- Email: [email protected]
Surgical treatment of mitral insufficiency secondary to coronary artery disease
J
THORAC CARDIOVASC SURG
79:12-18, 1980
Surgical treatment of mitral insufficiency secondary to coronary artery disease From 1970 to 1978, 61 patients were operated upon for mitral insufficiency secondary to coronary artery disease. These patients were between 44 and 71 years of age and all were in Class III or IV of the New York Heart Association Classification. The left ventricular end-diastolic pressure was 15 mm Hg or more in 32 of the 39 patients in whom it was measured. Twenty-four of 31 patients in whom right heart catheterization was performed had a systolic pulmonary artery pressure of 50 mm Hg or greater. All 61 patients had myocardial revasculariration, 52 had repair of the mitral valve, and nine had mitral valve replacement. There were five hospital deaths in these 61 patients. Among the nine patients with a preoperative ejection fraction of 0.1 to 0.2, there were two hospital deaths; among the 20 patients with a preoperative ejection fraction of 0.25 to 0.40, there were two hospital deaths; and among the 32 patients with a preoperative ejection fraction of 0.45 to 0.70, there was only I hospital death. For those patients with repair and revascularization, the survivability was 81 % at 7 years. In the patients with repair and myocardial revascularization, the incidence of peripheral embolization was 0.5% per patient-year.
Jerome Harold Kay, M.D., Pablo Zubiate, M.D. (by invitation), Michael A. Mendez, M.D. (by invitation), Neal Vanstrom, M.D. (by invitation), Taro Yokoyama, M.D. (by invitation), and Mohammad A. Gharavi, M.D. (by invitation), Los Angeles, Calif.
In
September, 1970, a patient had myocardial revascularization and mitral valve repair for mitral insufficiency secondary to coronary artery disease. From then through Dec. 31, 1977, 61 patients had mitral valve operation and revascularization for mitral insufficiency secondary to coronary artery disease. The patients operated upon during the year 1978 were not included to allow at least a 15 month follow-up period for all patients. There were 38 men and 23 women, ranging in age from 44 to 71 years. The patients were classified into the New York Heart Association Classification according to fatigue and dyspnea but not angina. Forty-eight of the patients had angina in addition to fatigue and dyspnea. There were 28 patients in Class III and 33 in Class IV. The left ventricular end-diastolic
From The Los Angeles Heart Institute at The St. Vincent Medical Center and the University of Southern California School of Medicine, Los Angeles, Calif. Aided by a grant from The Los Angeles Thoracic and Cardiovascular Foundation. Read at the Fifty-ninth Annual Meeting of The American Association for Thoracic Surgery, Boston, Mass., April 30 to May 2, 1979. Address for reprints: Jerome Harold Kay, M.D., 123 South Alvarado St., Los Angeles, Calif. 90057.
12
pressure was determined in 39 patients. Thirty-two of these patients (82%) had an end-diastolic pressure of 15 or greater (mean 22.7 mm Hg). The pulmonary artery pressure was measured in 31 of the 61 patients. Twenty of these 31 patients (65%) had a pulmonary artery systolic pressure of 50 mm Hg or greater. The amount of mitral insufficiency was calculated as follows. The cardiac output was determined by means of the Fick principle and the degree of mitral insufficiency was determined after calculating the left ventricular output by radiographic technique. If the blood ejected into the left atrium during systole equaled the blood ejected into the aorta, the mitral insufficiency was graded as 2/6; if the blood ejected into the left atrium was two times the volume ejected into the aorta, this was graded as 4/6; and if the amount ejected into the left atrium was triple that ejected into the aorta, this was graded as 6/6. One of the patients had Grade 1/6 mitral insufficiency, 12 patients had Grade 2/6, 27 had Grade 3/6, 19 had Grade 4/6, one had Grade 5/6, and in one patient the amount of insufficiency was not calculated. Nine patients had a preoperative ejection fraction of 0.1 to 0.2, 20 had a preoperative ejection fraction of 0.25 to 0.40, and 32 had a preoperative ejection fraction of 0.45 to 0.70.
0022-5223/80/010012+07$00.70/0 © 1980 The C. V. Mosby Co.
Volume 79
Mitral insufficiency secondary to coronary disease
Number 1 1980
January.
I3
Ant leailel An nulus
-:1""-::
.'
1s t.
,
sul ur' g
,
Mu;:' a l
l e Zlo fl e t
~.
A
B
Fig. I. In order to obliterate the insufficiency, the greatly dilated. sagging mural anulus is decreased considerably in size . This can be accomplished by decreasing the length of the mural anulus to about one fourth and thereby bringing it up into apposition with the anterior leaflet. The anulus of the anterior leaflet must not be encroached upon . The repair is accomplished with No.2 silk sutures as illustrated.
