Pulmonary site bioprostheses: morphologic findings in 40 cases. (2024)

Although many previous studies (1-11) have reported findings inbioprosthetic heart valves, very few have focused on right-sided valves.Herein, we report a series of pulmonary-site tissue valves explantedbetween January 1995 and June 2003, with an emphasis on morphologicfindings.

The Toronto General Hospital and Toronto's Hospital for SickChildren have a long history of pulmonary valve replacement surgery forcongenital heart disease, with a total of 1081 patients receiving 1335pulmonary valve implants (including valves in conduits), as allograftsor bioprostheses. There were 514 living, adult patients (age >18years) who underwent valve replacement from January 1995 to June 2003 atthese facilities. Fifty-four (10.5%) of these adults requiredreoperation and valve replacement because of failed prosthetic valves.Forty (74.1%) of the explanted valves were bioprostheses, and 14 (25.9%)were allografts.

MATERIALS AND METHODS

Data were obtained by review of pathologic, clinical, and surgicalrecords. All pulmonary valve bioprostheses explanted from January 1995to June 2003 at Toronto General Hospital have been included in thecurrent study. Most of these prosthetic heart valves had not beentreated with antimineralization agents.

All explanted prostheses were examined radiologically, grossly andhistologically, by one author (J.B.). The biologic parts of the valveswere removed from the superstructure (plastic-metal-fabric framework)and decalcified. Sections were then submitted for histologic processing.Paraffin-embedded sections were cut at 4 to 5 [micro]m and stained withhematoxylin-eosin. A Movat pentachrome stain (for collagen and elastictissue) was obtained. Additional sections for microorganisms (Gram stainand Gomori methenamine silver) were obtained if infective endocarditiswas suspected. Histologic examination was performed to document theseverity and location (cusps and/or porcine aorta) of pannus,calcification, thrombus, inflammation, and tissue degeneration. Thesections were examined using a Leica DMRB microscope (Leica Systems,Toronto, Ontario, Canada), and photomicrographs were obtained using aLeica digital camera (DC 500, Leica).

Cusp tears were classified according to Ishihara et al. (12) Type Itears involved the free edges of the cusps. Type II lesions were linearperforations along the base of the cusps parallel to the sewing ring.Type III lesions were large perforations in the center of the cusps, andtype IV lesions were pinholelike perforations in the cusps.

Host-tissue overgrowth, or pannus, was classified according toButany et al. (13) In this system, pannus covering part of thecircumference of the valve sewing ring was classified as mild. Pannuscovering the whole width of the ring and up to 2 mm of the cuspal tissuewas considered moderate, and severe pannus was indicated with coverageof greater than 2 mm of the basal portion of a valve cusp.

[FIGURE 1 OMITTED]

Fisher exact test was used for contingency analysis of categoricvariables. Student t test was used for analysis of continuous variables.All statistical tests were performed at a significance level of P <.05.

RESULTS

The 40 bioprosthetic valves were explanted from 19 (47.5%) womenand 21 (52.5%) men. They were implanted for a mean of 14.3 [+ or -] 5.2years (range, 2 to 26 years). The average patient age at the time ofvalve implantation was 16.7 [+ or -] 10.3 years (range, 3 months to 53years). For many patients, the explanted valve was not their firstprosthetic valve in that position, and hence, the average age ofpatients at the time of the first implant would be somewhat younger,10.3 [+ or -] 8.0 years. For all cases, significant valvular dysfunctionwas seen on echocardiography, before explantation.

The primary indication for valve implantation was tetralogy ofFallot (n = 22; 55.0%) followed by transposition of the great arteries(n = 7; 17.5%), pulmonary atresia (n = 5;12.5%), double outlet rightventricle (n = 4;10.0%), and truncus arteriosus (n = 2; 5.0%). Nineteen(47.5%) of the explanted bioprostheses were of the Hanco*ck variety(Medtronic Heart Valve Division, Irvine, Calif), 13 (32.5%) wereIonescu-Shiley valves (Shiley Heart Valve Research Center, Irvine,Calif), and 8 (20%) were Carpentier-Edwards bioprostheses (BaxterHealthcare Corporation, now Edwards Life Sciences, Irvine, Calif; Figure1).

