Premature closure of prosthetic mitral valves as a consequence of gravity

Catheterization and Cardiovascular Diagnosis 8:131-136 (1982)

Premature Closure of Prosthetic Mitral Valves as a Consequence of Gravity Ulrich W. Busch, MD, Leonard W. Pechacek, RDMS, Efrain Garcia, MD, Virendra S. Mathur, MD, and Robert J. Hall, MD To determine the influence of gravity on prematureclosure of prosthetic mitral valves, we studied 17 patients In whom this phenomenon had been observed during routine examination. Ail patients were in atrial fibrillation and none had aortic incompetence. Patients were studied In multiple positions by means of simultaneous echocardie graphy, phonocardiography, and cinefluoroscopy. i n all patients premature closure wss observed when the atrial side of the prosthesls was below the ventricular side, resulting in a downward motion of the occluder inside the valve cage. When patients were studied in positions in which the atrial side of the valve was higher than the ventricular side, premature closure never occurred, even during extremely prolonged diastolic periods. Since minor positional changes, which were found to determine whether premature closure occurred or not, are unlikely to produce significant alterations in pressure and flow across the mitral orifice during diastole, we conclude that position-dependent premature closure of prosthetlc mitral valves in patients with atrial fibrillation is best explained by the effect of gravlty on the prosthetic occluder. Examination of such patients in multiple positions should be helpful in distinguishing premature valve closure caused by aortic regurgitation from gravity-related presystolic closure. Inability to produce prematureclosure in patients in whom it had previously been demonstrated in the presenceof similar R-R intervals may even prove useful in diagnosing new oriflce obstruction.

Key words: cineradiography, echocardlography, mitral valve, phonocardiography

From the Clayton Foundation for ResearchCardiovascular Laboratories, St. Luke’s Episcopal Hospital and the Texas Heart Institute, Houston. Presented in part at the 28th Annual Scientific Sessions of the American College of Cardiology, Anaheim, California, March 7, 1978. Dr. Busch is now at the Klinikum rechts der Isar, Kardiologie, Ismaninger Str. 22, 8000 Miinchen 80, Federal Republic of Germany. Address reprint requests to Efrain Garcia, MD, Texas Heart Institute, P.O. Box 20269. Houston, TX 11025.

Received January 6, 1981, accepted August 10, 1981.

0098-6569/82/0802.0131$02.00 0 1982 Alan R. Liss, Inc.

132

Busch e l al

INTRODUCTION Premature closure of prosthetic mitral valves has been observed in patients with a variety of valve types. Related echocardiographic, phonocardiographic, and cinefluorographic findings also have been described [ 1-12]. In several cases, a causal relationship has been established, or suggested, between premature closure and concomitant abnormalities such as aortic regurgitation [4,10], atrial flutter [13], or the presence of first degree atrioventricular block [6,8,9,12]. Most frequently, however, premature closure is seen in patients with atrial fibrillation during prolonged diastolic periods in the absence of aortic regurgitation [ 11,12,14]. Here, the cause of premature closure thus far has not been clarified, although it has been suspected that gravity might be the driving force [5,11]. The present study was undertaken to determine the role of gravity in the occurrence of premature closure of normally functioning prosthetic valves in the presence of atrial fibrillation.

MATERIAL AND METHODS Seventeen patients underwent noninvasive evaluation of their prosthetic mitral valve. Twelve patients had a Cooley-Cutter mitral prosthesis containing a biconical, hollow pyrolitic carbon occluder; two, an early Cooley-Cutter model with a discoid pyrolitic occluder; two, a Starr-Edwards prosthesis with a silastic ball; and one, a Beall disc valve. In all prosthetic valve types examined, the specific gravity of the occluder exceeded that of blood (> 1.05 gm/cm3). All patients were in atrial fibrillation, and premature closure had been observed in all during routine echocardiographic examination. None had aortic incompetence and there was no evidence of prosthetic malfunction. To test the hypothesis that premature closure in these patients results from gravitational forces acting on the occluder, the patients were studied in various positions including supine, prone, sitting, as well as right and left decubitus positions of various degrees, thus allowing for different orientation of the prosthetic cage and occluder with respect to gravity. Patients were studied by means of simultaneous echocardiography, phonocardiography, and cinefluoroscopy. Optimal visualization of the poppet and its excursions were achieved with the x-ray beam directed parallel to the plane of the valve ring. Cine films were obtained at a speed of 60 frameshec. For simultaneous cinefluoroscopy and echocardiography, the ultrasound transducer was held in position by a specially designed, radiolucent holder [ 151. Superimposition of a QRS marker on the cine film and a cine frame marker on the echocardiogram permitted direct comparison of synchronously recorded events. Echocardiography of the prosthetic valve was performed with a 2.25 MHz, 13 mm diameter, focused transducer placed at the apex in a manner described previously [16]. A single-lead electrocardiogram along with a high frequency phonocardiogram were also obtained. Data were recorded at a paper speed of 100 mm/sec. Prosthetic valve closure was considered premature if partial or complete closure occurred prior to the onset of the QRS complex.