Operative procedure At the time of operation, 23 patients had a dilated mitral anulus only; however, 38 patients also had a ruptured or infarcted papillary muscle or torn chordae tendineae as well as the dilated anulus. All patients with mitral insufficiency have a dilated mitral anulus. The dilatation always is located at the area of the mural anulus and never at the area of the anulus of the anterior leaflet. The fibrous trigone prevents stretching of the anulus of the anterior leaflet. Although mitral valve repair was deemed to be superior to replacement, in nine patients replacement was performed. Replacement was performed when the surgeon felt repair was not possible or early in the series, when we did not realize that the mitral valve that was insufficiency secondary to infarction could be repaired . Not being able to repair the valve was not related to ejection fraction . Five patients had replacement with a Kay-Shiley disc valve with muscle guard and four had replacement with a Hancock glutaraldehyde-treated porcine valve . Mitral valve repair was performed as follows. Careful search was made to determine if there were torn chordae tendineae or torn papillary muscle. If the chordae tendineae and papillary muscles were intact, then the repair consisted of decreasing the anulus of the mitral valve by doing away with approximately 70% to 75% of the anulus of the mural leaflet. The anulus was decreased mainly at the posteromedial commissural area, where two or three interrupted figure-of-eight sutures of No.2 silk were placed to do away with 45% to 50% of the anulus. Sutures were also placed to decrease the mural anulus of the anterolateral commissural area
by another 20% to 25%. The sutures at the anterolateral commissural area were placed superficially enough so as not to encircle the circumflex coronary artery. All the sutures were tied and only 25% to 30% of the mural anulus remained . There was then a snug two-fingerbreadth opening (Fig. I). If on opening the left atrium there were torn chordae tendineae or torn papillary muscle, this area of the valve was sutured down to the nearest papillary muscle head (Fig . 2). In dealing with a ruptured head or chordae tendineae of the anterior leaflet, the surgeon was careful to be sure that there was adequate anterior leaflet to billow against the mural leaflet in correcting the insufficiency. This was not a problem with suturing the mural leaflet to papillary muscle when there were torn chordae tendineae or papillary muscle of the mural leaflet, since most of the anulus of the mural leaflet was done away with by the repair . After the torn chordae tendineae or torn papillary muscle is repaired, the mural anulus is decreased to 25% to 30% of its original length, since there is always annular dilatation secondary to torn chordae tendineae or torn papillary muscle. Prior to the start of bypass, the proximal ends of the vein grafts are sutured into the aorta. After the start of bypass, the distal ends of the grafts are sutured into the coronary arteries with the aorta cross-clamped. The cross-clamp is released for 5 to 10 minutes and the heart is allowed to beat, the cross-clamp is again applied, and the mitral operation is performed . Twenty-six of the patients had one vein graft inserted, 20 had two vein grafts, 10 had three vein grafts, two had four vein grafts , and one had five vein grafts .
The Journal of
14
Thoracic and Cardiovascular Surgery
Kay et al.
, To',.n he""d of \ pZLP\1lZ5.r'y mUB. to '" mur 5 1 leaf l e t
A
B
~.
Fig. 2. The area of the leaflet involved with tom chordae tendineae or tom papillary muscle is repaired by placing interrupted figure-of-eight sutures of 2-0 Tevdek through the head of the nearer papillary muscle and then bringing this suture up through the leading edge of the involved area of the leaflet. Usually two separate sutures are used at each area . For ease of exposure. all sutures are placed prior to tying .
Three patients also had anastomosis of the left internal mammary artery to the left anterior descending coronary artery.
Results There were two hospital deaths in the nine patients with replacement and myocardial revascularization and three hospital deaths in the 52 patients undergoing repair and myocardial revascularization. The five deaths were as follows: One patient with an ejection fraction of 0.2 developed a bacteremia and died of a cardiac arrest 37 days after mitral repair. Another patient with disc valve replacement and an ejection fraction of 0.2 died of an embolus to the left anterior descending coronary artery 4 days after operation. One patient with a normal ejection fraction and with mitral replacement with a Kay-Shiley disc valve died of a myocardial infarct 9 days following operation. A patient with an ejection fraction of 0.4 and mitral repair died of a pulmonary embolus 10 days following operation. The fifth hospital death was in a patient with an ejection fraction of 0.4 who entered the hospital in cardiogenic shock and renal failure. This patient was operated upon as an emergency and myocardial revascularization and repair were performed. She continued to do poorly after operation and died on the fourteenth postoperative day. There were I I late deaths (Table I). Table II reveals the hospital and late mortality rates as related to ejection fraction. The lower the ejection fraction, the higher the
hospital mortality rate and the less likelihood of longterm survival. There was one peripheral embolic event resulting in hemiplegia in the 52 patients with repair followed for 229 patient-years, or 0.5% per patient year . One other patient with repair died from a cerebral hemorrhage 7 months after operation. This patient was continued on sodium warfarin by her family physician for this period of time, and at the time of the cerebral hemorrhage the prothrombin concentration was 10% of normal . Anticoagulant therapy is advised for only 3 months following mitral repair with the prothrombin concentration maintained at 20%. There were nine patients with mitral replacement. Five had disc valve replacement and four had replacement with the Hancock glutaraldehyde-treated porcine valve . One hospital death was due to an embolus to the left anterior descending coronary artery and another was due to myocardial infarction in the five patients with a disc valve . There was one late death owing to a cerebral embolus 27 months after operation following replacement with a disc valve. In the four patients with a porcine valve there was one cerebral embolus, which caused the death of the patient I I months following operation. Table III reveals the postoperative change in New York Heart Association Classification in the 45 living patients 15 months or longer following operation. An attempt was made to have all patients restudied post-
Volume 79 Number 1
I5
Mitral insufficiency secondary to coronary disease
January, 1980
Table I. Late deaths and cause of death E.F.
No. of patients
0.25 0.60 0.18 0.46 0.40 0.35 0.68 0.15 0.15 0.25 0.25
Cause of death
Repair or replacement
Cerebral emboli Cerebral hemorrhage Cardiac arrest Pulmonary emboli Pulmonary emboli Mediastinitis Unknown Cerebral emboli Unknown, diedinsleep Unknown Unknown
Replacement - disc Repair Repair Repair Repair Repair Repair Replacement - porcine Repair Repair Repair
Postop time (mo) 27 7 8 3 5 3 18 11 11
23 43
Legend: E.F., Ejection fraction.
operatively, but if the patient was doing well, neither the patient nor the family physician wanted the patient restudied. For this reason, only 12 patients were restudied 5 months to 3 years after operation. Of 25 grafts inserted in these 12 patients, 20 (80%) were patent. The mitral regurgitation had decreased from Grade 3/6 (mean) to Grade 0.8/6 (mean) (p < 0.001), and the ejection fraction had increased from 0.40 (mean) preoperatively to 0.48 (mean) postoperatively (p < 0.01) in the 12 patients having postoperative studies. Fig. 3 represents the survivability for those patients at 7 years with replacement (33%), repair (81%), and the entire group (74%).
