotubular junction dynamics were different in each root. While in the aortic root, expansion at commissural and sinotubular junction levels was significantly different (63.7 +/- 3.6% versus 37.0 +/- 2.1%), in the pulmonary root, they were similar (29.0 +/- 1.3% versus 27.7 +/- 1.4%). Expansion of the three sinuses was also different (P<0.001). In the aortic root: the right expanded more than the left and more than the non-coronary sinus. In the pulmonary root: the right sinus expanded more than the anterior more than the left. CONCLUSIONS: Dynamic differences might explain the global pulmonary root dilatation when subjected to systemic pressure, particularly at the level of the sinotubular junction which might result in the autograft failure. Differences in the asymmetrical expansion of the aortic and pulmonary roots should be considered for the implantation of the pulmonary autograft in the most physiological position.

Hemodynamic evaluation of a new stentless autologous pericardial mitral valve.

BACKGROUND: There is no satisfactory mitral valve prosthesis. An ideal mitral valve substitute should be biologic, nonantigenic, and anatomically correct. METHODS: We developed a stentless, chordal-supported (including anterior basal stay chords) mitral valve made with glutaraldehyde-treated autologous pericardium. Eight such prostheses were implanted in sheep. RESULTS: Seven animals survived the operation and were studied postoperatively immediately, at 1 week, and at 1 month. Simultaneous left ventricular and left atrial pressures showed peak and mean transvalvular pressure gradients of 6+/-2 mm Hg and 1+/-1 mm Hg, respectively. Echocardiography performed intraoperatively and then 1 week and 1 month postoperatively showed normal valve leaflet movements. Color and pulsed Doppler echocardiography showed no sign of transvalvular stenosis or regurgitation. Effective orifice area was 5.39+/-0.35 cm2 at intraoperative, 5.51+/-0.29 cm2 1 week after operation (n = 5), and 5.51+/-.28 cm2 1 month after operation (n = 3). Three animals were sacrificed at 19 days and at 1 and 3 months. One animal is alive at 10 months. CONCLUSIONS: This new stentless pericardial mitral valve performed satisfactorily with low gradients and no regurgitation. Possible advantages of this pericardial valve are excellent hemodynamics, ease of construction and implantation, lack of immunogenicity, and low cost. Similarly designed valves but with a shorter nonglutaraldehyde treatment time have been used in 3 sheep monitored for more than 3 months.

Pathogenic mechanisms in rheumatic carditis: focus on valvular endothelium.

To clarify immune-mediated mechanisms in rheumatic heart disease caused by group A streptococcal infection, valve tissues from rheumatic patients with valvular heart disease who required valve replacement were studied for reactivity with monoclonal anti-CD4 or anti-CD8 monoclonal antibodies or anti-vascular cell adhesion molecule-1 (VCAM-1). At the valve surface, CD4(+) and CD8(+) T lymphocytes were adherent to valve endothelium and penetrated through the subendothelial layer. T cell extravasation into the valve through the surface valvular endothelium appeared to be an important event in the development of rheumatic heart disease. VCAM-1 was expressed on the valvular endothelium in rheumatic valves. Evidence suggested that the pathogenesis of rheumatic heart disease involved the activation of surface valvular endothelium with the expression of VCAM-1 and the extravasation of CD4(+) and CD8(+) lymphocytes through the activated endothelium into the valve. Lymphocytic infiltration through the valve surface endothelium has not been appreciated as a potential initiating step in disease pathogenesis.

A new method for the preservation of aortic valve homografts.

