- Email: [email protected]
Effect of cryopreservation on the presence of endothelial cells on human valve allografts
Effect of cryopreservation on the presence of endothelial cells on human valve allografts Cryopreserved human allograft valves are useful in a variety of cardiac operations. The presence or absence of endothelial cells on allografts may be important in determining immunogenicity and ultimate graft longevity. The purpose of this study was to determine whether endothelial cells are present on cryopreserved human allografts. Portions of cryopreserved allografts (35 valve leaflets, 96 pieces of arterial wall) not used at operation were studied. For comparison, untreated tissues (44 valve leaflets, 46 pieces of arterial wall) were obtained from structurally normal hearts and lungs removed or inserted at the time of transplantation and from pathologic tissues obtained during operations for congenital heart defects. A monolayer of cells from the luminal surface of each specimen was harvested by means of a Hautchen preparation. The monolayer was stained with fluorescein isothiocyanate-Iabeled Ulex europaeus I, a lectin with strong affinity for human endothelium. Positive staining with fluorescein was considered to be evidence for the presence of human endothelium. Endothelial cells were observed on 21 of 131 (16 %) cryopreserved allograft specimens and on 70 of 90 (78 %) untreated tissues (p < 0.001). These results show that cryopreservation typically results in the loss of endothelium from aortic and pulmonary valve allografts. These findings may have important implications for the immunologic response of the host to allograft implantation. (J 'fHORAC CARDloVASC SURG 1993;106:912-7)
Flavian M. Lupinetti, MD, Thomas T. Tsai, James M. Kneebone, MS, and Edward L. Bove, MD, Ann Arbor, Mich.
Human allograft valves have been widely applied for the treatment of acquired and congenital cardiac diseases. When these tissues are implanted at a time remote from their acquisition, attention must be paid to the methods of allograft harvesting, sterilization, and storage that are used. These factors are important in assuring allograft safety and durability and in providing adequate From the Department of Surgery, Section of Thoracic Surgery, University of Michigan School of Medicine, Ann Arbor, Mich. Presented in part at the Forty-eighth Annual Sessions of the Forum on Fundamental Surgical Problems, 1992 Clinical Congress of the American College of Surgeons, New Orleans, La., October 11-16, 1992. Supported by grant ROi HL42426 from the National Heart, Lung, and Blood Institute. Received for publication Nov. 3,1992. Accepted for publication Jan. 28, 1993. Address for reprints: Flavian M. Lupinetti, MD, Division of Cardiovascular Surgery, Children's Hospital and Medical Center, 4800 Sand Point Way N.E., Seattle, WA 98105. Copyright w 1993 by Mosby-Year Book, Inc. 0022-5223/93 $1.00 +.10
912
12/1/46378
quantities of these scarce resources. In the United States, cryopreservation ( -196 0 C) has been the chief method of allograft preservation. A number of centers have described good clinical results with cryopreserved valve allografts. Indeed, it has been suggested that cryopreserved valve allografts may exhibit durability superior to that of fresh grafts. 1 The cellular properties of cryopreserved tissues, however, have not been well characterized. The purpose of this investigation was to evaluate cryopreserved human aortic and pulmonary valve allografts for the presence of endothelial cells. Patients and methods At the C. S. Mott Children's Hospital and the University of Michigan Medical Center from July 1991 through September 1992, a total of 93 patients underwent cardiac surgical procedures that included insertion of a cryopreserved aortic or pulmonary valveallograft, or both. Each allograft was processed by Cryolife, Inc. (Marietta, Ga.) and rapidly thawed at the time of use according to the processor's instructions. During the trimming of the graft to accommodate the recipient anatomy, unnecessary pieces of the graft were removed and placed immediately in 10% formalin. In very few cases was the entire graft used, and this study therefore includes an almost com-
The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 5
9 13
Lupinetti et al.
Table I. Diagnoses in patients receiving allografts
Table II. Diagnoses in donors of untreated tissues
Diagnosis
No.
