Reoperation for persistant outflow obstruction in hypertrophic cardiomyopathy

Reoperation for Persistent Outflow Obstruction in Hypertrophic Cardiomyopathy Charles Stewart Roberts, MD, Charles L. McIntosh, MD, PhD, Paul S. Brown, Jr, MD, Richard 0. Cannon 111, MD, S. David Gertz, MD, PhD, and Richard E. Clark, MD Surgery, Cardiology, and Pathology Branches, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland

This study compares results of a second left ventricular myotomy and myectomy (M+M) with those of mitral valve replacement (MVR) as reoperative procedures for persistent left ventricular outflow obstruction after M+M in hypertrophic cardiomyopathy. Comparison of the second M+M group (n = 12)with the MVR group (n = 11)disclosed significant differences ( p < 0.05) in mean age at the initial operation (29 f 11 years versus 40 f 8 years), interval between operations (46 2 57 months versus 18 f 13 months), and age at reoperation (33 2 10 years versus 42 f 8 years); and insignificant differences in mean preoperative functional class, cardiac index, left ventricular outflow gradients at rest or with provocation, and hospital mortality at reoperation (2/12versus 1/11). At 6 months after reoperation, comparison of results of a second M+M with MVR showed that mean functional

class, cardiac index, and left ventricular outflow gradient at rest were similarly improved, but the outflow gradient with provocation was significantly higher in the second M+M group (57 & 44 mm Hg versus 14 f 9 mm Hg, p < 0.05). Total follow-up was 108 patient-years (100% complete) with an average of 5.9 years per patient in the second M+M group and 3.4 years per patient in the MVR group. Actuarial survival, including hospital mortality, at 3 and 5 years was 83% and 76%, respectively, after the second M+M, which was similar to 92% and 77% after MVR. Thus, either a second M+M or MVR is effective in relieving the hemodynamic obstruction and decreasing symptoms, but a second M-M is preferable because complications of anticoagulation and substitute valves are avoided. (Ann Thoruc Surg 2991;52:455-60)

0

analysis as was 1 patient who underwent combined coronary artery bypass and a second M+M. The indication for reoperation in the remaining 23 patients was persistent left ventricular outflow obstruction with or without mitral regurgitation. Myotomy and myectomy was the initial operation in all 23 patients. The initial operation was performed at the National Institutes of Health in 16 of the 23 patients and at other medical centers in 7 patients. The reoperative procedures were all performed at the National Institutes of Health. Twelve of the 23 patients underwent a second M+M and the other 11 patients underwent MVR. A second M+M was performed in all 7 patients in whom the initial M+M was performed at a different institution. Likewise, all patients who underwent MVR at reoperation had undergone the previous M+M at the National Institutes of Health. The operative techniques used in M+M were described by Morrow (31 in 1978, and those used in MVR were described by McIntosh and associates [5] in 1988. The excised mitral valves from the 11 patients who underwent MVR were reexamined. Follow-up after reoperation was routinely achieved by yearly outpatient clinic visits, the findings from which were reviewed. The patients were also contacted by telephone after January 1, 1990, to update their records. The study period extended from 1959 to 1990. Total postoperative follow-up was 108 patient-years (100% of potential follow-up) with a mean of 5.9 years per patient

perative treatment of hypertrophic cardiomyopathy was initiated in 1958 by Cleland and associates [l, 21, who performed transaortic left ventricular myotomy. Numerous authors since then have reported results of various operations to relieve left ventricular outflow obstruction in hypertrophic cardiomyopathy [3, 41. At the National Heart, Lung, and Blood Institute, 535 patients have undergone 551 operations for hypertrophic cardiomyopathy from 1959 to 1990. Reoperation for persistent left ventricular outflow obstruction was performed in 23 patients, 12 of whom had a second left ventricular myotomy and myectomy (M+M) and 11 of whom had mitral valve replacement (MVR). This report compares results of these two procedures at reoperation.

Material and Methods The clinical records of the Surgery Branch, National Heart, Lung, and Blood Institute were searched by computer to identify all patients who underwent reoperation for hypertrophic cardiomyopathy. Twenty-seven patients were identified and their records were reviewed. Three of the 27 patients underwent reoperation for pure mitral regurgitation without associated left ventricular outflow obstruction, and they were excluded from further Accepted for publication Nov 27, 1990. Address reprint requests to Dr Roberts, Department of Surgery, Medical University of South Carolina, 171 Ashley Ave, Charleston, SC 294252270.