RESULTS Recordings of premature closure of the prosthetic valve were obtained in all patients. The interval between the preceding R wave and onset of premature closure

Prosthetic Valve Closure

133

ranged from 620msec to 1130 msec. Onset of premature closure did not correlate with the length of the preceding R-R interval. Most importantly, premature closure was dependent upon the spatial orientation of the valve. Premature closure could be observed when the atrial side of the prosthesis was lower than the ventricular side; for example, when the patient was supine or in a right decubitus position (Fig. 1). Premature closure was never observed with the patient prone, sitting upright or bent forward, or in the steep left decubitus position (during which the atrial side of the prosthesis was always higher than the ventricular side) even during extremely long diastolic periods (Fig. 2A). Stepwise rotation of the patient from supine to steep left decubitus or gradual elevation of the patient’s back to the sitting position always prevented premature closure when the atrial side of the prosthesis began to tilt upward. With horizontal orientation of the valve (ie, with the atrial and ventricular sides facing laterally) premature closure, although never complete, could also be observed. Floating and tilting of the occluder during premature closure contrasted with the usual rapid and uniform closure at onset of systole. The degree of premature closure was variable, as was the intensity of the closing sound of the prosthesis (Fig. 2B). There was no closing click at onset of systole with complete premature closure.

Fig. 1. Representative cineangiographicframe of a prosthetic mltral valve (Cooiey-Cutter) obtained with lateral camera. The patient is in a left decubitusposition with his head to the left of the photograph. Since the atrial side of the prosthesis is lower than the ventricular side with respect to gravity, early closure of the disc occurs during a prolonged diastolic period.

134

Buschet el

Fig. 2. A. Echocardiogram of a prosthetic mitrai valve (Cooky-Cutter) with the patient sitting upright. Premature closure does not occur, even during long diastolic periods. Note the loud closing sound of the prosthesls on the high frequency phonocardiogram. A cine frame marker appears at the bottom of the record. B. Echocardlographicrecording from the same patient but in a right decubitus position. With long R-R intervals the occiuder begins to close prematurely (PC) during late diastole. At the onset of subsequent systole complete closure results in a markedly diminished closlng sound (small arrows) when compared to usual closure.

Prosthetic Valve Closure

135

DISCUSSION Opening and closure of prosthetic mitral valves is normally attributed to reversal of pressure and flow across the mitral orifice during early diastole and early systole [3]. Therefore, the observed premature closure of prosthetic mitral valves in patients who have concomitant severe aortic regurgitation is readily explained by the rise of left ventricular diastolic pressure above left atrial pressure [4]. Similarly, premature prosthetic mitral closure that is occasionally seen in patients with first degree atrioventricular block [9,12] or atrial flutter [13] can possibly result from a fall of left atrial pressure below left ventricular end-diastolic pressure with atrial relaxation. Presystolic closure of prosthetic mitral valves in patients with atrial fibrillation and little or no aortic regurgitation, however, can hardly be attributed to the same mechanism. The observed dependence of premature closure upon valve position in our study can best be explained by the effect of gravity. The weight of the occluder promotes premature closure only when the atrial side of the prosthesis is below the ventricular side, provided of course that the occluder is heavier than blood, as in the valve models studied. In several of our patients with disc occluders, we observed instances of partial premature closure when the atrial and ventricular sides of the disc were in a horizontal orientation (ie, facing laterally). We speculate that such closure was also induced by gravity when the labile equilibrium of the occluder was disturbed. It should be noted that not all prosthetic valve types were examined in our study, namely tilting disc valves and those with occluders having a specific gravity less than or equal to blood. Preliminary in vitro studies of the St. Jude bileaflet prosthesis mounted in a mock circulatory system have provided additional evidence of gravity-related premature closure [17,18]. Using low mock ventricular rates with reduced ejection velocity to simulate the effects of atrial fibrillation, premature closure of whichever pyrolitic leaflet was oriented posteriorly (toward the direction of gravity) was consistently produced. The leaflet oriented away from the direction of gravity never closed prematurely. Since the weight and shape of occluders currently in use provide good-to-excellent response to flow of blood across the valve, it becomes clear that gravity-related premature closure will only occur when there is no, or insignificant, flow across the valve. From this it follows that factors which contribute to diastolic cessation of mitral flow, such as R-R prolongation (certainly the most frequent cause), low cardiac output or large valvular orifice size, will favor the occurrence of gravity-related premature closure. Conversely, factors that provide sufficient pandiastolic transvalvular flow will prevent premature closure, like shortening of the R-R interval, increased cardiac output, or small effective valvular orifice size (owing to either a small valve itself or secondary orifice obstruction by thrombus formation or fibrotic ingrowth). Our study shows that positiondependent premature closure of prosthetic mitral valves in patients with atrial fibrillation is a normal phenomenon produced by the effects of gravity. Absence of gravity-related premature closure in a patient in whom it had previously been demonstrated, in the presence of similar R-R intervals, may even be a helpful sign in diagnosing new orifice obstruction. We have recently observed such a case in a patient with atrial fibrillation found to have a 20 mm Hg gradient across his Cooley-Cutter mitral valve. Previously noted premature closure could not be induced even during R-R intervals greater than one second despite multiple changes

