Abstract
OBJECTIVES
Aortic homografts are an alternative to mechanical or biological valve prostheses. Homografts are generally not transplanted ABO-compatible while this policy is still under debate. The purpose of this study was to investigate whether ABO compatibility impacts on long-term outcomes or not.
METHODS
Between 1992 and 2009, 363 adult patients with a mean age of 52 years received homografts in aortic position. Donor and acceptor blood groups could be obtained for 335 patients. Sixty-three percent received blood group-compatible (n = 212) (Group iso) and 37% non-blood group-compatible allografts (n = 123) (Group non-iso).
RESULTS
The overall event-free survival (freedom from death or reoperation) was 55.5% (n = 186). In the iso group, the event-free survival was 84.1% at 5 years and 63.3% at 10 years. In the non-iso group, the event-free survival was 79.4% at 5 years and 51.8% at 10 years. 28.5% of patients (n = 35) with ABO-incompatible and 25.5% (n = 54) with ABO-compatible grafts required reoperation. The mean time to reoperation in the iso group was 97.3 vs 90 months in the non-iso group.
CONCLUSIONS
In 17 years of research, we have not yet found a statistical significant difference in blood group incompatibility regarding overall event-free survival. In our opinion, there is no need to use ABO-compatible homografts for aortic valve replacement in adults. Histological and immunohistochemical assays are mandatory to confirm our results.
Keywords: Aortic homografts, Blood group incompatibility, Reoperation
INTRODUCTION
Aortic valve replacement generally has good results but the perfect valve that combines life-long durability with good haemodynamic performance has yet to be found. All prostheses have appropriate haemodynamic function in aortic position while at the same time all valves have certain disadvantages. The use of homografts is accepted worldwide especially in determinate subgroups; however, some doubts on their compatibility and durability exist.
The impact of blood group incompatibility as a risk factor for homograft failure has been discussed in the literature controversially. As for the use of homografts for right ventricular outflow tract reconstruction, there are several studies confirming [1] or confounding [2–5] an association between ABO-mismatch and accelerated homograft failure. Christenson et al. [6, 7] found a relationship between ABO-mismatch and homograft failure. In comparison, less data exist for homografts in aortic position. Immunohistochemical studies [8, 9] have shown that there is a lack of ABO-antigen expression on fresh and cryopreserved homograft valve leaflets. Human leucocyte antigen expression in vessels and valves does not exist and the chemical and cryopreservation processes help in destroying the majority of immunogenicity; however, a strong expression on the medial and adventitial vasa vasorum of cryopreserved pulmonary artery homografts still remains. Other studies found no association between ABO-mismatch and valve dysfunction [10–12]. Our aim was to investigate the impact of ABO-mismatched human aortic valve homografts on long-term outcomes.
PATIENTS AND METHODS
Between June 1992 and August 2009, 363 adult patients received aortic or pulmonary homografts in aortic position. Donor and acceptor blood groups could be obtained for 335 patients (mean age 52 years, range 18–68). Sixty-three percent received blood group-compatible (n = 212) (Group iso) and 37% non-blood group-compatible allografts (n = 123) (Group non-iso). Blood group matching was not performed routinely at our centre and the choice of a homograft valve was surgeon related. Homografts were chosen by availability and size. Primary diagnoses were stenosis (41.5%, n = 139), regurgitation (40.3%, n = 135) or mixed aortic valve disease (18.2%, n = 61). Degenerative valve disease occurred in 80.6% (n = 270) and endocarditis in 19.4% (n = 65). 37.6% (n = 126) of patients required additional surgical procedures: coronary artery bypass grafting (CABG), replacement of the ascending aorta and mitral valve surgery.
Systemic hypertension and diabetes were defined by medical treatment of the patient. Twenty-one patients had a previous aortic valve replacement with a mechanical prosthesis, tissue valve or allograft. The need for reoperation and echocardiographic valve failure were evaluated during follow-up. The preoperative patient profile is shown in Table 1.
