Ann Thorac Surg 2001;72:1509-1514
© 2001 The Society of Thoracic Surgeons
Original article: cardiovascular
Replacement of degenerated mitral and aortic bioprostheses without explantation
Alexander S. Geha, MD, MS*a,
Malek G. Massad, MDa,
Norman J. Snow, MDa
a Division of Cardiothoracic Surgery, University of Illinois at Chicago Medical Center, Chicago, Illinois, USA
Accepted for publication June 1, 2001.
* Address reprint requests to Dr Geha, Division of Cardiothoracic Surgery, University of Illinois at Chicago Medical Center, 840 S Wood St, MC 958 Suite 417 CSB, Chicago, IL 60612, USA
e-mail: ageha{at}uic.edu
 |
Abstract
|
|---|
Background. The most common indication for reoperation in patients with a bioprosthetic valve is primary tissue failure. Explantation of the bioprosthesis is time consuming, and for a mitral valve, may be complicated by cardiac rupture at the atrioventricular junction or the posterior left ventricular wall where a strut is imbedded, injury to the circumflex artery, and late perivalvular leak; for an aortic valve, annular disruption and perivalvular leak may complicate explantation. A new approach to simplify these procedures and avoid these complications, by excising only the bioprosthetic tissue and attaching a bileaflet mechanical valve to the intact stent, was developed in 1991 and was evaluated over a 9-year period in 50 patients who had had one (34), two (10), three (4), or four (2) previous open cardiac operations.
Methods. Since 1991, we have replaced degenerated mitral bioprostheses in 34 patients (25 to 84 years of age; 12 male, 22 female) by preserving the stent and suturing a St. Jude or Carbomedics bileaflet valve to the atrial side of the bioprosthetic cuff; the mitral valve was exposed through a median sternotomy in 21 patients and through a right anterolateral thoracotomy in 13. Using a similar approach, starting in 1995, 16 additional patients (55 to 73 years of age; 11 male, 5 female) with degenerated aortic bioprostheses had the aortic valve replaced by excising the bioprosthetic tissue and amputating the struts, then suturing a Carbomedics valve to the aortic side of the bioprosthetic cuff. This allows the use of a bileaflet valve similar in size to the bioprosthesis with exact matching of the orifices.
Results. Bypass time averaged 61 ± 14 minutes and aortic cross-clamp time 43 ± 12 minutes. There has been no operative mortality. Three late deaths occurred at 9, 37, and 58 months, and were not valve related. No gradients of hemodynamic significance have been detected on transesophageal echocardiographic follow-up.
Conclusions. Leaving the bioprosthetic cuff intact eliminates the need for extensive dissection, thus shortening and simplifying the procedure and diminishing its attendant mortality and morbidity. This valve-on-valve approach also allows replacement of a degenerated bioprosthesis with a bileaflet valve of comparable size rather than a smaller one jammed into the orifice of the bioprosthetic stent, thus avoiding undue trauma to the bileaflet valve and maintaining excellent hemodynamic function.
|
Introduction
|
|---|
Although several causes would lead to reoperation in patients with previously implanted cardiac bioprosthetic valves, the most common cause is structural failure of these valves [1–4]. Bioprosthetic valves are particularly vulnerable to structural degeneration, which occurs more frequently in valves implanted in the mitral position than in the aortic position, especially when used in patients younger than 65 to 70 years of age [2–4]. Reoperations for bioprosthetic cardiac valve failure are associated with significant mortality and morbidity, and can pose formidable technical challenges [5, 6]. The conventional surgical approach to dealing with these failed bioprosthetic valves consists of explanting the valve and replacing it with another bioprosthesis or mechanical valve. Patients with bioprosthetic structural valvular dysfunction tend to be older and sicker than those having primary valve replacement. In addition to the potential complications inherent to repeat cardiac surgery, reoperations on the mitral valve can be associated with cardiac rupture at the atrioventricular junction or posterior ventricular wall where a strut may be embedded; they also can result in damage to the left circumflex coronary artery during explantation of the degenerated bioprosthesis and insertion of a new one [7]. In the case of reoperation on the aortic valve, annular disruption may result when a previously oversized valve had been implanted. With explantation and replacement of either mitral or aortic valve, late perivalvular leaks may also develop [8].