456

ROBERTS ET AL REOPERATION IN HYPERTROPHIC CARDIOMYOPATHY

in the second M+M group and 3.4 years per patient in the MVR group. Preoperative and postoperative data within each group were compared using paired Student's t tests. Data between each group were compared using unpaired Student's t tests and the Mantel-Haenszel test. Average values in each group are expressed as a mean k standard error of the mean.

Ann Thorac Surg 1991;51:45540

(Table 1, patient 1) died 18 months later of congestive heart failure, 52 months (4.3 years) after the second M+M. Only 1 other patient (Table 1, patient 11) died during the follow-up period, and the cause was a ventricular arrhythmia. An actuarial survival curve of the 12 patients who underwent a second M+M is presented in Figure 1. The acturial survival at 3 and 5 years was 83% and 76%, respectively, including the two operative deaths.

Results Repeat Myotomy and Myectomy

Mitral Valve Replacement

Clinical observations on the 12 patients are presented individually in Table 1. The ages of the patients at the initial operation ranged from 8 to 45 years (mean, 29 years); 7 were men and 5 were women. Left ventricular outflow obstruction of varying degrees was the indication for the initial operation in all patients. Only 1 patient (Table 1, patient 2) had severe mitral regurgitation in addition to left ventricular outflow obstruction. Four of the 12 patients had reports of preoperative echocardiograms, which indicated in each a maximal ventricular septal thickness greater than 18 mm and systolic anterior motion of the anterior leaflet of the mitral valve. The amount of ventricular septum resected was unknown in 7 patients, 6 of whom had M+M at a medical center other than the National Institutes of Health. The interval from initial M+M to a second M+M ranged from 6 days to 168 months (mean, 46 months). Cardiac catheterization was performed during the interval in the 12 patients, and persistent left ventricular outflow obstruction was evident in each patient. Two patients (Table 1, patients 6 and 12) also had severe mitral regurgitation. The ages of the patients at reoperation ranged from 22 to 48 years (mean, 33 years). Fatal hemorrhage occurred during reoperation from laceration of the posterior wall of the ascending aorta before the planned M+M in 2 of the 12 patients (Table 1, patients 2 [1976] and 9 [1977]). The amount of ventricular septum resected in the other 10 patients ranged from 0.6 to 3.5 g (mean, 1.9 g). These 11 patients survived a second M+M and were discharged from the Clinical Center. At 6 to 12 months after reoperation, 7 patients were symptomatically improved, 2 were unchanged, and 1was symptomatically worse. Mean functional class in the 10 hospital survivors was significantly improved from that preoperatively (2.0 versus 3.0, p < 0.05). Cardiac catheterization was performed in 9 of the 10 patients, and the left ventricular outflow gradient in each was either decreased or absent. The mean outflow gradient at rest and with provocation (use of isoproterenol or other maneuver noted in Table 1) postoperatively was significantly lower than that preoperatively although mean cardiac indices were unchanged. Four of the 11 patients had significant obstruction (> 50 mm Hg) during provocation with intravenous isoproterenol or Valsalva maneuver. The mitral regurgitation was mild postoperatively in the 2 patients in whom it was previously severe. Two of the 10 hospital survivors subsequently underwent MVR for persistent left ventricular outflow obstruction (Table 1, patients 1 and 10). Both patients survived this third operation, but 1