Table 1:
Preoperative patient profile
| ABO- compatible | ABO- incompatible | P-value | |
|---|---|---|---|
| Number of patients | 212 | 123 | |
| Demographic data | |||
| Age (years) | 52 ± 12 | 53 ± 12 | 0.47 |
| Gender | |||
| Male | 145 | 84 | 1 |
| Female | 67 | 39 | |
| Clinical data | |||
| Aetiology of valve disease | |||
| Degenerative | 171 | 99 | 1 |
| Infectious | 41 | 24 | |
| NYHA class | |||
| I | 14 | 15 | 0.25 |
| II | 107 | 60 | |
| III | 78 | 38 | |
| IV | 13 | 10 | |
| Valve disease | |||
| Stenosis | 91 | 40 | 0.38 |
| Regurgitation | 102 | 44 | |
| Combination | 34 | 33 | |
| Previous aortic valve replacement | 14 | 7 | 0.82 |
| Mechanical prosthesis | 10 | 4 | |
| Xenograft | 2 | 3 | |
| Allograft | 2 | ||
| Hypertension | 92 | 49 | 0.57 |
| Diabetes mellitus | 10 | 6 | 1 |
| Atrial fibrillation | 17 | 9 | 1 |
Allograft data
All the 335 homografts were obtained from our in-house homograft bank, established since the beginning of homograft surgery at our centre. The percentage of valves that were explanted from heart-beating donors (HBD) was 91.6% (n = 307), mainly heart transplant recipients or rejected donor hearts, and 8.4% (n = 28) from cadaveric valve donors (non-HBD).
The donor hearts were transported to the homograft bank in a sterile plastic container containing Medium 199 (Biochrom AG, Berlin, Germany). The grafts were dissected and trimmed under sterile conditions as quickly as possible and then either cryopreserved in liquid nitrogen (91.0%, n = 305) or stored as fresh-wet valves in Medium 199 at 4°C (9%, n = 30) until implantation. Sterilization with an antibiotic cocktail, containing Ciprofloxacin (3 mg/100 ml), Amicacin (2 mg/100 ml), Vancomycin (2 mg/100 ml), Metronidazol (2 mg/100 ml) and Amphotericin B (0.5 mg/100 ml) was carried out for valves from non-HBD or in case of fresh-wet storage. Cryopreserved valves were not routinely sterilized with antibiotics. Cryopreservation followed a standardized protocol: first the valves were placed into a cryoprotectant medium containing 60 ml Medium 199 and 10% dimethyl sulfoxide as crystallization-protection (WAK-Chemie, Steinbach, Germany), then immediately controlled-rate frozen until a temperature of −150°C was attained. Contraindication for allograft use was: donor seropositive for HIV or viral hepatitis, age >60 years, active endo-/myocarditis and time between death and harvest >24 h. Allograft data are shown in Table 2.
Table 2:
Allograft profile
| ABO- compatible | ABO- incompatible | P-value | |
|---|---|---|---|
| Number of patients | 212 | 123 | |
| Donors | |||
| HBD | 192 | 115 | 0.42 |
| Non-HBD | 20 | 8 | |
| Graft preservation | |||
| Fresh-wet homograft | 22 | 8 | 0.32 |
| Cryopreserved homograft | 190 | 115 | |
| Graft | |||
| Pulmonary homograft in aortic position | 24 | 16 | 0.73 |
| Aortic homograft in aortic position | 188 | 107 | |
HBD: heart-beating donors.
Surgical technique
The operation was performed via a median sternotomy, using standard cardiopulmonary bypass techniques and Bretschneider's solution (1500–2500 ml) in moderate or deep hypothermia, depending on the need for additional circulatory arrest. After intraoperative reassessment of the anatomy, a size-matched homograft was thawed at 40°C and trimmed if necessary. Two techniques were used for implantation, as described earlier [13]: subcoronary implantation in 19.1% (n = 64) and miniroot replacement in 80.9% (n = 271) of cases. The surgical procedure was switched to the miniroot technique in 1995 because of two consecutive perioperative graft insufficiencies using the subcoronary technique. In the iso group, associated procedures were performed as follows: CABG in 9.4% (n = 20), mitral valve surgery in 1.9% (n = 4) and replacement of the ascending aorta in 21.2% (n = 45).
In the non-iso group, CABG was performed in 7.3% (n = 9), mitral valve surgery in 5.7% (n = 7) and replacement of the ascending aorta in 17.1% (n = 21).
Other associated procedures were performed in 6% (n = 20) of all patients. Mean bypass time was 168 min for both groups (±60 min for non-iso patients and ±49 min for iso patients) and mean aortic cross-clamping time was 123 min for iso (±30 min) and 121 min for non-iso grafts (±37 min). During weaning from extracorporeal circulation, normal homograft function was confirmed.
The surgical profile is shown in Table 3.
