Three-dimensional echocardiography [1], sonomicrometry

Transcription

1 The Effect of Regional on Mitral Valve Annular Saddle Shape Joseph H. Gorman III, MD, Benjamin M. Jackson, MD, Yoshiharu Enomoto, MD, and Robert C. Gorman, MD Harrison Department of Surgical Research, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania Background. The mitral valve annulus has a distinctive saddle shape. Recent finite element analysis indicates this shape may contribute to normal valve function by increasing leaflet curvature and reducing leaflet stress. This study tests the hypothesis that acute ischemic mitral regurgitation (AIMR) is associated with loss of annular saddle shape. Methods. Sonomicrometry array localization (SAL) measured the three-dimensional geometry of the mitral annulus in 6 sheep before and after 30 min of posterior ischemia that produced severe AIMR. Using this SAL data the annular height to commissural width ratio (AHCWR), a measure of annular saddle shape, was calculated throughout the cardiac cycle and reported as a percentage. Results. The normal mitral annulus accentuated its saddle shape rapidly during isovolemic contraction: AHCWR increased from 11.6% 1.1% 13.9% 1.6% (p < 0.001). ejection AHCWR remained relatively constant ranging from a minimum of 14.1% 1.5% to a maximum of 14.9% 1.3%. ischemia AHCWR was found to be significantly smaller (p < 0.05) during isovolemic contraction, ejection, and isovolemic relaxation, but not during diastolic filling. Whereas ischemia did not affect AHCWR at end diastole (11.6% 2.8%), the isovolemic accentuation of the saddle shape was lost. Conclusions. The normal mitral annulus accentuates its saddle shape during systole. This accentuation is eliminated during ischemia that causes AIMR. These data suggest an association between annular saddle shape and valve competency. (Ann Thorac Surg 2004;77:544 8) 2004 by The Society of Thoracic Surgeons Three-dimensional echocardiography [1], sonomicrometry [2], and marker radiography [3, 4] have been used to demonstrate that the mitral valve annulus has a distinctly nonplanar saddle shape. A recent theoretical finite element model has shown that this annular shape imposes leaflet curvature that in combination with leaflet bulging reduces stress on valve components during systole [5]. As a result annular saddle shape may contribute to valve competence. This concept has implications for the surgical repair of mitral regurgitation secondary to ischemic or degenerative etiologies. Most repair strategies rely on ring annuloplasty to restore normal geometry and decrease tension on suture lines [6, 7]. All currently available annuloplasty devices are essentially flat. When implanted the rigid and semirigid devices only restore the annular geometry in two dimensions. The height of the annulus is obliterated diminishing leaflet curvature and potentially placing increased stress and subsequent strain on the repair. The effects of annular flattening on ventricular remodeling are unknown but could be significant. We used an ovine model and sonomicrometry array localization (SAL) imaging to test the hypothesis that annular flattening is associated with and may contribute Accepted for publication July 3, Address reprint requests to Dr Joseph H. Gorman, Department of Surgery, 6 Silverstein, Hospital of the University of Pennsylvania, 3400 Spruce St, Philadelphia, PA 19104; gormanj@uphs.upenn.edu. to the development of acute ischemic mitral regurgitation (AIMR). Material and Methods In compliance with guidelines for humane care (National Institutes of Health Publication No. 85 to 23, revised 1985) six Dorsett sheep (35 45 kg) were induced with sodium thiopental (10 15 mg/kg iv), intubated, and anesthetized with isofluorane (1.5% 2%) and oxygen. The surface electrocardiogram and arterial blood pressure were monitored. Through a sterile left lateral thoracotomy snares were placed around the proximal second and third obtuse marginal and posterior descending branches of the circumflex coronary artery (OM2, OM3, and PDA) [8, 9]. cardiopulmonary bypass six 2 mm hemispherical PZT-5A piezoelectric transducers (Sonometrics, London, Ontario, Canada) were placed around the mitral valve annulus in each sheep as described previously [2, 10]. Figure 1 shows the relationship of the annular transducers, leaflet anatomy, coronary anatomy, and regional ischemia. The posterior commissure (PC) and the anterior commissure (AC) transducers were placed at the clefts between the anterior and posterior leaflets. The AC transducer was in all animals very near to the central fibrous body. The aortic crystal (Ao) was placed on the aortic-mitral continuity at the posterior trigone. The 2004 by The Society of Thoracic Surgeons /04/$30.00 Published by Elsevier Inc doi: /s (03)