Diagnosis
Hypoplastic left-heart syndrome Right ventricular outflow tract obstruction Congenital aortic stenosis and/or insufficiency Tetralogyof Fallot Truncusarteriosus Endocarditis Total
33 26 16 9
Reconstructive operation for congenital heart defect Congenital aortic stenosis and/or insufficiency Tetralogyof Fallot Coarctationof the aorta Tricuspid atresia Right ventricularoutflow tract obstruction Transposition. of the great arteries Endocarditis Other Heart-lung transplant recipient Heart transplant recipient Cardiomyopathy Structural congenital heart defect Heart donor Lung transplant recipient Pulmonaryhypertension, secondary to congenital heart defect Pulmonaryhypertension, primary Emphysema Lung donor Total
6 3 93
pletely consecutive series of allografts. Diagnoses in these patients are shown in Table I. No tissues obtained at autopsy were included in this investigation. Hautchen ("thin skin") preparations were used to obtain a confluent en face monolayer of luminal cells after the method of Hirsch and colleagues.? Arterial walls and valve leaflets were separated when both components were available for study, and each component was individually examined. Each sample was pinned to a paraffin plate and immersed in 10% formalin for at least 24 hours at 4 0 C. The tissues were washed in cold running water for 30 to 60 minutes, or until the odor of formalin disappeared. Specimens were dehydrated by immersion in 30%, 70%, 95%, and 100% ethanol for 15 minutes each and remained in 100% ethanol for at least 24 hours. A strip of double-faced No. 137 tape (3M Company, St. Paul, Minn.) was attached to an ethanol-cleaned microscope slide. The sample was removed from ethanol, gently dried, and placed in acetone for 5 to 10 seconds. The specimen was then gently blotted and pressed endothelium-side down against the tape on the microscope slide. When the acetone evaporated, the slide was covered with 10% glycerol in distilled water for 2 to 3 minutes. The slide was then cooled by spraying dichlorodifluoromethane on the underside. The edge of the specimen was then grasped and gently stripped away. The monolayer that remained on the slide consisted of endothelial cells or other cellular and noncellular material that composed the luminal surface of the allograft. Staining was performed in a 37 0 C humidity chamber. The monolayer was covered with Ulex europaeus I lectin (Pierce, Rockford, Ill.) for 30 minutes. The slide was then immersed in two changes of phosphate-buffered saline for 3 minutes each, followed by fluorescein-labeled avidin (Pierce) for 30 minutes. Slides were again rinsed with phosphate-buffered saline 0.1 moljL for 3 minutes, washed with distilled water, and inspected. Specimens were studied at 50X and 250x with a Nikon Labophot II microscope (Garden City, N.Y.) with a B2A epifluorescence filter system using an excitation filter of 450 to 490 nm. To establish the specificity of this staining method, that is, for a negative control, we submitted some additional portions of tissues shown to be positive for endothelium to the same staining process but without the primary antibody. Specimens were considered to demonstrate evidence of endothelium if they contained characteristic polygonal cells that stained with the fluorescent marker. Specimens containing no such cells were considered to be negative for endothelium. All examinations were performed by two independent observers who were blinded as to the source of the specimen. Any disagreements between observers «10% of instances) were reconciled after simultaneous inspection. For comparison, and to assess the accuracy of our methods, we examined corresponding untreated aortic valve leaflets, aor-
No.
37 II 8 4 3
2 I
1 7 2 10 6 4 5 8 4
2 2 4 66
tic walls, pulmonary valve leaflets, and pulmonary artery walls. All untreated tissues were removed in the course of operations for heart and lung transplantation or during reconstructive operations performed for congenital cardiac anomalies. These control specimens included pathologic tissues as well as structurally normal tissues. After excision, all tissues were placed in 10% formalin as for usual pathologic studies. No autopsy tissues were included among the allograft specimens. Diagnoses in the 66 patients whose untreated tissues were examined are shown in Table 11).
Results Representative examples of positive and negative specimens are shown in Figs. 1 and 2. Positive specimens typically exhibited extensive and confluent endothelial cells that fluoresced brightly against the dark background. In rare instances, only a few fluorescein-positive cells were identified, but these tissues were also classified as positive. Negative specimens displayed an amorphous appearance, with no distinct cells in most cases, or at most a few cellular structures that did not fluoresce. Of the entire series of 131 allograft specimens, 21 (16%) displayed endothelial cells. Of the 90 untreated specimens, 70 (78%) were positive for endothelium. This difference was statistically significant (p < 0.001). No cryopreserved aortic valve leaflets were obtained, because all of these were completely implanted. Seventeen of 20 (85%) untreated aortic valve leaflets had endothelial cells. Of aortic wall specimens, 4 of 29 (14%) cryopreserved tissues and 15 of 18 (83%) untreated tissues contained
The Journal of Thoracic and Cardiovascular Surgery November 1993
9 1 4 Lupinettiet al.
Fig. 1. Hautchen preparation of untreated pulmonary valve leaflet stained with fluorescein-labeled Ulex europaeus I (I 270X). Numerous positively staining cells are evident.
Table III. Frequency of endothelial cellpresence Source
No. positive Percentpositive No. for endothelium for endothelium
Untreated tissues 20 Aortic valve leaflet Aortic wall 18 Pulmonary valve leaflet 24 Pulmonary artery wall 28 All untreated tissues 90 Cryopreserved allografts Aortic valve leaflet 0 Aortic wall 29 Pulmonary valve leaflet 35 Pulmonary artery wall 67 All allograft tissues 131
17 IS
21 17 70 4 11 6
21
85 83* 88* 61* 78* 14 31 9 16
.p < 0.001 versus corresponding cryopreserved allograft tissues.
endothelium (p < 0.00 1). Of pulmonary valve leaflets, 11 of 35 (31%) cryopreserved tissues and 21 of 24 (88%) untreated tissues contained endothelium (p < 0.001). Of pulmonary artery wall specimens, 6 of 67 (9%) cryopreserved tissues and 17 of 28 (61 %) untreated tissues contained endothelium (p < 0.001). Among the untreated tissues, there was only one diagnostic category that appeared to differ from the others regarding the presence of endothelium. The 6 pulmonary artery walls obtained from patients with primary or sec-
ondary pulmonary hypertension all failed to demonstrate endothelium. Untreated pulmonary artery walls in other patients were somewhat less likely to display endothelium than other untreated structures, but this difference was not significant.
