HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Ole Lund Rashid Mazhar Asghar Khaghani Magdi H. Yacoub Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lund, O. Articles by Yacoub, M. H. Search for Related Content PubMed PubMed Citation Articles by Lund, O. Articles by Yacoub, M. H.

J Thorac Cardiovasc Surg 1999;117:77-91
© 1999 Mosby, Inc.

SURGERY FOR ADULT CARDIOVASCULAR DISEASE

PRIMARY AORTIC VALVE REPLACEMENT WITH ALLOGRAFTS OVER TWENTY-FIVE YEARS: VALVE-RELATED AND PROCEDURE-RELATED DETERMINANTS OF OUTCOME

From the Academic Department of Cardiac Surgery^a and Department of Cardiology,^b Harefield Hospital, Middlesex, United Kingdom.

This study was supported by a grant from Baxter Healthcare Corporation, Santa Ana, Calif.

Read at the Seventy-eighth Annual Meeting of The American Association for Thoracic Surgery, Boston, Mass, May 3-6, 1998.

Received for publication May 8, 1998. Revisions requested Aug 5, 1998. Revisions received Aug 31, 1998. Accepted for publication Sept 18, 1998. Address for reprints: Professor Sir Magdi Yacoub, FRCS, Royal Brompton & Harefield Hospital, Uxbridge, Middlesex, UB9 6JH, United Kingdom.J Thorac Cardiovasc Surg 1999;117:77-91

	Abstract

Top Abstract Introduction Patients and methods Results Discussion Appendix: Discussion References

 
Objectives: Allografts offer many advantages over prosthetic valves, but allograft durability varies considerably.
Methods: From 1969 through 1993, 618 patients aged 15 to 84 years underwent their first aortic valve replacement with an aortic allograft. Concomitant surgery included aortic root tailoring (n = 58), replacement or tailoring of the ascending aorta (n = 56), and coronary artery bypass grafting (n = 87). Allograft implantation was done by means of a "freehand" subcoronary technique (n = 551) or total root replacement (n = 67). The allografts were antibiotic sterilized (n = 479), cryopreserved (n = 12), or viable (unprocessed, harvested from brain-dead multiorgan donors or heart transplant recipients, n = 127). Maximum follow-up was 27.1 years.
Results: Thirty-day mortality was 5.0%, and crude survival was 67% and 35% at 10 and 20 years. Ten- and 20-year rates of freedom from complications were as follows: endocarditis, 93% and 89%; primary tissue failure, 62% and 18%; and redo aortic valve replacement, 81% and 35%. Multivariable Cox analyses identified several valve- and procedure-related determinants: rising allograft donor age and antibiotic-sterilized allograft for mortality; donor more than 10 years older than patient for endocarditis; rising donor age minus patient age, rising implantation time (from harvest to aortic valve replacement), and donor age more than 65 years for tissue failure; and rising donor age minus patient age, young patient age, rising implantation time, and subcoronary implantation preceded by aortic root tailoring for redo aortic valve replacement. Estimated 10- and 20-year rates of freedom from tissue failure for a 70-year-old patient with a viable valve from a 30-year-old donor and no other risk factors were 91% and 64%; the figures were 71% and 20% if the donor age was 65 years. The rates of freedom from tissue failure for a 30-year-old patient with a 30-year-old donor were 82% and 39%; the figures were 49% and 3% with a 65-year-old donor. Beneficial influences of a viable valve were largely covered by short harvest time (no delay for allografts from brain dead organ donors or heart transplant recipients) and short implantation time.
Conclusions: Primary allograft aortic valve replacement can give acceptable results for up to 25 years. The late results can be improved by the use of a viable allograft, by matching patient and donor age, and by more liberal use of free root replacement with re-implantation of the coronary arteries rather than tailoring the root to accommodate a subcoronary implantation.