Clinical observations on each of the 11 patients who underwent MVR after M+M are presented in Table 2. The ages of the patients at the initial operation ranged from 29 to 58 years (mean, 40 years); 5 were men and 6 were women. Left ventricular outflow obstruction was the indication for initial M+M in all patients. One patient (Table 2, patient 8) also had severe mitral regurgitation. All patients underwent preoperative echocardiography, which disclosed in each a maximal ventricular septal thickness of 18 mm or more and a systolic anterior motion of the anterior mitral leaflet. Myotomy and myectomy had been performed at the National Institutes of Health in all patients, and the amount of ventricular septum resected ranged from 0.9 to 6.0 g (mean, 2.4 g). The interval between M+M and MVR ranged from 1 day to 47 months (mean, 18 months). In 2 patients (Table 2, patients 8 and 9), MVR was performed within 7 days of the M+M. Cardiac catheterization was performed during the interval in all 11patients, each of whom had persistent left ventricular outflow obstruction. None had severe mitral regurgitation. The ages of the patients at reoperation ranged from 30 to 60 years (mean, 42 years). A Bjork-Shiley prosthesis was inserted in 8 patients, and a St. Jude Medical prosthesis in 3 patients. Each of the excised mitral valves was focally but extensively thickened by fibrous tissue without calcific deposits. Each valve was typical of that observed in patients with hypertrophic cardiomyopathy [6]. All 11 patients survived reoperation, but 1 patient (Table 2, patient 9) died on the first postoperative day of congestive heart failure. In this patient, the interval from M+M to MVR had been 1 day, during which time he had remained in congestive heart failure. The other 10 patients were discharged from the Clinical Center. At 6 to 12 months postoperatively, 8 of the 10 hospital survivors were symptomatically improved and 2 were unchanged. The mean functional class after reoperation was significantly improved from that preoperatively (1.9 versus 3.3, p < 0.05). Eight of the 10 patients underwent postoperative cardiac catheterization, and in each of them, the left ventricular outflow obstruction was absent or trivial. The mean left ventricular outflow gradients at rest and with provocation (use of isoproterenol or other maneuver noted in Table 2) were significantly decreased from preoperative values whereas mean cardiac indices were unchanged. Two of the 10 hospital survivors died during the follow-up period: 1 of congestive heart failure 4.2 years postoperatively (Table 2, patient 10) and 1 of cancer after 4.9 years (Table 2, patient 4). Further mitral

3

F

33

8

'

2.8

3

...

29/13

34/6

26/14

... ... ...

75

100

86

... ...

50

55

120

...

___ 29/18

62

60

6

150 2+

4.5

3.8

1+

0

1.1 3+

2.5

2.0

... 111 2.7

2+

3

1.9

2

1.4 3

4

3

...f

...f ...f

3

. . . 2.3 1+

...f

4

. . . . . .f

. . . . . . 1+ ......... .........

91h 3.8

2

3

3

3

1.0

1.3

0e.f

3 3

1.3

...

...f

140 2.2 . . . .........

136

...

80

... 80

31/15 65

80

55

55/35

40

...

65

75

42

90

50

75

80

45

1.8

4.2

2.5

2.7

3.0

1.9

2.2

1+

3+

...

2+

1+

1+

1+

2.0

3.4

125

75

2.7

... ......

115

160

3+

1+

1+

1+

. . . 4.7 2+

78'

100

120

140

205

120

155

33

45

43

48

37

39

34

40

26

22

24

24

22

46

<1

10

77

17

78

38

69

19

4

43

37

168

3'

2.5

1

2

3b

5"'

2

2

2

1

3

1

5

25/11

36/21

13

20

0

... 30

.

0

...

40

20

15

0

...

.

0

...

23/13

1

...

...

32/14

...

1719

W8

...

2618

2.8

1+

57

40

0

90

2.9

...

3.4

2.9

2+ 1+

1+

. _ . ......

85

1.9

1.6 2.0h

2.0

Od

1.6

1.4

0.6

3.5h

3.5

Od

. _ ___ _

1+

...

1+

...

1.5

2.9

2.2

3.4

3.6

0.7

. _ .......

949

40

40

09

... ......

127

Septum ProMR Resected Resting voked" CI (0-3) (g)

LV-A0 Gradients

Catheterization Data 6 Months After Second LVMM

Interval Age at to Second Second ProMR LVMM LVMM FC PA Resting voked" CI (0-3) (y) (mo) (14) s/d LV-A0 Gradients

30120

30110

50120

24/10

30115

...