Table 3:
Surgical profile and associated surgical procedures
| ABO- compatible | ABO- incompatible | P-value | |
|---|---|---|---|
| Number of patients | 212 | 123 | |
| Allograft valve procedure | |||
| Subcoronary implantation | 34 | 30 | 0.08 |
| Miniroot implantation | 178 | 93 | |
| Associated procedures | |||
| Yes | 84 | 42 | 0.35 |
| Coronary artery bypass | 20 | 9 | |
| Mitral valve surgery | 4 | 7 | |
| Ascending aorta procedure | 45 | 21 | |
| Other | 15 | 5 | |
| No | 128 | 81 | |
| Cross-clamp time (min) | 123 ± 30 | 121 ± 37 | 0.08 |
| Bypass time (min) | 168 ± 49 | 168 ± 60 | 0.16 |
Follow-up
After discharge from hospital, the patients attended outpatient appointments 6 and 12 months postoperatively, followed by annual routine examinations. The follow-up included transthoracic echocardiographic measurement of transvalvular gradient for stenosis (mild from 30 to 50 mmHg and severe >50 mmHg) and insufficiency (0: no insufficiency, 1: mild, 2: moderate, 3: moderate to severe, 4: severe) with semiquantitative colour Doppler, as well as left ventricular ejection fraction (EF, %).
The median follow-up was 98 months, ranging from 0 days to 217 months. The mean echocardiographic follow-up was 90 months, ranging from 4 to 204 months. Echocardiographic follow-up was complete in 288 cases (86%), 47 patients were lost to the echocardiographic follow-up. Echocardiographic data were obtained until reoperation.
Statistical analysis
Association between categorical variables was tested with Fisher's exact test, while the Mann–Whitney test was used to compare metric variables in two groups. The log-rank test was used to compare a censored time-to-event in several groups.
Longitudinal analysis of stenosis and EF was conducted on the basis of linear mixed models with time, squared time, predictor of interest and interaction as fixed effects and random intercept and random slope. The linear mixed models were fitted using the R function ‘lme’. Longitudinal analysis of insufficiency was performed using the generalized linear mixed model with binary dependent variable with time, the predictor of interest and their interaction as fixed effects and a random intercept. The models were fitted by maximum likelihood and numerical integration via Gauss–Hermite quadrature using the R function glmmML from the package of the same name. Within the (generalized) linear mixed models, significance was assessed on the basis of the Wald test. P-values smaller than 0.05 were considered significant. All statistical analyses were conducted using the statistical platform R 2.10 (www.r-project.org).
RESULTS
Mortality
The overall 30-day mortality was 6.3% (n = 21). In the iso group, 5.7% (n = 12) of patients died in the first 30 days (low cardiac output (n = 4), pneumonia (n = 1), endocarditis (n = 2) and other non-valve related causes (n = 5)), whereas in the non-iso group, the 30-day mortality was 8.1% (n = 9) (sepsis (n = 3), endocarditis (n = 1), myocardial infarction (n = 3) and low cardiac output (n = 2)).
Five-year mortality was 10.3% for the iso group vs 17.3% for the non-iso group and 14.4 vs 24.6% at 10 years (iso vs non-iso) and overall survival for the complete follow-up was 81.1% (n = 172, iso group) vs 74% (n = 91, non-iso group). Differences between Kaplan–Meier estimates of overall actuarial survival in the iso group vs non-iso group did not reach statistical significance (log-rank test) (Fig. 1).
Figure 1:
Kaplan–Meier estimates of overall actuarial survival in patients with aortic homograft and separated into blood group-compatible and blood group-incompatible grafts.
Event-free survival
The overall event-free survival (freedom from death or reoperation) was 55.5% (n = 186). In the iso group, actuarial event-free survival was 84.1% at 5 years and 63.3% at 10 years, respectively; in the non-iso group event-free survival was 79.4% at 5 years and 51.8% at 10 years, respectively. 28.5% of patients (n = 35) with ABO-incompatible and 25.5% (n = 54) with ABO-compatible grafts required reoperation. Of those 59 received a biological prosthesis (24 from ABO-incompatible and 35 from ABO-compatible) and 30 a mechanical device (11 from ABO-incompatible and 19 from ABO-compatible). Mean time to reoperation in the iso group was 97.3 months (median 100 months, range 4.4–162.4 months) vs 90 months (median 94.1 months, range 2 days–210.2 months) in the non-iso group. Twelve patients (3.58%) had two ‘events’: they required reoperation for aortic valve replacement and then died in the postoperative period: 1.89% (n = 4) of patients with compatible allograft vs 6.5% (n = 8) with incompatible grafts. Cumulative survival analysis could not identify ABO incompatibility as a significant predictor for event-free survival (Fig. 2). In univariate analyses (log-rank test) donor age, type of graft (pulmonary grafts) and operation in case of active endocarditis were significantly associated with event-free survival.
