Aortic Valve Debridement by Ultrasonic Surgical Aspirator: A Word of Caution Joseph M. Craver, MD Division of Cardio-Thoracic Surgery, Joseph B. Whitehead Department of Surgery, Emory University School of Medicine, Atlanta, Georgia
Aortic stenosis was relieved in 11 patients by ultrasonic debridement of the valve and annulus, while 102 other patients underwent valve replacement for aortic stenosis during 1988. Debridement was selectively applied based on findings of small annulus size (19 mm or less) and extensive calcification. Additional patient characteristics were mean transvalvular gradient of 78 mm Hg, advanced age, and marked left ventricular hypertrophy. Six patients had no residual gradient and 5 others a mean gradient less than 10 mm Hg. There were no complications related to the debridement process. Intraoperative transesophageal Doppler echocardiography demonstrated improved leaflet mobility and elimination of the gradient in all patients and elimination of associated
M
echanical debridement of the aortic valve was applied as treatment to relieve aortic stenosis for many years before successful valve substitutes were available [l, 21. The recurrence of aortic stenosis in a high percentage of patients within 5 years as well as the emergence of successful prosthetic valve devices led to the abandonment of techniques of valve repair [%5].
For editorial comment, see page 689. Recently, valve debridement using ultrasonic surgical aspiration devices has been reported as an effective alternative to valve replacement [6-81. This report presents an experience using ultrasonic surgical debridement of the aortic valve in 11 patients to correct aortic stenosis and avoid valve replacement.
Material a n d Methods Eleven select patients out of a total of 113 patients undergoing operation for aortic stenosis underwent debridement of their aortic valve using ultrasonic surgical aspiration techniques. Nine of the 11 had severe aortic stenosis, with a mean gradient of 94 mm Hg (range, 80 to 128 mm Hg), and sclerotic trileaflet valves that were heavily calcified, with a small, calcified valve annulus (19 mm). Two patients were primarily undergoing coronary artery bypass grafting for ischemic symptoms but also had calcified Presented at the Thirty-sixth Annual Meeting of the Southern Thoracic Surgical Association, Scottsdale, AZ, Nov 9-11, 1989. Address reprint requests to Dr Craver, The Emory Clinic, 1365 Clifton Rd, NE, Atlanta, GA 30322.
0 1990 by The Society of Thoracic Surgeons
valvular insufficiency in 2 patients. Follow-up echocardiography demonstrated late onset of new valvular regurgitation in 5 patients that was progressive and required reoperation in 3. Thickened, hardened, and retracted valve leaflets with loss of central coaptation were found in all 3 patients who underwent reoperation. Ultrasonic debridement can effectively relieve aortic stenosis, provide an excellent immediate hemodynamic result, and decrease operative time. However, the early occurrence of aortic insufficiency in a high percentage of patients makes it an unacceptable alternative to valve replacement, and the technique should be abandoned as a treatment for severe calcific aortic stenosis. (Ann Thorac Surg 1990;49:746-53)
aortic valves with moderate stenosis (mean gradient, 40 mm Hg; range, 36 to 45 mm Hg). Intraoperative election to use the ultrasonic debridement technique was based on the desire to avoid implantation of a mechanical valve prosthesis in 9 elderly patients (mean age, 78 years) with severe aortic stenosis and a small calcified aortic root and annulus. Improvement in valve function without valve replacement was the indication in the 2 other patients with moderately stenotic valves who were primarily having coronary artery bypass grafting. Valve debridement as an alternative to valve replacement was discussed with all patients before operation. Ultrasonic debridement was accomplished with a handheld probe that applied high-speed vibration at 23 kHz, a frequency above the human aural range, to the calcified atherosclerotic valve leaflets and annulus. Ultrahighfrequency vibration impacts the harder elements of the sclerotic leaflet lesions and causes them to be dislodged from the softer tissues, which can absorb the vibration forces. Ultrasonic vibration fragments and disintegrates the calcific deposits, which are then aspirated from the remaining leaflet tissue. The ultrasonic vibrator probe was cooled by a continuous external irrigation of saline solution over the tip. Fragmented debris was aspirated from the operative field through the hollow core of the probe tip. Figure 1 demonstrates the debridement of a heavily calcified stenotic valve with resultant wide leaflet mobility and increase in orifice size. Debridement was performed using the Cavitron ultrasonic surgical aspirator (Cavitron Surgical Systems, Inc, Stamford, CT) in 9 patients and the Bovie ultrasonic surgical aspirator (MDT Castle, Inc, Rochester, NY) in 2 patients. 0003-4975/90/$3.50
Ann Thorac Surg 1990:49:74&53
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
747
Fig 1. (A) Calcified aortic valve. ( B ) Debridement by ultrasonic technique in process. (C) Leaflets after completion of debridement process; note widely opened orifice area.
Effectiveness of the valve debridement was assessed by direct inspection and by simultaneous central aortic and intraventricular pressure measurements after the patient had been separated from cardiopulmonary bypass and hemodynamic stability had been obtained. Additional evaluation of leaflet mobility and the presence of any valve insufficiency was made by intraoperative transesophageal color flow Doppler echocardiographic comparison of preoperative and postoperative valve function and status. Later follow-up assessment was obtained by clinical examination and Doppler echocardiography, and by cardiac catheterization where needed. In 3 other patients ultrasonic debridement was initiated but aborted when transleaflet calcification was found to be present and fenestration of a leaflet was produced (1 patient) or seemed likely to occur (2 patients). These 3
patients were treated with valve replacement and did well. Anesthesia and techniques of cardiopulmonary bypass were identical to those employed for all patients undergoing aortic valve operations. Myocardial protection involved systemic hypothermia to 24°C and ischemic arrest with cold oxygenated crystalloid cardioplegia administered antegrade via the aortic root, coronary ostia, and/or coronary artery bypass grafts in 7 patients (mean, two grafts per patient).