[FIGURE 2 OMITTED]

The morphologic features associated with structural valvedeterioration (SVD) included collagen degeneration, calcification,prolapse, increased cusp stiffness/cusp immobilization, fluidinsudation, fibrosis and cusp tears (Figures 2 through 4). A total of 39(97.5%) of the valves showed evidence of SVD (97.5%). The only valvethat did not show evidence of SVD was a Hanco*ck valve that had been inplace for 2 years (the shortest duration in this series) and wasexplanted because of severe pannus-causing valvular stenosis. Evidenceof stenosis was found in all 40 (100.0%) of the cases. Valvularincompetence was determined based on cusp tears and cusp immobilization.In total, 28 (70.0%) of the valves were incompetent. The average implantduration of incompetent valves was 13.2 [+ or -] 4.6 years, and theaverage implant duration of nonincompetent valves was 16.8 [+ or -] 6.0years. This difference was statistically significant (P = .04). Overall,27 (67.5%) of the valves showed all 3: SVD, stenosis, and incompetence.One valve in this series was explanted because of infectiveendocarditis. This valve had been in place for 12 years, and also showedSVD, stenosis and incompetence. Morphologic findings are summarized inthe Table.

Calcification

Calcification was present in 32 (80.0%) of the valves (Figure 3).The calcification was severe and diffuse in 22 (55.0%) of the cases. In2 (5.0%) of the cases, the calcification was so severe that it formed aplate or platelike layer over the valve cusps. A third case showed areasof both platelike and nodular calcification in separate areas.

Valves with calcification had been implanted for a mean of 14.7 [+or -] 4.9 years (range, 7 to 26 years). Among the 9 (22.5%) prosthesesimplanted for less than 10 years, calcification was severe and diffusein 3 (33.3%) of the cases, whereas among the 31 (77.5%) prosthesesimplanted for more than 10 years, calcification was severe and diffusein 19 (61.3%) (P = .25).

[FIGURE 3 OMITTED]

Pannus

Pannus was present on 39 (97.5%) of the explanted specimens. Theone case where there was no evidence of pannus was a conduit in whichmost of the cuspal tissue had been destroyed. Severe pannus was found on35 (87.5%) of the specimens, often covering both the flow and non-flowsurfaces, with extensions to the free margins. Of the remaining valves,1 (2.5%) demonstrated moderate pannus, 1 (2.5%) of the valves displayedmild pannus, and 2 (5.0%) of the cases could not be categorized becauseof partial destruction of cusp tissue.

Cusp Immobilization

Significant pannus led to valve cusps immobilization in 22 (55.0%)of the specimens. All 3 cusps were immobilized in 17 (42.5%) of thecases, and all 17 of these had calcification in addition to pannus. Nine(52.9%) of these 17 specimens had all 3 cusps immobilized in a partiallyopen position. Pannus extending onto both the flow and non-flow cuspalsurfaces led to retraction of collagen and shortening of the cusp(Figure 2). In 4 (10.0%) of the 40 cases, all 3 cusps were immobilizedin the fully open position. There were 2 (5.0%) of the cases with all 3cusps immobilized in a closed position.

Tears

Tears were present in 18 (45.0%) of the valves. These valves werein place for an average of 14.3 [+ or -] 4.3 years. Nine (22.5%) of thevalves showed multiple tears. Overall, there was one type III tear, withthe rest of the tears being type I. The type III tear was an ovalperforation at the center of one cusp of a Carpentier-Edwards porcinevalve, measuring 0.4 cm by 0.2 cm. This valve had been in place for 18years and was heavily calcified, with all cusps immobilized in the openposition.

Infective Endocarditis

In 1 (2.5%) of the cases, the primary mode of valve failure wasinfective endocarditis. The explanted device was a conduit with aCarpentier-Edwards porcine bioprosthesis that had been in place for 12years. Gram positive cocci were present in thrombi on the cusps. Thecuspal tissue was not calcified. The wall of the conduit was acutelyinflamed and showed signs of necrosis.

Other Changes

In 3 (7.5%) of the cases, structural deterioration was so severethat little cuspal tissue remained at explantation. These valves wereimplanted for 26, 19, and 9 years. At the time of bioprosthetic valveexplantation, these patients were 32, 25, and 24 years old,respectively.

Evidence of thrombi (generally small) was discovered in 11 (27.5%)of the explanted valves. These valves were implanted for 13.1 [+ or -]4.6 years (range, 7 to 21 years). Cusp prolapse was present in 5 (12.5%)of the cases, of which 2 also demonstrated calcification. In 3 (60.0%)of the 5 cases, only one cusp was prolapsed. All of these valves hadtears. In 1 (20.0%) of the cases, 2 cusps were prolapsed, and there wasextensive calcification and pannus formation. All 3 cusps were prolapsedin 1 (20.0%) of the cases, and this valve also showed broad pannusformation and extensive calcification.

There was only one case of paravalvular leak. This valve was inplace for 13.5 years in a conduit. Two tears were present in this valve,measuring 0.6 cm and 0.3 cm in length, respectively. Despite thisfinding, there was no evidence of any calcification. The valve diddisplay evidence of thrombus and severe pannus.

