Pathogenesis of ischemic mitral insufficiency

Pathogenesis of ischemic mitral insufficiency We developed a new animal model of ischemic mitral insufficiency in sheep and used it to test the hypothesis that the combination of posterior papillary muscle infarction and left ventricular dilatation was required to produce mitral regurgitation after acute inferior myocardial infarction of moderate size. In 12 sheep, ligation of the first two circumflex marginal coronary arteries infarcted 23 % of the left ventricular mass, increased left ventricular cavitary area from 13.2 ± 1.2 cm2 to 20.0 ± 2.7 cm2 by 8 weeks and did not produce ischemic mitral regurgitation. In 13 sheep, ligation of the second and third circumflex marginal arteries infarcted 21 % of the left ventricular mass and, in 11 of these sheep, the posterior papillary muscular mass as well. When the papillary muscle was included, this infarction produced progressively severe mitral regurgitation over 8 weeks, as left ventricular cavitary area increased from 12.5 ± 2.6 cm2 to 22.8 ± 3.8 cm2• We conclude that neither posterior papillary muscle infarction nor left ventricular dilatation alone produces ischemic mitral regurgitation after moderatesized inferior wall infarction, but that the combination does. (J THORAC CARDIOVASC SURe 1993;105: 439-43)

Mario R. L1aneras, MD (by invitation), Michael L. Nance, MD (by invitation), James T. Streicher, MSE (by invitation), Philip L. Linden, MD (by invitation), Stephen W. Downing, MD (by invitation), Joao A. C. Lima, MD (by invitation), Radu Deac, MD, and L. Henry Edmunds, Jr., MD, Philadelphia. Pa.

ApproXimatelY 1.5 million Americans have an acute myocardial infarction each year, and 19% experience acute ischemic mitral regurgitation (MR); 16% of these patients experience progressive, moderate or severe (MR).1,2 In the absence of papillary muscle or chordal rupture, the mechanism by which severe MR develops is unclear. Acute infarction may produce papillary muscle dysfunction with or without change in chordal length, left ventricular (LV) enlargement, or both. The relative importance of each of these changes in the development of ischemic MR is poorly understood; consequently, surgical reparative procedures are based on empiric methods. This study introduces a new large animal model of

From the Division of Cardiothoracic Surgery, Harrison Department of Surgery, Hospital of the University of Pennsylvania, Philadelphia, Pa. Supported by grant HL 36308 from the National Heart Lung and Blood Institute, National Institutes of Health. Read at the Seventy-second Annual Meeting of The American Association forThoracic Surgery, LosAngeles, Calif., April 26-29, 1992. Address forreprints: L. HenryEdmunds, Jr., MD, Department ofSurgery,4Silverstein, Hospital ofthe University ofPennsylvania, 3400 Spruce St., Philadelphia, PA 19104. Copyright 1993 by Mosby-Year Book, Inc. 0022-5223/93 $1.00 + .10 12/6/42669

ischemic MR. We used this model to test the hypothesis that both posterior papillary muscle (PPM) infarction and LV dilatation are necessary to cause progressive ischemic MR.

Materials and methods Anesthesia was induced intravenously (IV) with thiopentothal sodium (0.5 ml/kg) in 28 Dorsett sheep (30 to 40 kg, Ovine Biotechnologies Inc., N.J.); they were intubated and their lungs were ventilated at IS ml/kg with a 2% isofluorane and oxygen mixture. All animals were studied in compliance with the "Principles of Laboratory Animal Care" formulated by the National Society for Medical Research and the "Guide for the Care and Use of Laboratory Animals" published by the National Institutes of Health (NIH Publication No. 86-23, revised 1985). All animals received one dose of glycopyrolate (0.4 mg IV) and prophylactic chloramphenicol (I gm IV) I hour before incision and one dose of antibiotic postoperatively. The surface electrocardiogram and arterial pressure were monitored. Each sheep received procainamide hydrochloride (15 rng/kg IV) and lidocaine (3 rug/kg IV followed by 2 mg/rnin) 20 minutes before infarction. Either the first and second obtuse marginal branches (OM I,OM2) of the left circumflex artery or the second and third (OM3) were ligated through a left thoracotomy. Ligation of OM I and OM2 caused an infarction of approximately 23% of the LV mass but did not cause an infarction ofthePPM (group I, n = 12). Ligation ofOM2 and OM3 caused an infarction of 21% of the LV and completely infarcted the PPM (group 2, n = 16) (Fig. I). In both groups, transepicardial color flow Doppler and two-

439

440

The Journal of Thoracic and Cardiovascular Surgery

Llaneras et al.

March 1993

A

B

Fig. 1. Five-millimeter LV transverse slices after infusion of triphenyltetrazolium chloride, 3 hours after ligation of aMI and OM2 (A) and OM2 and OM3 (B) coronary arteries. Infarcted myocardium is unstained. PPM is infarcted only after OM2 and OM3 ligations. APM, Anterior papillary muscle.