Results Operative Results and Follow-up The transvalvular gradient was reduced from a mean of 94 mm Hg to a mean of less than 10 mm Hg (range, 0 to 20
748
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
-
200 182-
LV
Ao
Pre-Op CI-2.0
80
-
0
Post Repair
(3-2.4 0 -
Fig 2 . Aortic (Ao) and intraventricular (LV) pressure curves before and after valve debridement. (CI = cardiac index in liters per minute.)
mm Hg) in the 9 patients with calcified atherosclerotic trileaflet valves. In the 2 patients with moderately stenotic valves it was reduced from a mean of 40 mm Hg to a mean
Ann Thorac Surg 1990;49:746-53
of 4 mm Hg (range, 0 to 7 mm Hg). Figure 2 presents aortic and ventricular pressure curves for a patient treated with ultrasonic debridement, demonstrating the effectiveness of ultrasonic debridement in eliminating the gradients associated with valvular stenosis. Intraoperative transesophageal color flow Doppler echocardiography demonstrated improved leaflet motion and freedom from serious valvular insufficiency in all patients. Two patients who had moderate aortic insufficiency preoperatively due to failure of effective leaflet coaptation were shown to have no insufficiency on the postoperative study. All patients survived the operative procedure and did well hemodynamically. One patient died during the initial hospitalization of complications arising from a postoperative colon perforation. A Doppler echocardiographic study of her debrided valve before her death showed no gradient and no insufficiency. Ten patients were discharged and available for further follow-up. Follow-up of the 2 patients with moderately stenotic valves after discharge has demonstrated improved leaflet motion and freedom from insufficiency. They continue to do well 2 years postoperatively. Follow-up data on the 8 discharged patients who had severely stenotic trileaflet atherosclerotic valves treated by ultrasonic debridement reveals that 3 patients have re-
Fig 3. Serial postoperative color pow Doppler echocardiographic study (parasternal-long axis view) of dehrided aortic value showing progressive aortic insufficiency: (A) 1 week, (5) 6 weeks, (C) 7 months, and ( D ) 1 year. (Ao = aorta; LV = left ventricle.)
Ann Thorac Surg 1990;49:746-53
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
749
Fig 4. (A) Section of undebrided aortic valve leaflet showing pan-leaflet degeneration process of sclerosis myxoid degeneration and dystrophic calcification. (B)Section of partially debrided leaflet showing similar histology, but reduction in thickness.
quired reoperation and aortic valve replacement for progressive valve insufficiency. Two patients, aged 65 and 77 years, manage a quiet ambulatory existence without symptoms at daily activity but with mild valvular insufficiency documented at 19 and 13 months postoperatively. The 3 remaining patients (mean age, 80 years), all with chronic congestive heart failure preoperatively, have died 5 to 14 months postoperatively due to chronic respiratory insufficiency (1 patient) or sudden death (2 patients). The 3 patients who have required reoperation and valve replacement were documented to have progressive aortic insufficiency on serial follow-up echocardiographic studies (Fig 3). They also exhibited failure to improve clinically, persistent or enlarging cardiomegaly on chest roentgenogram, and the presence of a new and increasing murmur of aortic insufficiency. Cardiac catheterization documented severe aortic insufficiency in all with mild or
no aortic stenosis (maximum peak gradient, less than 30 mm Hg). Valve replacement required a 19-mm mechanical bileaflet prosthesis in each patient after further maximum annular debridement. Aortic root enlargement procedures were not felt advisable due to extensive annular calcification, age of the patients, and magnitude of the procedures. All patients did well, were discharged on warfarin anticoagulation, and remain clinically improved.
Histopathological Observations The valve leaflets of several patients were available for study. Figure 4 presents histological sections of aortic leaflets from a patient in whom debridement was performed but had to be aborted due to fenestration. A nondebrided leaflet (Fig 4A) demonstrates histologically that the atherosclerotic calcified lesion of senescent aortic stenosis is a pan-leaflet process producing a thickened
750
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
Ann Thorac Surg 1990;49:74653
Fig 5. (A) Aortic valve with progressive insuficiency 1 year after ultrasonic debridement. ( B ) Excised thickened, shrunken, and retracted leaflets. (C) Histological section of excised leaflet showing marked fibrosis with sclerosis.
and sclerotic leaflet with focal areas of myxoid degeneration and extensive deposition of calcium within the fibrosed areas. It is intrinsic degeneration of the leaflet substance rather than a deposition on the leaflet surface. No normal leaflet anatomy is observed except the thin endothelial cell surface layers. Figure 48 shows a pathological section of a partially debrided leaflet that is histologically similar to the nondebrided leaflet. Sclerosis and calcification with focal myxoid degeneration is evident although the leaflet is thinner as a result of removal of one endothelial surface and central leaflet substances. Figure 5A is an operative view of a valve in a patient with progressive valvular insufficiency who required reoperation and valve replacement 1 year after ultrasonic debridement. Grossly, the valve leaflets appear thickened, hardened, shrunken, and retracted (Fig 5B). Microscopic study of this leaflet tissue shows marked fibrosis with sclerosis and calcification. Neovascularization is prominent in areas but there is minimal myxoid degeneration (Fig 5C).
Comment Early surgical procedures to correct symptomatic aortic stenosis were direct efforts to regain orifice area and improve leaflet mobility and thereby reduce the gradient without inducing regurgitation [9-111. Before the advent of cardiopulmonary bypass and satisfactory valve substitutes, closed methods were used in the early 1950s by pioneering surgeons to relieve this critical lesion. Because of the rapid deterioration seen in patients with aortic stenosis once atrial fibrillation was added to the symptom triad of angina, syncope, or left ventricular failure, Harken and associates [9] advocated valvotomy by digital manipulation of the stenotic valve by way of an "operating tunnel graft" sewn to the ascending aorta. Muller and Hyman [lo] and others reported some success using a transventricular mechanical dilator. These early efforts proved to be of limited success and associated with discouragingly high risks. The limitations of these "blind" techniques led Mulder
Ann Thorac Surg 1990:4974&53