Discussion Functions of vascular endothelium include resistance to thrombosis, maintenance of hemostasis, and modulation of vascular smooth muscle activity. Mediation of immunologic and inflammatory responses is another function of endothelium that is particularly important for allograft valves. Vascular endothelium constitutively expresses class I antigens and is induced to express class II antigens when exposed to an allogeneic milieu. Accordingly, vascular endothelium is considered to be the most immunostimulatory component of whole organ allografts.' Allograft valves are immunogenic and capable of sensitizing the recipient," and this property may not be markedly altered by cryopreservation.v 6 Immunohistochemical staining with the lectin from the seeds of the gorse, Ulex europaeus, has permitted identification of human endothelial cell with a high degree of sensitivity.' Ulex has a strong affinity for L-fucose, which is abundant in the plasma membranes of endothelial cells of normal blood vessels, vessels of benign and malignant
The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 5
Lupinetti et al.
9I5
Fig. 2. Hautchen preparation of pulmonary valve leaflet from a cryopreserved allograft stained with fluoresceinlabeled Ulex europaeus I (l270X). No endothelial cells are observed.
neoplasms.v? and vessels in tissues affected by a variety of diseases. 10,II Ulex also labels human endothelium in fetal tissues'? and isolated endothelial cells in culture.P Only in paraffin-embedded tissues have endothelial cells been reported not to stain with Ulex, and more recent reports consistently describe positive Ulex staining of paraffin-embedded endothelium." Thus it is improbable that the failure of most cryopreserved allograft tissue to stain for Ulex is attributable to artifact resulting from the cryopreservation process. It is more likely that the absence of Ulex staining indicates the absence of endothelial cells. Other methods of identification of endothelial presence have been used with varying degrees of success. Scanning electron microscopy is one such method but necessitates extensive specimen preparation and highly specialized equipment. Furthermore, because scanning microscopy requires small specimens, it is difficult to examine large areas thoroughly and it may not be consistently possible to distinguish between endothelium and underlying structures.l' Staining with silver salts outlines the borders between endothelial cells and may improve the accuracy of scanning microscopic inspection, but requires even more specimen preparation. Intravenous injection of Evans blue dye is an excellent method for gross assessment of endothelial denudation.l? This technique, however, is not applicable to an investigation of unimplanted
allografts. Fluorescent light microscopic examination for endothelial cells is quick and reliable and requires minimal equipment. In addition to Ulex staining, other methods of identification with light microscopy include stains for factor VIII-related antigen.!? Factor VIII-related antigen does not appear to be as sensitive as Ulex, however, at least for some vascular endothelium.P' 19 The presence and viability of endothelium on allograft valves has not been as extensively studied as have the more numerous fibroblasts. Yankah and associates/" were the first to examine the presence and viability of allograft valve endothelium in a rat model. They discovered not only that cryopreserved valves contained endothelial cells, but also that these cells were only 8% to 20% less viable than those of untreated valve grafts. 2 I Yankah and colleagues'? have also described 70% to 80% endothelial cell viability in cryopreserved human allografts. The discrepancy between the findings of Yankah and those described in this report suggest that specific methods of valve processing should be closely examined. It is possible that subtle differences in methods of valve preservation may explain important differences in allograft structure. It cannot be assumed that all methods of cryopreservation are identical. Harvesting of valves from a multiple organ donor or from the excised heart of a transplant recipient is necessarily associated with surgical
9 I 6 Lupinetti et al.
trauma and a variable period of warm and cold ischemia. Sterilization of allografts requires the use of strong antibiotics. Some antibiotic combinations previously used for this purpose are now known to have adverse effects on cellular viability.P The freezing process used for cryopreserved valves requires the use of dimethyl sulfoxide to prevent intracellular and intercellular ice crystal formation, but this may also affect the endothelial cells. Controlled-rate freezing, subsequent rapid thawing, and surgical trauma during implantation are other factors that may contribute to the loss of endothelial cellularity. The high frequency of endothelial cells in surgically excised specimens may reflect a sampling bias attributable to the patient population. Most patients in this series had congenital cardiac anomalies. Such patients are likely to be a good control group because of the absence of known pathologic effects on the endothelium of the excised tissues. Patients with atherosclerotic or other degenerative cardiac abnormalities are known to have abnormal endothelial structure and function and may have displayed a low frequency of Ulex staining. It is also important to note that no autopsy specimens were used in this investigation. Although autopsy specimens may have exhibited a similarly high frequency of Ulex staining, a low frequency could have been attributable to postmortem changes. It is interesting to note the results in the untreated pulmonary artery walls of patients undergoing operation for primary or secondary pulmonary hypertension. This was the only subgroup of untreated tissues that had a low frequency of Ulex staining. This low frequency may reflect the considerable pathologic changes of pulmonary hypertension on large arteries. The importance of endothelial cells to the long-term fate of allograft valves is unknown. Loss of endothelium may contribute to enhanced graft longevity by reducing the host immune response, which may contribute to valve degeneration. On the other hand, loss of endothelium may increase the capacity for thrombus formation and adversely affect the underlying fibroblasts, therebyaccelerating graft deterioration. Thus it cannot yet be determined whether the loss of endothelial cells has a positive, negative, or neutral effect on the long-term structure and function of aortic and pulmonary valve allografts. REFERENCES 1. O'Brien MF, McGiffin DC, Stafford EG, et al. Allograft aortic valve replacement: long-term comparative clinical analysis of the viable cryopreserved and antibiotic 4° C stored valves. 1 CardiacSurg 1991;6(Suppl):534-43. 2. Hirsch EZ, Martino W, Orr CH, White H, Chisolm GM III. A simple rapid method for the preparation of en face endothelial (Hautchen) monolayers from rat and rabbit aortas. Atherosclerosis 1980;37:539-48.