BACKGROUND AND AIM OF THE STUDY: Aortic valve homografts were treated with 50% ethanol and glycerol followed by freeze-drying (D-Hydro). Comparative results of fresh versus D-Hydro-treated aortic roots implanted for up to nine months in the descending aorta of sheep with induced aortic regurgitation (AR) are reported. METHODS: Six fresh and six D-Hydro valves were implanted in 12 sheep for three, six and nine months, and echocardiography and pressures were taken at surgery and sacrifice. Tissue sections were stained with hematoxylin and eosin, von Kossa, Masson's trichrome, Movat's pentachrome, von Willebrand factor, CD3 (a T-cell marker) and smooth muscle alpha-actin. RESULTS: No grafts had increased gradients after implantation, or at sacrifice. At explantation, fresh homografts showed early pannus formation followed by thrombus, annular dilatation and wall calcification. Leaflets were thickened and progressively retracted. All had severe AR. The appearance of D-Hydro-treated homografts was normal, except for mild leaflet retraction in three, resulting in AR (in two animals the induced AR had healed). Histologically, a T-cell-mediated reaction was evident in the fresh homografts, and collagen distortion was noted. Calcification was present in all fresh specimens and was severe at nine months. D-Hydro roots showed only minor calcification in the six-month samples. Normal collagen, and a complete layer of von Willebrand factor-stained cells were present at three months. At nine months, cell rehabitation extended for two-thirds of the leaflets (alpha-actin +). The inflammatory reaction was very mild, with CD3+-stained cells absent in most samples. CONCLUSION: Aortic valve homografts treated with the D-Hydro freeze-drying method performed better than fresh homografts due to the absence of thrombus and annulus dilatation, limited calcification, and rehabitation of the aortic wall and parts of the leaflet by myofibroblasts, as well as the presence of a complete endothelial layer on the aortic wall and leaflet.

Activation of myocardial angiogenesis and upregulation of fibroblast growth factor-2 in transmyocardial-revascularization-treated mice.

OBJECTIVE: To evaluate the growth factor responses associated with myocardial angiogenesis. DESIGN: Mice were treated with transmyocardial revascularization (TMR) and evaluated for angiogenic and growth factor responses. METHODS: TMR was performed via thoractomy with a 27 g needle. At 2, 5, and 7 days post-treatment, hearts were removed from the TMR treated and control groups, then assayed for angiogenesis, fibroblast growth factor (FGF)-2 expression and vascular endothelial cell growth factor (VEGF) expression. RESULTS: TMR caused an angiogenic reaction in the myocardial blood vessels at 7 days post-TMR treatment. Elevated FGF-2 corresponded with increased TMR related angiogenesis. VEGF levels only increased in hearts that were prewounded then TMR treated. CONCLUSIONS: The data show that TMR stimulates myocardial angiogenesis. The angiogenic reaction is mediated by FGF-2 which increased in most experimental treatment groups. The VEGF response was more specific, requiring prewounding then TMR treatment for a VEGF increase.

A study of the junction between glutaraldehyde-treated allogeneic aorta and host aorta.

BACKGROUND AND AIM OF THE STUDY: Stentless aortic valve bioprostheses have become popular because of their superior hemodynamics and expected increased durability. However, the stentless bioprosthesis differs from stented valves in that glutaraldehyde (GA)-treated tissue is implanted in direct contact with the native aorta. The effect of GA-treated tissue on host tissue has not been reported. METHODS: In order to analyze the effect of GA in the healing process, sheep descending aortic conduits treated with 0.625% GA were inserted in the descending thoracic aorta of 10 adult sheep. The implants were removed after 4, 5, 10, 12, 15, 25, 30, 32, 60 and 120 days. The upstream and downstream junctions were evaluated macro- and microscopically, and by immunohistology for smooth muscle cell alpha-actin and von Willebrand factor. RESULTS: By day 60 of implantation, the GA-treated conduits were calcified. By days 60 and 120, the calcification had spread to the host aorta, and was seen as foci of calcification in the junctional area. Acellular areas were also seen in the host aorta near the anastomosis. A fibrotic layer spanning the abluminal aspect of the junction between the implant and host aorta was present at day 4 and continued through 120 days. This layer was characterized by a progressive increase in collagenous matrix and cellularity, as well as new blood vessel formation. The luminal aspect of the junction had a neointimal layer of variable thickness containing alpha-actin-expressing cells covered by a monolayer of von Willebrand factor-expressing cells, seen at 15-30 days and present through 120 days. CONCLUSION: In our model, implanting GA-fixed tissue in direct contact with living tissues resulted in cell death and calcification of host tissue within 60 days. The integrity of the junction did not appear to be compromised. This may be of interest in light of the increased popularity of the stentless aortic bioprosthesis.