and Winfield [11 to advocate direct-vision commissurotomy using systemic hypothermia and three to nine minutes of aortic occlusion. However, they observed that "anatomic opening of the commissure is readily accompanied with some increase in valve orifice, but the heavy calcific deposits allow no improved valvular mobility." (This finding was recently repeated and confirmed by interventional cardiology colleagues using percutaneous aortic balloon valvoplasty.) This observation led Mulder and Winfield to advocate both direct commissurotomy and mechanical debridement of the calcified encrusted material from the leaflets under direct vision employing extracorporeal circulation and anoxic arrest. Three cases were presented in their original report in 1960 in which the "gradient was completely abolished and dramatic clinical improvement achieved." This led to a wide application of mechanical debridement for calcified aortic stenosis. Even though this experience took place in the infancy years of cardiac surgery and extracorporeal circulation, patient clinical improvement was clearly substantial and operative risks, high by current standards, were felt to be acceptable at the time given the high mortality of symptomatic aortic stenosis. Unfortunately, the initial beneficial results were not always found to be lasting [2, 3, 121. Reports [Z, 51 demonstrated that restenosis occurs in the majority of patients within 5 years although clinical improvement was observed to be longer lasting in a minority of patients. Scannell and Austen [4] reported in 1966 that mechanical debridement compared favorably with aortic valve replacement at 3 years but not after 5 years. These and other reports of the improved and lasting results using new prosthetic valve substitutes led to the abandonment of mechanical debridement for the treatment of aortic stenosis. Mindich and associates [13] reintroduced mechanical aortic valvoplasty in 1986 as an alternative to prosthetic valve implantation in select patients. Their initial report consisted of 23 patients with an average age of 75 years, all with acquired aortic stenosis, small aortic roots, and relative contraindications to anticoagulation. Valvoplasty was felt to be successful and an effective alternative to valve replacement if an orifice area greater than 1.1 cm could be achieved without producing more than mild residual aortic insufficiency. Only 2 patients required reoperation within 6 months to 6 years of follow-up: 1 for endocarditis and 1 for late valve "restenosis." Two patients died of unrelated diseases and 19 remained asymptomatic and free of thromboembolic complications; none required anticoagulation. In 1988 Mindich and co-workers [6] expanded their valvoplasty techniques to utilize ultrasonic surgical aspirator debridement to decalcify stenotic aortic valves. Fifteen patients were reported who underwent aortic valve decalcification by the same methods and using the same equipment as in the present report. Although the characteristics of their patients were not defined specifically, our current population resembled very closely the elderly group with high gradients and small aortic roots that Mindich and co-workers reported in their 1986 series [13] having mechanical debridement. Their 1988 data on ultra-
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
751
sonic debridement related that the preoperative gradient in 35 patients was 47.5 & 25 mm Hg, which was about one half that observed preoperatively in our series. Gradient reduction to 6.4 4 mm Hg and a significant increase in valve area to 1.6 f 0.3 cm with no increase in insufficiency by Doppler echocardiography were similar to the immediate postoperative data for our series. No follow-up data beyond 6 months were given in an additional 1989 report by Mindich and co-workers [6, 81. However, they advocated ultrasonic debridement as a possible safe and effective new treatment for aortic stenosis. They believed that it could be considered instead of valve replacement whenever severe valvular deformity or excessive insufficiency were not present. Ultrasonic debridement of the stenotic aortic valve had been the subject of an earlier article by Brown and Davis in 1972 [14]. They modified the power and frequency of vibration of a dental device and demonstrated that it could effectively remove calcified material from excised aortic valve leaflet tissues. Two elderly patients were subsequently successfully treated by debridement of their stenotic aortic valves using the ultrasonic device. Both patients did well with excellent clinical early results, but only a 2-month follow-up was available in the initial report. Freeman and colleagues [7] reported 15 patients with an average age of 78 years who had aortic valve decalcification for senescent aortic valve stenosis by ultrasonic surgical aspiration. Their evaluation again documented the effectiveness of the technique in relieving the gradient and improving leaflet mobility and coaptation. Their early data were very similar to those of the present series. However, no late follow-up data were available. Without question ultrasonic surgical aspiration can debride calcified valve cusps and relieve aortic stenosis. However, the occurrence of regurgitation in 5 of 10 patients, including 3 patients with progressive regurgitation requiring reoperation and valve replacement, in the present series make its use most questionable. This is particularly of concern because the pathological mechanism of the insufficiency was seen to be inflammatory thickening and retraction of the debrided leaflets. An accompanying aortitis observed in 1 of the 3 patients makes one postulate that the ultrasonic technique elicits an intense "healing" process in the thin, flexible debrided leaflet that results in leaflet thickening and shrinkage as any other soft tissue wound would experience. It may be that the 2 patients with moderate aortic stenosis who have had a good result after valve debridement in this series have gained their "good" results from the fact that they needed commissurotomy and less extensive debridement. After removal of calcified deposits in these less diseased valves, the subsequent inflammatory reaction may have been less intense. Until further data are available regarding late results of the current technique of ultrasonic debridement or until adjunctive measures are provided to eliminate the late occurrence of progressive insufficiency due to leaflet shrinkage, this technique should be held in reserve. This is unfortunate because it is a procedure that, once learned, can quickly debride fibrous, calcified leaflets
+
752
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
responsible for aortic stenosis with an excellent initial hemodynamic result. If the occurrence of progressive postoperative insufficiency could be eliminated, it could be very useful i n dealing w i t h critically ill patients with severe aortic stenosis, small aortic roots, and contraindications to anticoagulation.
References 1. Mulder DG, Winfield ME. Valvuloplasty for acquired aortic stenosis. Ann Surg 1960;151:2, 20S15. 2, Enright LP, Hancock EW, Shumway NE, Aortic debridement: lone-term follow-uu. Circulation 1971:43. . . 44(Suuu1 1): " 68-72. 3. Hurley PJ, Lowe JB, Barratt-Boyes BG. Debridementvalvotomy for aortic stenosis in adults. Thorax 1967;22:31&9. 4. Scannell JG, Austen WG. Replacement versus reconstruction of the aortic valve. Circulation 1966;33, 34(Suppl 1):133. 5. Hill DG. Long-term results of debridement valvotomy for calcific aortic stenosis. J Thorac Cardiovasc Surg 1973;65:5, 7 0 ~ 1 . 6. Mindich BP, Guarino T, Krenz H, et al. Aortic valve salvage \
1 1
Ann Thorac Surg 1990;49:746-53
utilizing high frequency vibratory debridement [Abstract]. J Am Coll Cardiol 1988;11:3A. 7. Freeman WK, Schaff HV, Orszulak TA. Ultrasonic aortic valve decalcification: serial Doppler echocardiographic follow-up [Abstract]. Circulation 1988;78:379. 8. Mindich BP, Guarino T, Fisher EA, et al. Is ultrasonic debridement of aortic stenosis a reliable alternative to valve replacement [Abstract]? J Am Coll Cardiol 1989;13:123A. 9. Harken DE, Black H, Taylor WJ, Thrower WB, Soroff HS. The surgical correction of calcific aortic stenosis in adults. J Thorac Surg 1958;36:759-76. 10. Muller WH, Hyman M. Valvulotomy for the surgical relief of aortic stenosis. Surg Gynecol Obstet 1954;99:587-94. 11. Baker CG, Campbell M. The results of valvotomy for aortic stenosis. Lancet 1956;1:171-5. 12. Rees JR, Holswade GR, Lillehei CW, Glenn F. Aortic valvuloplasty for stenosis in adults: late results. J Thorac Cardiovasc Surg 1974;67:39&4. 13. Mindich BP, Guarino T, Goldman M. Aortic valvuloplasty for acquired aortic stenosis. Circulation 1986;74(Suppl 1):13G5. 14. Brown AH, Davies PG. Ultrasonic decalcification of calcified cardiac valves and annuli. Br Med J 1972;3:27&7.