The Journal of Thoracic arid Cardiovascular Surgery November 1993
3. Pober lS, CollinsT, Gimbrone MA lr, Libby P, ReissCS. Inducible expression of class II major histocompatibility complex antigens and the immunogenicity of vascular endothelium. Transplantation 1986;41:141-6. 4. Heslop BF, Wilson SE, Hardy BE. Antigenicity of aortic valve allografts. Ann Surg 1973;177:301-6. 5. Cochran RP, Kunzelman KS. Cryopreservation does not alter antigenic expression of aortic allografts. 1 Surg Res 1989;46:597-9. 6. El Khatib H, Lupinetti FM. Antigenicityof fresh and cryopreserved rat valve allografts. Transplantation 1990; 49:765-7. 7. Holthofer H, Virtanen I, KariniemiAL, Hormia M, Linder E, Miettinen A. Ulex europaeus I lectin as a marker for vascular endothelium in human tissues. Lab Invest 1982;47:60-6. 8. Gonzalez-Campora R, Montero C, Martin-Lacave I, Galera H. Demonstration of vascular endothelium in thyroid carcinomausing Ulex europaeus I agglutinin.Histopathology 1986;10:261-6. 9. Yonezawa S, Nakamura T, Tanaka S, Sato E. Glycoconjugate with Ulex europaeus agglutinin-l-binding sites in normal mucosa, adenoma, and carcinoma of the human large bowel. 1 Natl Cancer Inst 1982;69:777-85.. 10. Petrovic LM, Burroughs A, Scheuer Pl. Hepatic sinusoidal endothelium: Ulex lectin binding. Histopathology 1989;14:233-43. 11. Shoda E, Watanabe M, Hirohata K, Urano Y. Histochemical study of the vascular endothelium of the synovial membrane of rheumatoid arthritis and osteoarthritis by Ulex europaeus agglutinin I (UEA-I). J Rheumatol1985; 12:841-5. 12. Holthofer H. Vascularization of the embryonic kidney: detection of endothelial cells with Ulex europaeus I lectin. Cell Differ 1987;20:27-31. 13. Jackson Cl, Garbett PK, Nissen B, Schrieber L. Binding of human endothelium to Ulex europaeus I-coated Dynabeads: application to the isolationof microvascular endothelium. J Cell Sci 1990;96:257-62. 14. OrdonezNG, BrooksT, ThompsonS, BatsakislG. Use of Ulex europaeus agglutinin I in the identification of lymphatic and blood vessel invasion in previously stained microscopic slides. Am 1 Surg Pathol1987;11:543-50. 15. Davies PF, Bowyer RE. Scanning electron microscopy: arterial endothelial integrity after fixation at physiological pressure. Atherosclerosis 1975;21:463-9. 16. Stemerman MB, Spaet TH, Pitlick F, Cintron J, Lejnieks I, Tiell ML. Intimal healing: the pattern of reendothelialization and intimal thickening. Am 1 Pathol1977;87:12542; 17. Little D, Said rw, Siegel Rl, Fealy M, Fishbein Me. Endothelial cell markers in vascular neoplasms: an immunohistochemical study comparing factor VIII-related antigen, bloodgroup specific antigens, 6-keto-PGF I alpha and Ulex europaeus I lectin. 1 Pathol 1986;149:89-95. 18. Leader M, CollinsM, Patel J, Henry K. Staining for factor VIII-related antigen and Ulex europaeus agglutinin I
The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 5
(UEA-I) in 230 tumours: an assessment of their specificity for angiosarcoma and Kaposi's sarcoma. Histopathology 1986;10:1153-62. 19. Ordonez NG, Batsakis JG. Comparison of Ulex europaeus I lectin and factor VIII-related antigen in vascular lesions. Arch Pathol Lab Med 1984;108:129-32. 20. YankahAC, RandzioG, WottgeHU, Bernhard A. Factors influencing endothelial-cell viability during procurement and preservation of valve allografts. In: Thiede A, DeItz E, Engemann R, Hamelmann H, eds. Microsurgical models in rats for transplantation research. Heidelberg: SpringerVerlag, 1985:107-11. 21. YankahAC, DreyerW, WottgeHU, Muller-RucholtzW,
Lupinetti et al. 9 1 7
Bernhard A. Kinetics of endothelial cells of preserved aortic valve allografts used for heterotopic transplantation in inbred rat strains. In: Bodnar E, Yacoub MH, eds. Biologic and bioprosthetic valves. New York: Yorke Medical Books, 1986:73-84. 22. Yankah AC, Wottge H-U, Muller-Hermelink HK, et al. Transplantation of aortic and pulmonary allografts, enhanced viability of endothelial cells bycryopreservation, importance of histocompatibility. J Cardiac Surg 1987;1 (Suppl):209-20. 23. Strickett MG, Barratt-Boyes BG, MacCulloch D. Disinfection of human heart valve allografts with antibiotics in low concentration. Pathology 1983;15:457-62.