Fig. 2. Color flow Doppler echocardiogram 8 weeks after ligation ofOM2 and OM3 coronary arteries. Regurgitant jet (MR JET, arrow) reaches and reverses flow in the pulmonary vein (PV). LV, Left ventricle; LA, left atrium; MV, mitral valve.

dimensional echocardiograms were obtained before infarction, 30 minutes after infarction, and during a second thoracotomy 8 weeks later. Both apical four-chamber views and crosssectional images at the high papillary muscle level were taken with a 5 MHz transducer (model SH-140, Toshiba America Medical Instruments, Tustin, Calif.). Images were recorded on videotape with a 1.27 em VHS recorder (Panasonic model NV-8950, Matsushita Electric, Osaka, Japan). After infarction, the thoracotomy was closed, and the animals were returned to an animal care facility for 8 weeks. In group 2 animals, left ventriculograms were obtained in the right anterior oblique projection before infarction, 30 minutes

after infarction, and 1,4, 6, and 8 weeks later. A 5.2F pigtail catheter was passed across the aortic valve from a sterile left or right carotid arterial cutdown. For the ventriculogram 45 ml of contrast material (Hypaque-76, Winthrop Pharmaceuticals, N.Y.) was injected over 5 seconds (at 40 PSI) with a power injector that was connected to a manifold. Fluoroscopic images (Flurocon; General Electric CGR USA, Milwaukee, Wis.) were recorded onto videotape with a 1.27 ern VHS recorder (Matsushita Electric). At the end of 8 weeks, each animal from both groups had a second left thoracotomy while under general anesthesia. After the data were recorded, the animals were killed with thiopen-

The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 3

Llaneras et al.

44 1

2 4+ D E

N=l1

G 3+ R E E 2+ 0

f

M R

1+

11 0

PRE

POST

1 WK 4WKS TIME

6WKS

8WKS

Fig. 3. Progressiveincrease in severityof mitral insufficiency from analysisof left ventriculogramsin II of 13sheep after ligation of OM2 and OM3 coronary arteries. tothal sodium(0.5 ml/kg IV) and potassiumchloride(80 mEq). Hearts were quickly excised,subvalvular structures were carefully inspected,and wall thicknesseswere measured at the high papillary muscle level. Data analysis. Any MR that was noted on color flow Doppler echocardiography was graded according to the following scale. If an MR jet was present, it wasgraded I+; if the jet went halfwayback to the posterioratrial wall,it wasgraded 2+; ifthe jet hit the posterior atrial wall, it was 3+; if it reversedflow in a pulmonary vein, it was graded 4+. 3 The two-dimensional short-axis echocardiographic images were digitized off-lineat 30 Hz with a video frame grabber (Computer Eyes RT, Dedham Mass.). Epicardial and endocardial contours from end-diastolic and end-systolic images were traced with proprietary imageanalysissoftware on a UNIX workstation (Sparcstation 2,SUN Microsystems, Mountainview,Calif.). End-diastolewas defined as the image coincidentwith the Q waveon the electrocardiogram. End-systole was defined as the echocardiographic framewiththe smallestvolume."We calculated the LV cavitary area from these tracings, in both groups, before infarction, 30 minutes after infarction, and 8 weeks after infarction. Ventriculograms from the group 2 animals with infarcted PPM werereviewed by twoexperts,and the amount of MR was graded according to the degree of left atrial opacification." Results Demonstration of ischemic MR

Group 1: Viable PPM. Two of 12 sheep died before 8 weeks and were excluded from further analysis. None of the 12 sheep had spontaneous MR before infarction, and none had significant MR 30 minutes after infarction. At 8 weeks, two sheep exhibited 1+ (mild) MR; the remainder had no MR. Group 2: Infarcted PPM. None of the 16 sheep had MR before infarction. Three of the 16 sheep had 4+ MR