DISCUSSION DR THOMAS A. ORSZULAK (Rochester, MN): I think this is an important piece of information in the puzzle of aortic valve decalcification. In terms of this report, we could criticize its size and its narrowness but should recognize that with your critical selectivity in choosing patients in hopes of having it succeed, it still failed. If the results were superb we could at least say that ultrasonic debridement would work some of the time. It did not. We have since had a further follow-up (beyond 9 months) for patients that you mentioned in your talk in whom we performed decalcification. We had a total of 61 patients. Twenty-six percent had severe aortic insufficiency at late follow-up, 30% had moderate insufficiency, and 14% required reoperation. Our original procedure was done with a mortality rate of 8.8%, and we have echoed your feelings in terms of abandoning the procedure. I have three questions, and I have one challenge for you in terms of the report. Can you identify a line between too much and too little decalcification and perhaps find some niche for this procedure for aortic stenosis? Is the microscopic pathology perhaps a burn from the ultrasonic tip rather than just a healing process? And could platelet deposition and thromboxane interaction have anything to do with this scarring phenomenon? The challenge is to address your comment in the report regarding mechanical valves being used in small roots due to the magnitude of the root-enlargement procedure, age of the patient, and extensive annular calcification. We have found pericardial patch enlargement of the aortic root always allows at least one, and rarely two, size increases in prostheses. With modest annular debridement of calcium, which may be facilitated with an ultrasonic surgical aspirator, closure of the aortotomy is enhanced with autologous pericardium with no aneurysm formation at 20-year follow-up. Our elderly population has a very high risk with the addition of Coumadin, and annular debridement enhances our choice of tissue valves. DR STUART H. HARRIS JR (Lynchburg, VA): I enjoyed this report, particularly because I was privileged to be a resident with Dr Harken at Brigham in 1958 and 1960 and watched him do the finger valvoplasty. I then went down to Charlottesville and assisted Dr Muller using his transventricular valvulotome. I saw
these procedures work, and I think it is a real tribute to these pioneers that they did. I also saw both of them open up the aorta, particularly after Dr Harken had obtained his pump, and do manual debridement of the aortic valve. I have also seen the valve tear when there was not an adequate valve to replace it. It was like being in an airplane when the motors went out and there was no parachute. I do think it is a tribute that in those days it did work at times. It is very interesting to see the nice advancement your report represents. DR JOHN L. OCHSNER (New Orleans, LA): I think this is an important report and I am only sorry that it was not presented 3 years ago, because at the Ochsner Clinic we have operated on 18 patients from May 1987 to October 1988. Two of the patients died, late deaths unrelated to the procedure, but only 4 patients are without evidence of serious aortic insufficiency. Aortic insufficiency, as you know, is seen in varying degrees and is not predictable. I think it is probably related to the amount of redundancy that one has in the free coapting edges of the leaflet when one completes the debridement. It is very analogous to the old fascia lata valves we did some 25 years ago. We found out that if we made the valve a little too big, and the coapting edges had somewhat of a serpentine configuration and were redundant, then it took a long time for the scarring and retraction to occur to the point where it became insufficient. I believe we obtained better results in those valves in which we had a lot of valvular tissue left; however, all of them will eventually contract to the point of insufficiency. I have to agree with Dr Craver that the procedure should be abandoned except for the use of the apparatus to debride the valve to put in a mechanical bioprosthesis or homograft. DR HENDRICK B. BARNER (St. Louis, MO): My associates, Andrew Fiore and Larry McBride, have reported our experience with ultrasonic decalcification of the aortic valve to the Western Thoracic Surgical Association. In 1988, we performed ultrasonic decalcification on 22 patients with a valve area of less than 0.8 cm2 (mean, 0.7 cm2), which increased to 1.42 cm2 1week postoperatively and was 1.29 cm2 6
Ann Thorac Surg 1990;49:74&53
months later. The peak valve gradient decreased from 74 to 25 mm Hg and was 31 mm Hg 6 months postoperatively. Of concern to us has been the occurrence of aortic insufficiency, which was present in 50% of patients preoperatively and 87% of patients at 6 months. Four patients had new onset of aortic insufficiencyand 3 patients had worsening of their aortic insufficiency, but no patient had more than 2+ regurgitation on a scale of 0 to 4. Since analyzing these data, we have greatly curtailed our use of ultrasonic decalcification, but we have not abandoned the procedure totally. We believe that results are operator and patient dependent. At this point, we do not have enough data to analyze the influence of these variables on outcome, but we hope to do so by continued echocardiographic follow-up of our patients. DR CRAVER Dr garner, we will be interested in the continuing follow-up of your patients. We will see if you can figure out some better method of patient selection or an adjuvant treatment for these debrided leaflets that would make the technique successful and the result long-lasting. Certainly it is a very gratifying procedure that relieves aortic stenosis. We just do not have any way to stabilize the leaflet membrane once it has been debrided and no model to take to the laboratory of the particular diseased tissue. If you could either treat the valve, increase the cross-links, or in some way keep the valve from shrinking, much as you treat pericardium with glutaraldehyde, it might well work. Dr Ochsner, I appreciate your comments and would agree with what you are saying. In general, the amount of coaptation surface is the key to valve competency. And as you can see, in these senescent valves there seems to be quite a good area, usually 2 to 4 mm of leaflet coaptation, without annular dilatation. The insufficiency that we observed postoperatively is primarily a matter of progressive leaflet shrinkage, which produces inadequate leaflet coaptation rather than annular dilatation. This is supported by the elimination of aortic insufficiency in several patients who had it preoperatively after the valve leaflets were debrided and their coaptability restored. Dr Harris, it was good to hear your comments. That must have been a fascinating time to live. I trained also with Dr Muller and I know that he is a man of great skill and bravery. To be a pioneer in cardiac surgery took great courage and persistence when the operative mortality in the early stages was 40%, because you
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
753
knew that the patient mortality otherwise was 100%. And I have nothing but admiration for him and for the other pioneers who went through that era without parachutes that are employed today. Dr Orszulak, thanks for bringing the experience from the Mayo Clinic. You certainly have had one of the widest experienceswith this. I am sorry to see that your late insufficiency rate is also as high as the one we have observed. I do not think it is a matter of taking out too much of the disease process, because you are still left with the same underlying process. I think it is rather a matter of new scar formation, edema, and retraction. We looked carefully at those patients after the first one needed to be reoperated on to see if there was any indication of thickening of the leaflets by echocardiography, which may suggest a local thrombotic process or platelet deposition. We could see only the same degree of motion and leaflet thickness on the following echoes as we had before. So we still come back to scarring as the cause. We have used root enlargement procedures for a decade and a half now and advocate those where needed. With this debridement technique, I had hoped to work out a way to expedite that process in very elderly patients. To debride the annulus, incise the noncoronary sinus, and insert a pericardial patch adds about 30 minutes to the operation, with some potential for bleeding. If the patients have any significant reason not to be able to take Coumadin, enlarging the aortic annulus and putting an appropriate sized porcine valve in is a valid alternative procedure. Because you can move up a from a size 19 easily to a 21 or 23, this gives the patient a quite good hemodynamic result. Using the ultrasonic surgical aspirator just to debride the annulus is expensive. The disposable items that come with the Cavitron unit cost our hospital about $900 per case. If you start trying to reuse them multiple times, you get into problems of tip fracture in the device and burr formation on it. Also, I am just too frugal to spend that much money to do something I think can be done effectively with standard surgical methods. Our overall mortality for aortic valve replacement in the group older than 75 years is 8.9%, taking all comers. So, I think using a root enlargement when you have to or a mechanical bileaflet valve when you cannot is probably preferable in managing patients with the very small aortic annulus with calcified aortic stenosis.