16112

2316

34/20

2818

Septum ProMR Resected FC PA (g) (14) d d Resting voked' CI (0-3)

28/16

28/6

50125

18/10

PA dd

LV-A0 Gradients

Catheterization Data Before Second LVMM

a Gradient measured during infusion of isoproterenol unless otherwise noted. Mitral valve replacement (without hospital death) was later performed for persistent left ventricular outflow obstruction. ' Patient died before postoperative catheterization. Fatal hemorrhage occurred before left ventricular myotomy and myectomy could be accomplished. Myotomy not myectomy was performed. First operation not performed at the National Institutes of Health. g Gradient measured during Valsalva maneuver. Suture plication of the anterior mitral leaflet was performed in addition to left ventricular myotomy and myectomy. Gradient measured during inhalation of amyl nitrite. A 0 = aortic pressure (mmHg); BS = Bjork-Shiley (tilting disc); CI = cardiac index (L . min-' . m-'); LV F = female; FC = functional class (New York Heart Association); H = Hancock; = left ventricular pressure (mm Hg); LVMM = left ventricular myotomy and myectomy; M = male; MR = mitral regurgitation; MVR = mitral valve replacement; PA = pulmonary artery pressure (mm Hg); s/d = systoliddiastolic; SJM = St. Jude Medical.

29

45

12

Mean

F

42

11

M

M 4

3 3

M

35 42

9 10

3

F

31

7

3

30

6 3

25

5

M

3

M

22

4

M

2

F

20

3

2 3

19

2

F M

8

1

Age at First Patient LVMM FC No. (Y) Sex ( 1 4 )

Catheterization Data Before First LVMM

Table 1. Clinical Findings in 12 Patients With Hypertrophic Cardiomyopathy Who Underwent LVMM Followed by a Second LVMM

3m

%

n

ii

%z

m

1

5

458

ROBERTSET AL REOPERATION IN HYPERTROPHIC CARDIOMYOPATHY

Fig I. Actuarial survival curves, including hospital deaths, after reoperation for persistent left ventricular outflow obstruction after previous left ventricular myotomy and myectomy (LVh4M) in hypertrophic cardiomyopathy; 12 patients underwent repeat LVMM and 11 underwent mitral valve replacement (MVR).

Ann Thorac Surg 1991;51:455-60

loo+

0-OLVMM;_. A-ALVMM-LVMM

2

. . , .

....,.

.

, ,

.,.

,

..

-,

10

Years Postoperation

0 Number of Patients 0 at Risk 1

2

3

4

5

6

LVMM-MVR

7

6 7

6

3

0

7

6

5

LVMM-LVMM

operations were required in 2 patients: 1 had closure of a perivalvular leak 7 months after MVR (Table 2, patient 10) and 1 had replacement of the prosthetic valve because of thrombosis 4.1 years after insertion (Table 2, patient 3). An actuarial survival curve of the 11 patients who underwent MVR after M+M is presented in Figure 1. Including the one hospital death, actuarial survival at 3 and 5 years was 92% and 77%, respectively.

Comment This analysis of 23 patients who underwent reoperation for persistent left ventricular outflow obstruction after M+M in hypertrophic cardiomyopathy indicates that either a second M+M or MVR is effective in relieving the hemodynamic obstruction and in decreasing symptoms. Comparison of the second M+M group with the MVR group disclosed a significantly younger mean age at the initial operation (29 +- 11 years versus 40 +- 8 years, p < 0.05) and at reoperation (33 f 10 years versus 42 ? 8 years, p < 0.05), a significantly longer time interval between operations (46 2 43 months versus 18 13 months, p < 0.05), and a significantly higher mean left ventricular outflow gradient with provocation (57 2 44 mm Hg versus 14 -t- 8 mm Hg, p < 0.05) after reoperation. Both groups showed similar improvement in functional class and in relief of left ventricular outflow obstruction at rest. Differences in hospital and late mortality after reoperation were insignificant. Previous studies have compared the effectiveness of M+M and MVR at the initial operation for obstructive hypertrophic cardiomyopathy. Krajcer and associates [7] from Houston compared results of M+M in 127 patients with results of MVR in 58 patients. Early and late postoperative mortality were similar. Mitral valve replacement significantly lowered the mean left ventricular enddiastolic pressure, whereas M+M did not. Long-term symptomatic and hemodynamic improvement, however, were similar. Four of the 58 patients who had MVR had undergone previous M+M, but no description of outcome

