Mitral Valve Abnormalities in Hypertrophic Cardiomyopathy: Echocardiographic Features and Surgical Outcomes

Transcription

1 Mitral Valve Abnormalities in Hypertrophic Cardiomyopathy: Echocardiographic Features and Surgical Outcomes Ryan K. Kaple, BS, Ross T. Murphy, MD, Linda M. DiPaola, BA, Penny L. Houghtaling, MS, Harry M. Lever, MD, Bruce W. Lytle, MD, Eugene H. Blackstone, MD, and Nicholas G. Smedira, MD Departments of Thoracic and Cardiovascular Surgery, Cardiovascular Medicine, and Quantitative Health Sciences, The Cleveland Clinic, Cleveland, Ohio ADULT CARDIAC Background. Functional and intrinsic mitral valve (MV) abnormalities are common in hypertrophic cardiomyopathy (HCM); however, morphologic characteristics constituting indications for surgical intervention are incompletely defined. This study was conducted to define the echocardiographic features of MV pathology in patients with HCM and relate these to repairability of the MV, MV procedures performed, durability of repair, and survival. Methods. From 1986 to 2003, 851 patients with HCM underwent operation, and 115 had a concomitant MV procedure. Detailed analysis of their 784 transthoracic and transesophageal echocardiograms, performed intraoperatively and postoperatively, was conducted. Outcomes were assessed by cross-sectional follow-up. Results. Sixty-seven patients (58%) underwent MV repair, and 48 (42%) had MV replacement. The mean left ventricular outflow tract peak gradient was mm Hg. Systolic anterior motion was present in 95%. Valve abnormalities were degenerative in 36 (31%), myxomatous in 23 (20%), papillary muscle in 23 (20%), restrictive chordal in 22 (19%), restrictive leaflet in 80 (70%), and long leaflet in 64 (56%). Patients undergoing MV repair had higher prevalence of long leaflets and degenerative MV pathology. The anterior mitral leaflet was cm in the repair group vs cm in the replacement group (p ). MV replacement patients were older, more symptomatic, and had more renal dysfunction and lower hematocrits. By 3 years, 91% of patients with a repair were free of reoperation. Conclusions. Intrinsic MV pathology is frequently observed in HCM patients with symptomatic obstruction who undergo myectomy. Echocardiography can identify MV features predictive of successful valve repair. Repair, although durable, is feasible in only about half of patients. (Ann Thorac Surg 2008;85: ) 2008 by The Society of Thoracic Surgeons Hypertrophic cardiomyopathy (HCM) is a genetic disorder resulting in unexplained hypertrophy of the left ventricle. In up to 60% of patients, altered geometry of the left ventricular outflow tract (LVOT), septal hypertrophy, and systolic anterior motion (SAM) of the mitral valve (MV) lead to an outflow tract gradient and symptomatic hypertrophic obstructive cardiomyopathy (HOCM) [1]. Hypertrophic obstructive cardiomyopathy is also associated with a variety of intrinsic abnormalities of the MV [2], including anomalous mitral papillary muscles or chordae [3], direct papillary insertion into the mitral leaflet [4], chordal rupture [5], and MV prolapse. One autopsy study suggested that HCM was associated with increased mitral leaflet area in 70% of cases [6]. Several surgical series suggest that concomitant MV surgery is required in 11% to 20% of patients [3, 7]; however, indications for this are incompletely defined, as Accepted for publication Jan 15, Address correspondence to Dr Smedira, Kaufman Center for Heart Failure, Department of Thoracic and Cardiovascular Surgery, The Cleveland Clinic, 9500 Euclid Ave/F24, Cleveland, OH 44195; smedirn@ccf.org. are echocardiographic predictors of the type of procedure required. This study aimed to define echocardiographic characteristics of intrinsic MV abnormalities leading to surgical intervention in patients with HCM and relate these to valve repairability, durability of repair, and survival. Patients and Methods Patients From January 1986 to January 2003, 851 symptomatic patients underwent operation for HCM at The Cleveland Clinic. These 851 were established after personal review (B.W.L., N.G.S.) of all patients undergoing operation for HCM for any reason during this period, using consensus criteria [8]. Of these, 115 (14%) underwent a concomitant MV procedure, with 102 undergoing a concomitant septal myectomy, leaving 13 patients with HOCM who underwent MV replacement alone. Valve repairs were done in 67 patients (58%), and 48 (42%) had valve replacement with a mechanical valve (n 39) or bioprosthesis (n 9). The mean age was years, and 56% were men by The Society of Thoracic Surgeons /08/$34.00 Published by Elsevier Inc doi: /j.athoracsur