Aortic stenosis was relieved in 11 patients by ultrasonic debridement of the valve and annulus, while 102 other patients underwent valve replacement for aortic stenosis during 1988. Debridement was selectively applied based on findings of small annulus size (19 mm or less) and extensive calcification. Additional patient characteristics were mean transvalvular gradient of 78 mm Hg, advanced age, and marked left ventricular hypertrophy. Six patients had no residual gradient and 5 others a mean gradient less than 10 mm Hg. There were no complications related to the debridement process. Intraoperative transesophageal Doppler echocardiography demonstrated improved leaflet mobility and elimination of the gradient in all patients and elimination of associated
M
echanical debridement of the aortic valve was applied as treatment to relieve aortic stenosis for many years before successful valve substitutes were available [l, 21. The recurrence of aortic stenosis in a high percentage of patients within 5 years as well as the emergence of successful prosthetic valve devices led to the abandonment of techniques of valve repair [%5].
For editorial comment, see page 689. Recently, valve debridement using ultrasonic surgical aspiration devices has been reported as an effective alternative to valve replacement [6-81. This report presents an experience using ultrasonic surgical debridement of the aortic valve in 11 patients to correct aortic stenosis and avoid valve replacement.
Material a n d Methods Eleven select patients out of a total of 113 patients undergoing operation for aortic stenosis underwent debridement of their aortic valve using ultrasonic surgical aspiration techniques. Nine of the 11 had severe aortic stenosis, with a mean gradient of 94 mm Hg (range, 80 to 128 mm Hg), and sclerotic trileaflet valves that were heavily calcified, with a small, calcified valve annulus (19 mm). Two patients were primarily undergoing coronary artery bypass grafting for ischemic symptoms but also had calcified Presented at the Thirty-sixth Annual Meeting of the Southern Thoracic Surgical Association, Scottsdale, AZ, Nov 9-11, 1989. Address reprint requests to Dr Craver, The Emory Clinic, 1365 Clifton Rd, NE, Atlanta, GA 30322.
0 1990 by The Society of Thoracic Surgeons
valvular insufficiency in 2 patients. Follow-up echocardiography demonstrated late onset of new valvular regurgitation in 5 patients that was progressive and required reoperation in 3. Thickened, hardened, and retracted valve leaflets with loss of central coaptation were found in all 3 patients who underwent reoperation. Ultrasonic debridement can effectively relieve aortic stenosis, provide an excellent immediate hemodynamic result, and decrease operative time. However, the early occurrence of aortic insufficiency in a high percentage of patients makes it an unacceptable alternative to valve replacement, and the technique should be abandoned as a treatment for severe calcific aortic stenosis. (Ann Thorac Surg 1990;49:746-53)
aortic valves with moderate stenosis (mean gradient, 40 mm Hg; range, 36 to 45 mm Hg). Intraoperative election to use the ultrasonic debridement technique was based on the desire to avoid implantation of a mechanical valve prosthesis in 9 elderly patients (mean age, 78 years) with severe aortic stenosis and a small calcified aortic root and annulus. Improvement in valve function without valve replacement was the indication in the 2 other patients with moderately stenotic valves who were primarily having coronary artery bypass grafting. Valve debridement as an alternative to valve replacement was discussed with all patients before operation. Ultrasonic debridement was accomplished with a handheld probe that applied high-speed vibration at 23 kHz, a frequency above the human aural range, to the calcified atherosclerotic valve leaflets and annulus. Ultrahighfrequency vibration impacts the harder elements of the sclerotic leaflet lesions and causes them to be dislodged from the softer tissues, which can absorb the vibration forces. Ultrasonic vibration fragments and disintegrates the calcific deposits, which are then aspirated from the remaining leaflet tissue. The ultrasonic vibrator probe was cooled by a continuous external irrigation of saline solution over the tip. Fragmented debris was aspirated from the operative field through the hollow core of the probe tip. Figure 1 demonstrates the debridement of a heavily calcified stenotic valve with resultant wide leaflet mobility and increase in orifice size. Debridement was performed using the Cavitron ultrasonic surgical aspirator (Cavitron Surgical Systems, Inc, Stamford, CT) in 9 patients and the Bovie ultrasonic surgical aspirator (MDT Castle, Inc, Rochester, NY) in 2 patients. 0003-4975/90/$3.50
Ann Thorac Surg 1990:49:74&53
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
747
Fig 1. (A) Calcified aortic valve. ( B ) Debridement by ultrasonic technique in process. (C) Leaflets after completion of debridement process; note widely opened orifice area.
Effectiveness of the valve debridement was assessed by direct inspection and by simultaneous central aortic and intraventricular pressure measurements after the patient had been separated from cardiopulmonary bypass and hemodynamic stability had been obtained. Additional evaluation of leaflet mobility and the presence of any valve insufficiency was made by intraoperative transesophageal color flow Doppler echocardiographic comparison of preoperative and postoperative valve function and status. Later follow-up assessment was obtained by clinical examination and Doppler echocardiography, and by cardiac catheterization where needed. In 3 other patients ultrasonic debridement was initiated but aborted when transleaflet calcification was found to be present and fenestration of a leaflet was produced (1 patient) or seemed likely to occur (2 patients). These 3
patients were treated with valve replacement and did well. Anesthesia and techniques of cardiopulmonary bypass were identical to those employed for all patients undergoing aortic valve operations. Myocardial protection involved systemic hypothermia to 24°C and ischemic arrest with cold oxygenated crystalloid cardioplegia administered antegrade via the aortic root, coronary ostia, and/or coronary artery bypass grafts in 7 patients (mean, two grafts per patient).
Results Operative Results and Follow-up The transvalvular gradient was reduced from a mean of 94 mm Hg to a mean of less than 10 mm Hg (range, 0 to 20
748
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
-
200 182-
LV
Ao
Pre-Op CI-2.0
80
-
0
Post Repair
(3-2.4 0 -
Fig 2 . Aortic (Ao) and intraventricular (LV) pressure curves before and after valve debridement. (CI = cardiac index in liters per minute.)
mm Hg) in the 9 patients with calcified atherosclerotic trileaflet valves. In the 2 patients with moderately stenotic valves it was reduced from a mean of 40 mm Hg to a mean
Ann Thorac Surg 1990;49:746-53
of 4 mm Hg (range, 0 to 7 mm Hg). Figure 2 presents aortic and ventricular pressure curves for a patient treated with ultrasonic debridement, demonstrating the effectiveness of ultrasonic debridement in eliminating the gradients associated with valvular stenosis. Intraoperative transesophageal color flow Doppler echocardiography demonstrated improved leaflet motion and freedom from serious valvular insufficiency in all patients. Two patients who had moderate aortic insufficiency preoperatively due to failure of effective leaflet coaptation were shown to have no insufficiency on the postoperative study. All patients survived the operative procedure and did well hemodynamically. One patient died during the initial hospitalization of complications arising from a postoperative colon perforation. A Doppler echocardiographic study of her debrided valve before her death showed no gradient and no insufficiency. Ten patients were discharged and available for further follow-up. Follow-up of the 2 patients with moderately stenotic valves after discharge has demonstrated improved leaflet motion and freedom from insufficiency. They continue to do well 2 years postoperatively. Follow-up data on the 8 discharged patients who had severely stenotic trileaflet atherosclerotic valves treated by ultrasonic debridement reveals that 3 patients have re-
Fig 3. Serial postoperative color pow Doppler echocardiographic study (parasternal-long axis view) of dehrided aortic value showing progressive aortic insufficiency: (A) 1 week, (5) 6 weeks, (C) 7 months, and ( D ) 1 year. (Ao = aorta; LV = left ventricle.)