and died in the early postinfarction period as a result of ligation of a prominent terminal branch of the left circumflex artery and OM2 and OM3; this caused a much larger (greater than 30%) infarction. These animals were excluded. Eleven of 13 sheep had 1+ MR, as determined by both color Doppler and left ventriculography 30 minutes after infarction. At 8 weeks, color flow Doppler imaging showed 3+ MR in 9 of 13 sheep and 4+ MR in 2 of 13 sheep (Fig. 2). Serial left ventriculography (performed in group 2 sheep only) showed increasing MR duringthe8 weeks after infarction (Fig. 3). Two sheep did not experience MR 30 minutes after infarction or at 8 weeks. In these two animals, two-dimensional echocardiography showed marked systolic thickening of the PPM, which indicated a viable papillary muscle. Postmortem examination confirmed that the majority of the PPM in these two sheep was not infarcted. Demonstration of LV dilatation. Two-dimensional short-axis echocardiographic images were used to assess LV dilatation in both groups. LV end-diastolic and endsystolic cavitary areas were compared before infarction, 30 minutes after infarction, and 8 weeks later. LV enddiastolic cavitary area in group 1 increased from 13.2 ± 1.2 ern- to 14.7 ± 2.1 ern? after 30 minutes (p = NS*) and to 20.0 ± 2.7 ern- at 8 weeks (p < 0.001). In group 2 animals, end-diastolic cavitary area increased from 12.5 ± 2.6 ern- to 13.8 ± 3.5 em- by 30 minutes (p = NS) and to 22.8 ± 3.8 em- at 8 weeks (p < 0.001). *NS

= Not significant.

442

The Journal of Thoracic and Cardiovascular Surgery March 1993

Llaneras et al.

Discussion Selective ligation of the OM2 and OM3 branches of the ovine circumflex coronary artery produces a moderate-sized myocardial infarction, which usually includes the PPM. Over time, this infarction produces significant MR and mimics the clinical course of many patients who experience chronic postinfarction MR. This sheep model is superior to acute models in which leaflets are perforated'' or chordae are cut 7 and is superior to previously published canine models, which are not based on the pathophysiology of ischemic MR. 8- 10 Coronary arterial anatomy is quite consistent in sheep, and collateral vessels do not develop to any significant degree in response to ischemia in the sheep. Thus the model is reproducible, predictable, efficient in terms of low attrition, and similar to chronic ischemic MR in patients. End-diastolic LV cavitary area was not significantly increased in either group 30 minutes after infarction, and no sheep had appreciable MR. Myocardial ischemia stops contraction within 5 minutes; therefore the PPM did not thicken or contract during echocardiograms that were obtained at 30 minutes in 11 of 13 group 2 animals. After a moderate infarct, papillary muscle dysfunction alone did not cause MR. However, over the course of 8 weeks, LV end-diastolic cavitary area increased 55% and MR occurred. End-diastolic cavitary area also increased in group 1 animals, but MR did not develop. The PPM did scar and atrophy during the 8 weeks of follow-up. This likely contributed to the development of MR. It is possible that papillary muscle fibrosis and atrophy alone can cause MR without concomitant LV dilatation. Clinically, however, a small selective infarction of the PPM is unlikely, and therefore this speculation is academic. From these studies, we conclude that the combination of papillary muscle infarction and LV dilatation is necessary to produce MR after moderate-sized (20% to 25%) infarctions. REFERENCES 1. Hickey MS, Smith LR, Muhlbaier LH, et ai. Current prognosis of ischemic mitral regurgitation: implications for future management. Circulation 1988;78(Pt 2):151-9. 2. Gahl K, Sutton R, Pearson M, Caspau P, Lairet A, McDonald L. Mitral regurgitation in coronary heart disease. Br Heart J 1977;39: 13-8. 3. Helmcke F, Nanda N, Hsiung M, et al. Color Doppler assessment of mitral regurgitation with orthogonal planes. Circulation 1987;75:175-83. 4. Lima JAC, Becker LC, Melin JA, et al. Impaired thickening of nonischemic myocardium during acute regional ischemia in the dog. Circulation 1985;71:1048-59. 5. Grossman WS. Profiles in valvular heart disease. In: Grossman WS, Baim DS, eds. Cardiac catheterization,

6.

7.

8.

9.

10.

angiography and intervention. 4th ed. Philadelphia: Lea & Febiger, 1991 :565-6. Hoit BD, Jones M, Eidbo EE, Elias W, Sahn DJ. Sources of variability for Doppler color flow mapping of regurgitant jets in an animal model of mitral regurgitation. J Am Coli CardioI1989;13:1631-6. Kleaveland JP, Kussmaul WG, Vinciguerra T, Diters R, Carrabello BA. Volume overload hypertrophy in a close chested model of mitral regurgitation. Am J Physiol 1988;254:H I034-41. Miller GE, Cohn KE, Kerth W J, Selzer A, Gerbode F. Experimental papillary muscle infarction. J THORAC CARD10VASC SURG 1968;56:611-6. Mittal AK, Langston M, Cohn KE, Selzer A, Kerth WJ. Combined papillary muscle and left ventricular wall dysfunction as a cause of mitral regurgitation: an experimental study. Circulation 1971;44: 174-80. Tsakiris AG, Rastelli GC, Amorim Ds, Titus JL, Wood EH. Effect of experimental papillary muscle damage on mitral valve closure in intact anesthetized dogs. Mayo Clin Proc 1970;45:275-85.