2 ADULT CARDIAC 1528 KAPLE ET AL Ann Thorac Surg MITRAL VALVE ABNORMALITIES IN HCM 2008;85: Table 1. Characteristics of Patients with Hypertrophic Cardiomyopathy Undergoing Concomitant Mitral Valve Surgery Characteristic Total Repair Mitral Valve Procedure Replacement p Value Patients, No Demographics Age, mean SD y Male, No. (%) 64 (56) 44 (66) 20 (42) 0.01 Symptoms Canadian angina class, No. (%) (28) 22 (34) 9 (20) 1 44 (40) 15 (23) 29 (63) 2 25 (23) 18 (28) 7 (15) 3 7 (6.4) 6 (9.4) 1 (2.2) 4 3 (2.7) 3 (4.7) 0 (0) LVOT, mean SD mm Hg Gradient Resting peak a Provoked peak b Septal thickness, mean SD cm c Comorbidity Pre-op AF or flutter, No. (%) 5 (4.3) 1 (1.5) 4 (8.3) 0.16 d CrCl, mean SD ml/min e Hematocrit, mean SD % a Data available for 107 patients (62 repairs, 45 replacements). b Data available for 66 patients (37 repairs, 29 replacements). c Data available for 111 patients (66 repairs, 45 replacements). d Fisher exact test. e Calculated. AF atrial fibrillation; CrCL creatinine clearance; LVOT left ventricular outflow tract; SD standard deviation. These and other patient characteristics are presented in Table 1. Data were extracted from the Cardiovascular Information Registry (CVIR), a database maintained concurrently with patient care, and were approved for use in research by the Institutional Review Board (IRB), with patient consent waived. Conduct of Operation Mitral valve morphology, presence of SAM, and associated posteriorly directed mitral regurgitation jet, non SAM-associated mitral regurgitation, and magnitude of LVOT gradient were assessed systematically by intraoperative transesophageal echocardiography (TEE) in all 115 patients. In patients with SAM and severe intrinsic MV pathology and dysfunction, some surgeons went directly to valve replacement without myectomy. One patient who presented early in the experience with severe shock, refractory LVOT obstruction, and a normal MV also underwent isolated MV replacement. After valve replacement, 1 surgeon routinely examined the LVOT obstruction to ensure absence of strut obstruction of the LVOT and complete mobility of mechanical valve leaflets (Table 2). A limited myectomy was often performed. Mitral valve repair was the preferred therapy for intrinsic MV disease. Echocardiographic Analysis All available transthoracic echocardiograms (TTEs) and TEEs for each patient were obtained and analyzed. Digital software was used to perform all echocardiograms in accordance with recommendations of the American Society of Echocardiography [9, 10]. Systolic anterior motion of the MV at rest or on provocation was noted. Severity of mitral regurgitation was characterized by semi-quantitative assessment (grades 0 to 4), and jet number and direction were noted. Mitral regurgitation was characterized as SAM-associated or not SAMassociated. Maximum peak LVOT pressure gradient at rest and on provocation by amyl nitrate or the Valsalva maneuver was determined using continuous-wave Doppler echocardiography and the modified Bernoulli equation [11, 12]. The thickness of the interventricular septum was noted during diastole at the point of leaflet septal contact during systole. A segmental description of MV morphology and pathology, including anulus, leaflets, and subvalvar apparatus, was made from the echocardiographic review, and the surgeon s gross inspection was recorded in the operative note. Each valve was coded by 2 independent, experienced observers and then organized into 6 nonmutually exclusive groups (Fig 1; Table 2). The length of the anterior and posterior MV leaflets was measured on intraoperative TEE when available. Follow-Up Clinical follow-up was obtained by chart review and by cross-sectional follow-up using an IRB-approved questionnaire or telephone script with patient consent and

3 Ann Thorac Surg KAPLE ET AL 2008;85: MITRAL VALVE ABNORMALITIES IN HCM Table 2. Mitral Valve Morphology in Patients with Hypertrophic Cardiomyopathy Undergoing Mitral Valve Surgery Morphologic Categories Criteria Total, a No. (%) b Mitral Valve Procedure Repair, No. Replacement, (%) c No. (%) c 1529 p Value ADULT CARDIAC Degenerative Leaflet prolapse Chordae ruptured 36 (31) 24 (67) 12 (33) 0.04 Chordae elongated Myxomatous Leaflet myxomatous 23 (20) 10 (43) 13 (57) 0.5 Papillary muscle abnormalities Attached to lateral leaflet Attached to ventricular wall 23 (20) 10 (43) 13 (57) 0.5 Hypertrophied papillary muscle Restrictive chordal abnormalities Chordae calcified Chordae shortened Chordae abnormalities of anterior leaflet 22 (19) 14 (64) 8 (36) 0.2 Chordae abnormalities of posterior leaflet Restrictive leaflet abnormalities Leaflet calcified Leaflet thickened 80 (70) 40 (50) 40 (50) 1.0 Leaflet tethered/restricted Long leaflet AML 2.5 cm or PML 2.0 cm Surgeon s observation 64 (56) 47 (73) 17 (27) a Morphologic categories not mutually exclusive. b Percentage of 115. c Percentage of morphologic category. AML anterior mitral leaflet; PML posterior mitral leaflet. was supplemented for vital status by the Social Security Death Index [13, 14]. Mean follow-up among survivors was years (median, 2.5 years; range, 1 day to 14 years); 10% were followed up more than 8 years. Follow-up was 93% complete, with 417 patient-years of data available for analysis. Postoperative echocardiograms or echocardiographic reports from The Cleveland Clinic and outside institutions were reviewed. Among 67 patients undergoing repair, 140 echocardiograms were available for review in 63 patients. Fifty percent of echocardiograms were obtained within 2 weeks of operation, and 25% more than a year later (Appendix Fig 1*). Data Analysis Analyses were directed to the associations of echocardiographic and observed MV morphology with valve repairability, durability of repair, and survival. REPAIRABILITY. Multivariable