Flavian M. Lupinetti, MD, Thomas T. Tsai, James M. Kneebone, MS, and Edward L. Bove, MD, Ann Arbor, Mich.
Human allograft valves have been widely applied for the treatment of acquired and congenital cardiac diseases. When these tissues are implanted at a time remote from their acquisition, attention must be paid to the methods of allograft harvesting, sterilization, and storage that are used. These factors are important in assuring allograft safety and durability and in providing adequate From the Department of Surgery, Section of Thoracic Surgery, University of Michigan School of Medicine, Ann Arbor, Mich. Presented in part at the Forty-eighth Annual Sessions of the Forum on Fundamental Surgical Problems, 1992 Clinical Congress of the American College of Surgeons, New Orleans, La., October 11-16, 1992. Supported by grant ROi HL42426 from the National Heart, Lung, and Blood Institute. Received for publication Nov. 3,1992. Accepted for publication Jan. 28, 1993. Address for reprints: Flavian M. Lupinetti, MD, Division of Cardiovascular Surgery, Children's Hospital and Medical Center, 4800 Sand Point Way N.E., Seattle, WA 98105. Copyright w 1993 by Mosby-Year Book, Inc. 0022-5223/93 $1.00 +.10
912
12/1/46378
quantities of these scarce resources. In the United States, cryopreservation ( -196 0 C) has been the chief method of allograft preservation. A number of centers have described good clinical results with cryopreserved valve allografts. Indeed, it has been suggested that cryopreserved valve allografts may exhibit durability superior to that of fresh grafts. 1 The cellular properties of cryopreserved tissues, however, have not been well characterized. The purpose of this investigation was to evaluate cryopreserved human aortic and pulmonary valve allografts for the presence of endothelial cells. Patients and methods At the C. S. Mott Children's Hospital and the University of Michigan Medical Center from July 1991 through September 1992, a total of 93 patients underwent cardiac surgical procedures that included insertion of a cryopreserved aortic or pulmonary valveallograft, or both. Each allograft was processed by Cryolife, Inc. (Marietta, Ga.) and rapidly thawed at the time of use according to the processor's instructions. During the trimming of the graft to accommodate the recipient anatomy, unnecessary pieces of the graft were removed and placed immediately in 10% formalin. In very few cases was the entire graft used, and this study therefore includes an almost com-
The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 5
9 13
Lupinetti et al.
Table I. Diagnoses in patients receiving allografts
Table II. Diagnoses in donors of untreated tissues
Diagnosis
No.
Diagnosis
Hypoplastic left-heart syndrome Right ventricular outflow tract obstruction Congenital aortic stenosis and/or insufficiency Tetralogyof Fallot Truncusarteriosus Endocarditis Total
33 26 16 9
Reconstructive operation for congenital heart defect Congenital aortic stenosis and/or insufficiency Tetralogyof Fallot Coarctationof the aorta Tricuspid atresia Right ventricularoutflow tract obstruction Transposition. of the great arteries Endocarditis Other Heart-lung transplant recipient Heart transplant recipient Cardiomyopathy Structural congenital heart defect Heart donor Lung transplant recipient Pulmonaryhypertension, secondary to congenital heart defect Pulmonaryhypertension, primary Emphysema Lung donor Total
6 3 93
pletely consecutive series of allografts. Diagnoses in these patients are shown in Table I. No tissues obtained at autopsy were included in this investigation. Hautchen ("thin skin") preparations were used to obtain a confluent en face monolayer of luminal cells after the method of Hirsch and colleagues.? Arterial walls and valve leaflets were separated when both components were available for study, and each component was individually examined. Each sample was pinned to a paraffin plate and immersed in 10% formalin for at least 24 hours at 4 0 C. The tissues were washed in cold running water for 30 to 60 minutes, or until the odor of formalin disappeared. Specimens were dehydrated by immersion in 30%, 70%, 95%, and 100% ethanol for 15 minutes each and remained in 100% ethanol for at least 24 hours. A strip of double-faced No. 137 tape (3M Company, St. Paul, Minn.) was attached to an ethanol-cleaned microscope slide. The sample was removed from ethanol, gently dried, and placed in acetone for 5 to 10 seconds. The specimen was then gently blotted and pressed endothelium-side down against the tape on the microscope slide. When the acetone evaporated, the slide was covered with 10% glycerol in distilled water for 2 to 3 minutes. The slide was then cooled by spraying dichlorodifluoromethane on the underside. The edge of the specimen was then grasped and gently stripped away. The monolayer that remained on the slide consisted of endothelial cells or other cellular and noncellular material that composed the luminal surface of the allograft. Staining was performed in a 37 0 C humidity chamber. The monolayer was covered with Ulex europaeus I lectin (Pierce, Rockford, Ill.) for 30 minutes. The slide was then immersed in two changes of phosphate-buffered saline for 3 minutes each, followed by fluorescein-labeled avidin (Pierce) for 30 minutes. Slides were again rinsed with phosphate-buffered saline 0.1 moljL for 3 minutes, washed with distilled water, and inspected. Specimens were studied at 50X and 250x with a Nikon Labophot II microscope (Garden City, N.Y.) with a B2A epifluorescence filter system using an excitation filter of 450 to 490 nm. To establish the specificity of this staining method, that is, for a negative control, we submitted some additional portions of tissues shown to be positive for endothelium to the same staining process but without the primary antibody. Specimens were considered to demonstrate evidence of endothelium if they contained characteristic polygonal cells that stained with the fluorescent marker. Specimens containing no such cells were considered to be negative for endothelium. All examinations were performed by two independent observers who were blinded as to the source of the specimen. Any disagreements between observers «10% of instances) were reconciled after simultaneous inspection. For comparison, and to assess the accuracy of our methods, we examined corresponding untreated aortic valve leaflets, aor-
No.