2 Ann Thorac Surg GORMAN ET AL 2004;77:544 8 ISCHEMIC CHANGES IN 3D ANNULAR SHAPE 545 Fig 1. The relationship of the sonomicrometry transducers, annular anatomy, coronary anatomy, and ischemic region: The nomenclature for the transducers is as follows: AC anterior commissure; AO aortic; PC posterior commissure; P1 anterior portion of mural annulus; P2 mid portion of mural annulus; and P3 posterior portion of mural annulus. OM2 and OM3 mark the second and third obtuse marginal branches of the circumflex coronary artery, respectively. PDA marks the posterior descending coronary artery. Note that transducers P2, P3, and PC are most closely associated with the ischemic region. Also note that AC and AO are closely related to the anterior and posterior trigones of the heart s fibrous skeleton, respectively. annular transducers marked P1, P2, and P3 were centered over the anterior, middle, and posterior scallops of the mural leaflet, respectively. wound closure the coronary snares were placed in a subcutaneous position. Seven days later sheep were again sedated with thiopental, placed supine, intubated, anesthetized with isofluorane, and mechanically ventilated. A high fidelity double pressure transducer (SPC-350, Millar Instruments Inc., Houston, TX) for simultaneous measurements of left ventricular (LV) and aortic root pressures was passed percutaneously via a femoral artery. A Swan-Ganz catheter (Baxter, Irvine, CA, model 131-HF7) was placed via the left internal jugular vein for measurement of pulmonary artery pressures, central venous pressure, and thermodilution cardiac output. Surface electrocardiogram (ECG), LV, and aortic root pressures were monitored continuously. Subdiaphragmatic echocardiographic color flow Doppler velocity maps (E-CFDVM) to assess the degree of mitral regurgitation (model 77020A, Hewlett-Packard Inc., Santa Clara, CA) were obtained via a sterile upper midline laparotomy as previously described [11]. The severity of mitral regurgitation (MR) was assessed quantitatively by the area of the regurgitant jet as a percentage of left atrial area in the apical four-chamber view. The following grading was used: grade 1 20%; grade 2 20% 40%, grade 3 40% 60%, and grade 4 60% [12]. Transducer wires were connected to a Sonometrics Series 5001 Digital Sonomicrometer (Sonometrics, London, Ontario, Canada). Ventilation was suspended during sonomicrometry measurements. All distances between the six transducers were measured every 5 ms during a 5 s data run [2]. The ECG, LV, and aortic root pressures were recorded simultaneously with the sonomicrometry data. Coronary snares were exteriorized. Before ischemia each animal received bolus lidocaine (15 mg/kg) followed by an infusion (2 mg/min). Normal saline containing 2 gm MgSO 4 and 40 meq KCl per liter was infused at 70 ml/h. Snares, occluding OM2, OM3, and PDA, were sequentially tightened over a 5-min period. Thirty minutes after the last snare was tightened SAL data were again recorded and MR reassessed by color flow Doppler echocardiography. Animals were euthanized with 1 gm of thiopental and 80 meq of KCl. Hearts were removed and opened to verify the placement of the sonomicrometry transducers. Data Analysis As described previously [2] sonomicrometry distance data were used to determine the three-dimensional (3D) coordinates of each transducer every 5 ms throughout the cardiac cycle. The saddle shape of each annulus was quantified using an annular height to commissural width ratio (AHCWR). AHCWR is reported as a percentage. A least-squares plane was fitted to all six annular transducers, then AHCWR was calculated (as previously described) [2, 5] by dividing the height of the annulus perpendicular to this plane by the intercommissural distance (Fig 2). Fig 2. Image of a normal ovine mitral annulus at end systole: a three-dimensional (3D) surface has been fit using the 3D coordinates of six annular sonomicrometry transducers. The surface is not meant to represent the leaflets but rather to allow visualization of the 3D shape of the annulus. Note the saddle shape. Again, the nomenclature for the transducers is as follows: AC anterior commissure; Ao aortic; H annular height; PC posterior commissure; P1 anterior portion of mural annulus; P2 mid portion of mural annulus; P3 posterior portion of mural annulus; W intercommissural distance. (A) Annulus as viewed from posterior annulus to aorta. (B) Annulus as viewed from anterior to posterior commissure. Also illustrated is the definition of the annular height to commissural width ratio (AHCWR), a parameter used to quantify the degree of mitral annular nonplanarity. The shading of the images represents the vertical displacement of the transducers and surface relative to a best fit annular reference plane; lighter shading represents points closer to the ventricular apex.