136

Buschet al

in patient positioning. Several reports have attributed premature closure of prosthetic mitral valves to the presence of aortic regurgitation [4,10,19]. Our study shows, however, that early closure of a mitral prosthesis is not specific for aortic regurgitation in the presence of atrial fibrillation. In such circumstances, examination of the patient in appropriate positions should prevent confusion of gravity-related premature closure with presystolic closure produced by abnormal hemodynamics. ACKNOWLEDGMENTS The authors are grateful to American Edwards Laboratories, to Medical Engineering Corporation, and t o Mr. Carlos Martin of the Texas Heart Institute for providing technical specifications of the various valve types examined. Mr. John Jurgens provided invaluable technical assistance. REFERENCES I. Hultgren HN, Hubis H: A phonocardiographic study of patients with the Starr-Edwards mitral valve prosthesis. Am Heart J 69:306-319, 1965 2. Boicourt OW, Bristow JD, Starr A, Griswold HE: A phonocardiographic study of patients with multiple Starr-Edwards prosthetic valves. Br Heart J 28531-538, 1966. 3. Winters WL Jr, Gimenez J, Soloff LA: Clinical application of ultrasound in the analysis of prosthetic ball valve function. Am J Cardiol 19:97-107,1%7. 4. Agnew TM, Carlisle R: Premature valve closure in patients with a mitral Starr-Edwards prosthesis and aortic incompetence. Br Heart J 32:436-439, 1970. 5. Siggers DS, Srivongse SA, Deuchar D: Analysis of dynamics of mitral Starr-Edwards valve prosthesis using reflected ultrasound. Br Heart J 33:401-408,1971. 6. Brown DF: Decreased intensity of closure sound in a normally functioning Starr-Edwards mitral valve prosthesis: Observations on presystolic mitral valve closure. Am J Cardiol 31:93-97, 1973. 7. Hamby RI, Aintablian A, Wisoff BG: Mechanism of closure of the mitral prosthetic valve and the role of atrial systole: Phonocardiographic and cinefluorographic study. Am J Cardiol31:616-622,1973. 8. Gibson TC, Starek PJK, Moos S, Craige E: Echocardiographic and phonocardiographic characteristics of the Lillehei-Kaster mitral valve prosthesis. Circulation 49:434-440, 1974. 9. Stefadouros MA, Smith WS, Fraser RC, Hitch W, Rubin JW: Firstdegree atrioventricular block: A cause of false malfunction of a mitral disc-valve prosthesis. J Thorac Cardiovasc Surg 69:776-780, 1975. 10. Sands MJ, Kreulen TH, McDonough MT, Fadali MA, Spann JF: Pseudomalfunction of a Beall mitral valve prosthesis in the presence of paravalvular aortic regurgitation. Am J Cardiol 36:88-90, 1975. 11. Berndt TB, Goodman DJ. Popp RL: Echocardiographic and phonocardiographic confirmation of suspected caged mitral valve malfunction. Chest 70:221-230, 1976. 12. Busch U, Pechacek LW, Garcia E, Hall RJ: Acoustic changes in normally functioning mitral valve prostheses: Echophonocardiographic observations. Cardiovasc Dis Bull Texas Heart Inst 5: 107-1 11, 1978. 13. Greenberg MA, Herman LS, Cohen MV: Mitral valve closure in atrial flutter. Circulation 59902-909, 1979. 14. Pechacek LW, Zarrabi A, Massumkhani A, Garcia E, De Castro CM Jr, Hall RJ: Echophonocardiographic characteristics of the normally functioning Cooky-Cutter mitral valve prosthesis. Cardiovasc Dis Bull Texas Heart Inst 4:390-399, 1979. 15. Pechacek LW, Busch U, Garcia E: Radiolucent holder for echocardiographic recordings during fluoroscopy and exercise. J Clin Ultrasound 6:355-356,1978. 16. Johnson ML, Holmes JH, Paton BC: Echocardiographic determination of mitral disc valve excursion. Circulation 47:1274-1280. 1973. 17. Pechacek LW, Montiel-Amoroso G , Edelman SK, Garcia E, de Castro C: Echophoncardiographic assessment of St. Jude Medical mitral prosthesis: In vitro and in vivo studies. Circulation 62(Suppl 111):132, 1980 (abstr). 18. Feldman H, Halpern S , Gray R, Shapira J, Kraus R, Allen H, Chaux A, Conklin C, Maltoff J: In vitro echocardiographic validation of clinical St. Jude mitral valve images. Clin Res 28:5A, 1980. 19. Castaneda-Zuniga N, Nicoloff D, Jorgensen C, Nath PH, Zollikofer C, Amplatz K: In vivo radiographic appearance of the St. Jude valve prosthesis. Radiology 134:775-776, 1980.