Beginning in 1991, we developed and evolved a surgical technique for replacement of degenerated bioprosthetic valves without explantation [8, 9]. This approach minimizes the hazards associated with removal of the entire bioprosthesis by leaving the stent and support of the bioprosthesis intact. Removal of only the degenerated leaflet tissue shortens and simplifies the operation and allows an adequate orifice for the attachment of a mechanical bileaflet prosthesis. This approach was used for mitral bioprostheses during the first 4 years, and further extended to aortic bioprostheses since 1995. The purpose of this communication is to describe the approach briefly and analyze the results obtained with it in a series of 50 consecutive patients requiring replacement of a bioprosthetic cardiac valve because of structural degeneration.
|
Material and methods
|
|---|
Patient population
We evaluated the results of replacement of a failed bioprosthetic valve with a bileaflet mechanical valve in a series of 50 consecutive patients who underwent this procedure by the senior author (ASG) between May 1991 and December 1999. Table 1 depicts the clinical profile of the patients, divided into two groups according to the valve replaced without explantation. The 23 male and 27 female patients had an age range of 25 to 84 years (mean 72 years) at the time of the reoperation. Thirty-four patients (19 mitral, 15 aortic) had undergone one previous cardiac operation, 10 patients (9 mitral, 1 aortic) had undergone two previous operations; 4 mitral patients had undergone three, and 2 other mitral patients had undergone four previous cardiac operations. Concomitant cardiac procedures were performed in 19 of the 50 patients: 3 mitral and 2 aortic patients had concomitant coronary revascularization, and another 14 mitral patients had concomitant tricuspid valve repair.
Operative technique
Cardiac exposure was obtained through a repeat sternotomy in all 16 aortic patients and in 21 of the 34 mitral patients. A right thoracotomy approach was used in the other 13 mitral patients. Cardiopulmonary bypass was instituted at a flow rate of 2.2 L · m-2 · m-1 and a temperature of 28°C to 32°C. Myocardial protection was achieved with cold-blood cardioplegia antegradely and retrogradely during the first 4 years of the study and with warm-blood cardioplegia delivered similarly since 1995. Exposure of the mitral bioprosthesis was achieved by an incision in the left atrium posterior and parallel to the interatrial groove in 15 patients and through the interatrial septum in 19 patients. The aortic bioprosthesis was exposed through a repeat aortotomy. The previously implanted bioprosthesis was inspected to ascertain the magnitude of structural degeneration and to rule out the presence of perivalvular leak.
Mitral valve replacement procedure
After inspecting the valve, only the degenerated valve cusps are excised (Fig 1). The cloth-covered stent including the sewing ring of the bioprosthesis is left intact within the original mitral valve annulus; the atrial aspect of the sewing ring is cleaned of any fibrous ingrowth, and a mitral bileaflet valve (St. Jude Medical, Inc, St. Paul, MN [26 patients], or Carbomedics, SulzerMedica, Austin, TX [8 patients]) with an external diameter only 2 mm smaller than that of the mitral bioprosthesis is implanted in a "valve-on-valve" approach (Figs 1, 2) as described previously [8, 9]. Horizontal mattress sutures without pledgets are placed through the sewing ring of the mitral bioprosthesis and subsequently passed through the sewing ring of the bileaflet valve, which is oriented in a perpendicular axis to the horizontal axis of the native mitral valve (Fig 2). A red rubber catheter is inserted through the open leaflets into the ventricle to keep the valve incompetent during closure of the left atriotomy or septotomy and subsequent evacuation of air. After all air has been removed, the left atrial suture line is tied and the patient is weaned off cardiopulmonary bypass.
View larger version (50K):
[in this window]
[in a new window]
|
Fig 1. Principles of replacement of degenerated mitral bioprosthesis. (Left) All leaflet tissue is removed, and a bileaflet mechanical valve is apposed to the inflow aspect of the bioprosthetic stent. In this case, an inverted St. Jude aortic valve with extended cuff is used, and the valve holder is shown pointing to the ventricular side of the valve. (Right) Horizontal mattress sutures are taken through the latter and passed through the cuff of the bileaflet valve. Reprinted, with permission, from Geha AS, Lee JH. Evolution of the surgical approach for replacement of degenerated mitral bioprostheses. Surgery 1995;118:662–8.
|
|
View larger version (55K):
[in this window]
[in a new window]
|
Fig 2. (Left) Schematic drawing of the valve-on-valve mitral replacement; the bileaflet valve (poppets removed for clarity) sits on the atrial aspect of the bioprosthetic stent. (Top right) The cuffs of the bileaflet valve and bioprosthesis are sealed together by sutures, and the bileaflet internal orifice matches that of the bioprosthesis in the atrial view (bottom right). Reprinted, with permission, from Geha AS, Lee JH. Evolution of the surgical approach for replacement of degenerated mitral bioprostheses. Surgery 1995;118:662–8.
|
|
Aortic valve procedure
Beginning in 1995, we adapted this valve-on-valve approach to the replacement of degenerated aortic valve bioprostheses without explantation. To allow the implantation of a bileaflet valve with an internal orifice similar to that of the bioprosthesis, it is necessary to excise the degenerated valves cusps as well as the struts of the bioprosthesis (Fig 3) in order to assure a flat surface onto which the bileaflet valve can be implanted (Fig 4). The struts are removed with the help of a wire cutter, a smooth and flat outflow plane is assured, and the bileaflet valve is attached to the cloth cover of the bioprosthetic stent in a fashion similar to that used for mitral replacement (Fig 5); mattress sutures are placed in the bioprosthetic cuff to close any gap in the cloth at the site of strut removal (Fig 5B). We used a Carbomedics "top hat" aortic valve for this purpose in all 16 patients, because it ensures that adequate space is left between the valve and the coronary ostia (Fig 4). Furthermore, its flat inflow surface avoids the difficulty and potential danger of pushing the semicircular inflow "guards" of a St. Jude valve into the cuff of the bioprosthesis. The patient is then weaned off cardiopulmonary bypass after evacuation of air from the left side of the heart.
View larger version (33K):
[in this window]
[in a new window]
|
Fig 3. Principles of replacement of degenerated aortic bioprosthesis. (Left) The tissue leaflets are excised at the junction with the stent. (Right) The struts are then removed with a wire cutter and scissors, converting the scalloped outflow aspect of the bioprosthetic stent to a flat leveled one.
|
|
View larger version (46K):
[in this window]
[in a new window]
|
Fig 4. Schematic drawing of the valve-on-valve aortic replacement. The bileaflet mechanical valve is seated on the aortic (outflow) surface of the bioprosthetic stent, below the coronary ostia, having sutured the two cuffs together with mattress sutures (lower left).
|
|
View larger version (27K):
[in this window]
[in a new window]
|
Fig 5. Anchoring sutures may be placed by passing only one needle from each pair in a circumferential direction through the cuff of the bioprosthesis (A). Alternatively, each needle from every pair can be passed from the orifice side to the periphery of the bioprosthetic cuff (B). The sutures are then passed through the cuff of the bileaflet valve, as shown in Figures 1 and 4.
|
|
Patient follow-up
Follow-up was obtained through clinic visits or by telephone contact with the patients, their families, or their primary physicians. Complete follow-up was available on all patients.
|
Results
|
|---|
There was a preponderance of female patients among those having mitral valve re-replacement, and of male patients among those having aortic valve re-replacement (Table 1). Duration of cardiopulmonary bypass averaged 61 ± 14 minutes, and aortic cross-clamp time averaged 43 ± 12 minutes. There was no significant difference in bypass time and cross-clamp time between the mitral and aortic groups. Duration of follow-up varies from 7 to 117 months (mean 54 months). Successful attachment of the bileaflet valve to the incorporated old bioprosthetic valve cuff was achieved in all 50 patients.
Hospital mortality and late deaths
No operative mortality, defined as death occurring while in the hospital or within 30 days after operation, was observed among these 50 patients. Three late deaths have occurred during follow-up, at 9, 37, and 58 months. All three were due to cancer (lung, colon, prostate), and there have been no cardiac- or valve-related deaths.
All patients are maintained on warfarin to achieve an international normalized ratio level of 2.5 to 3.5. There have been no late thromboembolic or hemorrhagic complications. Transesophageal two-dimensional echocardiography was performed on all patients during their cardiological follow-up. No gradients outside the expected range for the size of the implanted bileaflet valve have been detected on transesophageal echocardiography examination. All surviving patients presently are in New York Heart Association (NYHA) functional class I or II.
|
Comment
|
|---|
Structural degeneration of bioprosthetic valves, particularly in younger patients and especially in the mitral position, remains the major deficiency of these valves [3, 10, 11]. Although improvements in bioprosthetic valves with anticalcification treatment may extend their life potential, it is expected that patients who have had a stent-mounted bioprosthetic valve, especially in the mitral position, will continue to come back with structural degeneration and dysfunction requiring a repeat operation. Although the risk of reoperative valvular surgery has gradually decreased [12–14], explantation of a bioprosthetic valve remains technically difficult, particularly when there is calcification of the tissue ingrowth, strut incorporation, or valve-to-annulus size mismatch at the primary operation. Furthermore, we had previously observed the occurrence of late perivalvular dehiscence after complete explantation of mitral bioprostheses and their stents, despite a tight fit of the bileaflet valves used to replace them, as well as transesophageal echocardiographic documentation of the absence of any perivalvular leak immediately after operation [8]. This may be due to resorption of the fibrous annular tissue that had been tightly sandwiched between the Teflon pledgets and the mitral bileaflet valve cuff at the time of replacement of the explanted mitral bioprosthesis.