3 546 GORMAN ET AL Ann Thorac Surg ISCHEMIC CHANGES IN 3D ANNULAR SHAPE 2004;77:544 8 Table 1. Hemodynamic Summary HR MAP LVEDP PAD CVP CO (l/min) MR Pre minutes of NS p 0.05 p 0.05 p 0.05 NS NS Comparison of hemodynamics and mitral regurgitation before and during ischemia. Statistical significance was evaluated using Student s t-test. Cardiac output (CO), central venous pressure (CVP), Heart rate (HR), left ventricular end diastolic pressure (LVEDP), mean arterial blood pressure (MAP), mitral regurgitation grade (MR), diastolic pulmonary artery pressure (PAD). AHCWR was calculated every 5 ms throughout the cardiac cycle before and 30 min after the on set of ischemia. To compare changes in AHCWR the time points of end diastole (ED), end isovolemic contraction (EIVC), end systole (ES), and end isovolemic relaxation (EIVR) were determined as described previously [2]. Every data set was then normalized in time by means of linear interpolation such that heart rate equaled 120 bpm, isovolemic contraction (from ED to EIVC) 75 ms, systole (ED to ES) 295 ms, isovolemic relaxation (ES to EIVR) 50 ms, and diastole (ES to ED) 205 ms. Mean AHCWR (average of all 6 animals) was then calculated and plotted versus (normalized) time throughout the cardiac cycle before and during ischemia. Thus, the cardiac cycle was divided into four segments: isovolemic contraction, ejection, isovolemic relaxation, and diastolic filling. AHCWR values at base line and during ischemia were compared using one-way analysis of variance (ANOVA) for all four segments of the cardiac cycle. Within the segments of the cardiac cycle found to be significantly different by ANOVA each time point was compared using paired Student s t test. 30 min of ischemia AHCWR did not significantly change from its end diastolic value during the entire cardiac cycle ranging from 11.0% 1.8% 12.2% 0.4% (NS). Figure 4 illustrates the degree of annular flattening associated with AIMR for a representative sheep at ES. In Table 2 annular height, intercommissural width, and AHCWR are presented at ED, EIVC, mid systole (MS), and EIVC. ischemia the intercommissural distance showed an increasing trend that did not reach statistical significance (p ). Annular height was found to decrease significantly at both EIVC and MS after ischemia. Results All animals had competent mitral valves before ischemia. After 30 min of ischemia 2 3 MR developed in all animals. Table 1 summarizes the hemodynamics before and after 30 min of ischemia. Figure 3 demonstrates the dynamic nature of the normal mitral annular shape in sheep during the cardiac cycle and how it is affected by ischemia of the posterior wall which causes AIMR. Before ischemia AHCWR increased rapidly during isovolemic contraction from 11.6% 1.1% 13.9% 1.6% (p 0.001) at EIVC. ejection AHCWR remained constant ranging from 14.1% 1.5% 14.9% 1.3% (NS). In diastole AHCWR decreased from 14.1% 0.7% 11.6% 1.1% (p 0.007) ischemia AHCWR was found to be significantly different (p 0.05) during isovolemic contraction, ejection, and isovolemic relaxation, but not during diastolic filling (Fig 3). Whereas ischemia did not affect AHCWR at ED (11.6% 2.8%) the increase in AHCWR during isovolemic contraction was eliminated. In fact after Fig 3. Changes in annular height to commissural width ratio (AHCWR) throughout the cardiac cycle before (black circles) and during (white circles) ischemia causing acute ischemic mitral regurgitation (AIMR). Each graph represents an average of the same six animals. Differences in AHCWR were compared for all four portions of the cardiac cycle individually using one-way analysis of variance (ANOVA). ischemia AHCWR was found to be significantly different (p 0.05) during isovolemic contraction, ejection, and isovolemic relaxation, but not during diastolic filling. Within the portions of the cardiac cycle found to be significantly different by ANOVA each time point was compared using paired Student s t test. Each significantly different time point (p 0.05) is marked with a white or black dot. Note that the normal systolic accentuation of the saddle shape (increasing AHCWR) is eliminated during ischemia and AIMR. (EIVC end isovolemic contraction; EIVR end isovolemic relaxation; and ES end systole. End diastole is defined as time 0.)