Figure 2:
Kaplan–Meier estimates of event-free survival (freedom from death or reoperation) in patients with aortic homografts and separated into blood group-compatible and blood group-incompatible grafts.
Echocardiographic allograft function
Differences between both groups regarding grade of insufficiency, stenosis and global pump function were not significant in the first early postoperative echocardiographic evaluation (Figs 3–5).
Figure 3:
First early echocardiographic evaluation of postoperative ejection fraction (EF, %). Patients with blood group-compatible or blood group-incompatible aortic homografts.
Figure 4:
First early echocardiographic evaluation of postoperative transvalvular gradient. Patients with blood group-compatible or blood group-incompatible aortic homografts.
Figure 5:
Postoperative valve insufficiency (0 = no insufficiency, 1 = mild, 2 = moderate) in first early echocardiographic evaluation. Patients with blood group-compatible or blood group-incompatible aortic homografts.
One patient in the iso group had a high postoperative gradient of 39 mmHg with no insufficiency (0°). Overall homograft function was normalized in this case with a 10-year postop gradient of 6 mmHg and an insufficiency of 1°. One patient in the non-iso group with a postoperative gradient of 30 mmHg required reoperation 6 years after homograft implantation because of infectious endocarditis and received a mechanical prosthesis.
Echocardiographic follow-up was similar in both groups (mean 96.8 months, ±49.2 months for iso and 88.6 months, ±51.2 months for non-iso patients, P = 0.12).
In the (generalized) linear mixed models with time, ABO incompatibility and their interaction as main effects, neither ABO incompatibility nor its interaction with (squared) time was significant (P-values of the interaction with time for stenosis: 0.85; for insufficiency: 0.59 and for EF: 0.36).
Discussion
Biological prostheses have the calculated risk of structural failure and subsequent need for reoperation. Mechanical valves, on the other hand, are expected to show longer durability combined with the need for life-long anticoagulation and the risk of complications such as thromboembolism or bleeding [14, 15].
Homografts in aortic position are an alternative to mechanical or tissue valve prostheses. However, these grafts also have limited durability because of structural failure. Owing to the invention of homografts, several disadvantages could be eliminated. The preservation technique was developed from fresh-wet valves to cryopreserved homografts that underwent sterilization with an antibiotic cocktail [4]. The implantation technique was also improved from the subcoronary implantation to root replacement with reimplantation of the coronary arteries [16]. The use of pulmonary homografts in aortic position showed an impaired outcome in the long-term follow-up leading to the termination of using those grafts [17].
Potential risk factors for early and late mortality like reoperation and valve failure detected by echocardiographic follow-up, as shown in previous studies [18, 19], did not reveal any statistically significant differences in the ABO-compatible iso group and the ABO-incompatible non-iso group (Tables 1–3). Thus, no relevant impact of ABO incompatibility on echocardiographic homograft function was detected in our data. Plausibility check of echocardiographic data showed a high level of stenosis and a higher proportion of severe insufficiency prior to reoperation, whereas the level of EF showed no association to reoperation or death.
Distribution of valve storage factors (fresh-wet storage vs cryopreservation) as well as associated surgical procedures did not show any correlation with the differences in outcome. Preoperative recipient comorbidity (New York Heart Association status, diabetes and hypertension) and donor/recipient gender also did not show any statistical findings.
The purpose of this study was to investigate whether ABO compatibility has an impact on long-term outcomes, because homografts were generally not transplanted ABO-compatible.
Owing to the early findings from Balch and Karp in 1975 [2] aortic valve homografts were not matched to blood groups. In fact, the tissue procurement techniques and the storage in liquid nitrogen are expected to be rigorous enough to eliminate blood group antigens on the implanted valves.
With respect to Balch and Karp and their 46 patients, we describe an extensive population, the most extensive publication on this topic at the moment.
Furthermore, the absence of expression of carbohydrate antigen on valvular endothelium suggests that blood group incompatibility does not play a significant role in homograft degeneration [8]. Nevertheless, the immunological response evaluated in our institution by Fischlein et al. showed an immune response in the first line, but no significant effect on the mid-term outcome [20, 21].