Discussion There are very little data in the literature regarding the operative treatment of mitral insufficiency secondary to coronary artery disease. Buch and associates! reported on a group of 23 patients who had mitral valve disease with coronary artery disease. Eleven had coronary artery disease and nonrheumatic mitral regurgitation. Six of these 23 patients (26%) died within the first 30 days following replacement of the patient's mitral valve with a porcine valve. Three more patients died after discharge from the hospital. There were therefore nine deaths in this group of 23 patients with mitral replacement and coronary artery disease. Fourteen of these 23 patients had aorta-coronary bypass grafts. In a 1977 report on 243 patients with isolated replacement of the mitral valve with the Hancock glutaraldehydepreserved porcine aortic valve xenografts, Stinson and colleagues stated: MVR [mitral valve replacement] patients with associated coronary artery disease accounted for a substantial portion of both early and late deaths. For example, in 54 patients the diagnosis of coronary artery disease was established by coronary angiog-
Table II. Hospital and late deaths by ejection fraction Ejection fraction 0./-0.20
Total patients Hospital deaths Latedeaths Total deaths
0.45-0.70
9 2(22%) 3 5(56%)
32 1(3%) 3 4(13%)
20 2(10%) 5 7(35%)
Table III. New York Heart Association Functional Classification: Follow-up of 45 patients
_I From (18 patients) IV (27
patients)
III
To IV
III
1I
2
8
I
10
/
8 16
II I
raphy, electrocardiographic evidence of previous myocardial infarction, or postmortem examination within one year after operation. Operative mortality rate in this group was 18.5 per cent, and the survival rate at 2 years was 56 (::±: 7.8) per cent. Patients undergoing simultaneous coronary artery bypass grafting for angiographically defined coronary artery lesions fared similarly. * Among our 61 patients operated upon for mitral insufficiency secondary to coronary artery disease, there were nine patients who had the mitral valve replaced along with revascularization, with two hospital deaths and two deaths following discharge from the hospital. Among the 52 patients with mitral repair and rev as*Stinson EB, Griepp RB, OyerPE, Shumway NE: Long-term experience with porcine aortic valve xenografts. J THoRAc CARDtOVASC SURG
73:54-63, 1977.
16
The Journal of Thoracic and Cardiovascular Surgery
Kay et al.
100 81% 74%
.c
as > >
...
33%
::l
fI)
o
20
MITRAL REPAIR & REVASC
52 PT. 181% 7 VRS,)
[] MITRAL REPLACEMENT & REVASC 9 PT. 133% 7 VRS,) • ENTIRE GROUP
61 PT. (74% 7 VRS.)
OL-_ _.l.-_ _. L -_ _.L...-_ _.L...-_ _L..-_----IL..-_----J
o
1
2 3 4 5 Time Interval (years)
6
7
Fig. 3. Survivability, percent at 7 years. cularization, there were three hospital deaths (6%) and nine late deaths. For our entire series of 61 patients, there was a survivability of 74% at 7 years. However, if one excludes the nine patients who had replacement and a survivability of only 33% at 7 years, there was a survivability of 81 % at 7 years for the 52 patients undergoing repair and revascularization. There is a striking improvement in survivability for those patients with a good ejection fraction prior to operation compared to those with a poor ejection fraction. Among the nine patients with a preoperative ejection fraction of 0.1 to 0.2, there were four (44%) long-term survivors, as compared to 13 of 20 (65%) long-term survivors among those patients with a preoperative ejection fraction of 0.25 to 0.40 and 28 of 32 (87%) long-term survivors among those patients with a preoperative ejection fraction of 0.45 to 0.70. In the 52 patients who underwent repair, there was only one case of peripheral embolus for 229 patientyears, or 0.5% per patient-year. This figure compares favorably with a reported incidence of 5.2% per patient-year as noted by Salomon and co-workers! in 1976 with the Hancock glutaraldehyde-treated porcine valve. Whenever mitral valve operation is required for mitral insufficiency, it would appear advisable to repair the mitral valve if possible." Repair is more difficult than replacement; however, the morbidity and longterm mortality would appear less with repair than replacement. REFERENCES Buch WS, Pipkin RO, Hancock WO, Fogarty TJ: Mitral valve replacement with the Hancock stabilized glutaralde-
hyde valve. Clinical and laboratory evaluation. Arch Surg 110:1408-1415, 1975 2 Salomon NW, Stinson EB, Griepp RB. Shumway NE: Surgical treatment of degenerative mitral regurgitation. Am J Cardiol 38:463-468, 1976 3 Kay JH, Zubiate P, Mendez MA, Vanstrom N, Yokoyama T: Mitral valve repair for significant mitral insufficiency. Am Heart J 96:253-262, 1978
Discussion DR. JOHN W. KIRKLIN Birmingham. Ala.