Ann Thorac Surg 1990;49:746-53
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
749
Fig 4. (A) Section of undebrided aortic valve leaflet showing pan-leaflet degeneration process of sclerosis myxoid degeneration and dystrophic calcification. (B)Section of partially debrided leaflet showing similar histology, but reduction in thickness.
quired reoperation and aortic valve replacement for progressive valve insufficiency. Two patients, aged 65 and 77 years, manage a quiet ambulatory existence without symptoms at daily activity but with mild valvular insufficiency documented at 19 and 13 months postoperatively. The 3 remaining patients (mean age, 80 years), all with chronic congestive heart failure preoperatively, have died 5 to 14 months postoperatively due to chronic respiratory insufficiency (1 patient) or sudden death (2 patients). The 3 patients who have required reoperation and valve replacement were documented to have progressive aortic insufficiency on serial follow-up echocardiographic studies (Fig 3). They also exhibited failure to improve clinically, persistent or enlarging cardiomegaly on chest roentgenogram, and the presence of a new and increasing murmur of aortic insufficiency. Cardiac catheterization documented severe aortic insufficiency in all with mild or
no aortic stenosis (maximum peak gradient, less than 30 mm Hg). Valve replacement required a 19-mm mechanical bileaflet prosthesis in each patient after further maximum annular debridement. Aortic root enlargement procedures were not felt advisable due to extensive annular calcification, age of the patients, and magnitude of the procedures. All patients did well, were discharged on warfarin anticoagulation, and remain clinically improved.
Histopathological Observations The valve leaflets of several patients were available for study. Figure 4 presents histological sections of aortic leaflets from a patient in whom debridement was performed but had to be aborted due to fenestration. A nondebrided leaflet (Fig 4A) demonstrates histologically that the atherosclerotic calcified lesion of senescent aortic stenosis is a pan-leaflet process producing a thickened
750
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
Ann Thorac Surg 1990;49:74653
Fig 5. (A) Aortic valve with progressive insuficiency 1 year after ultrasonic debridement. ( B ) Excised thickened, shrunken, and retracted leaflets. (C) Histological section of excised leaflet showing marked fibrosis with sclerosis.
and sclerotic leaflet with focal areas of myxoid degeneration and extensive deposition of calcium within the fibrosed areas. It is intrinsic degeneration of the leaflet substance rather than a deposition on the leaflet surface. No normal leaflet anatomy is observed except the thin endothelial cell surface layers. Figure 48 shows a pathological section of a partially debrided leaflet that is histologically similar to the nondebrided leaflet. Sclerosis and calcification with focal myxoid degeneration is evident although the leaflet is thinner as a result of removal of one endothelial surface and central leaflet substances. Figure 5A is an operative view of a valve in a patient with progressive valvular insufficiency who required reoperation and valve replacement 1 year after ultrasonic debridement. Grossly, the valve leaflets appear thickened, hardened, shrunken, and retracted (Fig 5B). Microscopic study of this leaflet tissue shows marked fibrosis with sclerosis and calcification. Neovascularization is prominent in areas but there is minimal myxoid degeneration (Fig 5C).
Comment Early surgical procedures to correct symptomatic aortic stenosis were direct efforts to regain orifice area and improve leaflet mobility and thereby reduce the gradient without inducing regurgitation [9-111. Before the advent of cardiopulmonary bypass and satisfactory valve substitutes, closed methods were used in the early 1950s by pioneering surgeons to relieve this critical lesion. Because of the rapid deterioration seen in patients with aortic stenosis once atrial fibrillation was added to the symptom triad of angina, syncope, or left ventricular failure, Harken and associates [9] advocated valvotomy by digital manipulation of the stenotic valve by way of an "operating tunnel graft" sewn to the ascending aorta. Muller and Hyman [lo] and others reported some success using a transventricular mechanical dilator. These early efforts proved to be of limited success and associated with discouragingly high risks. The limitations of these "blind" techniques led Mulder
Ann Thorac Surg 1990:4974&53
and Winfield [11 to advocate direct-vision commissurotomy using systemic hypothermia and three to nine minutes of aortic occlusion. However, they observed that "anatomic opening of the commissure is readily accompanied with some increase in valve orifice, but the heavy calcific deposits allow no improved valvular mobility." (This finding was recently repeated and confirmed by interventional cardiology colleagues using percutaneous aortic balloon valvoplasty.) This observation led Mulder and Winfield to advocate both direct commissurotomy and mechanical debridement of the calcified encrusted material from the leaflets under direct vision employing extracorporeal circulation and anoxic arrest. Three cases were presented in their original report in 1960 in which the "gradient was completely abolished and dramatic clinical improvement achieved." This led to a wide application of mechanical debridement for calcified aortic stenosis. Even though this experience took place in the infancy years of cardiac surgery and extracorporeal circulation, patient clinical improvement was clearly substantial and operative risks, high by current standards, were felt to be acceptable at the time given the high mortality of symptomatic aortic stenosis. Unfortunately, the initial beneficial results were not always found to be lasting [2, 3, 121. Reports [Z, 51 demonstrated that restenosis occurs in the majority of patients within 5 years although clinical improvement was observed to be longer lasting in a minority of patients. Scannell and Austen [4] reported in 1966 that mechanical debridement compared favorably with aortic valve replacement at 3 years but not after 5 years. These and other reports of the improved and lasting results using new prosthetic valve substitutes led to the abandonment of mechanical debridement for the treatment of aortic stenosis. Mindich and associates [13] reintroduced mechanical aortic valvoplasty in 1986 as an alternative to prosthetic valve implantation in select patients. Their initial report consisted of 23 patients with an average age of 75 years, all with acquired aortic stenosis, small aortic roots, and relative