In neonates and children younger than 3 years of age, non-blood group-compatible homografts had a significantly higher early reoperation rate than blood group-compatible homografts. This was not the case in children older than 3 years [7]. This observation was made in right ventricular outflow tract reconstruction. The evaluation of 174 patients who received homografts for right ventricular outflow tract reconstruction from the European Homograft Bank showed no impact of ABO mismatch on allograft failure [4].
The structural valve deterioration can be indirectly evaluated with an echocardiographic control as shown by Banbury et al. [22] on the pericardial aortic valve. As part of this protocol, we also recorded the grade of insufficiency, gradients and EF. These results had shown no statistical significance between the two groups and a good long-term function with a low incidence of reoperation rate.
In summary, there was no statistical significant difference in blood group incompatibility regarding survival and freedom from reoperation or death. As a result, we suggest that blood group matching is not necessary in human aortic allografts. It is still unclear whether there is an immunological benefit of ABO matching or not.
Limitation of this study
Our study lacks immunohistological data to describe and differentiate the rate of structural valve deterioration of the explanted homografts by patient age and time of homograft implantation in the iso and the non-iso homograft patients. As to the few explanted homografts that underwent histologic examination, our study does not cover these. It is not a randomized study and includes the limitation of heterogeneity between the two groups that could potentially influence the outcome of this study. However, as shown in Tables 1 and 2 about the patient and the allograft profiles, the groups do not show any statistical difference for the variable analysed.
Conflict of interest: none declared.
REFERENCES
- 1.Baskett R, Nanton M, Warren A, Ross D. Human leukocyte antigen-DR and ABO mismatch are associated with accelerated homograft valve failure in children: implications for therapeutic interventions. J Thorac Cardiovasc Surg. 2003;126:232–8. doi: 10.1016/s0022-5223(03)00210-1. [DOI] [PubMed] [Google Scholar]
- 2.Balch CM, Karp RB. Blood group compatibility and aortic-valve allotransplatation in man. J Thorac Cardiovasc Surg. 1975;70:256–9. [PubMed] [Google Scholar]
- 3.Gerestein CG, Takkenberg JJ, Oei FB, Cromme-Dijkhuis AH, Spitaels SE, van Herwerden LA, et al. Right ventricular outflow tract reconstruction with an allograft conduit. Ann Thorac Surg. 2001;71:911–7. doi: 10.1016/s0003-4975(00)02440-1. [DOI] [PubMed] [Google Scholar]
- 4.Jashari R, Daenen W, Meyns B, Vanderkelen A. Is ABO group incompatibility really the reason of accelerated failure of cryopreserved allografts in very young patients?—Echography assessment of the European Homograft Bank (EHB) cryopreserved allografts used for reconstruction of the right ventricular outflow tract. Cell Tissue Bank. 2004;5:253–9. doi: 10.1007/s10561-004-1442-z. [DOI] [PubMed] [Google Scholar]
- 5.Bando K, Danielson GK, Schaff HV, Mair DD, Julsrud PR, Puga FJ. Outcome of pulmonary and aortic homografts for right-ventricular outflow tract reconstruction. J Thorac Cardiovasc Surg. 1995;109:509–18. doi: 10.1016/S0022-5223(95)70282-2. [DOI] [PubMed] [Google Scholar]
- 6.Christenson J, Vala D, Sierra J, Beghetti M, Kalangos A. Blood group incompatibility and accelerated homograft fibrocalcifications. J Thorac Cardiovasc Surg. 2004;127:242–50. doi: 10.1016/j.jtcvs.2003.07.047. [DOI] [PubMed] [Google Scholar]
- 7.Christenson JT, Sierra J, Colina Manzano NE, Jolou J, Beghetti M, Kalangos A. Homografts and xenografts for right ventricular outflow tract reconstruction: long-term results. Ann Thorac Surg. 2010;90:1287–93. doi: 10.1016/j.athoracsur.2010.06.078. [DOI] [PubMed] [Google Scholar]