Significant increase of aortic root volume and commissural area occurs prior to aortic valve opening.

BACKGROUND AND AIM OF THE STUDY: The increased use of autologous, homologous or heterologous aortic root demands a detailed knowledge of its anatomy and function. The advent of 3-D digital sonomicrometry offered the opportunity to acquire precise information on the root and leaflet movements during the cardiac cycle. METHODS: Under cardiopulmonary bypass, sonomicrometry crystals were implanted in the aortic root and valve of eight sheep. Crystals were sutured at each commissure (n = 3), the top of the sinotubular junction (n = 3), lowest point of the annulus (n = 3), and leaflet tip (n = 3). 3-D coordinates of each crystal were recorded, together with left ventricular and aortic root pressures and ECG. When the animal had returned to a stable hemodynamic condition, the maximum and minimum distances between two crystals, and areas between three crystals, were calculated. Changes in root volume and leaflet position were time-related to the pressure changes. RESULTS: The most significant change between maximum and minimum distance between crystals during the cardiac cycle occurred at the commissural level. Similarly, the triangle defined by the three commissural crystals showed the greatest change in area (47%). The root volume increased by an average of 22%; about 40% of this increase occurred during the isovolumic phase. The aortic leaflets began to open before ejection. CONCLUSION: We postulate that aortic valve opening is initiated by the outward pull of the commissures. These findings should impact on aortic root surgery.

Behavior of vital and killed autologous pericardium in the descending aorta of sheep.

OBJECTIVES: Cardiovascular implants of fresh autologous pericardium produced mixed results including fibrosis with retraction or thinning and dilatation. The reasons for these differences are unknown but may involve activation of cells intrinsic to the tissue implant. To better understand the behavior of autologous pericardial implants, we studied the outcomes of vital pericardium (fresh) versus ethanol-killed pericardium. METHODS: Fresh and ethanol-killed autologous pericardium was transplanted as a patch, a conduit, or a rectangular flap bisecting the lumen in the descending aorta of sheep. The implants, recovered at 1, 5, 10, 15, and 30 days, were evaluated macroscopically and microscopically and by immunohistologic studies. RESULTS: Fresh implants showed good preservation with fibrin deposition on day 15. Microscopically, cells positive for alpha-actin and von Willebrand-related antigen appeared in the fibrin by day 10. By day 30 the flap was fibrotic and retracted whereas the patch and conduit retained their original appearance on the luminal aspect. An endothelium-like layer expressing von Willebrand-related antigen was present in the patch and conduit but absent in the flap. In contrast, the ethanol-killed implants were free of fibrin by day 10. By day 30, there were no signs of fibrosis or retraction, and a surface layer of cells expressing von Willebrand-related antigen, characteristic of endothelial cells, was present on all implants. All ethanol-killed implants were repopulated by host cells. CONCLUSION: The transluminal flap is an interesting model for studying the behavior of intraluminal autologous pericardial cardiovascular implants. Killing of the pericardial implants alleviated the fibrosis and tissue retraction observed with fresh flap implants.

Expanding indications for the Ross procedure.

The pulmonary autograft procedure has been shown to provide excellent hemodynamic results in young patients with aortic pathology. However, the use of this procedure in those with more complex aortic disease has not been extensively evaluated. The purpose of this report is to present the application of the Ross procedure in a 21-year-old man with extensive acquired aortic root pathology, both subannular and supraannular, and prosthetic valve dysfunction after two previous procedures.

Aortic root geometry: pattern of differences between leaflets and sinuses of Valsalva.