I pay a good deal of attention to what Dr. Kay says because I remember 5, 6, 7 years ago he was reporting very good results in coronary artery bypass grafting in patients with poor left ventricular function, and many of us were skeptical. Today his position in that regard has been fully justified, and he simply understood how to do that before we did. For my colleagues and me, the experience he has discussed today is also quite remarkable because we have felt that we nearly always had to replace the mitral valve in patients whose mitral incompetence was secondary to ischemic heart disease. I hope the discussers will let us know their thoughts about this matter. May I ask Dr. Kay a couple of questions. First, do you anticoagulate the patients after the annuloplastyand, if so, for how long? Second, do you have any information about the regression of mitral incompetence after satisfactory coronary artery bypass grafting in patients whose valve was not altered? Is that any problem in interpreting the results of annuloplasty? Finally, could you tell us something about residual mitral incompetence in these patients? DR. M. LAXMAN KAMATH Milwaukee. Wis,
Over the years we have seen many cases of coronaryartery disease with associated mitral insufficiency. Our preliminary
Volume 79 Number 1
January, 1980
observation is that such patients can be categorized into four groups. In the first group, the mitral insufficiency associated with angina can mean insufficiency secondary to poor myocardial blood flow. In such patients, total myocardial revascularization alone improves the mitral insufficiency. In this group of patients, we have seen a left atrial "V" wave as high as 55 mm Hg that disappeared completely after revascularization. Patients likely to have this kind of response are those with severe angina. The left ventricular contractions are generally excellent, and they do not have any scarring of myocardium during the operation. The second group of patients has a combination of mitral insufficiency with a massive scar along the posterolateral wall of the left ventricle. With this group, mitral valve repair or replacement is usually necessary to improve mitral insufficiency. The third group of patients are those with aneurysm of the left ventricle. When the aneurysm involves the inferior wall, during closure ofthe aneurysm, the closing sutures are passed from the lateral wall beneath the papillary muscles and through the base of the septum. When the papillary muscles are scarred, the sutures are passed through the papillary muscles and the aneurysm is closed in such a way as to pull the papillary muscles toward the apex. In such patients, we have never replaced or repaired the mitral valve for insufficiency. The patients in the fourth group are those with an anterolateral aneurysm of the left ventricle. In these patients, again, the aneurysm is closed by pulling the base of the anterior papillary muscles toward the apex. Unlike results with posterior wall aneurysm, we do not have uniform success with anterior wall aneurysm in relieving the mitral insufficiency. The majority of these patients must have replacement of the mitral valve. I would like to ask the authors to describe their experience in patients with mitral insufficiency associated with aneurysm of the left ventricle. DR. J. A. MIJANGOS Rochester, N. Y.
I would like to describe our experience with a special group of patients-those with a large scar secondary to a myocardial infarction either in the anterior or diaphragmatic wall of the left ventricle or with multiple infarctions that have produced fibrosis or rupture of the posterior papillary muscle. On these patients, our results with plastic repairs of the mitral valve have not been as good as the ones described by Dr. Kay. We believe that the complete approach to these patients should consist of three points: (I) myocardial revascularization, (2) correction of the ventricular geometry, and (3) mitral valve replacement in order to correct mitral incompetence. [Slide] In the first slide you can see the description of some of our patients with the important catheterization data. Some of these patients have a cardiac index and ejection fraction that does not reflect reality because they were catheterized during intra-aortic balloon counterpulsation or during the administration of intravenous nitroglycerin or dopamine. Our
Mitral insufficiency secondary to coronary disease
I7
series, to date, comprises approximately 20 patients. Both the ejection fractions and the cardiac indices are fairly low. In these patients, our approach to mitral valve replacement has been a previously described ventricular approach. We believe that this approach greatly simplifies the operation. The exposure through the ventricle is usually excellent, the anulus is well demarcated, the aortic valve and septum can easily be exposed, and injury to these structures can be avoided. In this group of patients, it is usually difficult to expose the mitral valve through a fairly small atrium. [Slide] This is an exposure of the mitral ring through an incision on an anteroseptal left ventricular aneurysm. The mitral valves were replaced with a Bjork-Shiley prosthesis. [Slide] This is the exposure of an anulus after a diaphragmatic infarction had been opened and resected, the valve excised, and the stitches placed on the anulus. In our series of 20 cases, the mortality rate is approximately 0% to date. The complications incurred are heart block in one patient, cardiac arrhythmia (supraventricular arrhythmia) in one patient, and frequent premature ventricular contractions in another patient, easily controlled with procainamide (Pronestyl). All of our patients were considered to be Cardiac Class IV preoperatively. Postoperatively, 75% of our patients were in Cardiac Class I and 30% in Cardiac Class II. DR. KAY (Closing) In answer to Dr. Kirklin's questions, following mitral repair for mitral insufficiency, a patient receives anticoagulant therapy for 3 months. With our 20 years' experience with repairing the mitral valve, we have noted that if the patients are not treated with an anticoagulant, they may have peripheral emboli. The anticoagulant need not be continued after 3 months because by then the areas of repair and incision are healed. In answer to Dr. Kirklin's second question: Would the mitral insufficiency regress in our patients if only revascularization were performed? If the patient has a relatively good ejection fraction and the mitral insufficiency is Grade 1/6 to 2/6, the insufficiency is not corrected at the time of operation. With a poor ejection fraction, such as 0.1 or 0.2, Grade 1/6 to 2/6 mitral insufficiency is extremely important and the valve must be repaired, or replaced if repair is not possible. We see many patients with what we consider minimal mitral insufficiency, and we do nothing about the mitral insufficiency at the time of revascularization because we are concerned that we may not be able to repair the valve and then it would have to be replaced. The decision is made prior to operation when the left ventriculogram is reviewed and the ejection fraction is determined. I would say that for every patient on whom we repair or replace the mitral valve, another nine patients are deemed to have too little insufficiency. Of our 61 patients reported today, 38 had tom chordae tendineae or ruptured papillary muscle, and therefore the mitral insufficiency was certainly quite significant for these patients. It was also significant in 21 of the remaining 23 patients in this series. As to residual mitral incompetence in these patients, there
18
Kay et al.
are three patients who still have significant mitral insufficiency, and they are the 3 patients of the 45 living patients who are in Class III. Most of the remaining 42 patients do not have residual murmurs. The patient with a residual murmur usually has a short systolic murmur. In these patients the heart is smaller than it was preoperatively.
The Journal of Thoracic and Cardiovascular Surgery
Regarding the treatment of an aneurysm associated with mitral insufficiency, if we had a patient with severe mitral insufficiency associated with a ventricular aneurysm, the mitral valve would be repaired and the aneurysm resected. We leave small aneurysms alone. We feel that small aneurysms should not be resected at the time of revascularization.
Information for authors Most of the provisions of the Copyright Act of 1976 became effective on January I, 1978. Therefore, all manuscripts must be accompanied by the following written statement, signed by one author: "The undersigned author transfers all copyright ownership of the manuscript (title of article) to The C. V. Mosby Company in the event the work is published. The undersigned author warrants that the article is original, is not under consideration by another journal, and has not been previously published. I sign for and accept responsibility for releasing this material on behalf of any and all co-authors." Authors will be consulted, when possible, regarding republication of their material.