- 8.Kadner A, Chen RH, Mitchell RN, Adams DH. Homograft crossmatching is unnecessary due to the absence of blood group antigens. Ann Thorac Surg. 2001;71:S349–52. doi: 10.1016/s0003-4975(01)02500-0. [DOI] [PubMed] [Google Scholar]
- 9.Feingold B, Wearden P, Morell V, Galvis D, Galambos C. Expression of A and B blood group antigens on cryopreserved homografts. Ann Thorac Surg. 2009;87:211–4. doi: 10.1016/j.athoracsur.2008.09.073. [DOI] [PubMed] [Google Scholar]
- 10.Yacoub M, Rasmi NR, Sundt TM, Lund O, Boyland E, Radleysmith R, et al. 14-year experience with homovital homografts for aortic-valve replacement. J Thorac Cardiovasc Surg. 1995;110:186–94. doi: 10.1016/S0022-5223(05)80025-X. [DOI] [PubMed] [Google Scholar]
- 11.Dignan R, O'Brien M, Hogan P, Passage J, Stephens F, Thornton A, et al. Influence of HLA matching and associated factors on aortic valve homograft function. J Heart Valve Dis. 2000;9:504–11. [PubMed] [Google Scholar]
- 12.Yap CH, Skillington PD, Matalanis G, Davis BB, Tait BD, Hudson F, et al. Human leukocyte antigen mismatch and other factors affecting cryopreserved allograft valve function. Heart Surg Forum. 2008;11:E42–5. doi: 10.1532/HSF98.20071104. [DOI] [PubMed] [Google Scholar]
- 13.Gulbins H, Kreuzer E, Reichart B. Homografts: a review. Expert Rev Cardiovasc Ther. 2003;1:533–9. doi: 10.1586/14779072.1.4.533. [DOI] [PubMed] [Google Scholar]
- 14.Rahimtoola SH. Choice of prosthetic heart valve for adult patients. J Am Coll Cardiol. 2003;41:893–904. doi: 10.1016/s0735-1097(02)02965-0. [DOI] [PubMed] [Google Scholar]
- 15.Bach DS. Choice of prosthetic heart valves: update for the next generation. J Am Coll Cardiol. 2003;42:1717–19. doi: 10.1016/j.jacc.2003.06.009. [DOI] [PubMed] [Google Scholar]
- 16.Kilian E, Fries F, Kowert A, Vogt F, Kreuzer E, Reichart B. Homograft implantation for aortic valve replacement since 15 years: results and follow-up. Heart Surg Forum. 2010;13:E238–42. doi: 10.1532/HSF98.20091160. [DOI] [PubMed] [Google Scholar]
- 17.Vogt F, Kowert A, Beiras-Fernandez A, Oberhoffer M, Kaczmarek I, Reichart B, et al. Pulmonary homografts for aortic valve replacement: long-term comparison with aortic grafts. Heart Surg Forum. 2011;14:E237–41. doi: 10.1532/HSF98.20101162. [DOI] [PubMed] [Google Scholar]
- 18.Lund O, Chandrasekaran V, Grocott-Mason R, Elwidaa H, Mazhar R, Khaghani A, et al. Primary aortic valve replacement with allografts over twenty-five years: valve-related and procedure-related determinants of outcome. J Thorac Cardiovasc Surg. 1999;117:77–90. doi: 10.1016/s0022-5223(99)70471-x. [DOI] [PubMed] [Google Scholar]
- 19.Palka P, Harrocks S, Lange A, Burstow DJ, O'Brien MF. Primary aortic valve replacement with cryopreserved aortic allograft—an echocardiographic follow-up study of 570 patients. Circulation. 2002;105:61–6. doi: 10.1161/hc0102.101357. [DOI] [PubMed] [Google Scholar]
- 20.Fischlein T, Schuetz A, Haushofer M, Frey R, Uhlig A, Detter C, et al. Immunologic reaction and viability of cryopreserved homografts. Ann Thorac Surg. 1995;60:122–6. doi: 10.1016/0003-4975(95)00257-l. [DOI] [PubMed] [Google Scholar]
- 21.Schutz A, Fischlein T, Breuer M, Haushofer M, Uhlig A, Detter C, et al. Cytoimmunological monitoring after homograft valve-replacement. Eur J Cardiothorac Surg. 1994;8:609–12. doi: 10.1016/1010-7940(94)90045-0. [DOI] [PubMed] [Google Scholar]
- 22.Banbury MK, Cosgrove DM, 3rd, Thomas JD, Blackstone EH, Rajeswaran J, Okies JE, et al. Hemodynamic stability during 17 years of the Carpentier–Edwards aortic pericardial bioprosthesis. Ann Thorac Surg. 2002;73:1460–5. doi: 10.1016/s0003-4975(02)03445-8. [DOI] [PubMed] [Google Scholar]