4 Ann Thorac Surg GORMAN ET AL 2004;77:544 8 ISCHEMIC CHANGES IN 3D ANNULAR SHAPE 547 Fig 4. End-systolic images of a representative ovine mitral annulus before and after a posterior ischemia causing acute ischemic mitral regurgitation (AIMR). Images were constructed as in Fig 2. Note the relative flattening of the posterior annulus near the posterior commissure (darker shading) during ischemia. (AC anterior commissure; Ao aortic; PC posterior commissure; P1 anterior portion of mitral annulus; P2 mid portion of mitral annulus; P3 posterior portion of mitral annulus.) Comment Levine first described the distinctive nonplanar shape of the human mitral annulus in 1989 [1] and we showed that the shape is preserved in several mammalian species [5]. Using a finite element mathematical modeling technique our group demonstrated that the annular saddle shape likely contributes to leaflet stress reduction during systole by increasing leaflet curvature. To what extent this annular shape and its putative effect on leaflet stress reduction contribute to valve competence requires further elucidation. The current study provides further evidence that the annular saddle shape is physiologically important. ejection, the period of greatest stress, the annular saddle shape is accentuated. After acute ischemia that results in immediate 2 3 mitral regurgitation the systolic increase in AHCWR is lost. A recent report studying the change in shape of the ovine mitral annulus in a model of chronic ischemic mitral regurgitation (CIMR) demonstrated that the annulus was flattened in the chronic disease process as well [4]. A human 3D echocardiographic study comparing the mitral annular shape in nine normal subjects and 8 patients suffering different degrees of CIMR corroborated the findings of this animal study [13]. The authors of the study found that CIMR was associated with both and an increase in intercommissural distance and decrease in annular height. They did not report the AH- CWR parameter we have used but the curves presented in Fig 5 were calculated from their data. Notice the similarity between this clinical data and the data produced by the current experimental study. In normal humans AHCWR increases in early systole and remains relatively constant during ejection. As in the sheep model this increase is severely blunted in patients with CIMR. Considering that the mechanisms of chronic [4, 14] and acute ischemic MR [15 17] are quite different with regard to annular-papillary muscle dynamics, it is interesting that the saddle shape of the annulus is effected similarly in both diseases. Whereas the aforementioned clinical and experimental studies do not definitively prove that annular saddle shape contributes to valve competence they do demonstrate an association. This association between mitral incompetence and decreased AHCWR brings into question the use of flat rings in mitral valve repair and suggests that preservation or restoration of the normal saddle shape may have a salutary effect. Over the past thirty years Carpentier has pioneered and standardized surgical techniques that allow reliable repair of valves with all types of leaflet, chordal, and annular deformities [6]. The widespread use of these techniques has produced good results at centers all over the world [18 23]. Although the durability of these repairs has been acceptable it is becoming apparent that there is a significant long-term failure rate especially in those patients with ischemic pathology. Gillinov and colleagues from the Cleveland Clinic reported a 5-year reoperative rate of 9% (average follow-up was 5 years) for repairs addressing MR of ischemic etiology. This is a conservative estimate because death was a strongly competing endpoint and recurrent MR not undergoing reop- Table 2. Annular Height, Intercommissural Width and AHCWR Time Point Intercommissural Width (mm) Annular Height (mm) AHCWR (%) End Diastole End Isovolemic Contraction a a Mid Systole a a End Systole b a p 0.05, b 0.06; when compared with baseline values using Student s t-test.

5 548 GORMAN ET AL Ann Thorac Surg ISCHEMIC CHANGES IN 3D ANNULAR SHAPE 2004;77:544 8 Fig 5. Average changes in mitral annular height to commissural width ratio (AHCWR) throughout the cardiac cycle in nine humans with normally functioning valves (black circles) and 8 patients with chronic ischemic mitral regurgitation (MR) of varying degrees (white circles). This figure was constructed from data in reference 13. The AHCWR values were not presented in the paper but annular height and intercommissural distance were reported. All cardiac cycles are normalized to ten systolic and ten diastolic time points. Note the systolic accentuation of the saddle shape in humans similar to that seen in sheep. In addition chronic ischemic MR in humans seems to be associated with loss of the