THORAC CARDIOVASC SURG
79:12-18, 1980
Surgical treatment of mitral insufficiency secondary to coronary artery disease From 1970 to 1978, 61 patients were operated upon for mitral insufficiency secondary to coronary artery disease. These patients were between 44 and 71 years of age and all were in Class III or IV of the New York Heart Association Classification. The left ventricular end-diastolic pressure was 15 mm Hg or more in 32 of the 39 patients in whom it was measured. Twenty-four of 31 patients in whom right heart catheterization was performed had a systolic pulmonary artery pressure of 50 mm Hg or greater. All 61 patients had myocardial revasculariration, 52 had repair of the mitral valve, and nine had mitral valve replacement. There were five hospital deaths in these 61 patients. Among the nine patients with a preoperative ejection fraction of 0.1 to 0.2, there were two hospital deaths; among the 20 patients with a preoperative ejection fraction of 0.25 to 0.40, there were two hospital deaths; and among the 32 patients with a preoperative ejection fraction of 0.45 to 0.70, there was only I hospital death. For those patients with repair and revascularization, the survivability was 81 % at 7 years. In the patients with repair and myocardial revascularization, the incidence of peripheral embolization was 0.5% per patient-year.
Jerome Harold Kay, M.D., Pablo Zubiate, M.D. (by invitation), Michael A. Mendez, M.D. (by invitation), Neal Vanstrom, M.D. (by invitation), Taro Yokoyama, M.D. (by invitation), and Mohammad A. Gharavi, M.D. (by invitation), Los Angeles, Calif.
In
September, 1970, a patient had myocardial revascularization and mitral valve repair for mitral insufficiency secondary to coronary artery disease. From then through Dec. 31, 1977, 61 patients had mitral valve operation and revascularization for mitral insufficiency secondary to coronary artery disease. The patients operated upon during the year 1978 were not included to allow at least a 15 month follow-up period for all patients. There were 38 men and 23 women, ranging in age from 44 to 71 years. The patients were classified into the New York Heart Association Classification according to fatigue and dyspnea but not angina. Forty-eight of the patients had angina in addition to fatigue and dyspnea. There were 28 patients in Class III and 33 in Class IV. The left ventricular end-diastolic
From The Los Angeles Heart Institute at The St. Vincent Medical Center and the University of Southern California School of Medicine, Los Angeles, Calif. Aided by a grant from The Los Angeles Thoracic and Cardiovascular Foundation. Read at the Fifty-ninth Annual Meeting of The American Association for Thoracic Surgery, Boston, Mass., April 30 to May 2, 1979. Address for reprints: Jerome Harold Kay, M.D., 123 South Alvarado St., Los Angeles, Calif. 90057.
12
pressure was determined in 39 patients. Thirty-two of these patients (82%) had an end-diastolic pressure of 15 or greater (mean 22.7 mm Hg). The pulmonary artery pressure was measured in 31 of the 61 patients. Twenty of these 31 patients (65%) had a pulmonary artery systolic pressure of 50 mm Hg or greater. The amount of mitral insufficiency was calculated as follows. The cardiac output was determined by means of the Fick principle and the degree of mitral insufficiency was determined after calculating the left ventricular output by radiographic technique. If the blood ejected into the left atrium during systole equaled the blood ejected into the aorta, the mitral insufficiency was graded as 2/6; if the blood ejected into the left atrium was two times the volume ejected into the aorta, this was graded as 4/6; and if the amount ejected into the left atrium was triple that ejected into the aorta, this was graded as 6/6. One of the patients had Grade 1/6 mitral insufficiency, 12 patients had Grade 2/6, 27 had Grade 3/6, 19 had Grade 4/6, one had Grade 5/6, and in one patient the amount of insufficiency was not calculated. Nine patients had a preoperative ejection fraction of 0.1 to 0.2, 20 had a preoperative ejection fraction of 0.25 to 0.40, and 32 had a preoperative ejection fraction of 0.45 to 0.70.
0022-5223/80/010012+07$00.70/0 © 1980 The C. V. Mosby Co.
Volume 79
Mitral insufficiency secondary to coronary disease
Number 1 1980
January.
I3
Ant leailel An nulus
-:1""-::
.'
1s t.
,
sul ur' g
,
Mu;:' a l
l e Zlo fl e t
~.
A
B
Fig. I. In order to obliterate the insufficiency, the greatly dilated. sagging mural anulus is decreased considerably in size . This can be accomplished by decreasing the length of the mural anulus to about one fourth and thereby bringing it up into apposition with the anterior leaflet. The anulus of the anterior leaflet must not be encroached upon . The repair is accomplished with No.2 silk sutures as illustrated.