logistic regression analysis was used to identify factors distinguishing patients undergoing MV replacement rather than repair. Bootstrap bagging was used for variable selection [15, 16], with automated analysis of 1000 resampled data sets. Factors occurring in 50% or more of these analyses were considered reliably identified at p DURABILITY OF MITRAL REPAIR. Among patients undergoing MV repair, longitudinal ordinal logistic regression for repeated measurements (SAS PROC GENMOD, SAS Inc, Cary, NC) was used to investigate the temporal pattern of the regurgitation grade at follow-up TTE. This method accounted for multiple records per patient. Only 10 electrocardiograms revealed 3 or 4 mitral regurgitation and so were collapsed into a single category for analyses. Because of the limited ability of PROC GEN- MOD to explore multivariable relations, ordinary multivariable logistic regression was initially used to screen variables under the assumption of independence. These variables were then investigated by manual backward elimination, and those with p 0.05 retained. SURVIVAL. Overall and stratified nonparametric survival estimates were obtained by the Kaplan-Meier method. A parametric method was used to resolve the number of phases of instantaneous risk of death (hazard function) and to estimate shaping parameters [17]. Continuous data are summarized as means standard deviations, or as 15th, 50th (median), and 85th percentiles when distributions were skewed. Wilcoxon rank sum tests and t tests were used to investigate group differences. Categoric data are summarized as frequencies and percentages, and group comparisons were made using 2 tests (Fisher exact test where appropriate). Proportions are accompanied by asymmetric 68% confidence limits (CL) equivalent to 1 standard error. Results Morphologic Features Echocardiography documented restrictive leaflet abnormalities in 80 patients and elongated leaflets in 64 (Table 2). Degenerative and myxomatous pathologies were present in 59 (51%). Subvalvar chordal and papillary muscle abnormalities were found in 22 and 23 patients, * See note at end of article.

4 ADULT CARDIAC 1530 KAPLE ET AL Ann Thorac Surg MITRAL VALVE ABNORMALITIES IN HCM 2008;85: Fig 1. Mitral valve morphology. (A) Degenerative: thin leaflets with excessive mobility. This is a classic example of torn chordae to the P 2 segment of the posterior leaflet with resultant anteriorly directed jet and systolic anterior motion of anterior leaflet with a posteriorly directed jet. (B) Myxomatous: similar to degenerative (A), with thickened redundant leaflets. (C) Papillary muscle: most common variant, with large anteromedial muscle inserting directly into A 1 segment of mitral leaflet. (D) Restrictive chordae: single or multiple (shown) secondary chordae restricting anterior leaflet mobility and tenting leaflet into left ventricular outflow tract. (E) Restrictive leaflet: thickened leaflet often associated with anular calcification. (F) Long leaflet: defined by anterior leaflet length 2.5 cm and posterior leaflet length 2.0 cm; one or both can be elongated. respectively. Nearly all of the myxomatous valves had severe MR but the lowest LVOT gradients (Table 3); conversely, restricted chordal abnormalities were associated with the least prevalence of severe MR and the highest LVOT gradients. Septal thickening was similar for all morphologic categories. Systolic anterior motion was present at rest in 107 of the 113 patients (95%) in whom it could be accurately assessed (Appendix Table 1*). Of 6 patients who did not have SAM at rest, it developed in 1 patient in provocative testing, did not develop in 3, and was not noted in 2. All patients had mitral regurgitation at rest, with most in grade 3 (30%) or 4 (47%). Myxomatous valves had the most variable regurgitant jet directions; jet direction was predominately posterior and central in all other morphologic categories, including those with long leaflet morphology (Appendix Table 2*). Surgical Features In 35% of patients, direct inspection, testing of the valve, or intraoperative TEE demonstrated features of MV pathology typical of degenerative and myxomatous disease and, to a lesser extent, rheumatic disease (Table 4). A severe intrinsic disease process, most not easily characterized as degenerative, was present in 28%, accompanied by SAM; all cases were treated by valve replace- Table 3. Mitral Regurgitation, Septal Thickness, and Left Ventricular Outflow Tract Gradient According to Mitral Valve Morphology Morphology a 3 or 4 MR, No. (%) b Septal Thickness, Mean SD (Median) LVOT Gradient, Mean SD (Median) Degenerative 29 (80) (2.3) (79) Myxomatous 22 (96) (2.1) (64) Papillary muscle abnormalities 15 (65) (2.2) (68) Restrictive chordal abnormalities c 14 (64) (2.5) (100) Restrictive leaflet abnormalities 64 (81) (2.2) (77) Long leaflet 48 (75) (2.3) (64) a Categories not mutually exclusive. b Percentage of morphologic category. c This group was associated with greater septal thickness (p 0.02) and larger LVOT gradient (p 0.02) in multivariable regression analysis. LVOT left ventricular outflow tract; MR mitral regurgitation; SD standard deviation. * See note at end of article.