systolic increase in AHCWR as in our acute ischemic MR model. eration was not reported [24]. Would saddle shape annuloplasty improve these results? This study represents a first step in establishing the relationship between annular shape and valve function. Further work is necessary to understand the impact annular shape has on leaflet shape, stress distribution, and postinfarction left ventricular remodeling. A better understanding of these relationships may allow a more rational design of annuloplasty rings to improve the results of mitral valve repair. This work was supported by HL63954 and HL71137 from the National Heart Lung Blood Institute, National Institutes of Health, Bethesda, MD, and grants from the Mary L. Smith Charitable Trust and W. W. Smith Charitable Trust, both of Newtown Square, PA. References 1. Levine RA, Handschumacher MD, Sanfilippo AJ, et al. Three-dimensional echocardiographic reconstruction of the mitral valve, with implications for the diagnosis of mitral valve prolapse. Circulation 1989;80: Gorman JH III, Gupta KB, Streicher JT, et al. Dynamic three-dimensional imaging of the mitral valve and left ventricle by rapid sonomicrometry array localization. J Thorac Cardiovasc Surg 1996;112: Glasson JR, Komeda M, Daughters GT, et al. Most ovine mitral annular three-dimensional size reduction occurs before ventricular systole and is abolished with ventricular pacing. Circulation 1997;96(Suppl. II): Tibayan FA, Rodriguez F, Langer F, et al. Annular geometric remodeling in chronic ischemic mitral regurgitation. Ann Thorac Surg 2003;76: Salgo IS, Gorman JH III, Gorman RC, Jackson BM, Bowen FW, Plappert T, St John Sutton MG, Edmunds LH Jr. Effect of annular shape on leaflet curvature in reducing mitral leaflet stress. Circulation 2002;106: Carpentier A. Cardiac valve surgery the French correction. J Thorac Cardiovasc Surg 1983;86: Carpentier A. Reconstructive valvuloplasty: A new technique of mitral valvuloplasty. Presse Medicale 1969;77: Llaneras MR, Nance ML, Streicher JT, et al. Pathogenesis of ischemic mitral insufficiency. J Thorac Cardiovasc Surg 1993; 105: Llaneras MR, Nance ML, Streicher JT, et al. Large animal model of ischemic mitral regurgitation. Ann Thorac Surg 1994;57: Gorman RC, McCaughan JS, Ratcliffe MB, et al. Pathogenesis of acute ischemic mitral regurgitation in three dimensions. J Thorac Cardiovasc Surg 1995;109: Kelley ST, Malekan R, Jackson BM, et al. Restraining infarct expansion preserves left ventricular geometry and function after acute anteroapical infarction. Circulation 1999;99: Miyatake K, Izumi S, Okamoto M, et al. Semiquantitative grading of severity of mitral regurgitation by real-time two-dimensional Doppler flow imaging technique. J Am Coll Cardiol 1986;7: Kaplan SR, Bashein G, Sheehan FH, et al. Threedimensional echocardiographic assessment of annular shape changes in the normal, and regurgitant mitral valve. Am Heart J 2000;139: Gorman JH III, Gorman RC, Jackson BM, et al. Annuloplasty Ring Selection for Chronic Ischemic Mitral Regurgitation. Lessons From the Ovine Model. Ann Thorac Surg 2003;76: Gorman JH III, Gorman RC, Jackson BM, et al. Distortions of the mitral valve in acute ischemic mitral regurgitation. Ann Thorac Surg 1997;64: Gorman JH III, Jackson BM, Gorman RC, et al. Papillary muscle discoordination rather than increased annular area facilitates mitral regurgitation after acute posterior myocardial infarction. Circulation 1997;96(Suppl.):II Glasson JR, Komeda M, Daughters GT, Bolger AF, Karlsson MO, Foppiano LE, Hayase M, Oesterle SN, Ingels NB Jr, Miller DC. Early systolic mitral leaflet loitering during acute ischemic mitral regurgitation. J Thorac Cardiovasc Surg 1998;116: Braunberger E, Deloche A, Berrebi A, et al. Very long-term results (more than 20 years) of valve repair with Carpentier s techniques in nonrheumatic mitral valve insufficiency. Circulation 2001;104(Suppl I):I Chauvaud S, Fuzellier JF, Berrebi A, et al. Long-term (29 years) results of reconstructive surgery in rheumatic mitral valve insufficiency. Circulation 2001;104(Suppl I): Gillinov AM, Cosgrove DM, Blackstone EH, et al. Durability of mitral valve repair for degenerative disease. J Thorac Cardiovasc Surg 1998;116: Cohn LH, Couper GS, Aranki SF, Rizzo RJ, Kinchla NM, Collins JJ Jr. Long-term results of mitral valve reconstruction for regurgitation of the myxomatous mitral valve. J Thorac Cardiovasc Surg 1994;107: David TE, Armstrong S, Sun Z, Daniel L. Late results of mitral valve repair for mitral regurgitation due to degenerative disease. Ann Thorac Surg 1993;56: Gillinov AM, Blackstone EH, White J, et al. Durability of combined aortic and mitral valve repair. Ann Thorac Surg 2001;72: Gillinov AM, Wierup PN, Blackstone EH, et al. Is repair preferable to replacement for ischemic mitral regurgitation? J Thorac Cardiovasc Surg 2001;122: