- Email: [email protected]
Morphologic observations on heart valve prostheses made of fascia lata
Morphologic observations on heart valve prostheses made of fascia lata A cooperative, international study permitted morphologic observations on 58 fascia lata heart valve prostheses. Thirty-nine had been in position less than a year, the remainder for periods up to 50 months. The fascial cusps, no matter whether free or stent mounted, underwent changes that included death of the original fascia and vacuolar, degenerative changes if a valve had been in situ more than 3 years. Formation of a pseudointima on a cusp's surface caused it to shrink. The rate of progression and severity of the changes varied from prosthesis to prosthesis. Severe cusp shrinkage caused valve incompetence. That seemed particularly likely in mitral and tricuspid prostheses and explains why their use has been abandoned. Clinically, fascia lata aortic valves function satisfactorily for several years. However, their long-term value is still uncertain. Morphologically, the specter of late calcification and other degenerative changes haunt their use until more, from long-term survivors, are available for study.
Malcolm D. Silver, M.D., Reginald E. B. Hudson, M.D., and Alan S. Trimble, M.D., Toronto, Ontario, Canada, and Shoreham-by-Sea, England
he
impetus for this study came from the First International Roundtable on Fascia Lata Heart Valves held at Toronto, Ontario, Canada, in 1970. 1 One of the conclusions of those attending was as follows: "Because of the limited pathologic data available . . . a Central Registry be organized in Toronto, so that specimens from around the world could be collected and evaluated by the two participating pathologists." This Registry was organized, and the following morphologic observations have been made on 58 valves contributed to it. Organization and techniques Letters were sent to all surgeons known to have inserted fascia lata prostheses asking that they forward valves excised at surgery or recovered at autopsy to the RegisFrom the Departments of Pathology and Cardiovascular Surgery, Toronto General Hospital and University of Toronto, Toronto, Ontario, Canada. Supported by grants from the Medical Research Council of Canada and the Ontario Heart Foundation.
Received for publication March 6, 1975.
360
try. The contributors completed a form listing data about each patient, the method of valve preparation, and the reason for its recovery. In addition, black or white sutures were placed in specific valve cusps to assist orientation. On arrival in Toronto, the valves, which had been fixed in 10 per cent formaldehyde solution, were numbered and presented without clinical data to a pathologist. Each valve was photographed and a gross description made. Then, each cusp was sectioned, with the section passing from the free margin of the cusp to the base at the midpoint. Any other lesion on a cusp, such as vegetations of infective endocarditis, was also sectioned. Fascia lata is a tough and difficult tissue to cut after ordinary paraffin embedding. The histologic study was facilitated by initially soaking each cusp section 24 hours in a 50/50 solution of strong ammonia in 70 per cent ethyl alcohol. Next, it was soaked 8 hours in a 5 per cent phenol solution made up in 70 per cent ethyl alcohol. Finally, it was dehydrated through graded alcohol solu-
Volume 70 Number 2 August. 1975
Fascia lata valve prostheses
361
tions, each containing 4 per cent phenol, before being embedded in paraffin. Duplicate histologic sections were stained with hematoxylin and eosin, a combined Verhoeff's elastic-Masson's trichrome stain and with the World Health Organization stain (combined Masson and alcian green). Initially, the histologic material was examined independently by the pathologists (M. D. S. and R. E. B. H.). Subsequently, they compared findings and reached a consensus. Finally, clinicopathological correlations were made. Results
Table I lists the length of time that each valve was in situ and its anatomic position. It should be noted that four of the valves were not stented but were free fascial grafts. Their morphologic changes will be included in this review and commented upon later. All other valves had been placed on stents by means of the technique initially described by Ionescu and Ross." Macroscopic observations. Prostheses recovered soon after operation had a deposit of platelet/fibrin material on their surfaces (Fig. 1). The subsequent organization of this material led to cusp thickening. The process occurred on the cusps in all prostheses and on exposed cloth surfaces of stent-rnounted prostheses. It produced a gray-white surface layer of connective tissue that presumably became coated with endothelium. The newly formed connective tissue thickened the cusps in their basal half or third. Frequently, the thickening was distributed across the surface of a cusp in a crescentic fashion, with the horns of the crescent extending toward the commissural areas. The latter change, if marked, could cause some fusion of adjacent sides of cusps at the commissures. The thickening also increased from the free margin of the cusp toward its base, with an accentuation where the cusp met the stent. The deposits at the latter site produce a smooth junction between the two. A comparable smoothing occurred at the junction between prosthesis and host tissue.
Fig. 1. Fascia lata aortic valve recovered 1 week after insertion. Note slight redundancy of valve cusps forming a sinuous line of closure; flecks of fibrin on cusp surface. Scale indicates 1 em.
Table I Duration in situ (mo.)
Tricuspid
0-6 7-12 13-18 19-24 25-30 31-36 37-42 43-48 49-54
3 0 I 0 I 0 0 0 0
12 3 2 I 2 2 0 0 0
20* 1* 2 2* 0 2 0 3 1
35 4 5 3 3 4 0 3 I
5
22
31
58
Totals
Valve position
I Mitral I Aortic
Total
·Four free fascial grafts were removed from the aortic area 10 weeks and 5, 9, and 21 months after insertion.
Within 1 to 2 months of insertion, the deposit of newly formed connective tissue caused a rigidity of the previously thin, pliable fascial cusp and led to some cusp shrinkage (Fig. 2). If one or two of the cusps became markedly shrunken, the prosthesis developed a fixed triangular orifice and became incompetent (Fig. 3). The changes affected prostheses removed from all valve sites and did not seem to be confined to any particular cusp. The thin, free margin of a cusp from a
The Journal 01
362
Silver, Hudson, Trimble
Fig. 2. Aortic valve excised after 50 months in situ because of incompetence. The cusps are slightly shrunken and did not coapt completely. Their edges are fr ayed . (Valve provided courtesy of Dr. H. Bahnson.) Scale indicates I em.
prosthesis recovered soon after surgery might be somewhat frayed . Later, with the organization of the plat elet/fibrin deposits , a similar cusp would be thickened. Retraction of the thickened margin could impair accurate cusp apposition and lead to prosthesis incompetence. Tiny filiform tags often appeared on the cusp surface, close to its free margin, at the line of apposition. They resembled LambI's excrescences, which can appear along the free margins of any thickened heart valve and probably are formed in the same manner." No macroscopic calcification was detected in any prosthesis. Apart from the process of platele t/fibrin deposition, no exuberant thrombus formation was seen in the absence of infection. Microscopic observations. Normal fascia lata consists of at least three layers of relatively acellular, dense collagen. The collagen fibers in each layer run parallel but at an angle to an adjacent layer. Small blood vessels, sometimes surrounded by mononuclear cells, are seen between the collagen
Thoracic ond Cardiovascular Surgery
Fig. 3. Ventricular aspect of fascia lata mitral prosthesis in situ for 5 months. Note two retracted cusps (arrows) producing a fixed incompetent orifice. Scale indicates 1 em .
layers and have elastic fibrillas related to them . After preparation for surgical insertion, one surface of the fascia is smooth, shiny, and virtually devoid of areolar tissue. The other is roughened and may have tags of areolar tissue, containing small blood vessels, attached to it. The earliest change after valve insertion appeared to be some swelling of the fascia due to edema . In the following weeks to months, the number of original cell nuclei in the fascia diminished and then increased. The increase was due to an invasion of fibroblasts and macrophages that accompanied small blood vessels. The latter traversed the fascia from one surface to another, mainly toward the base of a cusp. After a prosthesis had been in place for many months up to a year , the number of cell nuclei found in the fascia again diminished so that large areas of collagen become faintly acidophilic and acellular. Vascular channels too were less obvious (Fig. 4). The rate of occurrence of these changes and their severity varied somewhat from one valve to another. A vacuolar degeneration of the acellular collagen was observed in the cusps
Volume 70 Number 2 August, 1975
of one prosthesis in situ for 5 months and in those of four prostheses in situ longer than 3 years (Fig. 5). Tiny platelet/fibrin deposits were found on the surface of the fascial cusps within hours of insertion. They increased in number and thickness, especially toward the bases of the cusps. Eventually, they covered almost all of the exposed surface of each cusp. With time, they were organized to form a connective tissue layer that coated both surfaces of a cusp and thickened it (Fig. 4). Initially, the cells of this layer were loosely arranged, being separated by an abundant amount of mucopolysaccharide material; later , the layer seemed to condense and the amount of mucopolysaccharide material was much reduced . Distinct smooth muscle cells could be found in the layer after a prosthesis had been in situ from 12 to 36 months. Again, the extent of the changes described varied from one valve to another. Small blood vessels and foreign bodies , often associated with giant cells and having the morphologic characteristics of glove powder, could be observed at the junction between fascia and overlying connective tissue. Early calcification was seen in one prosthesis removed after 5 months. It appeared as fine granules in the fibrous tissue deposit covering the cusps, especially in areas where this deposit was excessive. Microscopic calcification was also observed in the original fascia in cusps of one other prosthesis recovered at autopsy after 43 months (Fig . 6) . Tiny collections of foam cells were apparent in the connective tissue layer covering the cusps in one prosthesis removed after 3 years. Complicating factors. Valve dysfunction and incompetence was caused by a cusp separating from the pillars of a strut in two instances (valves removed after 3 and 36 months). Progressive weakening in the area of the suture line used to form the initial cylinder may have been a fault that caused separation at 3 months . The collagen of the
Fascia lata valve prostheses
363
Fig. 4. Prosthesis in situ for 12 months. Note paucity of cell nuclei in original fascia and pseudointima (P) thickening the cusp surfaces. The luminal surface is to the left. (Hem atoxylin and eosin x20.)
cusps of the other prosthesis showed severe vacuolar degeneration. No other cause for separation was obvious. In another valve removed after 6 months, there was a hole in the belly of a cusp. The perforation was not caused by an infection but was probably due to a weakening of the tissue at the site of perforation of a nutrient blood vessel. In five prostheses, infection had caused valve incompetence either as a result of cusp ulceration or perforation (Fig. 7). These complications presented between 5 weeks and 45 months after operation. The infections were caused by Aspergillus in one instance, an unidentified fungus morphologically like Candida in another, by Staph ylococcus a/bus in two others, and by a grampositive coccus, probably a Staphylococcus, in the fifth case. Clinical data. Surgeons and pathologists from nine centers contributed the 58 prostheses to the Registry . They came from 51 patients. Of these, 29 had undergone single aortic valve replacement, 16 single mitral valve replacement, 5 had had double valve
The Journo l of
364
Thorocic ond Cordiovosculo r
Silver, Hudson , Trimble
Fig. S. Vacuolar degeneration in acellular fascia of an aortic prosthesis in situ for 44 months. (Hematoxylin and eosin x50.)
replacements ( 4 mitral and tricuspid; 1 mitral and aortic), and 1 patient had had a triple valve replacement. In preparation, the fascia of all valves was washed either in a saline solution or in a saline/antibiotic mixture. The valves recovered from 22 patients had functioned normally. The majority of these individuals had died relatively early after surgery from other complicating factors, 16 patients in the first month. Two of the others died 5 weeks after valve insertion, 1 each died 6 and 10 weeks later, and 1 each died 5 and 19 months after insertion. The - prosthetic valves from the other 29 patients were incompetent. Unfortunately, data are incomplete as to whether the incompetence developed soon after surgery or later. Valve incompetence could be attributed to one of the three factors listed in Table II. In the majority of instances, one or more cusps had shrunk to produce a fixed, incompetent orifice. In three, technical faults associated with construction were presumed to be a causative factor, and in
Surgery
Fig. 6. Calcific ation (black granules) in dead fascia of aortic prosthesis inserted 43 months previously. The clefts in the fascia in this section are artifacts. (Hematoxylin and eosin x50.)
five values infective endocarditis was the cause. The pathological findings in these latter valves have been described previously. Discussion
Observations on some of the prostheses included in this study have been reported elsewhere.':" Those studies dealt either with valves in situ for short periods or reported on a few valves that were in position a longer time. The Registry had the advantages of a larger number of specimens and an unbiased review by the two pathologists. However, a certain bias exists in the study, in that malfunction or early death was the most frequent reason for the submission of the prostheses. We have defined certain consistent gross and microscopic changes in stent-mounted fascia lata heart prosth eses that permit some conclusions, but we emphasize that this is an interim report. For example, only 19 of the prostheses studied had been in place longer than 12 months. Thus an assessment of the ultimate fate of this type of pro sthesis
Volume 70 Number 2
Fascia lata valve prostheses
365
August, 1975
and of its worth in the treatment of valvular heart disease must continue. The original fascia does not survive indefinitely, even though Willen and his colleagues' demonstrated that fascial tissue in situ for up to 41 months was able to synthesize ribonucleic acid. Rather, it acts as a framework upon which connective tissue forms as a result of the organization of thrombus. This fibrous buttressing seems to maintain the strength of the cusp for a number of years. Nevertheless, degenerative changes take place in the collagen of the fascia after prolonged insertion (3 or more years). Senning," a pioneer in the use of this type of valve, recently reported: "A patchy necrosis appears in the inside of the valve, apparently as a result of malnutrition."!" Such degenerative changes could explain the stretching of cusps, with resultant poor coaptation and aortic valve incompetence, reported by Schumacher," and the cusp perforation and separation of cusp from strut pillars observed as late complications by others. H. ]2 Our observations indicate that there is inevitable thickening of the fascia lata cusp tissue resulting from the organization of surface thrombus, even though a valve may function satisfactorily for years. In the longterm period, and as with homologous aortic valve replacements," a gradual increase in this tissue may eventually interfere with cusp function. Indeed, in Senning's'v opinion this problem increases the longer a fascial valve is in position. Shrinkage of the fibrous coating of the fascia lata cusp is unavoidable, but its degree depends upon the thickness of the fibrous tissue; this, in turn, may depend upon the skill of the surgeon preparing and inserting the prosthesis. With experience, surgeons performing these operations have tended to mold a more redundant valve to allow for the shrinkage that occurs during the first 3 months in situ. Very careful cleaning of the areolar tissue has reduced the degree of superficial thrombosis and ultimate thickening by connective tissue, and perfect approximation of free margins and commissures has led to a totally com-
Fig. 7. Vegetation (V) of infective endocarditis affecting aortic prosthesis placed in position 5 weeks earlier. Scale indicates 1 em.
Table II. Causes of valve incompetence Cause
No. ofpts. affected
Cusp retraction Technical molding Infective endocarditis
21 3 5
petent valve. Whether this attention to detail in molding will affect the morphologic changes and the ultimate durability of these valves remains to be seen. The cusps of the four free fascial grafts studied showed changes similar to those of stent-mounted prostheses. Calcification was not a serious problem in the valves studied, but one should be cautious on this point. Late calcification has proved a problem in the dead tissue of aortic homografts."- 14 Again, Senning!" has reported stenosis and calcification of valves in position more than 8 years. Exuberant thrombus formation does not occur on fascia lata valves in the absence of infection. Therefore, thromboembolism should not be a serious complication. This conclusion has been supported by clinical experience in some series,": 10 although not in that of Joseph and colleagues."
366
Silver, Hudson, Trimble
Conclusions The cusps of heart valve prostheses made of fascia lata, no matter whether free or stent mounted, undergo a series of recognizable morphologic changes. These include death of the original fascia with vacuolar degenerative changes if valves have been in situ more than 3 years, the formation of a pseudointima on the surface of the cusps, and resultant shrinkage. The rate at which these changes occur and their severity vary from prosthesis to prosthesis. Cusp shrinkage has been so extensive in tricuspid and mitral prostheses that their use has been abandoned. This type of valve has been shown to function satisfactorily in the aortic area for a number of years," but its value over a long period is still uncertain. The specter of late calcification and other late degenerative changes haunts its use until detailed, morphologic studies of long-term survivors are available. We are deeply grateful to Drs. H. T. Bahnson, Pittsburgh, Pa.; C. Cheanvechai, L. Groves, and D. Effler, Cleveland, Ohio; W. T. Dubiel, Uppsala, Sweden; J. A. Helmsworth, Cincinnati, Ohio; M. I. Ionescu, Leeds, England; G. Rodewald and U. Kirsch, Hamburg, West Germany; H. Shurnacker, Indianapolis, Ind.; and C. Baker and J. K. Yao, Toronto, Ontario, Canada, for submitting material to the Registry. We thank Mrs. M. Lorber and Mr. P. Horne for technical assistance. REFERENCES Trimble, A. S. (Editor): First International Roundtable on Fascia Lata Heart Valves, Surgery 71: 10, 1972. 2 Ionescu, M. I., and Ross, D. N.: Heart-Valve Replacement With Autologous Fascia Lata, Lancet 2: 335, 1969.
The Journal of Thorocic ond Cardiovascu lor Surgery
3 Magarey, F. R.: On the Mode of Formation of Lambl's Excrescences and Their Relation to Chronic Thickening of the Mitral Valve, J. Pathol. 61: 203, 1949. 4 Dubiel, W. T.: Aortic Valve Replacement With Frame-Supported Autologous Fascia Lata, Acta Univ. Upsalensis 189: 1, 1974. 5 Dubiel, W. T., Johansson, L., and Willen, R.: Postoperative Changes in Autologous Fascia Lata Heart Valve Grafts: A Pathological Study, Ann. Thorac. Surg. 15: 140, 1973. 6 Ionescu, M. I., Pakrashi, B. C., Mary, D. A. S., Bartek, I. T., and Wooler, G. H.: Long-Term Evaluation of Tissue Valves, J. THORAc. CARDIOVASC. SURG. 68: 361, 1974. 7 Silver, M. D., and Trimble, A. S.: Structure of Autologous Fascia Lata Heart Valve Prostheses, Arch. Pathol. 93: 109. 1972. 8 Willen, R., Dubiel, W. T., and Johansson, L.: Viability and Surface Properties of Fascia Lata Heart Valve Grafts, Ann. Thorac. Surg. 18: 597, 1974. 9 Senning, A.: Fascia Lata Replacement of Aortic Valves, J. THoRAc. CARDIOVASC. SURG. 54: 465, 1967. 10 Senning, A.: Discussion of Ionescu et a1.6 : J. THORAc. CARDIOVASC. SURG. 68: 379, 1974. 11 Shumacker, H. B: Discussion of Ionescu et a1.6 : J. THORAc. CARDIOVASC. SURG. 68: 378, 1974. 12 Joseph, S., Somerville, J., Emanuel, R., Ross, D., and Ross, K.: Aortic Valve Replacement With Frame-Mounted Autologous Fascia Lata: Long-Term Results, Br. Heart J. 36: 760, 1974. 13 Hudson, R. E. B.: Pathology of Heart Valve Grafting, in Ross, D. N., and Wooler, G. H., editors: Biological Tissue in Heart Valve Replacement, London, 1972, Butterworth & Co., Ltd., p. 835. 14 Hudson, R. E. B.: Cardiovascular Pathology, London, Edward Arnold (Publishers) Ltd., 3: S-596, 1970. 15 Trimble, A. S., Gunstensen, J., Silver, M. D., Aldridge, H. E., Schwartz, L., and Morch, J. E.: Aortic Valve Replacement With Fascia Lata: An Encouraging Late Study, J. THoRAc. CARDIOVASC. SURG. 68: 219, 1974.
Malcolm D. Silver, M.D., Reginald E. B. Hudson, M.D., and Alan S. Trimble, M.D., Toronto, Ontario, Canada, and Shoreham-by-Sea, England
he
impetus for this study came from the First International Roundtable on Fascia Lata Heart Valves held at Toronto, Ontario, Canada, in 1970. 1 One of the conclusions of those attending was as follows: "Because of the limited pathologic data available . . . a Central Registry be organized in Toronto, so that specimens from around the world could be collected and evaluated by the two participating pathologists." This Registry was organized, and the following morphologic observations have been made on 58 valves contributed to it. Organization and techniques Letters were sent to all surgeons known to have inserted fascia lata prostheses asking that they forward valves excised at surgery or recovered at autopsy to the RegisFrom the Departments of Pathology and Cardiovascular Surgery, Toronto General Hospital and University of Toronto, Toronto, Ontario, Canada. Supported by grants from the Medical Research Council of Canada and the Ontario Heart Foundation.
Received for publication March 6, 1975.
360
try. The contributors completed a form listing data about each patient, the method of valve preparation, and the reason for its recovery. In addition, black or white sutures were placed in specific valve cusps to assist orientation. On arrival in Toronto, the valves, which had been fixed in 10 per cent formaldehyde solution, were numbered and presented without clinical data to a pathologist. Each valve was photographed and a gross description made. Then, each cusp was sectioned, with the section passing from the free margin of the cusp to the base at the midpoint. Any other lesion on a cusp, such as vegetations of infective endocarditis, was also sectioned. Fascia lata is a tough and difficult tissue to cut after ordinary paraffin embedding. The histologic study was facilitated by initially soaking each cusp section 24 hours in a 50/50 solution of strong ammonia in 70 per cent ethyl alcohol. Next, it was soaked 8 hours in a 5 per cent phenol solution made up in 70 per cent ethyl alcohol. Finally, it was dehydrated through graded alcohol solu-
Volume 70 Number 2 August. 1975
Fascia lata valve prostheses
361
tions, each containing 4 per cent phenol, before being embedded in paraffin. Duplicate histologic sections were stained with hematoxylin and eosin, a combined Verhoeff's elastic-Masson's trichrome stain and with the World Health Organization stain (combined Masson and alcian green). Initially, the histologic material was examined independently by the pathologists (M. D. S. and R. E. B. H.). Subsequently, they compared findings and reached a consensus. Finally, clinicopathological correlations were made. Results
Table I lists the length of time that each valve was in situ and its anatomic position. It should be noted that four of the valves were not stented but were free fascial grafts. Their morphologic changes will be included in this review and commented upon later. All other valves had been placed on stents by means of the technique initially described by Ionescu and Ross." Macroscopic observations. Prostheses recovered soon after operation had a deposit of platelet/fibrin material on their surfaces (Fig. 1). The subsequent organization of this material led to cusp thickening. The process occurred on the cusps in all prostheses and on exposed cloth surfaces of stent-rnounted prostheses. It produced a gray-white surface layer of connective tissue that presumably became coated with endothelium. The newly formed connective tissue thickened the cusps in their basal half or third. Frequently, the thickening was distributed across the surface of a cusp in a crescentic fashion, with the horns of the crescent extending toward the commissural areas. The latter change, if marked, could cause some fusion of adjacent sides of cusps at the commissures. The thickening also increased from the free margin of the cusp toward its base, with an accentuation where the cusp met the stent. The deposits at the latter site produce a smooth junction between the two. A comparable smoothing occurred at the junction between prosthesis and host tissue.
Fig. 1. Fascia lata aortic valve recovered 1 week after insertion. Note slight redundancy of valve cusps forming a sinuous line of closure; flecks of fibrin on cusp surface. Scale indicates 1 em.
Table I Duration in situ (mo.)
Tricuspid
0-6 7-12 13-18 19-24 25-30 31-36 37-42 43-48 49-54
3 0 I 0 I 0 0 0 0
12 3 2 I 2 2 0 0 0
20* 1* 2 2* 0 2 0 3 1
35 4 5 3 3 4 0 3 I
5
22
31
58
Totals
Valve position
I Mitral I Aortic
Total
·Four free fascial grafts were removed from the aortic area 10 weeks and 5, 9, and 21 months after insertion.
Within 1 to 2 months of insertion, the deposit of newly formed connective tissue caused a rigidity of the previously thin, pliable fascial cusp and led to some cusp shrinkage (Fig. 2). If one or two of the cusps became markedly shrunken, the prosthesis developed a fixed triangular orifice and became incompetent (Fig. 3). The changes affected prostheses removed from all valve sites and did not seem to be confined to any particular cusp. The thin, free margin of a cusp from a
The Journal 01
362
Silver, Hudson, Trimble
Fig. 2. Aortic valve excised after 50 months in situ because of incompetence. The cusps are slightly shrunken and did not coapt completely. Their edges are fr ayed . (Valve provided courtesy of Dr. H. Bahnson.) Scale indicates I em.
prosthesis recovered soon after surgery might be somewhat frayed . Later, with the organization of the plat elet/fibrin deposits , a similar cusp would be thickened. Retraction of the thickened margin could impair accurate cusp apposition and lead to prosthesis incompetence. Tiny filiform tags often appeared on the cusp surface, close to its free margin, at the line of apposition. They resembled LambI's excrescences, which can appear along the free margins of any thickened heart valve and probably are formed in the same manner." No macroscopic calcification was detected in any prosthesis. Apart from the process of platele t/fibrin deposition, no exuberant thrombus formation was seen in the absence of infection. Microscopic observations. Normal fascia lata consists of at least three layers of relatively acellular, dense collagen. The collagen fibers in each layer run parallel but at an angle to an adjacent layer. Small blood vessels, sometimes surrounded by mononuclear cells, are seen between the collagen
Thoracic ond Cardiovascular Surgery
Fig. 3. Ventricular aspect of fascia lata mitral prosthesis in situ for 5 months. Note two retracted cusps (arrows) producing a fixed incompetent orifice. Scale indicates 1 em .
layers and have elastic fibrillas related to them . After preparation for surgical insertion, one surface of the fascia is smooth, shiny, and virtually devoid of areolar tissue. The other is roughened and may have tags of areolar tissue, containing small blood vessels, attached to it. The earliest change after valve insertion appeared to be some swelling of the fascia due to edema . In the following weeks to months, the number of original cell nuclei in the fascia diminished and then increased. The increase was due to an invasion of fibroblasts and macrophages that accompanied small blood vessels. The latter traversed the fascia from one surface to another, mainly toward the base of a cusp. After a prosthesis had been in place for many months up to a year , the number of cell nuclei found in the fascia again diminished so that large areas of collagen become faintly acidophilic and acellular. Vascular channels too were less obvious (Fig. 4). The rate of occurrence of these changes and their severity varied somewhat from one valve to another. A vacuolar degeneration of the acellular collagen was observed in the cusps
Volume 70 Number 2 August, 1975
of one prosthesis in situ for 5 months and in those of four prostheses in situ longer than 3 years (Fig. 5). Tiny platelet/fibrin deposits were found on the surface of the fascial cusps within hours of insertion. They increased in number and thickness, especially toward the bases of the cusps. Eventually, they covered almost all of the exposed surface of each cusp. With time, they were organized to form a connective tissue layer that coated both surfaces of a cusp and thickened it (Fig. 4). Initially, the cells of this layer were loosely arranged, being separated by an abundant amount of mucopolysaccharide material; later , the layer seemed to condense and the amount of mucopolysaccharide material was much reduced . Distinct smooth muscle cells could be found in the layer after a prosthesis had been in situ from 12 to 36 months. Again, the extent of the changes described varied from one valve to another. Small blood vessels and foreign bodies , often associated with giant cells and having the morphologic characteristics of glove powder, could be observed at the junction between fascia and overlying connective tissue. Early calcification was seen in one prosthesis removed after 5 months. It appeared as fine granules in the fibrous tissue deposit covering the cusps, especially in areas where this deposit was excessive. Microscopic calcification was also observed in the original fascia in cusps of one other prosthesis recovered at autopsy after 43 months (Fig . 6) . Tiny collections of foam cells were apparent in the connective tissue layer covering the cusps in one prosthesis removed after 3 years. Complicating factors. Valve dysfunction and incompetence was caused by a cusp separating from the pillars of a strut in two instances (valves removed after 3 and 36 months). Progressive weakening in the area of the suture line used to form the initial cylinder may have been a fault that caused separation at 3 months . The collagen of the
Fascia lata valve prostheses
363
Fig. 4. Prosthesis in situ for 12 months. Note paucity of cell nuclei in original fascia and pseudointima (P) thickening the cusp surfaces. The luminal surface is to the left. (Hem atoxylin and eosin x20.)
cusps of the other prosthesis showed severe vacuolar degeneration. No other cause for separation was obvious. In another valve removed after 6 months, there was a hole in the belly of a cusp. The perforation was not caused by an infection but was probably due to a weakening of the tissue at the site of perforation of a nutrient blood vessel. In five prostheses, infection had caused valve incompetence either as a result of cusp ulceration or perforation (Fig. 7). These complications presented between 5 weeks and 45 months after operation. The infections were caused by Aspergillus in one instance, an unidentified fungus morphologically like Candida in another, by Staph ylococcus a/bus in two others, and by a grampositive coccus, probably a Staphylococcus, in the fifth case. Clinical data. Surgeons and pathologists from nine centers contributed the 58 prostheses to the Registry . They came from 51 patients. Of these, 29 had undergone single aortic valve replacement, 16 single mitral valve replacement, 5 had had double valve
The Journo l of
364
Thorocic ond Cordiovosculo r
Silver, Hudson , Trimble
Fig. S. Vacuolar degeneration in acellular fascia of an aortic prosthesis in situ for 44 months. (Hematoxylin and eosin x50.)
replacements ( 4 mitral and tricuspid; 1 mitral and aortic), and 1 patient had had a triple valve replacement. In preparation, the fascia of all valves was washed either in a saline solution or in a saline/antibiotic mixture. The valves recovered from 22 patients had functioned normally. The majority of these individuals had died relatively early after surgery from other complicating factors, 16 patients in the first month. Two of the others died 5 weeks after valve insertion, 1 each died 6 and 10 weeks later, and 1 each died 5 and 19 months after insertion. The - prosthetic valves from the other 29 patients were incompetent. Unfortunately, data are incomplete as to whether the incompetence developed soon after surgery or later. Valve incompetence could be attributed to one of the three factors listed in Table II. In the majority of instances, one or more cusps had shrunk to produce a fixed, incompetent orifice. In three, technical faults associated with construction were presumed to be a causative factor, and in
Surgery
Fig. 6. Calcific ation (black granules) in dead fascia of aortic prosthesis inserted 43 months previously. The clefts in the fascia in this section are artifacts. (Hematoxylin and eosin x50.)
five values infective endocarditis was the cause. The pathological findings in these latter valves have been described previously. Discussion
Observations on some of the prostheses included in this study have been reported elsewhere.':" Those studies dealt either with valves in situ for short periods or reported on a few valves that were in position a longer time. The Registry had the advantages of a larger number of specimens and an unbiased review by the two pathologists. However, a certain bias exists in the study, in that malfunction or early death was the most frequent reason for the submission of the prostheses. We have defined certain consistent gross and microscopic changes in stent-mounted fascia lata heart prosth eses that permit some conclusions, but we emphasize that this is an interim report. For example, only 19 of the prostheses studied had been in place longer than 12 months. Thus an assessment of the ultimate fate of this type of pro sthesis
Volume 70 Number 2
Fascia lata valve prostheses
365
August, 1975
and of its worth in the treatment of valvular heart disease must continue. The original fascia does not survive indefinitely, even though Willen and his colleagues' demonstrated that fascial tissue in situ for up to 41 months was able to synthesize ribonucleic acid. Rather, it acts as a framework upon which connective tissue forms as a result of the organization of thrombus. This fibrous buttressing seems to maintain the strength of the cusp for a number of years. Nevertheless, degenerative changes take place in the collagen of the fascia after prolonged insertion (3 or more years). Senning," a pioneer in the use of this type of valve, recently reported: "A patchy necrosis appears in the inside of the valve, apparently as a result of malnutrition."!" Such degenerative changes could explain the stretching of cusps, with resultant poor coaptation and aortic valve incompetence, reported by Schumacher," and the cusp perforation and separation of cusp from strut pillars observed as late complications by others. H. ]2 Our observations indicate that there is inevitable thickening of the fascia lata cusp tissue resulting from the organization of surface thrombus, even though a valve may function satisfactorily for years. In the longterm period, and as with homologous aortic valve replacements," a gradual increase in this tissue may eventually interfere with cusp function. Indeed, in Senning's'v opinion this problem increases the longer a fascial valve is in position. Shrinkage of the fibrous coating of the fascia lata cusp is unavoidable, but its degree depends upon the thickness of the fibrous tissue; this, in turn, may depend upon the skill of the surgeon preparing and inserting the prosthesis. With experience, surgeons performing these operations have tended to mold a more redundant valve to allow for the shrinkage that occurs during the first 3 months in situ. Very careful cleaning of the areolar tissue has reduced the degree of superficial thrombosis and ultimate thickening by connective tissue, and perfect approximation of free margins and commissures has led to a totally com-
Fig. 7. Vegetation (V) of infective endocarditis affecting aortic prosthesis placed in position 5 weeks earlier. Scale indicates 1 em.
Table II. Causes of valve incompetence Cause
No. ofpts. affected
Cusp retraction Technical molding Infective endocarditis
21 3 5
petent valve. Whether this attention to detail in molding will affect the morphologic changes and the ultimate durability of these valves remains to be seen. The cusps of the four free fascial grafts studied showed changes similar to those of stent-mounted prostheses. Calcification was not a serious problem in the valves studied, but one should be cautious on this point. Late calcification has proved a problem in the dead tissue of aortic homografts."- 14 Again, Senning!" has reported stenosis and calcification of valves in position more than 8 years. Exuberant thrombus formation does not occur on fascia lata valves in the absence of infection. Therefore, thromboembolism should not be a serious complication. This conclusion has been supported by clinical experience in some series,": 10 although not in that of Joseph and colleagues."
366
Silver, Hudson, Trimble
Conclusions The cusps of heart valve prostheses made of fascia lata, no matter whether free or stent mounted, undergo a series of recognizable morphologic changes. These include death of the original fascia with vacuolar degenerative changes if valves have been in situ more than 3 years, the formation of a pseudointima on the surface of the cusps, and resultant shrinkage. The rate at which these changes occur and their severity vary from prosthesis to prosthesis. Cusp shrinkage has been so extensive in tricuspid and mitral prostheses that their use has been abandoned. This type of valve has been shown to function satisfactorily in the aortic area for a number of years," but its value over a long period is still uncertain. The specter of late calcification and other late degenerative changes haunts its use until detailed, morphologic studies of long-term survivors are available. We are deeply grateful to Drs. H. T. Bahnson, Pittsburgh, Pa.; C. Cheanvechai, L. Groves, and D. Effler, Cleveland, Ohio; W. T. Dubiel, Uppsala, Sweden; J. A. Helmsworth, Cincinnati, Ohio; M. I. Ionescu, Leeds, England; G. Rodewald and U. Kirsch, Hamburg, West Germany; H. Shurnacker, Indianapolis, Ind.; and C. Baker and J. K. Yao, Toronto, Ontario, Canada, for submitting material to the Registry. We thank Mrs. M. Lorber and Mr. P. Horne for technical assistance. REFERENCES Trimble, A. S. (Editor): First International Roundtable on Fascia Lata Heart Valves, Surgery 71: 10, 1972. 2 Ionescu, M. I., and Ross, D. N.: Heart-Valve Replacement With Autologous Fascia Lata, Lancet 2: 335, 1969.
The Journal of Thorocic ond Cardiovascu lor Surgery
3 Magarey, F. R.: On the Mode of Formation of Lambl's Excrescences and Their Relation to Chronic Thickening of the Mitral Valve, J. Pathol. 61: 203, 1949. 4 Dubiel, W. T.: Aortic Valve Replacement With Frame-Supported Autologous Fascia Lata, Acta Univ. Upsalensis 189: 1, 1974. 5 Dubiel, W. T., Johansson, L., and Willen, R.: Postoperative Changes in Autologous Fascia Lata Heart Valve Grafts: A Pathological Study, Ann. Thorac. Surg. 15: 140, 1973. 6 Ionescu, M. I., Pakrashi, B. C., Mary, D. A. S., Bartek, I. T., and Wooler, G. H.: Long-Term Evaluation of Tissue Valves, J. THORAc. CARDIOVASC. SURG. 68: 361, 1974. 7 Silver, M. D., and Trimble, A. S.: Structure of Autologous Fascia Lata Heart Valve Prostheses, Arch. Pathol. 93: 109. 1972. 8 Willen, R., Dubiel, W. T., and Johansson, L.: Viability and Surface Properties of Fascia Lata Heart Valve Grafts, Ann. Thorac. Surg. 18: 597, 1974. 9 Senning, A.: Fascia Lata Replacement of Aortic Valves, J. THoRAc. CARDIOVASC. SURG. 54: 465, 1967. 10 Senning, A.: Discussion of Ionescu et a1.6 : J. THORAc. CARDIOVASC. SURG. 68: 379, 1974. 11 Shumacker, H. B: Discussion of Ionescu et a1.6 : J. THORAc. CARDIOVASC. SURG. 68: 378, 1974. 12 Joseph, S., Somerville, J., Emanuel, R., Ross, D., and Ross, K.: Aortic Valve Replacement With Frame-Mounted Autologous Fascia Lata: Long-Term Results, Br. Heart J. 36: 760, 1974. 13 Hudson, R. E. B.: Pathology of Heart Valve Grafting, in Ross, D. N., and Wooler, G. H., editors: Biological Tissue in Heart Valve Replacement, London, 1972, Butterworth & Co., Ltd., p. 835. 14 Hudson, R. E. B.: Cardiovascular Pathology, London, Edward Arnold (Publishers) Ltd., 3: S-596, 1970. 15 Trimble, A. S., Gunstensen, J., Silver, M. D., Aldridge, H. E., Schwartz, L., and Morch, J. E.: Aortic Valve Replacement With Fascia Lata: An Encouraging Late Study, J. THoRAc. CARDIOVASC. SURG. 68: 219, 1974.