*

11 12

10 8

7

of these 4 patients was provided. Krajcer and associates concluded that most patients who require operative therapy should undergo M+M to avoid the disadvantages of valvar prostheses. Although our analysis concerned patients requiring reoperation, not initial operative therapy, our conclusion regarding the preferable procedure, M+M, is similar to that of Krajcer and associates. A comparison of results of M+M and MVR was also recently reported by Walker and associates [4] from Edinburgh, Scotland. Eleven patients underwent M+M and 11 underwent MVR. One of the latter 11 patients had previously undergone M+M. Although both operations significantly reduced the left ventricular outflow gradient, the reduction was greater after MVR. Walker and associates also noted increased left ventricular cavity enlargement after MVR and suggested that this procedure was preferable in patients with mitral dysfunction or poor diastolic compliance. Our results at reoperation were similar in that the provokable left ventricular outflow gradient was significantly lower after MVR. Persistent left ventricular outflow obstruction after M+M most commonly occurs because the M+M trough in the ventricular septum is inadequate in length, depth, or location. Intraoperative epicardial echocardiography has been used at our institution to characterize the trough after M+M. The length of the trough toward the apex is determined in the long-axis view; the depth of the trough and its circumferential location in the outflow tract are determined in the short-axis view. Mitral-septa1 contact with outflow obstruction may persist after M+M if the trough does not extend far enough apically, if the trough is not deep enough in the septum, or if the trough lies in a different circumferential location than the point of mitral-septa1 contact in the outflow tract. From this retrospective analysis, we conclude that a second M+M is the preferred reoperative procedure in most patients with persistent left ventricular outflow obstruction because complications related to anticoagulation and substitute valves are avoided. Such complications occurred in 2 of the 11 patients who had MVR

37 38 39 40 41 41 41 53 58

3 4 5 6 7

8 9

10 11

3 4 3 3 3

F F M M F

3 3

M M 22/10

40115

40/19

38/25

27/12

25/12

30112

20110

3V16

20110

22/12

3.0 29/13

3

M

3

3

F

F

3

F

86

80

60

76

110

85

130

90

70

125

60

70

2.3

...

2.6

120

2.4

loog 1.4

120

2.8

1+

2+

1+

3+

. . . 2.2 110

1+

2+

1+

2+

1+

1+

2+

2.2

90

2.1

125

2.3

2.6

... 200

3.0

2.5

75

145

2.4

2.2

2.3

2.8

3.0

6.0

2.3

2.1

1.0

1.3

0.9

2.6

25/7

37122

40120

35/22

24/12

23/10

30116

29/13

2319

22/14

2V9

3.4 28114

3

4

4

4

2

4

4

3

3

3

3

Mi.d

44

50

50

40

140

30

10

40

10

60

35

28

2.4

2.6

1.8

1.8

2.8

3.0

3.1

3.5

1.5

92

2.5

809 2.6

90

...

... 2.4

75

108

85

140

100

60

90

1+

0

41

... ...

42

60

54

41

41

42

41

41

39

35

30

0

1+

2+

1+

1+

1+

2+

18

24

20

<1

<1

10

19

20

2.6

21

47

16

25

27

25

23

BS

BS

27

25f

25

SJh4 27 BS

2

3

2.1

1

2

5d-e

2

2

2

23b 3

SJM 25

BS

BS

BS

BS

SJM 23

BS

0

10

15

20

3.4

3.1

1+

1+

ProMR Resting vokeda CI (0-3)

0

5

0

20‘

20

0

2.8

3.8

2.2

1+

0

0

0

4

2.5

0

3246

2.5

10

2.9

...............

38/18

...............

30116

30116

48/20

...............

24/14

26/15

PA dd

LV-A0 Gradients

A 0 = aortic pressure (mHg); BS = Bjork-Shiley (tiltingdisc); CI = cardiac index (L . min-’ . m-’); = left ventricular pressure (mm Hg); LVMM = left ventricular myotomy and myectomy; M = male; pressure (mm Hg); s/d = systoliddiastolic; SJM = St. Jude Medical.

a

F = female; FC = functional class (New York Heart Association); H = Hancock; LV MR = mitral regurgitation; PA = pulmonary artery MVR = mitral valve replacement;

Gradient measured during infusion of isoproterenol unless otherwise noted. Repeat mitralvalve replacement for acute valve thrombosis was performed 4 years later. Gradient measured during inhalation of amyl nitrite. This patient died on the first postoperative day of congestive heart failure; the interval from left ventricular myotomy to mitralvalve replacement was 1 day. Patient died before postoperative catherization. Closure of a peribasilar leak was performed 7 months after mitrd valve replacement. Gradient measured during Valsalva maneuver.