contraindications to anticoagulation. Valvoplasty was felt to be successful and an effective alternative to valve replacement if an orifice area greater than 1.1 cm could be achieved without producing more than mild residual aortic insufficiency. Only 2 patients required reoperation within 6 months to 6 years of follow-up: 1 for endocarditis and 1 for late valve "restenosis." Two patients died of unrelated diseases and 19 remained asymptomatic and free of thromboembolic complications; none required anticoagulation. In 1988 Mindich and co-workers [6] expanded their valvoplasty techniques to utilize ultrasonic surgical aspirator debridement to decalcify stenotic aortic valves. Fifteen patients were reported who underwent aortic valve decalcification by the same methods and using the same equipment as in the present report. Although the characteristics of their patients were not defined specifically, our current population resembled very closely the elderly group with high gradients and small aortic roots that Mindich and co-workers reported in their 1986 series [13] having mechanical debridement. Their 1988 data on ultra-
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
751
sonic debridement related that the preoperative gradient in 35 patients was 47.5 & 25 mm Hg, which was about one half that observed preoperatively in our series. Gradient reduction to 6.4 4 mm Hg and a significant increase in valve area to 1.6 f 0.3 cm with no increase in insufficiency by Doppler echocardiography were similar to the immediate postoperative data for our series. No follow-up data beyond 6 months were given in an additional 1989 report by Mindich and co-workers [6, 81. However, they advocated ultrasonic debridement as a possible safe and effective new treatment for aortic stenosis. They believed that it could be considered instead of valve replacement whenever severe valvular deformity or excessive insufficiency were not present. Ultrasonic debridement of the stenotic aortic valve had been the subject of an earlier article by Brown and Davis in 1972 [14]. They modified the power and frequency of vibration of a dental device and demonstrated that it could effectively remove calcified material from excised aortic valve leaflet tissues. Two elderly patients were subsequently successfully treated by debridement of their stenotic aortic valves using the ultrasonic device. Both patients did well with excellent clinical early results, but only a 2-month follow-up was available in the initial report. Freeman and colleagues [7] reported 15 patients with an average age of 78 years who had aortic valve decalcification for senescent aortic valve stenosis by ultrasonic surgical aspiration. Their evaluation again documented the effectiveness of the technique in relieving the gradient and improving leaflet mobility and coaptation. Their early data were very similar to those of the present series. However, no late follow-up data were available. Without question ultrasonic surgical aspiration can debride calcified valve cusps and relieve aortic stenosis. However, the occurrence of regurgitation in 5 of 10 patients, including 3 patients with progressive regurgitation requiring reoperation and valve replacement, in the present series make its use most questionable. This is particularly of concern because the pathological mechanism of the insufficiency was seen to be inflammatory thickening and retraction of the debrided leaflets. An accompanying aortitis observed in 1 of the 3 patients makes one postulate that the ultrasonic technique elicits an intense "healing" process in the thin, flexible debrided leaflet that results in leaflet thickening and shrinkage as any other soft tissue wound would experience. It may be that the 2 patients with moderate aortic stenosis who have had a good result after valve debridement in this series have gained their "good" results from the fact that they needed commissurotomy and less extensive debridement. After removal of calcified deposits in these less diseased valves, the subsequent inflammatory reaction may have been less intense. Until further data are available regarding late results of the current technique of ultrasonic debridement or until adjunctive measures are provided to eliminate the late occurrence of progressive insufficiency due to leaflet shrinkage, this technique should be held in reserve. This is unfortunate because it is a procedure that, once learned, can quickly debride fibrous, calcified leaflets
+
752
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
responsible for aortic stenosis with an excellent initial hemodynamic result. If the occurrence of progressive postoperative insufficiency could be eliminated, it could be very useful i n dealing w i t h critically ill patients with severe aortic stenosis, small aortic roots, and contraindications to anticoagulation.
References 1. Mulder DG, Winfield ME. Valvuloplasty for acquired aortic stenosis. Ann Surg 1960;151:2, 20S15. 2, Enright LP, Hancock EW, Shumway NE, Aortic debridement: lone-term follow-uu. Circulation 1971:43. . . 44(Suuu1 1): " 68-72. 3. Hurley PJ, Lowe JB, Barratt-Boyes BG. Debridementvalvotomy for aortic stenosis in adults. Thorax 1967;22:31&9. 4. Scannell JG, Austen WG. Replacement versus reconstruction of the aortic valve. Circulation 1966;33, 34(Suppl 1):133. 5. Hill DG. Long-term results of debridement valvotomy for calcific aortic stenosis. J Thorac Cardiovasc Surg 1973;65:5, 7 0 ~ 1 . 6. Mindich BP, Guarino T, Krenz H, et al. Aortic valve salvage \
1 1
Ann Thorac Surg 1990;49:746-53
utilizing high frequency vibratory debridement [Abstract]. J Am Coll Cardiol 1988;11:3A. 7. Freeman WK, Schaff HV, Orszulak TA. Ultrasonic aortic valve decalcification: serial Doppler echocardiographic follow-up [Abstract]. Circulation 1988;78:379. 8. Mindich BP, Guarino T, Fisher EA, et al. Is ultrasonic debridement of aortic stenosis a reliable alternative to valve replacement [Abstract]? J Am Coll Cardiol 1989;13:123A. 9. Harken DE, Black H, Taylor WJ, Thrower WB, Soroff HS. The surgical correction of calcific aortic stenosis in adults. J Thorac Surg 1958;36:759-76. 10. Muller WH, Hyman M. Valvulotomy for the surgical relief of aortic stenosis. Surg Gynecol Obstet 1954;99:587-94. 11. Baker CG, Campbell M. The results of valvotomy for aortic stenosis. Lancet 1956;1:171-5. 12. Rees JR, Holswade GR, Lillehei CW, Glenn F. Aortic valvuloplasty for stenosis in adults: late results. J Thorac Cardiovasc Surg 1974;67:39&4. 13. Mindich BP, Guarino T, Goldman M. Aortic valvuloplasty for acquired aortic stenosis. Circulation 1986;74(Suppl 1):13G5. 14. Brown AH, Davies PG. Ultrasonic decalcification of calcified cardiac valves and annuli. Br Med J 1972;3:27&7.