Operative procedure At the time of operation, 23 patients had a dilated mitral anulus only; however, 38 patients also had a ruptured or infarcted papillary muscle or torn chordae tendineae as well as the dilated anulus. All patients with mitral insufficiency have a dilated mitral anulus. The dilatation always is located at the area of the mural anulus and never at the area of the anulus of the anterior leaflet. The fibrous trigone prevents stretching of the anulus of the anterior leaflet. Although mitral valve repair was deemed to be superior to replacement, in nine patients replacement was performed. Replacement was performed when the surgeon felt repair was not possible or early in the series, when we did not realize that the mitral valve that was insufficiency secondary to infarction could be repaired . Not being able to repair the valve was not related to ejection fraction . Five patients had replacement with a Kay-Shiley disc valve with muscle guard and four had replacement with a Hancock glutaraldehyde-treated porcine valve . Mitral valve repair was performed as follows. Careful search was made to determine if there were torn chordae tendineae or torn papillary muscle. If the chordae tendineae and papillary muscles were intact, then the repair consisted of decreasing the anulus of the mitral valve by doing away with approximately 70% to 75% of the anulus of the mural leaflet. The anulus was decreased mainly at the posteromedial commissural area, where two or three interrupted figure-of-eight sutures of No.2 silk were placed to do away with 45% to 50% of the anulus. Sutures were also placed to decrease the mural anulus of the anterolateral commissural area
by another 20% to 25%. The sutures at the anterolateral commissural area were placed superficially enough so as not to encircle the circumflex coronary artery. All the sutures were tied and only 25% to 30% of the mural anulus remained . There was then a snug two-fingerbreadth opening (Fig. I). If on opening the left atrium there were torn chordae tendineae or torn papillary muscle, this area of the valve was sutured down to the nearest papillary muscle head (Fig . 2). In dealing with a ruptured head or chordae tendineae of the anterior leaflet, the surgeon was careful to be sure that there was adequate anterior leaflet to billow against the mural leaflet in correcting the insufficiency. This was not a problem with suturing the mural leaflet to papillary muscle when there were torn chordae tendineae or papillary muscle of the mural leaflet, since most of the anulus of the mural leaflet was done away with by the repair . After the torn chordae tendineae or torn papillary muscle is repaired, the mural anulus is decreased to 25% to 30% of its original length, since there is always annular dilatation secondary to torn chordae tendineae or torn papillary muscle. Prior to the start of bypass, the proximal ends of the vein grafts are sutured into the aorta. After the start of bypass, the distal ends of the grafts are sutured into the coronary arteries with the aorta cross-clamped. The cross-clamp is released for 5 to 10 minutes and the heart is allowed to beat, the cross-clamp is again applied, and the mitral operation is performed . Twenty-six of the patients had one vein graft inserted, 20 had two vein grafts, 10 had three vein grafts, two had four vein grafts , and one had five vein grafts .
The Journal of
14
Thoracic and Cardiovascular Surgery
Kay et al.
, To',.n he""d of \ pZLP\1lZ5.r'y mUB. to '" mur 5 1 leaf l e t
A
B
~.
Fig. 2. The area of the leaflet involved with tom chordae tendineae or tom papillary muscle is repaired by placing interrupted figure-of-eight sutures of 2-0 Tevdek through the head of the nearer papillary muscle and then bringing this suture up through the leading edge of the involved area of the leaflet. Usually two separate sutures are used at each area . For ease of exposure. all sutures are placed prior to tying .
Three patients also had anastomosis of the left internal mammary artery to the left anterior descending coronary artery.
Results There were two hospital deaths in the nine patients with replacement and myocardial revascularization and three hospital deaths in the 52 patients undergoing repair and myocardial revascularization. The five deaths were as follows: One patient with an ejection fraction of 0.2 developed a bacteremia and died of a cardiac arrest 37 days after mitral repair. Another patient with disc valve replacement and an ejection fraction of 0.2 died of an embolus to the left anterior descending coronary artery 4 days after operation. One patient with a normal ejection fraction and with mitral replacement with a Kay-Shiley disc valve died of a myocardial infarct 9 days following operation. A patient with an ejection fraction of 0.4 and mitral repair died of a pulmonary embolus 10 days following operation. The fifth hospital death was in a patient with an ejection fraction of 0.4 who entered the hospital in cardiogenic shock and renal failure. This patient was operated upon as an emergency and myocardial revascularization and repair were performed. She continued to do poorly after operation and died on the fourteenth postoperative day. There were I I late deaths (Table I). Table II reveals the hospital and late mortality rates as related to ejection fraction. The lower the ejection fraction, the higher the
hospital mortality rate and the less likelihood of longterm survival. There was one peripheral embolic event resulting in hemiplegia in the 52 patients with repair followed for 229 patient-years, or 0.5% per patient year . One other patient with repair died from a cerebral hemorrhage 7 months after operation. This patient was continued on sodium warfarin by her family physician for this period of time, and at the time of the cerebral hemorrhage the prothrombin concentration was 10% of normal . Anticoagulant therapy is advised for only 3 months following mitral repair with the prothrombin concentration maintained at 20%. There were nine patients with mitral replacement. Five had disc valve replacement and four had replacement with the Hancock glutaraldehyde-treated porcine valve . One hospital death was due to an embolus to the left anterior descending coronary artery and another was due to myocardial infarction in the five patients with a disc valve . There was one late death owing to a cerebral embolus 27 months after operation following replacement with a disc valve. In the four patients with a porcine valve there was one cerebral embolus, which caused the death of the patient I I months following operation. Table III reveals the postoperative change in New York Heart Association Classification in the 45 living patients 15 months or longer following operation. An attempt was made to have all patients restudied post-
Volume 79 Number 1
I5
Mitral insufficiency secondary to coronary disease
January, 1980
Table I. Late deaths and cause of death E.F.
No. of patients
0.25 0.60 0.18 0.46 0.40 0.35 0.68 0.15 0.15 0.25 0.25
Cause of death
Repair or replacement
Cerebral emboli Cerebral hemorrhage Cardiac arrest Pulmonary emboli Pulmonary emboli Mediastinitis Unknown Cerebral emboli Unknown, diedinsleep Unknown Unknown
Replacement - disc Repair Repair Repair Repair Repair Repair Replacement - porcine Repair Repair Repair
Postop time (mo) 27 7 8 3 5 3 18 11 11
23 43
Legend: E.F., Ejection fraction.
operatively, but if the patient was doing well, neither the patient nor the family physician wanted the patient restudied. For this reason, only 12 patients were restudied 5 months to 3 years after operation. Of 25 grafts inserted in these 12 patients, 20 (80%) were patent. The mitral regurgitation had decreased from Grade 3/6 (mean) to Grade 0.8/6 (mean) (p < 0.001), and the ejection fraction had increased from 0.40 (mean) preoperatively to 0.48 (mean) postoperatively (p < 0.01) in the 12 patients having postoperative studies. Fig. 3 represents the survivability for those patients at 7 years with replacement (33%), repair (81%), and the entire group (74%).
Discussion There are very little data in the literature regarding the operative treatment of mitral insufficiency secondary to coronary artery disease. Buch and associates! reported on a group of 23 patients who had mitral valve disease with coronary artery disease. Eleven had coronary artery disease and nonrheumatic mitral regurgitation. Six of these 23 patients (26%) died within the first 30 days following replacement of the patient's mitral valve with a porcine valve. Three more patients died after discharge from the hospital. There were therefore nine deaths in this group of 23 patients with mitral replacement and coronary artery disease. Fourteen of these 23 patients had aorta-coronary bypass grafts. In a 1977 report on 243 patients with isolated replacement of the mitral valve with the Hancock glutaraldehydepreserved porcine aortic valve xenografts, Stinson and colleagues stated: MVR [mitral valve replacement] patients with associated coronary artery disease accounted for a substantial portion of both early and late deaths. For example, in 54 patients the diagnosis of coronary artery disease was established by coronary angiog-
Table II. Hospital and late deaths by ejection fraction Ejection fraction 0./-0.20
Total patients Hospital deaths Latedeaths Total deaths
0.45-0.70
9 2(22%) 3 5(56%)
32 1(3%) 3 4(13%)
20 2(10%) 5 7(35%)
Table III. New York Heart Association Functional Classification: Follow-up of 45 patients
_I From (18 patients) IV (27
patients)
III
To IV
III
1I
2
8
I
10
/
8 16
II I
raphy, electrocardiographic evidence of previous myocardial infarction, or postmortem examination within one year after operation. Operative mortality rate in this group was 18.5 per cent, and the survival rate at 2 years was 56 (::±: 7.8) per cent. Patients undergoing simultaneous coronary artery bypass grafting for angiographically defined coronary artery lesions fared similarly. * Among our 61 patients operated upon for mitral insufficiency secondary to coronary artery disease, there were nine patients who had the mitral valve replaced along with revascularization, with two hospital deaths and two deaths following discharge from the hospital. Among the 52 patients with mitral repair and rev as*Stinson EB, Griepp RB, OyerPE, Shumway NE: Long-term experience with porcine aortic valve xenografts. J THoRAc CARDtOVASC SURG
73:54-63, 1977.
16
The Journal of Thoracic and Cardiovascular Surgery
Kay et al.
100 81% 74%
.c
as > >
...
33%
::l
fI)
o
20
MITRAL REPAIR & REVASC
52 PT. 181% 7 VRS,)
[] MITRAL REPLACEMENT & REVASC 9 PT. 133% 7 VRS,) • ENTIRE GROUP
61 PT. (74% 7 VRS.)
OL-_ _.l.-_ _. L -_ _.L...-_ _.L...-_ _L..-_----IL..-_----J
o
1
2 3 4 5 Time Interval (years)
6
7
Fig. 3. Survivability, percent at 7 years. cularization, there were three hospital deaths (6%) and nine late deaths. For our entire series of 61 patients, there was a survivability of 74% at 7 years. However, if one excludes the nine patients who had replacement and a survivability of only 33% at 7 years, there was a survivability of 81 % at 7 years for the 52 patients undergoing repair and revascularization. There is a striking improvement in survivability for those patients with a good ejection fraction prior to operation compared to those with a poor ejection fraction. Among the nine patients with a preoperative ejection fraction of 0.1 to 0.2, there were four (44%) long-term survivors, as compared to 13 of 20 (65%) long-term survivors among those patients with a preoperative ejection fraction of 0.25 to 0.40 and 28 of 32 (87%) long-term survivors among those patients with a preoperative ejection fraction of 0.45 to 0.70. In the 52 patients who underwent repair, there was only one case of peripheral embolus for 229 patientyears, or 0.5% per patient-year. This figure compares favorably with a reported incidence of 5.2% per patient-year as noted by Salomon and co-workers! in 1976 with the Hancock glutaraldehyde-treated porcine valve. Whenever mitral valve operation is required for mitral insufficiency, it would appear advisable to repair the mitral valve if possible." Repair is more difficult than replacement; however, the morbidity and longterm mortality would appear less with repair than replacement. REFERENCES Buch WS, Pipkin RO, Hancock WO, Fogarty TJ: Mitral valve replacement with the Hancock stabilized glutaralde-
hyde valve. Clinical and laboratory evaluation. Arch Surg 110:1408-1415, 1975 2 Salomon NW, Stinson EB, Griepp RB. Shumway NE: Surgical treatment of degenerative mitral regurgitation. Am J Cardiol 38:463-468, 1976 3 Kay JH, Zubiate P, Mendez MA, Vanstrom N, Yokoyama T: Mitral valve repair for significant mitral insufficiency. Am Heart J 96:253-262, 1978
Discussion DR. JOHN W. KIRKLIN Birmingham. Ala.