40

31

2

Mean

29

1

Age Interval

Catheterization Data 6 Months After MVR

at LVMM Prosthesis ProMR MVR toMVR FC Resting vokeda CI (0-3) (y) (mo) Type Size (14)

LV-A0 Gradients

LV-A0 Gradients

Age at Septum FC PA ProMR Resected FC PA Patient LVMM No. (y) Sex (14) s/d Resting voked” CI (0-3) (g) (14) s/d

Catheterization Before MVR

Catheterization Data Before LVMh4

Table 2. Clinical Findings in 11 Patients With Hypertrophic Cardiomyopathy Who Underwent Left LVMM Followed by MVR for Persistent Left Ventricular Outflow Obstruction

$3

SE! Lc”

om

! i ag

-

3 ii

3 5z B

5

3 m

460

ROBERTS ET AL REOPERATION IN HYPERTROPHIC CARDIOMYOPATHY

(Table 2, patients 3 and 10). Mitral valve replacement as a reoperative procedure can be justified, however, in severely symptomatic patients in whom (1) left ventricular outflow obstruction is mild and mitral regurgitation is severe; (2) the ventricular septum at the site of the trough is relatively thin, increasing the risk of ventricular septal defect in a second M+M; (3) the anterior mitral leaflet contacts the septum within an already deep trough; or (4) more than one previous M+M failed to relieve outflow obstruction.

References 1. Goodwin JF, Hollman A, Cleland WP, Teare D. Obstructive cardiomyopathy simulating aortic stenosis. Br Heart J 1960; 22:40&14. 2. Cleland WP. The surgical management of obstructive cardiomyopathy. J Cardiovasc Surg 1963;4:489-91. 3. Morrow AG. Hypertrophic subaortic stenosis. Operative methods utilized to relieve left ventricular outflow obstruction. J Thorac Cardiovasc Surg 1978;76:42&30. 4. Walker WS, Reid KG, Cameron WJ, Walbaum PR, Kitchin AH. Comparison of ventricular septal surgery and mitral valve replacement for hypertrophic cardiomyopathy. Ann Thorac Surg 1989;48:528-35. 5. McIntosh CL, Greenberg CJ, Maron BJ, Leon MB, Cannon RO 111, Clark RE. Clinical and hemodynamic results after mitral valve replacement in patients with obstructive hypertrophic cardiomyopathy. Ann Thorac Surg 1989;47:23646.

Ann Thorac Surg 1991;51:45560

6. Roberts CS, Roberts WC. Morphologic features of hypertrophic cardiomyopathy. In: Zipes DP, Rowlands DJ, eds. Progress in cardiology 2/2, Philadelphia: Lea & Febiger, 1989:&32. 7. Krajcer Z, Leachman RD, Cooley DA, Coronado R. Septa1 myotomy-myectomy versus mitral valve replacement in hypertrophic cardiomyopathy. Ten-year follow-up in 185 patients. Circulation 1989;8O(Suppl 1):57-64. 8. Maron BJ, Epstein SE, Morrow AG. Symptomatic status and prognosis of patients after operation for hypertrophic obstructive cardiomyopathy: efficiency of ventricular septal myotomy and myectomy. Eur Heart J 1983;4(Suppl F): 175-85. 9. Schulte HD, Bircks W, Losse 8. Surgical treatment of hypertrophic obstructive cardiomyopathy (HOCM): early and late results. In: Zipes DP, Rowlands DJ, eds. Progress in cardiology 2/2. Philadelphia: Lea & Febiger, 1989:195-215. 10. Williams WG, Wigle ED, Rakowski H, Smallhorn J, LeBlanc J, Trusler GA. Results of surgery for hypertrophic obstructive cardiomyopathy. Circulation 1987;76(Suppl 5):104-8. 11. Mohr R, Schaff HV, Danielson GK, Puga FJ, Pluth JR, Tajik AJ. The outcome of surgical treatment of hypertrophic obstructive cardiomyopathy. Experience over 15 years. J Thorac Cardiovasc Surg 1989;97666-74. 12. Cooley DA, Wuksach DC, Leachman RD. Mitral valve replacement for idiopathic hypertrophic subaortic stenosis: results in 27 patients. J Cardiovasc Surg 1976;17:38&7. 13. McIntosh CL, Maron BJ. Current operative treatment of obstructive hypertrophic cardiomyopathy. Circulation 1988; 78:487-95.