37 II 8 4 3
2 I
1 7 2 10 6 4 5 8 4
2 2 4 66
tic walls, pulmonary valve leaflets, and pulmonary artery walls. All untreated tissues were removed in the course of operations for heart and lung transplantation or during reconstructive operations performed for congenital cardiac anomalies. These control specimens included pathologic tissues as well as structurally normal tissues. After excision, all tissues were placed in 10% formalin as for usual pathologic studies. No autopsy tissues were included among the allograft specimens. Diagnoses in the 66 patients whose untreated tissues were examined are shown in Table 11).
Results Representative examples of positive and negative specimens are shown in Figs. 1 and 2. Positive specimens typically exhibited extensive and confluent endothelial cells that fluoresced brightly against the dark background. In rare instances, only a few fluorescein-positive cells were identified, but these tissues were also classified as positive. Negative specimens displayed an amorphous appearance, with no distinct cells in most cases, or at most a few cellular structures that did not fluoresce. Of the entire series of 131 allograft specimens, 21 (16%) displayed endothelial cells. Of the 90 untreated specimens, 70 (78%) were positive for endothelium. This difference was statistically significant (p < 0.001). No cryopreserved aortic valve leaflets were obtained, because all of these were completely implanted. Seventeen of 20 (85%) untreated aortic valve leaflets had endothelial cells. Of aortic wall specimens, 4 of 29 (14%) cryopreserved tissues and 15 of 18 (83%) untreated tissues contained
The Journal of Thoracic and Cardiovascular Surgery November 1993
9 1 4 Lupinettiet al.
Fig. 1. Hautchen preparation of untreated pulmonary valve leaflet stained with fluorescein-labeled Ulex europaeus I (I 270X). Numerous positively staining cells are evident.
Table III. Frequency of endothelial cellpresence Source
No. positive Percentpositive No. for endothelium for endothelium
Untreated tissues 20 Aortic valve leaflet Aortic wall 18 Pulmonary valve leaflet 24 Pulmonary artery wall 28 All untreated tissues 90 Cryopreserved allografts Aortic valve leaflet 0 Aortic wall 29 Pulmonary valve leaflet 35 Pulmonary artery wall 67 All allograft tissues 131
17 IS
21 17 70 4 11 6
21
85 83* 88* 61* 78* 14 31 9 16
.p < 0.001 versus corresponding cryopreserved allograft tissues.
endothelium (p < 0.00 1). Of pulmonary valve leaflets, 11 of 35 (31%) cryopreserved tissues and 21 of 24 (88%) untreated tissues contained endothelium (p < 0.001). Of pulmonary artery wall specimens, 6 of 67 (9%) cryopreserved tissues and 17 of 28 (61 %) untreated tissues contained endothelium (p < 0.001). Among the untreated tissues, there was only one diagnostic category that appeared to differ from the others regarding the presence of endothelium. The 6 pulmonary artery walls obtained from patients with primary or sec-
ondary pulmonary hypertension all failed to demonstrate endothelium. Untreated pulmonary artery walls in other patients were somewhat less likely to display endothelium than other untreated structures, but this difference was not significant.
Discussion Functions of vascular endothelium include resistance to thrombosis, maintenance of hemostasis, and modulation of vascular smooth muscle activity. Mediation of immunologic and inflammatory responses is another function of endothelium that is particularly important for allograft valves. Vascular endothelium constitutively expresses class I antigens and is induced to express class II antigens when exposed to an allogeneic milieu. Accordingly, vascular endothelium is considered to be the most immunostimulatory component of whole organ allografts.' Allograft valves are immunogenic and capable of sensitizing the recipient," and this property may not be markedly altered by cryopreservation.v 6 Immunohistochemical staining with the lectin from the seeds of the gorse, Ulex europaeus, has permitted identification of human endothelial cell with a high degree of sensitivity.' Ulex has a strong affinity for L-fucose, which is abundant in the plasma membranes of endothelial cells of normal blood vessels, vessels of benign and malignant
The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 5
Lupinetti et al.
9I5
Fig. 2. Hautchen preparation of pulmonary valve leaflet from a cryopreserved allograft stained with fluoresceinlabeled Ulex europaeus I (l270X). No endothelial cells are observed.
neoplasms.v? and vessels in tissues affected by a variety of diseases. 10,II Ulex also labels human endothelium in fetal tissues'? and isolated endothelial cells in culture.P Only in paraffin-embedded tissues have endothelial cells been reported not to stain with Ulex, and more recent reports consistently describe positive Ulex staining of paraffin-embedded endothelium." Thus it is improbable that the failure of most cryopreserved allograft tissue to stain for Ulex is attributable to artifact resulting from the cryopreservation process. It is more likely that the absence of Ulex staining indicates the absence of endothelial cells. Other methods of identification of endothelial presence have been used with varying degrees of success. Scanning electron microscopy is one such method but necessitates extensive specimen preparation and highly specialized equipment. Furthermore, because scanning microscopy requires small specimens, it is difficult to examine large areas thoroughly and it may not be consistently possible to distinguish between endothelium and underlying structures.l' Staining with silver salts outlines the borders between endothelial cells and may improve the accuracy of scanning microscopic inspection, but requires even more specimen preparation. Intravenous injection of Evans blue dye is an excellent method for gross assessment of endothelial denudation.l? This technique, however, is not applicable to an investigation of unimplanted
allografts. Fluorescent light microscopic examination for endothelial cells is quick and reliable and requires minimal equipment. In addition to Ulex staining, other methods of identification with light microscopy include stains for factor VIII-related antigen.!? Factor VIII-related antigen does not appear to be as sensitive as Ulex, however, at least for some vascular endothelium.P' 19 The presence and viability of endothelium on allograft valves has not been as extensively studied as have the more numerous fibroblasts. Yankah and associates/" were the first to examine the presence and viability of allograft valve endothelium in a rat model. They discovered not only that cryopreserved valves contained endothelial cells, but also that these cells were only 8% to 20% less viable than those of untreated valve grafts. 2 I Yankah and colleagues'? have also described 70% to 80% endothelial cell viability in cryopreserved human allografts. The discrepancy between the findings of Yankah and those described in this report suggest that specific methods of valve processing should be closely examined. It is possible that subtle differences in methods of valve preservation may explain important differences in allograft structure. It cannot be assumed that all methods of cryopreservation are identical. Harvesting of valves from a multiple organ donor or from the excised heart of a transplant recipient is necessarily associated with surgical
9 I 6 Lupinetti et al.
trauma and a variable period of warm and cold ischemia. Sterilization of allografts requires the use of strong antibiotics. Some antibiotic combinations previously used for this purpose are now known to have adverse effects on cellular viability.P The freezing process used for cryopreserved valves requires the use of dimethyl sulfoxide to prevent intracellular and intercellular ice crystal formation, but this may also affect the endothelial cells. Controlled-rate freezing, subsequent rapid thawing, and surgical trauma during implantation are other factors that may contribute to the loss of endothelial cellularity. The high frequency of endothelial cells in surgically excised specimens may reflect a sampling bias attributable to the patient population. Most patients in this series had congenital cardiac anomalies. Such patients are likely to be a good control group because of the absence of known pathologic effects on the endothelium of the excised tissues. Patients with atherosclerotic or other degenerative cardiac abnormalities are known to have abnormal endothelial structure and function and may have displayed a low frequency of Ulex staining. It is also important to note that no autopsy specimens were used in this investigation. Although autopsy specimens may have exhibited a similarly high frequency of Ulex staining, a low frequency could have been attributable to postmortem changes. It is interesting to note the results in the untreated pulmonary artery walls of patients undergoing operation for primary or secondary pulmonary hypertension. This was the only subgroup of untreated tissues that had a low frequency of Ulex staining. This low frequency may reflect the considerable pathologic changes of pulmonary hypertension on large arteries. The importance of endothelial cells to the long-term fate of allograft valves is unknown. Loss of endothelium may contribute to enhanced graft longevity by reducing the host immune response, which may contribute to valve degeneration. On the other hand, loss of endothelium may increase the capacity for thrombus formation and adversely affect the underlying fibroblasts, therebyaccelerating graft deterioration. Thus it cannot yet be determined whether the loss of endothelial cells has a positive, negative, or neutral effect on the long-term structure and function of aortic and pulmonary valve allografts. REFERENCES 1. O'Brien MF, McGiffin DC, Stafford EG, et al. Allograft aortic valve replacement: long-term comparative clinical analysis of the viable cryopreserved and antibiotic 4° C stored valves. 1 CardiacSurg 1991;6(Suppl):534-43. 2. Hirsch EZ, Martino W, Orr CH, White H, Chisolm GM III. A simple rapid method for the preparation of en face endothelial (Hautchen) monolayers from rat and rabbit aortas. Atherosclerosis 1980;37:539-48.
The Journal of Thoracic arid Cardiovascular Surgery November 1993
3. Pober lS, CollinsT, Gimbrone MA lr, Libby P, ReissCS. Inducible expression of class II major histocompatibility complex antigens and the immunogenicity of vascular endothelium. Transplantation 1986;41:141-6. 4. Heslop BF, Wilson SE, Hardy BE. Antigenicity of aortic valve allografts. Ann Surg 1973;177:301-6. 5. Cochran RP, Kunzelman KS. Cryopreservation does not alter antigenic expression of aortic allografts. 1 Surg Res 1989;46:597-9. 6. El Khatib H, Lupinetti FM. Antigenicityof fresh and cryopreserved rat valve allografts. Transplantation 1990; 49:765-7. 7. Holthofer H, Virtanen I, KariniemiAL, Hormia M, Linder E, Miettinen A. Ulex europaeus I lectin as a marker for vascular endothelium in human tissues. Lab Invest 1982;47:60-6. 8. Gonzalez-Campora R, Montero C, Martin-Lacave I, Galera H. Demonstration of vascular endothelium in thyroid carcinomausing Ulex europaeus I agglutinin.Histopathology 1986;10:261-6. 9. Yonezawa S, Nakamura T, Tanaka S, Sato E. Glycoconjugate with Ulex europaeus agglutinin-l-binding sites in normal mucosa, adenoma, and carcinoma of the human large bowel. 1 Natl Cancer Inst 1982;69:777-85.. 10. Petrovic LM, Burroughs A, Scheuer Pl. Hepatic sinusoidal endothelium: Ulex lectin binding. Histopathology 1989;14:233-43. 11. Shoda E, Watanabe M, Hirohata K, Urano Y. Histochemical study of the vascular endothelium of the synovial membrane of rheumatoid arthritis and osteoarthritis by Ulex europaeus agglutinin I (UEA-I). J Rheumatol1985; 12:841-5. 12. Holthofer H. Vascularization of the embryonic kidney: detection of endothelial cells with Ulex europaeus I lectin. Cell Differ 1987;20:27-31. 13. Jackson Cl, Garbett PK, Nissen B, Schrieber L. Binding of human endothelium to Ulex europaeus I-coated Dynabeads: application to the isolationof microvascular endothelium. J Cell Sci 1990;96:257-62. 14. OrdonezNG, BrooksT, ThompsonS, BatsakislG. Use of Ulex europaeus agglutinin I in the identification of lymphatic and blood vessel invasion in previously stained microscopic slides. Am 1 Surg Pathol1987;11:543-50. 15. Davies PF, Bowyer RE. Scanning electron microscopy: arterial endothelial integrity after fixation at physiological pressure. Atherosclerosis 1975;21:463-9. 16. Stemerman MB, Spaet TH, Pitlick F, Cintron J, Lejnieks I, Tiell ML. Intimal healing: the pattern of reendothelialization and intimal thickening. Am 1 Pathol1977;87:12542; 17. Little D, Said rw, Siegel Rl, Fealy M, Fishbein Me. Endothelial cell markers in vascular neoplasms: an immunohistochemical study comparing factor VIII-related antigen, bloodgroup specific antigens, 6-keto-PGF I alpha and Ulex europaeus I lectin. 1 Pathol 1986;149:89-95. 18. Leader M, CollinsM, Patel J, Henry K. Staining for factor VIII-related antigen and Ulex europaeus agglutinin I
The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 5
(UEA-I) in 230 tumours: an assessment of their specificity for angiosarcoma and Kaposi's sarcoma. Histopathology 1986;10:1153-62. 19. Ordonez NG, Batsakis JG. Comparison of Ulex europaeus I lectin and factor VIII-related antigen in vascular lesions. Arch Pathol Lab Med 1984;108:129-32. 20. YankahAC, RandzioG, WottgeHU, Bernhard A. Factors influencing endothelial-cell viability during procurement and preservation of valve allografts. In: Thiede A, DeItz E, Engemann R, Hamelmann H, eds. Microsurgical models in rats for transplantation research. Heidelberg: SpringerVerlag, 1985:107-11. 21. YankahAC, DreyerW, WottgeHU, Muller-RucholtzW,
Lupinetti et al. 9 1 7
Bernhard A. Kinetics of endothelial cells of preserved aortic valve allografts used for heterotopic transplantation in inbred rat strains. In: Bodnar E, Yacoub MH, eds. Biologic and bioprosthetic valves. New York: Yorke Medical Books, 1986:73-84. 22. Yankah AC, Wottge H-U, Muller-Hermelink HK, et al. Transplantation of aortic and pulmonary allografts, enhanced viability of endothelial cells bycryopreservation, importance of histocompatibility. J Cardiac Surg 1987;1 (Suppl):209-20. 23. Strickett MG, Barratt-Boyes BG, MacCulloch D. Disinfection of human heart valve allografts with antibiotics in low concentration. Pathology 1983;15:457-62.