DISCUSSION DR THOMAS A. ORSZULAK (Rochester, MN): I think this is an important piece of information in the puzzle of aortic valve decalcification. In terms of this report, we could criticize its size and its narrowness but should recognize that with your critical selectivity in choosing patients in hopes of having it succeed, it still failed. If the results were superb we could at least say that ultrasonic debridement would work some of the time. It did not. We have since had a further follow-up (beyond 9 months) for patients that you mentioned in your talk in whom we performed decalcification. We had a total of 61 patients. Twenty-six percent had severe aortic insufficiency at late follow-up, 30% had moderate insufficiency, and 14% required reoperation. Our original procedure was done with a mortality rate of 8.8%, and we have echoed your feelings in terms of abandoning the procedure. I have three questions, and I have one challenge for you in terms of the report. Can you identify a line between too much and too little decalcification and perhaps find some niche for this procedure for aortic stenosis? Is the microscopic pathology perhaps a burn from the ultrasonic tip rather than just a healing process? And could platelet deposition and thromboxane interaction have anything to do with this scarring phenomenon? The challenge is to address your comment in the report regarding mechanical valves being used in small roots due to the magnitude of the root-enlargement procedure, age of the patient, and extensive annular calcification. We have found pericardial patch enlargement of the aortic root always allows at least one, and rarely two, size increases in prostheses. With modest annular debridement of calcium, which may be facilitated with an ultrasonic surgical aspirator, closure of the aortotomy is enhanced with autologous pericardium with no aneurysm formation at 20-year follow-up. Our elderly population has a very high risk with the addition of Coumadin, and annular debridement enhances our choice of tissue valves. DR STUART H. HARRIS JR (Lynchburg, VA): I enjoyed this report, particularly because I was privileged to be a resident with Dr Harken at Brigham in 1958 and 1960 and watched him do the finger valvoplasty. I then went down to Charlottesville and assisted Dr Muller using his transventricular valvulotome. I saw
these procedures work, and I think it is a real tribute to these pioneers that they did. I also saw both of them open up the aorta, particularly after Dr Harken had obtained his pump, and do manual debridement of the aortic valve. I have also seen the valve tear when there was not an adequate valve to replace it. It was like being in an airplane when the motors went out and there was no parachute. I do think it is a tribute that in those days it did work at times. It is very interesting to see the nice advancement your report represents. DR JOHN L. OCHSNER (New Orleans, LA): I think this is an important report and I am only sorry that it was not presented 3 years ago, because at the Ochsner Clinic we have operated on 18 patients from May 1987 to October 1988. Two of the patients died, late deaths unrelated to the procedure, but only 4 patients are without evidence of serious aortic insufficiency. Aortic insufficiency, as you know, is seen in varying degrees and is not predictable. I think it is probably related to the amount of redundancy that one has in the free coapting edges of the leaflet when one completes the debridement. It is very analogous to the old fascia lata valves we did some 25 years ago. We found out that if we made the valve a little too big, and the coapting edges had somewhat of a serpentine configuration and were redundant, then it took a long time for the scarring and retraction to occur to the point where it became insufficient. I believe we obtained better results in those valves in which we had a lot of valvular tissue left; however, all of them will eventually contract to the point of insufficiency. I have to agree with Dr Craver that the procedure should be abandoned except for the use of the apparatus to debride the valve to put in a mechanical bioprosthesis or homograft. DR HENDRICK B. BARNER (St. Louis, MO): My associates, Andrew Fiore and Larry McBride, have reported our experience with ultrasonic decalcification of the aortic valve to the Western Thoracic Surgical Association. In 1988, we performed ultrasonic decalcification on 22 patients with a valve area of less than 0.8 cm2 (mean, 0.7 cm2), which increased to 1.42 cm2 1week postoperatively and was 1.29 cm2 6
Ann Thorac Surg 1990;49:74&53
months later. The peak valve gradient decreased from 74 to 25 mm Hg and was 31 mm Hg 6 months postoperatively. Of concern to us has been the occurrence of aortic insufficiency, which was present in 50% of patients preoperatively and 87% of patients at 6 months. Four patients had new onset of aortic insufficiencyand 3 patients had worsening of their aortic insufficiency, but no patient had more than 2+ regurgitation on a scale of 0 to 4. Since analyzing these data, we have greatly curtailed our use of ultrasonic decalcification, but we have not abandoned the procedure totally. We believe that results are operator and patient dependent. At this point, we do not have enough data to analyze the influence of these variables on outcome, but we hope to do so by continued echocardiographic follow-up of our patients. DR CRAVER Dr garner, we will be interested in the continuing follow-up of your patients. We will see if you can figure out some better method of patient selection or an adjuvant treatment for these debrided leaflets that would make the technique successful and the result long-lasting. Certainly it is a very gratifying procedure that relieves aortic stenosis. We just do not have any way to stabilize the leaflet membrane once it has been debrided and no model to take to the laboratory of the particular diseased tissue. If you could either treat the valve, increase the cross-links, or in some way keep the valve from shrinking, much as you treat pericardium with glutaraldehyde, it might well work. Dr Ochsner, I appreciate your comments and would agree with what you are saying. In general, the amount of coaptation surface is the key to valve competency. And as you can see, in these senescent valves there seems to be quite a good area, usually 2 to 4 mm of leaflet coaptation, without annular dilatation. The insufficiency that we observed postoperatively is primarily a matter of progressive leaflet shrinkage, which produces inadequate leaflet coaptation rather than annular dilatation. This is supported by the elimination of aortic insufficiency in several patients who had it preoperatively after the valve leaflets were debrided and their coaptability restored. Dr Harris, it was good to hear your comments. That must have been a fascinating time to live. I trained also with Dr Muller and I know that he is a man of great skill and bravery. To be a pioneer in cardiac surgery took great courage and persistence when the operative mortality in the early stages was 40%, because you
CRAVER ULTRASONIC DEBRIDEMENT FOR AORTIC STENOSIS
753
knew that the patient mortality otherwise was 100%. And I have nothing but admiration for him and for the other pioneers who went through that era without parachutes that are employed today. Dr Orszulak, thanks for bringing the experience from the Mayo Clinic. You certainly have had one of the widest experienceswith this. I am sorry to see that your late insufficiency rate is also as high as the one we have observed. I do not think it is a matter of taking out too much of the disease process, because you are still left with the same underlying process. I think it is rather a matter of new scar formation, edema, and retraction. We looked carefully at those patients after the first one needed to be reoperated on to see if there was any indication of thickening of the leaflets by echocardiography, which may suggest a local thrombotic process or platelet deposition. We could see only the same degree of motion and leaflet thickness on the following echoes as we had before. So we still come back to scarring as the cause. We have used root enlargement procedures for a decade and a half now and advocate those where needed. With this debridement technique, I had hoped to work out a way to expedite that process in very elderly patients. To debride the annulus, incise the noncoronary sinus, and insert a pericardial patch adds about 30 minutes to the operation, with some potential for bleeding. If the patients have any significant reason not to be able to take Coumadin, enlarging the aortic annulus and putting an appropriate sized porcine valve in is a valid alternative procedure. Because you can move up a from a size 19 easily to a 21 or 23, this gives the patient a quite good hemodynamic result. Using the ultrasonic surgical aspirator just to debride the annulus is expensive. The disposable items that come with the Cavitron unit cost our hospital about $900 per case. If you start trying to reuse them multiple times, you get into problems of tip fracture in the device and burr formation on it. Also, I am just too frugal to spend that much money to do something I think can be done effectively with standard surgical methods. Our overall mortality for aortic valve replacement in the group older than 75 years is 8.9%, taking all comers. So, I think using a root enlargement when you have to or a mechanical bileaflet valve when you cannot is probably preferable in managing patients with the very small aortic annulus with calcified aortic stenosis.