[FIGURE 4 OMITTED]

Valve Type

In total, 24 (60.0%) of the valves were porcine aortic(Hanco*ck/Carpentier-Edwards) and 16 (40.0%) were bovine pericardial(Ionescu-Shiley low-profile). The porcine bioprostheses were in place15.3 [+ or -] 5.8 years, whereas the pericardial valves were in placefor 12.7 [+ or -] 3.6 years. Calcification was present in 20 (83.3%) ofthe porcine valves as compared with 12 (75.0%) of the pericardialvalves; severe calcification was seen in 12 (50.0%) versus 10 (62.5%) ofthe valves, respectively. Cusp immobilization was similar in both groupsas well, with 54.2% (n = 13) of the porcine valves having one cuspimmobilized versus 56.3% (n = 9) in the pericardial group. All 3 casesof severe SVD causing cusp destruction were from porcine valves,comprising 18.8% of the porcine group. Differences in the abovemorphologic changes of porcine versus bovine pericardial valves did notreach statistical significance.

Nineteen (47.5%) of the explanted pulmonary valves were inconduits. These valves were in place for 15.9 [+ or -] 5.0 years. Theremaining 21 (52.5%) bioprostheses were orthotopic implants. Theorthotopic implants were in place for 12.8 [+ or -] 4.9 years. Pannuswas severe in 15 (78.9%) of the conduits, and mild in 1 (5.3%), whereasthe remaining 3 (15.8%) of the valves in conduits could not be assessedbecause of extensive cusp destruction. In the orthotopically placedseries, pannus was severe in 20 (95.2%) of the 21 cases and moderate in1 (4.8%). Evidence of pannus extending to, or near, the free margin wasfound on 10 (52.6%) of the 19 valve conduits and 6 (28.6%) of the 21orthotopically placed valves. Fourteen (73.7%) of the 19 valve conduitswere calcified compared with 18 (85.7%) of the 21 orthotopically placedvalves. Differences in morphologic features of valves from conduitsversus orthotopic implants did not demonstrate statistical significance.

Nineteen (47.5%) of the 40 patients were younger than 30 years atthe time of valve explantation. Fourteen (73.7%) of these 19 patientshad immobilization of all 3 valve cusps. In those older than 30 years ofa*ge, 3 (14.3%) had immobilization of all 3 cusps, a statisticallysignificant difference (P < .001). In the younger age group, 11(57.9%) of the explanted valves showed severe and diffuse calcificationas did 11 (52.4%) of the valves in the older age group. Eight (42.1% and38.1%, respectively) of the valves from each age group showed evidenceof cusp tears. These differences did not reach statistical significance.

COMMENT

The 40 explanted valves in this series had a relatively long periodof bioprosthesis survival (14.3 [+ or -] 5.2 years) in a fairly youngpatient population. Although left-sided valve explants ranged induration from 2.2 to 8.5 years in various older studies, (1-3,11,14) werecently reported on a series of Carpentier-Edwards supraannular (BaxterHealthcare Corporation, now Edwards Life Sciences) and Hanco*ck-II(Medtronic Heart Valve Division) porcine bioprostheses that were inplace for 13.9 [+ or -] 3.9 and 10.0 [+ or -] 5.1 years, respectively.(13) In the current series of pulmonary explants, the lower pressureload in the right side of the heart likely enhanced the survival of somevalves, as 8 (20.0%) of our valve explants were in place for 20 or moreyears.

Infective endocarditis has been demonstrated as a significant causeof failure in aortic and mitral valve bioprostheses. The occurrence ofinfective endocarditis has been reported at a linearized rate of 0.6%per patient-year in Carpentier-Edwards porcine valves in the aortic andmitral positions. (15) However, endocarditis was the cause of valvefailure in only 1 (2.5%) of the valves in our study. In addition, theonly case of paravalvular leak in our patient group occurred in thispatient with infective endocarditis, consistent with the theory thatparavalvular leak often occurs as a complication of endocarditis. (10)Paravalvular leaks may, in part, be caused by ventricular closingpressures and are, therefore, more common on the left side.

We found pannus almost universally (97.5%) in our cases, withsevere pannus present in 87.5% of these explanted valves. These findingsare consistent with our recent study of porcine explants, in whichpannus was found in 93.9% of cases. (13) Exuberant pannus may beworsened by the longevity of implants, the relatively young age of thepatient population, and the narrower right ventricular outflow tract(especially in those with the tetralogy of Fallot) and may beexacerbated by the differences in hemodynamics in the right versus theleft side of the heart. Indeed, in early term analysis of pulmonaryconduits, a susceptibility to obstruction from pannus formation, inconjunction with the effects of SVD (such as calcification andthrombosis), has been observed. (16) Furthermore, the thickness ofpannus has been reported to increase dramatically in valved conduitscompared with nonvalved conduits. (17) In a large-scale study ofchildren younger than 18 years of age, more than 90% of prostheticpulmonary valve re-replacements were due to stenosis. (18) In a separatestudy of porcine-valved conduits for right ventricular outflow tractreconstruction, a small patient group (n = 16) was found to have a 60%prevalence of severe conduit stenosis at 9 years after implant. (19) Arelatively large bioprosthesis, when placed in a comparatively narrowright ventricular outflow tract, likely allows greater tissue-prosthesiscontact and more rapid and extensive host-tissue overgrowth. Clearly,the extent of host-tissue overgrowth in pulmonary-valved conduits andorthotopic pulmonary implants is problematic. The exact impact of ageand hemodynamic differences compared with the left side is, however,difficult to determine.

Another interesting finding in this series of explanted valves istheir low incidence of cusp tears, considering the longevity of thevalves and the high incidence of other features of structuraldeterioration. In our previous series of porcine bioprosthesis, tearswere found in 76.1% of explants, compared with 45.0% in the currentseries. In many cases, severe pannus entirely covered one or more cusps,possibly providing a coating or sheath to the bioprosthesis cusps, thusprotecting it from collagen fiber damage that would normally causetearing. The lower incidence of tearing may also be related to decreasedpressures on the right side of the heart compared with the left.

Calcification was found in 80.0% of the pulmonary bioprostheses,most often with severe deposits spread throughout the cuspal tissue. Theprogression of calcific deposits has been shown to increase with lengthof implantation. (1) Furthermore, calcification of bioprostheses hasbeen shown to be accelerated in patients younger than 20 years of age(4) and to be less extensive in patients older than 35 years of age.(20) The high degree and severity of calcification andcalcification-related changes seen in our study would, therefore, beexpected, considering the young patient group analyzed and therelatively long duration of valve implant. Whether or not the lowerpressures in the right ventricular outflow tract have any impact on thisis difficult to determine. Our study also found a marked increase incusp immobilization, with a significant rate of cusp immobilizationinvolving all 3 cusps in patients younger than 30 years old at the timeof valve explantation. The high incidence of calcification, pannus, andcusp immobilization would have led to clinically significant stenosis inthe majority of valves in this series.

CONCLUSION

We have presented data from an 8-year review of explanted pulmonarysite bioprostheses, implanted in a young population, mainly forcongenital heart disease. There were no statistically significantdifferences in morphologic findings between different valve types(porcine aortic vs. bovine pericardial). The explanted bioprostheses hada high rate of severe calcification and pannus over-growth, and thesechanges together led to valve orifice stenosis. Severe cuspimmobilization was also a frequent finding, with the immobilization ofall 3 cusps being more common in patients younger than 30 years of age(at the time of explantation). The valves in our series were in placefor considerably long periods of time. Host-tissue overgrowth is asignificant problem with bioprosthetic valves, particularly pulmonarysite implants, and needs further evaluation.

Accepted for publication August 19, 2008.

References

(1.) Butany J, Yu W., Silver MD, David TE. Morphologic findings inexplanted Hanco*ck II porcine bioprostheses. J Heart Valve Dis.1999;8:4-15.

(2.) Butany J, Vanlerberghe K, Silver MD. Morphologic findings andcauses of failure in 24 explanted lonescu-Shiley low-profile pericardialheart valves. Human Pathol. 1992;1 1:1224-1233.

(3.) Shoen FJ, Fernandez J, Gonzalez-Lavin L, et al. Causes offailure and pathologic findings in surgically removed lonescu-ShileyStandard bovine pericardial hear valve bioprostheses: emphasis onprogressive structural deterioration. Circulation. 1987;76:61 8-627.

(4.) Milano A, Bortolotti U, Talenti E, et al. Calcificdegeneration as the main cause of porcine bioprosthetic valve failure.Am J Cardiol. 1984;53:1066-1070.

(5.) Schoen FJ, Hobson CE. Anatomic analysis of removed prostheticheart valves: causes of failure of 33 mechanical valves and 58bioprostheses, 1980 to 1983. Hum Pathol. 1985;16:549-559.

(6.) Schoen FJ, Collins JJ Jr, Cohn LH. Long-term failure rate andmorphologic correlations in porcine bioprosthetic heart valves. Am JCardiol. 1983;51:957 964.