5 Ann Thorac Surg KAPLE ET AL ;85: MITRAL VALVE ABNORMALITIES IN HCM Table 4. Indications for Mitral Valve Surgical Intervention Intervention Indication Repair, No. (%) a Replace, No. (%) a ADULT CARDIAC Patients 67 (100) 48 (100) Degenerative, myxomatous, rheumatic disease 27 (70) 12 (30) Severe intrinsic disease, not characterized as above, with SAM 0 (0) 27 (100) Myectomy 0 (0) 17 (100) No myectomy 0 (0) 10 (100) Endocarditis 3 (60) 2 (40) Postmyectomy 6 (46) 7 (54) Residual SAM 5 (45) 6 (55) Residual MR with SAM 1 (50) 1 (50) Iatrogenic MV injury 1 (100) 0 (0) Indeterminate 1 (100) 0 (0) MV intervention with uncertain morphologic indication 29 (100) 0 (0) Leaflet height reduction with myectomy 11 (100) 0 (0) Interventricular septum 17 mm 6 (100) 0 (0) Interventricular septum 17 mm 5 (100) 0 (0) Anterior leaflet, chordal, or papillary muscle resection 9 (100) 0 (0) Edge-to-edge (Alfieri) repair 9 (100) 0 (0) a Percentage of indication. MR mitral regurgitation; MV mitral valve; SAM systolic anterior motion. ment. Mitral valve intervention in 25% was performed for a variety of reasons with uncertain morphologic indication; the SAM and LVOT gradient may have resolved with myectomy alone. Repairability Of the 115 patients undergoing a concomitant MV procedure, 67 (58%) underwent repair and 48 (42%) had replacement. Patients undergoing replacement were somewhat older, more symptomatic, and more likely to have comorbid conditions and valve calcification (Appendix Table 1*). All patients undergoing MV repair also had a septal myectomy. A number of repair techniques were used (Table 5). Only 11 repairs (16%) included an anuloplasty ring. Nine of the 48 valve replacements followed attempted repair. All 4 patients with abnormal papillary muscle insertion into the lateral ventricular wall underwent valve replacement. Durability of Mitral Valve Repair POSTOPERATIVE MITRAL REGURGITATION. Early after operation, the proportion of patients with perfect mitral competence (grade 0 regurgitation) declined to about 35% at 1 year, and those with severe mitral regurgitation (grade 3 or 4 ) rose to about 10% (Fig 2). The only risk factors for higher grade of mitral regurgitation, in addition to the temporal trend depicted, was older age at repair (p 0.04). VALVE REPAIR FAILURES. All patients with failed repairs underwent valve replacement. The hazard model for MV replacement after repair or initial repair attempt resolved to two phases: a steep early hazard phase resulting from intraoperative failure of 13 repairs, and a late phase of six reoperations resulting in replacement. The four early reoperations (within 30 days) were for persistent mitral regurgitation (two after repair and two after Alfieri stitch); the two late reoperations were for mitral regurgitation (3 years after Alfieri stitch) and mitral stenosis (14 months after Alfieri stitch). Freedom from MV replacement at 30 days, 1, 2, 3, and 5 years was 85%, 85%, 84%, 84%, and 81%, respectively (Fig 3). For patients thought to have had a successful repair, 91% were free of reoperation at 3 years. Table 5. Surgical Techniques Procedure Mitral Valve Procedure Repair, No. (%) Replacement, No. (%) Patients 67 (100) 48 (100) Myectomy 67 (100) 35 (73) Mitral valve procedure Replacement Mechanical prosthesis 39 (81) Bioprosthesis 9 (19) Repair Complete anuloplasty with ring 11 (16) 1 (2.1) Leaflet procedure 50 (75) 11 (23) Chordal procedure 14 (21) 3 (6.2) Papillary muscle procedure 3 (4.5) 2 (4.2) Concomitant procedure CABG 17 (25) 17 (35) Other 9 (13) 3 (6.2) CABG coronary artery bypass grafting. * See note at end of article.