I pay a good deal of attention to what Dr. Kay says because I remember 5, 6, 7 years ago he was reporting very good results in coronary artery bypass grafting in patients with poor left ventricular function, and many of us were skeptical. Today his position in that regard has been fully justified, and he simply understood how to do that before we did. For my colleagues and me, the experience he has discussed today is also quite remarkable because we have felt that we nearly always had to replace the mitral valve in patients whose mitral incompetence was secondary to ischemic heart disease. I hope the discussers will let us know their thoughts about this matter. May I ask Dr. Kay a couple of questions. First, do you anticoagulate the patients after the annuloplastyand, if so, for how long? Second, do you have any information about the regression of mitral incompetence after satisfactory coronary artery bypass grafting in patients whose valve was not altered? Is that any problem in interpreting the results of annuloplasty? Finally, could you tell us something about residual mitral incompetence in these patients? DR. M. LAXMAN KAMATH Milwaukee. Wis,
Over the years we have seen many cases of coronaryartery disease with associated mitral insufficiency. Our preliminary
Volume 79 Number 1
January, 1980
observation is that such patients can be categorized into four groups. In the first group, the mitral insufficiency associated with angina can mean insufficiency secondary to poor myocardial blood flow. In such patients, total myocardial revascularization alone improves the mitral insufficiency. In this group of patients, we have seen a left atrial "V" wave as high as 55 mm Hg that disappeared completely after revascularization. Patients likely to have this kind of response are those with severe angina. The left ventricular contractions are generally excellent, and they do not have any scarring of myocardium during the operation. The second group of patients has a combination of mitral insufficiency with a massive scar along the posterolateral wall of the left ventricle. With this group, mitral valve repair or replacement is usually necessary to improve mitral insufficiency. The third group of patients are those with aneurysm of the left ventricle. When the aneurysm involves the inferior wall, during closure ofthe aneurysm, the closing sutures are passed from the lateral wall beneath the papillary muscles and through the base of the septum. When the papillary muscles are scarred, the sutures are passed through the papillary muscles and the aneurysm is closed in such a way as to pull the papillary muscles toward the apex. In such patients, we have never replaced or repaired the mitral valve for insufficiency. The patients in the fourth group are those with an anterolateral aneurysm of the left ventricle. In these patients, again, the aneurysm is closed by pulling the base of the anterior papillary muscles toward the apex. Unlike results with posterior wall aneurysm, we do not have uniform success with anterior wall aneurysm in relieving the mitral insufficiency. The majority of these patients must have replacement of the mitral valve. I would like to ask the authors to describe their experience in patients with mitral insufficiency associated with aneurysm of the left ventricle. DR. J. A. MIJANGOS Rochester, N. Y.
I would like to describe our experience with a special group of patients-those with a large scar secondary to a myocardial infarction either in the anterior or diaphragmatic wall of the left ventricle or with multiple infarctions that have produced fibrosis or rupture of the posterior papillary muscle. On these patients, our results with plastic repairs of the mitral valve have not been as good as the ones described by Dr. Kay. We believe that the complete approach to these patients should consist of three points: (I) myocardial revascularization, (2) correction of the ventricular geometry, and (3) mitral valve replacement in order to correct mitral incompetence. [Slide] In the first slide you can see the description of some of our patients with the important catheterization data. Some of these patients have a cardiac index and ejection fraction that does not reflect reality because they were catheterized during intra-aortic balloon counterpulsation or during the administration of intravenous nitroglycerin or dopamine. Our
Mitral insufficiency secondary to coronary disease
I7
series, to date, comprises approximately 20 patients. Both the ejection fractions and the cardiac indices are fairly low. In these patients, our approach to mitral valve replacement has been a previously described ventricular approach. We believe that this approach greatly simplifies the operation. The exposure through the ventricle is usually excellent, the anulus is well demarcated, the aortic valve and septum can easily be exposed, and injury to these structures can be avoided. In this group of patients, it is usually difficult to expose the mitral valve through a fairly small atrium. [Slide] This is an exposure of the mitral ring through an incision on an anteroseptal left ventricular aneurysm. The mitral valves were replaced with a Bjork-Shiley prosthesis. [Slide] This is the exposure of an anulus after a diaphragmatic infarction had been opened and resected, the valve excised, and the stitches placed on the anulus. In our series of 20 cases, the mortality rate is approximately 0% to date. The complications incurred are heart block in one patient, cardiac arrhythmia (supraventricular arrhythmia) in one patient, and frequent premature ventricular contractions in another patient, easily controlled with procainamide (Pronestyl). All of our patients were considered to be Cardiac Class IV preoperatively. Postoperatively, 75% of our patients were in Cardiac Class I and 30% in Cardiac Class II. DR. KAY (Closing) In answer to Dr. Kirklin's questions, following mitral repair for mitral insufficiency, a patient receives anticoagulant therapy for 3 months. With our 20 years' experience with repairing the mitral valve, we have noted that if the patients are not treated with an anticoagulant, they may have peripheral emboli. The anticoagulant need not be continued after 3 months because by then the areas of repair and incision are healed. In answer to Dr. Kirklin's second question: Would the mitral insufficiency regress in our patients if only revascularization were performed? If the patient has a relatively good ejection fraction and the mitral insufficiency is Grade 1/6 to 2/6, the insufficiency is not corrected at the time of operation. With a poor ejection fraction, such as 0.1 or 0.2, Grade 1/6 to 2/6 mitral insufficiency is extremely important and the valve must be repaired, or replaced if repair is not possible. We see many patients with what we consider minimal mitral insufficiency, and we do nothing about the mitral insufficiency at the time of revascularization because we are concerned that we may not be able to repair the valve and then it would have to be replaced. The decision is made prior to operation when the left ventriculogram is reviewed and the ejection fraction is determined. I would say that for every patient on whom we repair or replace the mitral valve, another nine patients are deemed to have too little insufficiency. Of our 61 patients reported today, 38 had tom chordae tendineae or ruptured papillary muscle, and therefore the mitral insufficiency was certainly quite significant for these patients. It was also significant in 21 of the remaining 23 patients in this series. As to residual mitral incompetence in these patients, there
18
Kay et al.
are three patients who still have significant mitral insufficiency, and they are the 3 patients of the 45 living patients who are in Class III. Most of the remaining 42 patients do not have residual murmurs. The patient with a residual murmur usually has a short systolic murmur. In these patients the heart is smaller than it was preoperatively.
The Journal of Thoracic and Cardiovascular Surgery
Regarding the treatment of an aneurysm associated with mitral insufficiency, if we had a patient with severe mitral insufficiency associated with a ventricular aneurysm, the mitral valve would be repaired and the aneurysm resected. We leave small aneurysms alone. We feel that small aneurysms should not be resected at the time of revascularization.
Information for authors Most of the provisions of the Copyright Act of 1976 became effective on January I, 1978. Therefore, all manuscripts must be accompanied by the following written statement, signed by one author: "The undersigned author transfers all copyright ownership of the manuscript (title of article) to The C. V. Mosby Company in the event the work is published. The undersigned author warrants that the article is original, is not under consideration by another journal, and has not been previously published. I sign for and accept responsibility for releasing this material on behalf of any and all co-authors." Authors will be consulted, when possible, regarding republication of their material.