Implantation of a mechanical bileaflet valve within the orifice of the stent of the bioprosthesis after excision of the bioprosthetic material has been reported in the mitral [8, 15] and aortic positions [16]. The success of this valve-in-valve technique is very limited in view of the need for a mechanical bileaflet valve 6 to 8 mm smaller in diameter than the bioprosthesis itself. It can be applied only when the original bioprosthesis is extremely large, and it also involves the risk of injury to the bileaflet valve during its seating within the frame of the bioprosthesis. Stassano and associates reported an approach somewhat similar to our valve-on-valve approach in highly selected patients, with good early and mid-term hemodynamic performance of the mechanical valves on clinical and two-dimensional color-flow echo Doppler evaluations [17, 18].
Previous work in our laboratory showed that a mechanical bileaflet valve with an external diameter 2 mm smaller than that of the mitral porcine bioprosthesis has an orifice that matches perfectly that of the bioprosthesis. The fixation of one valve to the other is easy and simple, and could be effected rapidly after excision of the bioprosthetic tissue. In the case of an aortic bioprosthesis, there has been no previous report of a valve-on-valve approach. It took us some time after the initiation of the valve-on-valve approach for degenerated mitral bioprostheses to develop the approach for degenerated aortic bioprostheses. Clearly, the valve-in-valve approach to the aortic bioprosthesis would result in implanting unacceptably small bileaflet valves in the vast majority of patients. Excision of the struts and flattening of the outflow portion of the bioprosthetic stent, however, resolves this problem and allows the implantation of a bileaflet valve of equal size to that of the previously implanted bioprosthesis with exact matching of orifice sizes. The St. Jude valve, with its guards on the inflow aspect of the cuff, can be difficult to implant, especially when the degenerated aortic bioprosthetic valve is a pericardial one, because the sewing cuff of these pericardial valves is relatively scanty and lacks the sponginess of the sewing cuff of an aortic porcine bioprosthesis. We therefore favor the use of a Carbomedics valve, particularly of the "top-hat" design for the aortic position. This also allows adequate room between the valve housing and the coronary ostia to ensure unencumbered coronary perfusion.
A few important details must be observed in anchoring the bileaflet valve onto the bioprosthetic stent. The bioprosthetic ring must be carefully cleaned of all tissue covering it in order to secure the stitches well into the Dacron material. The sutures can be taken through the bioprosthetic stent either as simple sutures along the circumference of the stent (Fig 5A) or as mattress sutures, with each needle passed from the inside towards the periphery of the stent (Fig 5B). It is important that all sutures exit the bioprosthetic stent equidistantly from the inner orifice and, similarly, should be passed through the cuff of the bileaflet valve equidistantly from its inside orifice, to assure perfect matching and exact superimposition of the orifices of the two valves.
This technique of bioprosthetic valve replacement has some limitations; the use of this approach definitely hinges on the absence of any perivalvular leak around the degenerated bioprosthesis. This should be ascertained as soon as the bioprosthesis is exposed; once certain about the integrity of the incorporation of the bioprosthetic cuff into the native annulus, expeditious removal of the entire bioprosthetic tissue can be achieved with sharp dissection, the surface of the bioprosthetic stent can be prepared, and the insertion of the new valve can be expeditiously completed. Obviously, the patient must have no contraindication to anticoagulation, which is required after this procedure. In our experience, however, the majority of these patients had been chronically anticoagulated because of the high prevalence of atrial fibrillation, particularly among the mitral valve patients.
Leaving the old bioprosthetic cloth-covered stent in place eliminates the need for extensive dissection and results in minimal trauma to the tissue. It virtually eliminates the risk of injury to the heart or the occurrence of late perivalvular leaks. Thus, it shortens and simplifies the procedure and diminishes its attendant mortality and morbidity. The lack of any operative death in our relatively older patient series, despite the fact that some of these patients have had multiple previous cardiac procedures, and the excellent long-term results on clinical follow-up and echocardiographic evaluation, have led us to continue to use this approach as a method of choice in reoperations for replacement of degenerated bioprosthetic cardiac valves.
|
Acknowledgments
|
|---|
We acknowledge the help of Kymberly Estrada in the preparation of the manuscript.