6 ADULT CARDIAC 1532 KAPLE ET AL Ann Thorac Surg MITRAL VALVE ABNORMALITIES IN HCM 2008;85: Fig 2. Risk-unadjusted estimate by echocardiography of postoperative mitral regurgitation (MR) after MV repair. The lines depict results of an ordinal longitudinal analysis that accounted for repeated echocardiograms of the same patient. Symbols represent grouped data that do not account for repeated measures, graphed to lend credibility to the solid lines. Squares, no MR; open circles, 1 MR; filled circles, 2 MR; triangles, 3 or 4 MR. Based on patient age of 50 years. Survival Twenty patients died, 5 within the first 30 days of operation (4.6%; 68% CL, 3.1%, 7.0%), 3 of which were inhospital (2.6%; 68% CL, 1.2%, 5.0%). The hazard model resolved to two phases: a steep early hazard phase extending to 3 months (accounting for 8 deaths) and a late hazard phase thereafter (Appendix Fig 2*). Overall survival was 95% at 30 days, 92% at 6 months, 91% at 1 year, 81% at 5 years, and 66% at 10 years, respectively (Fig 4). Survival was similar in the MV repair and replacement groups (p 0.5 early, p 0.7 late; Fig 5), and no patient who required an early reoperation died within 30 days (earliest death was 1.2 years). Comment Mitral valve abnormalities are common in patients with HCM, but their frequency and impact on surgical therapies is poorly defined. Limitations of previously published studies include small cohort size, focus on individual abnormalities, and focus on autopsy series. Our goal was to use echocardiographic analysis and our large experience with surgical therapy for HCM to define the spectrum of MV abnormalities, evaluate how they affect surgical therapies, and investigate durability of modern repair techniques that are supplanting valve replacement as the surgical procedure of choice for intrinsic MV disease. Fig 3. Freedom from mitral valve replacement (MVR) after repair or attempted repair. Each circle represents a replacement, the vertical bars are 68% confidence limits (CL) equivalent to 1 standard error, and the numbers in parentheses represent patients at risk. The solid line is the parametric estimate enclosed within dashed 68% CLs. * See note at end of article.

7 Ann Thorac Surg KAPLE ET AL 2008;85: MITRAL VALVE ABNORMALITIES IN HCM 1533 Fig 4. Survival after mitral valve repair or replacement in patients with hypertrophic cardiomyopathy. Each circle represents a replacement, vertical bars are 68% confidence limits (CL) equivalent to 1 standard error, and numbers in parentheses represent patients at risk. The solid line is the parametric estimate enclosed within dashed 68% CLs. ADULT CARDIAC Principal Findings A wide range of intrinsic MV abnormalities frequently coexist in patients with septal hypertrophy and dynamic LVOT obstruction. Although MV morphology appeared to affect the severity of mitral regurgitation and magnitude of LVOT gradient, it was independent of the degree of septal hypertrophy. Patient and valve characteristics influenced the likelihood of valve replacement, and with the exception of the Alfieri stitch, valve repairs were reproducible and durable in the intermediate term. Echocardiography and Valve Morphology Almost all patients in this series had resting SAM with severe mitral regurgitation. The presence of severe regurgitation without SAM suggests intrinsic MV disease, and detailed evaluation of regurgitant jet direction and MV morphology may define the valve pathology. Anterior regurgitant jet direction suggests underlying degenerative or myxomatous posterior leaflet pathology. Anomalous papillary muscle insertions, tethering chordae, and long leaflets can be readily determined by echocardiography. Differentiating intrinsic MV regurgitation from dynamic obstruction can be particularly difficult in patients with restricted, calcified, and thickening leaflets with resting SAM. Intrinsic MV disease is often obscured by the uniform presence of dynamic mitral regurgitation, but presence of mitral regurgitation without SAM or presence of an anteriorly directed jet is strong evidence for intrinsic MV disease. Many small case series describe specific MV abnormalities in patients with HCM, but only one sizeable report has been published that examined the pathology of 94 excised MVs [6, 18, 19]. These valves tended to have a larger area, were longer, and frequently had direct insertion of the papillary muscle onto the anterior leaflet. Fig 5. Survival in patients undergoing operation for hypertrophic cardiomyopathy with accompanying valve repair (filled circles) or replacement (open circles) for mitral valve abnormalities. Vertical bars are 68% confidence limits (CL) equivalent to 1 standard error, and numbers in parentheses represent patients at risk. The solid line is the parametric estimate enclosed within dashed 68% CLs.