Discussion Dr. Alain F. Carpentier (Paris. France). In the past, we conducted a similar study, but it was not documented as well. We tried to obtain MR by infiltrating the PPM with formaldehyde. Mitral valve incompetence could never be produced when only the papillary muscle was infiltrated. To get mitral valve incompetence, it was also necessary to infiltrate the myocardial wall that supports the papillary muscle. This finding confirms what you yourself have found. However, I would like to point out that there are several mechanisms involved in ischemic mitral valve insufficiency, not only because of PPM dysfunction (type III mitral valve insufficiency) but also because of pure dilatation of the anulus (type I) or leaflet prolapse (type II). My question is this: Would it be possible for you to reproduce these different mechanisms by different types of ligation of various coronary arteries? It would be extremely useful to have an experimental model to duplicate what can be done in surgery. In summary, I wanted to bring attention to the fact that, in clinical practice, an ischemic mitral valve insufficiency may result from different mechanisms, as assessed by echocardiography. Can the fine experimental model presented by the authors reproduce not only type III mitral valve incompetence (that is, restricted leaflet motion) but also type II (leaflet prolapse) and type 1 (normal leaflet motion) just by ligation of different coronary artery branches? This would be important because each functional type corresponds to a different type of repair. Dr. Llaneras, I think that your conclusions are consistent with our data, but I also believe our conclusions are correct. I suspect that if we had measured two-dimensional long-axis echocardiograms we would have found that the distance between the papillary muscle and the anulus had increased. However, we did not make that measurement, so I cannot answer. Dr. Robert W. Frater. (Bronxville. N. Y) Between diastole and systole, two things must happen for the mitral valve to close. The anulus must shorten and the posterior LV wall must short-

The Journal of Thoracic and Cardiovascular Surgery Volume 105, Number 3

en; this brings the papillary muscles, which have held the cusps open down in the cavity, back toward the anulus to allow the cusps to meet. If the posterior LV wall fails to shorten between diastole and systole, this alone will keep the cusps open down in the cavity of the ventricle. Volume overloading of a ventricle like this, or a ventricle with a small posterior infarction that involves the papillary muscle, will aggravate that tendency for the papillary muscles to fail to move toward the anulus and allow the cusps to come into apposition. Any surgeon who has had the experience of observing the ventricle and mitral valve with a transesophageal echocardiogram while changing the preload, afterload, and inotropy of the ventriclewould have seen how dynamic this phenomenon is with a posterior wall that is either infarcted or ischemic. The good news is that if you overcompensate with an annuloplasty and produce an end-systolic dimension that is smaller than normal, you can bring the cusps together even when the posterior left ventricular wall has failed to shorten, and the papillary muscle has failed to move toward the anulus to allow them to meet. You have concluded that a combination of papillary muscle infarction in addition to ventricular dilatation is a cause of MR. My alternative conclusion is that a ventricular infarction that affects the papillary muscle so that it fails to move toward the anulus or moves away from the anulus during systole is the major mechanism of ischemic MR. This conclusion encompasses the processes of dilatation as well as papillary muscle infarction. Dr. Llaneras, We have reviewed the anatomy thoroughly but do not believe it is possible to selectively produce annular dilatation in this model. However, we have produced global LV dilatation, without MR. Dilatation of the mitral anulus can produce MR, but our data show that that may not be as important as previously believed in the pathogenesis of ischemic MR.

Llaneras et al. 4 4 3

Dr. James H. Oury (Missoula, Mont.). This study documents, for the first time in a large animal model, a mechanism for consistently and accurately producing MR that is similar to what we see clinically. This clinical anomaly, which, as the authors correctly identify, accounts for between 10% and 20% of the patients who require bypass grafting in this country, is often overlooked or ignored and therefore lacks a consistent approach to treatment. With the use of preliminary data presented before the Association I year ago, we have initiated a prospective multicenter study to address the natural history effects of surgical intervention and the clinical course in this important subset of patients. Nine institutions will provide a worldwide base for the collection of data in this ongoing study. We propose a classification that, in very broad strokes, identifies the critical subset of patients with functional mitral insufficiency(which is clinically similar to the sheep model described by the authors) and focuses on this group, excluding those patients with obvious associated or organic mitral valve disease. What could the authors postulate as the actual pathophysiology for the progressive MR in the sheep model? Would an intraoperative, provocative test to elicit MR in the sheep model identify those animals in which progressive MR would develop? On the basis of that particular intraoperative test, would then annuloplasty tend to prevent the onset and progression of the annular dilatation that we see clinically? Dr. L1aneras. I believe that the pathophysiology of progressive ischemic MR is due to misalignment of chordae and papillary muscle because of ventricular dilatation and remodeling. Global LV dilatation alone does not produce MR, probably because the mitral anulus does not dilate sufficiently. As for the second question, we have not evaluated intraoperative, acute, provocative tests that may induce MR.