BACKGROUND AND AIMS OF THE STUDY: Growing interest in aortic root replacement with the use of stentless auto-, homo- and xenografts, and new developments in aortic valve conservation demand a deeper understanding of the normal aortic root anatomy. METHODS: Ten cryopreserved human aortic roots were pressurized, fixed and measured directly (leaflet free edge and attachment) and using three-dimensional computed tomography imaging software (sinus of Valsalva height and volume). RESULTS: The mean of the measurements of all four parameters yielded a pattern in which the non-coronary sinus (N) structures had the greatest dimensions followed by the right (R) and then the left (L). Non-parametric ANOVA on each of these parameters also showed significant differences among the sinuses yielding a pattern of N > R > L. This pattern determined an angle of tilt between the plane at the base (annulus) and the plane intersecting the sinotubular junction with a mean value of 11 degrees. Linear regression indicated that this angle did not depend on the size of the base (annulus). CONCLUSIONS: The data showed a geometric pattern of the aortic root, with the structures of the non-coronary sinus being the largest followed by the right and then the left. The possible hemodynamic relevance and surgical implications of these findings need to be explored.

Surgical techniques for the repair of anterior mitral leaflet prolapse.

Myxomatous disease is the leading cause of mitral valve regurgitation in the developed world. Although posterior mitral leaflet (PML) prolapse is the most common cause of regurgitation and of repair, lesions of the anterior mitral leaflet (AML) are often considered beyond the possibilities of repair. The surgical anatomy and pathology of the AML and a precise terminology for the location of the lesions are described. The surgical maneuvers for commissural prolapse, chordal shortening, chordal replacement, leaflet resection, and annuloplasty are also described. The advantages and limitations of each maneuver are discussed. It is concluded that a flexible approach to the repair of the AML is necessary to adapt the best technique for each type of lesion. No particular technique can be applied systematically to all lesions. It is the author's contention that the majority of AML lesions can be repaired safely.

Multiplane transesophageal echocardiography: a roadmap for mitral valve repair.

Multiplane transesophageal echocardiography (TEE) is useful in providing a detailed anatomic map for successful mitral valve repair. This report describes an approach, developed over the past two to three years, which helps to delineate valve anatomy in specific detail. Mid-esophageal views are selected to view different segments of the valve leaflets. When correlated with surgical anatomy, this approach is found to be both practical and useful.

Critical analysis of the Ross procedure: do its problems justify wider application?

The Ross procedure has shown superior hemodynamic results in young patients with aortic root pathology. Wider application of the procedure is restricted by its technical complexity and potential associated problems. The mortality/morbidity associated with 130 consecutive patients who have had the Ross procedure using the root replacement implantation technique between October 29, 1990, and October 8, 1998 is summarized. New York Heart Association (NYHA) preoperatively was class I, 23.5%; class II, 64.7%; and class III, 11.8%; mean age was 36 years (range 3 to 67 years). Men accounted for 73.8% and women 26.2% of the series. Preoperative diagnosis was congenital, 80.7%; rheumatic, 5.3%; failed prosthesis, 7.0%; degenerative, 2.6%; and endocarditis, 4.4% with preoperative aortic insufficiency (AI) 7.9% 1+, 19.8% 2+, 29.7% 3+, and 42.6% 4+, respectively. At operation mean cross-clamp time was 201 minutes (range 102 to 280 minutes). Patient follow-up was 99.2% (1 patient lost to follow-up), and 94.4% were NYHA class I at follow-up and 5.6% class II. Postoperative AI was 0 to 1 + in 93.6% and 2+ or greater in 6.4%. Mean time to patient follow-up was 436 days (range, 20 days to 2,878 days). Thirty-day mortality rate was 1.5%; one patient died of mediastinal bleeding, and one from complications of acute pancreatitis. There was no late mortality. Early autograft explant occurred in one patient secondary to iatrogenic injury to the pulmonary autograft at the time of harvesting, and one late explant occurred secondary to proximal suture line dehiscence. Late autograft repair occurred in one patient secondary to a false aneurysm along the proximal suture line; one patient was reoperated for left main coronary stenosis relative to iatrogenic injury at the time of the procedure. Right ventricular outflow tract replacement has occurred in two patients. Postoperative morbidity and mortality for the Ross procedure, as shown in this series, remains low and supports broader application of the procedure.