8 ADULT CARDIAC 1534 KAPLE ET AL Ann Thorac Surg MITRAL VALVE ABNORMALITIES IN HCM 2008;85: Echocardiography provides substantially more detailed assessment of in vivo MV function and morphology, and our challenge was to synthesize this information for clinical decision making. Leaflet Morphology and Surgical Results Restrictive leaflets with valve calcification and thickening were the most common form of intrinsic MV disease identified, and calcification was a risk factor for valve replacement. The mechanism of valve thickening and calcification is unknown, but the trauma of repetitive septal contact could result in fibrosis, retraction, and regurgitation. Both biologic and mechanical prostheses were used to treat restrictive leaflet pathology. Fear of posterior or lateral ventricular wall perforation from a tall valve strut protruding into a small ventricular cavity has been reduced by development of biologic MVs with substantially reduced strut height. These new valves offer older patients an alternative to mechanical valve replacement but do not eliminate need for careful orientation of valve struts to avoid mechanical LVOT obstruction. A concomitant myectomy will still be necessary in most cases [7]. We have no experience treating restricted leaflets with leaflet area extension using glutaraldehydetreated autologous pericardium[20]. A wealth of experience has been accumulated assessing and treating patients with prolapse or flail leaflets from degenerative and myxomatous valve disease [21, 22]. Time-tested surgical techniques of leaflet segment resection and leaflet height reduction with sliding valvuloplasties were used in this series. Annuloplasties were infrequently used to avoid inducing SAM of the anterior leaflet, and this approach did not affect repair durability during the relatively short follow-up period. If an annuloplasty were used, an oversized incomplete band would appear to be safest. Leaflet elongation was common and could involve one or both leaflets. Whether leaflet elongation is an early form or variant manifestation of myxomatous MV disease, where excessively tall anterior or posterior leaflets have been long described, is unknown. Many elongated leaflets were thickened, as seen in myxomatous disease and as reported in the pathology series. Our impression was that many patients with elongated anterior leaflets had less septal hypertrophy; however, measured leaflet length correlated with degree of septal hypertrophy, refuting this hypothesis. Multiple repair techniques to reduce leaflet height have been described. Plicating the anterior leaflet along its long axis (tip of leaflet to base), whether in its central portion or leaflet edge, makes the most sense, is easily accomplished through the aortic valve, and is our preferred method of repair. Plicating perpendicular to the long axis has been used successfully [23]. Reports of subvalvar abnormalities of the MV have focused on insertion of the papillary muscle, most commonly the anterolateral, which Kanani and Anderson [24] note is more correctly described as superoposterior, directly onto the mitral leaflet [18]. In addition to papillary muscle anomalies, we also found chordal abnormalities resulting in tethering of the anterior mitral leaflet toward the LVOT. Wide mid-to-apical resection of the septum, papillary muscle mobilization off the lateral wall, and excision of a portion of the excessively hypertrophied papillary muscle have been described to manage this anomaly [25]. Tethering secondary chordae have been excised, and a very anteriorly displaced anterior papillary muscle chordae complex has been excised and function established by attaching polytetrafluoroethylene neochordae to the most posterior papillary muscle head of the anterolateral complex. This variability of subvalvar anatomy and papillary muscle alignment is most likely related to abnormal valve morphogenesis [26]. After 5 to 6 weeks of gestation, the LV septum undergoes involution, with both chordae and papillary muscles originating from septal muscle tissue. Failure of chordae to develop would explain the direct insertion of papillary muscle onto the leaflet. Whether abnormal blood flow, sarcomeric protein gene mutations found in HCM, or both, result in a malpositioned anterior papillary muscle and subvalvar abnormalities is unknown. Septal hypertrophy and abnormal MV morphogenesis may be conceptually similar to abnormal bicuspid valvulogenesis and its associated deficiencies in fibrillin, endothelium-derived nitric oxide synthase, and the ubiquitin fusion degradation 1 gene [27]. The repair techniques used in this series are evolving. We thought the simplicity of the Alfieri stitch and that it could be placed through the aortic valve at myectomy would render it useful for treating patients with hypertrophy and intrinsic MV disease; however, frequent recurrence of mitral regurgitation has led us to abandon this technique. If standard valve repair techniques are insufficient, then replacement is preferred. A competent repair is durable, with excellent short-term freedom from recurrent mitral regurgitation and SAM. Strengths and Limitations This is a single institution s consecutive experience with surgical management of MV pathology accompanying HCM. However, HCM and its treatment has been a particular focus of our institution for many years, and our experience is extensive. We did not analyze MV morphology in HCM patients in whom no operation for mitral disease was deemed necessary; thus, we do not know characteristics of more minor valve morphology. Echocardiographic follow-up was limited to that available from treating cardiologists rather than from a systematic protocol; thus, it is limited to short-term results. The frequency of MV abnormalities requiring surgical intervention is much higher than has been reported. Careful review of operative reports suggests that this reflects a referral bias rather than overtreatment. Clinical Implications Assessing patients with symptomatic left ventricular hypertrophy requires detailed echocardiographic evaluation. In addition to provocative maneuvers in patients without resting obstruction, a detailed assessment of MV anatomy and function must be completed before embarking on therapies. Many valves can be repaired with the expecta-

9 Ann Thorac Surg KAPLE ET AL 2008;85: MITRAL VALVE ABNORMALITIES IN HCM tion of lasting durability. Valve replacement with a lowprofile biologic prosthesis is now possible when indicated, without concerns for ventricular perforation. Our experience suggests that the clinical and research focus of hypertrophic obstructive cardiomyopathy should expand beyond septal hypertrophy and myocyte abnormalities and begin to examine the complex and little understood relationship of the septum, sarcomere proteins, and morphogenesis of MV leaflets and subvalvar apparatus [9, 10]. This study was supported in part by the Kenneth Gee and Paula Shaw, PhD, Chair in Heart Research, held by Dr Blackstone. We thank Michelle Miluk, RN, BSN, for data adjudication and checking; Diana Dolney, RN, MSN, for data collection and patient follow-up; Angela York for database management; and Jocelyn Piskach, Karen Mrazeck, Wanda Weaver, Patricia White, and Sherry El Sakr for completion of patient follow-up. References Maron MS, Olivotto I, Zenovich AG, et al. Hypertrophic cardiomyopathy is predominantly a disease of left ventricular outflow tract obstruction. Circulation 2006;114: Fix P, Moberg A, Soederberg H, Karnell J. Muscular subvalvular aortic stenosis. Abnormal anterior mitral leaflet possibly the primary factor. Acta Radiol Diagn (Stockh) 1964;2: Minakata K, Dearani JA, Nishimura RA, Maron BJ, Danielson GK. Extended septal myectomy for hypertrophic obstructive cardiomyopathy with anomalous mitral papillary muscles or chordae. J Thorac Cardiovasc Surg 2004;127: Klues HG, Roberts WC, Maron BJ. Anomalous insertion of papillary muscle directly into anterior mitral leaflet in hypertrophic cardiomyopathy. Significance in producing left ventricular outflow obstruction. Circulation 1991;84: Zhu WX, Oh JK, Kopecky SL, Schaff HV, Tajik AJ. Mitral regurgitation due to ruptured chordae tendineae in patients with hypertrophic obstructive cardiomyopathy. J Am Coll Cardiol 1992;20: Klues HG, Maron BJ, Dollar AL, Roberts WC. Diversity of structural mitral valve alterations in hypertrophic cardiomyopathy. Circulation 1992;85: Yu EH, Omran AS, Wigle ED, Williams WG, Siu SC, Rakowski H. Mitral regurgitation in hypertrophic obstructive cardiomyopathy: relationship to obstruction and relief with myectomy. J Am Coll Cardiol 2000;36: Maron BJ, McKenna WJ, Danielson GK, et al. American College of Cardiology/European Society of Cardiology clinical expert consensus document on hypertrophic cardiomyopathy. A report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents and the European Society of Cardiology Committee for Practice Guidelines. J Am Coll Cardiol 2003;42: Henry WL, DeMaria A, Gramiak R, et al. Report of the American Society of Echocardiography Committee on Nomenclature and Standards in Two-dimensional Echocardiography. Circulation 1980;62: Shanewise JS, Cheung AT, Aronson S, et al. ASE/SCA guidelines for performing a comprehensive intraoperative multiplane transesophageal echocardiography examination: recommendations of the American Society of Echocardiography Council for Intraoperative Echocardiography and the Society of Cardiovascular Anesthesiologists Task Force for Certification in Perioperative Transesophageal Echocardiography. J Am Soc Echocardiogr 1999;12: Sasson Z, Yock PG, Hatle LK, Alderman EL, Popp RL. Doppler echocardiographic determination of the pressure gradient in hypertrophic cardiomyopathy. J Am Coll Cardiol 1988;11: Stewart WJ, Schiavone WA, Salcedo EE, Lever HM, Cosgrove DM, Gill CC. Intraoperative Doppler echocardiography in hypertrophic cardiomyopathy: correlations with the obstructive gradient. J Am Coll Cardiol 1987;10: Boyle CA, Decoufle P. National sources of vital status information: extent of coverage and possible selectivity in reporting. Am J Epidemiol 1990;131: Newman TB, Brown AN. Use of commercial record linkage software and vital statistics to identify patient deaths. J Am Med Inform Assoc 1997;4: Breiman L. Bagging predictors. Machine Learning 1996;24: Blackstone EH. Breaking down barriers: helpful breakthrough statistical methods you need to understand better. J Thorac Cardiovasc Surg 2001;122: Blackstone EH, Naftel DC, Turner ME Jr. The decomposition of time-varying hazard into phases, each incorporating a separate stream of concomitant information. J Am Stat Assoc 1986;81: Maron BJ, Nishimura RA, Danielson GK. Pitfalls in clinical recognition and a novel operative approach for hypertrophic cardiomyopathy with severe outflow obstruction due to anomalous papillary muscle. Circulation 1998;98: Ohkado A, Kitamura M, Hachida M, et al. Hypertrophic obstructive cardiomyopathy with abnormalities of the mitral valve complex. J Heart Valve Dis 1997;6: van der Lee C, Kofflard MJ, van Herwerden LA, Vletter WB, ten Cate FJ. Sustained improvement after combined anterior mitral leaflet extension and myectomy in hypertrophic obstructive cardiomyopathy. Circulation 2003;108: Carpentier A. Cardiac valve surgery the French correction. J Thorac Cardiovasc Surg 1983;86: Gillinov AM, Cosgrove DM, Blackstone EH, et al. Durability of mitral valve repair for degenerative disease. J Thorac Cardiovasc Surg 1998;116: McIntosh CL, Maron BJ, Cannon RO 3rd, Klues HG. Initial results of combined anterior mitral leaflet plication and ventricular septal myotomy-myectomy for relief of left ventricular outflow tract obstruction in patients with hypertrophic cardiomyopathy. Circulation 1992;86:II Kanani M, Anderson RH. The anatomy of the mitral valve: a retrospective analysis of yesterday s future. J Heart Valve Dis 2003;12: Schoendube FA, Klues HG, Reith S, Flachskampf FA, Hanrath P, Messmer BJ. Long-term clinical and echocardiographic follow-up after surgical correction of hypertrophic obstructive cardiomyopathy with extended myectomy and reconstruction of the subvalvular mitral apparatus. Circulation 1995;92:II Sadler TW, Langman J. Langman s medical embryology. Philadelphia, PA: Lippincott Williams & Wilkins; 2004: Aboulhosn J, Child JS. Left ventricular outflow obstruction: subaortic stenosis, bicuspid aortic valve, supravalvar aortic stenosis, and coarctation of the aorta. Circulation 2006;114: ADULT CARDIAC *The Appendix Figures 1, 2 and Tables 1, 2 are available only online. To access them, please visit: and search for the article by Kaple, Vol. 85, pages e1 2.