|
References
|
|---|
-
Hammond G.L., Geha A.S., Kpof G.S., Hashim S.W. Biological versus mechanical valves. J Thorac Cardiovasc Surg 1987;93:182-198.[Abstract]
-
Burdon T.A., Miller D.C., Oyer P.E., et al. Durability of porcine valves at fifteen years in a representative North American population. J Thorac Cardiovasc Surg 1992;103:238-252.[Abstract]
-
Magilligan D.J., Lewis J.W., Stein P., Alam M. The porcine bioprosthetic heart valve: experience at 15 years. Ann Thorac Surg 1989;48:324-330.[Abstract]
-
Akins C.W., Carroll D.L., Buckley M.J., Daggett W.M., Hilgenberg A.D., Austen W.G. Late results with Carpentier-Edwards porcine bioprosthesis. Circulation 1990;82:65-74.
-
Pansini S., Ottino G., Forsennati P.G., et al. Reoperations on heart valve prostheses: an analysis of operative risks and late results. Ann Thorac Surg 1990;50:590-596.[Abstract]
-
Husebye D.G., Pluth J.R., Piehler J.M., et al. Reoperation on prosthetic heart valves. J Thorac Cardiovasc Surg 1983;86:543-552.[Abstract]
-
Virmani R., Chun P.K.C., Parker J., et al. Suture obliteration of the circumflex coronary artery in three patients undergoing mitral valve operation. J Thorac Cardiovasc Surg 1982;84:773-778.[Abstract]
-
Geha A.S., Lee J.H. Evolution of the surgical approach for replacement of degenerated mitral bioprostheses. Surgery 1995;118:662-668.[Medline]
-
Geha A.S., Lee J.H. New approach for replacement of degenerated mitral bioprostheses. Eur J Cardiothorac Surg 1996;10:1090-1096.[Abstract]
-
Gallo I., Nistal F., Blasquez R., Arbe E., Artinano E. Incidence of primary tissue valve failure in porcine bioprosthetic heart valves. Ann Thorac Surg 1988;45:66-70.[Abstract]
-
Foster A.H., Greenberg G.J., Underhill D.J., McIntosh C.L., Clark R.E. Intrinsic failure of Hancock mitral bioprostheses: 10–15 year experience. Ann Thorac Surg 1987;44:568-577.[Abstract]
-
Syracuse D.C., Bowman F.O., Malm J.R. Prosthetic valve reoperations: factors influencing early and late survival. J Thorac Cardiovasc Surg 1979;77:346-354.[Medline]
-
Guynn T., Najafi C., Alden T. Declining risk of reoperative valvular surgery. J Card Surg 1995;10:185-197.[Medline]
-
Vogt P.R., Bruner-LaRoccah, Sidler P., et al. Reoperative surgery for degenerated aortic bioprostheses: predictors for emergency surgery and reoperative mortality. Eur J Cardiothorac Surg 2000;17:134-139.[Abstract/Free Full Text]
-
Paterson H.S., Jacob A., Campanella C., Bloomfield P., Cameron E.W. Retention of bioprosthetic valve annulus in mitral prosthetic replacement. Eur J Cardiothorac Surg 1993;7:511-513.[Abstract]
-
Najafi H. Aortic valve replacement 13 years following Bentall procedure. J Card Surg 1995;10:52-54.[Medline]
-
Stassano P., Losi M.A., Golino A., et al. Bioprosthesis replacement with mechanical valve implantation on the bioprosthetic ring. Surgical and 2D echo considerations. Eur J Cardiothorac Surg 1993;7:507-510.[Abstract]
-
Stassano P., Musumei A., Losi M.A., Gagliardi C., Spampinato N. Midterm results of the valve-on-valve technique for bioprosthetic failure. Eur J Cardiothorac Surg 2000;18:453-457.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
L. A. Vricella and D. E. Cameron
Facile conversion from mechanical to bioprosthetic composite aortic root replacement
J. Thorac. Cardiovasc. Surg.,
February 1, 2007;
133(2):
565 - 567.
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. Kurisu, Y. Ochiai, T. Kajiwara, H. Kumeda, and R. Tominaga
A modified valve-on-valve approach for aortic root replacement
Ann. Thorac. Surg.,
December 1, 2003;
76(6):
2099 - 2101.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J.-P. Chang, C.-L. Kao, K.-T. Tsai, M.-J. Hsieh, and M. Fu
Conduit-on-valve replacement of a degenerated mitral bioprosthesis with a bioprosthesis
Ann. Thorac. Surg.,
June 1, 2003;
75(6):
1987 - 1989.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