(7.) Valente M, Bortolotti U, Arbustini E, et al.Glutaraldehyde-preserved porcine bioprosthesis: factors affectingperformance as determined by pathologic studies. Chest. 1983;83:607-611.

(8.) Valente M, Ius P, Bortolotti U, et al. Fathology of thepericarbon bovine pericardial xenograft implanted in humans. J HeartValve Dis. 1998;7:180-189.

(9.) Ius P, Thiene G, Minarini M, et al. Low-profile porcinebioprosthesis (Liotta): pathologic findings and mode of failure in thelong-term. J HeartValve Dis. 1996; 5:323-327.

(10.) Grabenwoger M, Grimm M, Leukauf C, et al. Failure mode of anew pericardial valve prosthesis (Sorin pericarbon): a morphologicalstudy. Eur J Cardiothorac Surg. 1994;8:470-476.

(11.) Isomura T, Yanai T, Akagawa H, et al. Late pathologicalchanges of Carpentier-Edwards porcine bioprostheses in the mitralposition. J Cardiovasc Surg (Torino). 1986;27:307-315.

(12.) Ishihara T, Ferrans VJ, Boyce SW, et al. Structure andclassification of cuspal tears and perforations in porcine bioprostheticcardiac valves implanted in patients. Am J Cardiol. 1981;48:665-678.

(13.) Butany J, Leong SW, Cunningham KS, et al. A 10-yearcomparison of explanted Hanco*ck-II and Carpentier-Edwards supraannularbioprostheses. Car diovasc Pathol. 2007;16:4-13.

(14.) Bernal JM, Rabasa JM, Lopez R, et al. Durability of theCarpentier-Edwards porcine bioprosthesis: role of age and valveposition. Ann Thorac Surg. 1995; 60(suppl):248-252.

(15.) Jamieson WRE, Burr LH, Munro AI, et al. Carpentier-Edwardsstandard porcine bioprosthesis: a 21-year experience. Ann Thorac Surg.1998;66:S40-43.

(16.) Agarwal KC, Edwards WD, Feldt RH, et al. Clinicopathologicalcorrelates of obstructed right-sided porcine-valved extracardiacconduits. J Thorac Cardiovasc Surg. 1981;81:591-601.

(17.) Fiore AC, Peigh PS, Robison RJ, et al. Valved and nonvalvedright ventricular-pulmonary arterial extracardiac conduits. J ThoracCardiovasc Surg. 1983; 86:490-497.

(18.) Caldarone CA, McCrindle BW, Van Arsdell GS, et al.Independent-factors associated with longevity of prosthetic pulmonaryvalves and valved conduits. J Thorac Cardiovasc Surg.2000;120:1022-1031.

(19.) Albert JD, Bishop DA, Fullerton DA, et al. Conduitreconstruction of the right ventricular outflow tract. J ThoracCardiovasc Surg. 1993;106:228-236.

(20.) Magilligan DJ Jr, Lewis JW Jr, Jara FM, et al. Spontaneousdegeneration of the porcine bioprosthetic valve. Ann Thorac Surg.1980;30:259-266.

Gursharan S. Soor, BSc; Shaun W. Leong, BSc; Jagdish Butany, MBBS,MS, FRCPC; Jonathan L. Shapero, MD; William G. Williams, MD

From the Departments of Pathology and Cardiovascular Surgery,Toronto General Hospital and University of Toronto (Messrs Soor andLeong and Drs Butany and Shapero); and the Division of CardiovascularSurgery, The Hospital for Sick Children, University of Toronto (DrWilliams), Toronto, Ontario. Dr Shapero is now with the Department ofDermatology, University of British Columbia, Vancouver, BritishColumbia.

The authors have no relevant financial interest in the products orcompanies described in this article.

Reprints: Jagdish Butany, MBBS, MS, FRCPC, Department of Pathology,11E-420, Toronto General Hospital, 200 Elizabeth St, Toronto, ON, CanadaM5G 2C4 (e-mail: [emailprotected]).

Morphologic Findings in Explanted Pulmonary Bioprosthetic ValvesMorphologic Finding Total (%)Stenosis 40 (100)Structural valve deterioration 39 (97.5)Pannus 39 (97.5) Severe and diffuse 35 (87.5)Incompetence 28 (70.0)Calcification 32 (80.0) Severe and diffuse 22 (55.0)Cusp immobilization 22 (55.0) Involving all 3 cusps 17 (42.5)Cusp tears 18 (45.0)Thrombus 11 (27.5)Cusp prolapse 5 (12.5)Infective endocarditis 1 (2.5)Paravalvular leak 1 (2.5)