10 Ann Thorac Surg 2008;85:1536.e1 2 KAPLE ET AL MITRAL VALVE ABNORMALITIES IN HCM 1535.e1 Appendix Fig 1. Number of patients (black bars) with echocardiograms available at and beyond various time points and number of echocardiographic measurements (gray bars) available for analysis. Appendix Fig 2. Instantaneous risk of death (hazard function) after mitral valve repair or replacement in patients with hypertrophic cardiomyopathy. Solid line is parametric estimate enclosed within 68% confidence limits.

11 1535.e2 KAPLE ET AL Ann Thorac Surg MITRAL VALVE ABNORMALITIES IN HCM 2008;85:1536.e1 2 Appendix Table 1. Mitral Valve Pathophysiology Mitral Valve Procedure Abnormality Total Repair Replacement p Value Patients, No SAM at rest, No. (%) a 107/113 (95) 64/66 (97) 43/47 (92) 0.2 Mitral regurgitation at rest, No. (%) (10) 9 (14) 2 (4.3) 2 15 (13) 11 (17) 4 (8.5) 3 34 (30) 18 (27) 16 (34) 4 53 (47) 28 (42) 25 (53) Leaflet abnormalities, No. (%) Calcified 44 (38) 16 (24) 28 (58) Tethered/restricted 9 (7.8) 6 (9.0) 3 (6.2) 0.7 b Prolapse 46 (40) 22 (33) 24 (50) 0.06 Myxomatous 23 (20) 10 (15) 13 (27) 0.11 Thickening 56 (49) 27 (40) 29 (60) 0.03 Redundant 31 (27) 21 (31) 10 (21) 0.2 Elongated 30 (26) 22 (33) 8 (17) 0.05 Leaflet length, mean SD cm Anterior c Posterior d Chordae abnormalities, No. (%) Calcified 5 (4.3) 2 (3.0) 3 (6.2) 0.6 b Ruptured 28 (24) 17 (25) 11 (23) 0.8 Elongated 12 (10) 11 (16) 1 (2.1) 0.01 Shortened 6 (5.2) 2 (3.0) 4 (8.3) 0.2 To anterior leaflet 12 (10) 10 (15) 2 (4.2) 0.06 To posterior leaflet 1 (0.87) 1 (1.5) 0 (0) 0.9 b To septum 4 (3.5) 2 (3.0) 2 (4.2) 0.9 b To posterior wall 1 (0.87) 0 (0) 1 (2.1) 0.2 Papillary muscle abnormalities Hypertrophied anterolateral muscle 15 (13) 6 (9.0) 9 (19) 0.12 Hypertrophied posterolateral muscle 13 (11) 6 (9.0) 7 (15) 0.4 Direct attachment to ventricular wall 4 (3.5) 0 (0) 4 (8.3) 0.03 b Direct attachment to leaflet 6 (5.2) 5 (7.5) 1 (2.1) 0.4 b Evidence of leaflet septal contact 36 (31) 26 (39) 10 (21) 0.04 a In 2 patients, accurate assessment was not possible. b Fisher exact test. c Data available for 77 patients (51 repairs, 26 replacements). d Data available for 64 patients (41 repairs, 23 replacements). SAM systolic anterior motion; SD standard deviation. Appendix Table 2. Preoperative Jet Direction and Mitral Valve Morphology in Patients With Hypertrophic Cardiomyopathy Undergoing Mitral Valve Surgical Procedures Jet Direction, No. (%) c Morphology a No. b None Anterior Anterior to Medial Posterior Posterior to Lateral Central Multiple Degenerative 34 1 (2.9) 6 (18) 1 (2.9) 11 (32) 2 (5.9) 8 (24) 5 (15) Myxomatous 23 0 (0) 6 (26) 1 (4.3) 3 (13) 0 (0) 6 (26) 7 (30) Papillary muscle abnormalities 21 1 (4.8) 1 (4.8) 0 (0) 10 (4.8) 0 (0) 8 (38) 1 (4.8) Restrictive chordal abnormalities 21 2 (9.5) 2 (9.5) 0 (0) 9 (43) 0 (0) 6 (29) 2 (9.5) Restrictive leaflet abnormalities 75 3 (4.0) 7 (9.3) 1 (1.3) 30 (40) 2 (2.7) 26 (35) 6 (8.0) Long leaflet 63 1 (1.6) 7 (11) 1 (1.6) 26 (41) 0 (0) 17 (27) 11 (17) a Categories not mutually exclusive. b Number with jet direction recorded. c Percentage of morphologic category.