	Introduction

Top Abstract Introduction Patients and methods Results Discussion Appendix: Discussion References

 
The allograft aortic valve was the first successful replacement valve used in the aortic position, ¹ and its advantages suggested in early series ^1-3 still stand: excellent hemodynamic function, low thrombogenicity, and resistance to infection. A few centers have consistently used the allograft since the 1960s, ^4-8 and it still is the valve of choice at our center. A current growing interest in stentless xenografts in aortic valve replacement (AVR) rests to a significant degree on accumulated experience with the allograft.

The main concern regarding the allograft is its durability, which to some degree is determined by source (living or diseased donor) ^5,7,9 and methods of sterilization and preservation. ¹⁰ Furthermore, as with any other biologic valve, primary (degenerative) tissue failure of the allograft accelerates more than 10 years after the operation, indicating that follow-up well beyond 15 years is needed to judge the true freedom from failure at 10 and 15 years.

The aim of the present study was to judge the true behavior of the allograft after first-time AVR in a large homogeneous patient series with a follow-up exceeding 25 years, that is, beyond the high hazard phase for failure between 10 and 20 years after the operation. Furthermore, we wanted to identify independent determinants of outcome related to the allograft and the surgical procedure.

	Patients and methods

Top Abstract Introduction Patients and methods Results Discussion Appendix: Discussion References

 
During the 25-year period 1969 through 1993, 618 patients aged 15 to 84 (mean 51 years), who underwent their first AVR at Harefield Hospital, had an aortic allograft implanted. Patients with concomitant surgical procedures on the aortic root or the ascending aorta, as well as coronary artery bypass grafting, were included; any other surgical procedure and age of 14 years or younger served as exclusion criteria. The allograft was chosen irrespective of age, poor left ventricular function, and emergency operation, subject only to availability. The present 618 patients constitute 74% of all (with similar inclusion/exclusion criteria) who had a primary AVR performed by or under the service of one senior surgeon. The primary indication for surgery included aortic stenosis (n = 285), aortic regurgitation (n = 146), combined stenosis and regurgitation (n = 100), active endocarditis (n = 52), and aortic aneurysm or dissection with aortic regurgitation (n = 33) or stenosis (n = 2). Heart catheterization was performed before the operation in the majority of patients, combined with echocardiography in 177, while 43 had an echocardiogram exclusively. Coronary angiography was performed in 292 patients, of whom 87 had significant coronary artery disease (luminal diameter narrowing 50% of at least one major epicardial artery or a first branch). Clinical left ventricular failure was recorded in case of pulmonary edema or vascular congestion within a year before the operation, ¹¹ kidney failure if the S-creatinine concentration was elevated above the reference level for our laboratory in at least two preoperative blood samples, systemic hypertension in case of current or previous medical treatment, and diabetes mellitus in case of treatment with oral antidiabetic drugs or insulin. The preoperative patient profile is given in Table I.

Follow-up
All patients attended an outpatient clinic appointment 4 to 7 weeks after the operation, and those living in England additionally attended annual outpatient clinic appointments at our hospital. Since 1978 the outpatient clinic investigation has included an echocardiogram. From August 1995 through April 1997, we interviewed (mailed questionnaire and telephone) all patients who were still alive, and we contacted the general practitioner (mailed questionnaire and telephone) of all patients (living as well as deceased). All hospital contacts were checked. At termination of the study, 31 patients had died within 30 days after the operation (early mortality), 263 patients had died later, 294 patients were still alive, and 30 (4.9%) could not be traced. A total of 6229 patient-years (mean 10.1, maximum 27.1 years; maximum 17.1 years for the viable allografts) had been accumulated. Follow-up with an allograft in place (first or redo) totaled 5719 patient-years, which was used in the analyses of valve-related complications; that is, a patient was censored at the time of a redo AVR if the replacement valve was a xenograft or mechanical prosthesis.

Valve-related complications
Registration of valve-related complications followed internationally accepted guidelines ¹⁶ as previously detailed. ^5,17,18 Embolism was recorded if a systemic arterial vascular event could not be proved to be thrombotic or hemorrhagic. Bleeding was recorded if the patient received anticoagulant (or antiplatelet) treatment. Embolic and hemorrhagic events were recorded as minor if symptoms subsided within 48 hours (eg, transitory cerebral ischemia) or as major if they did not. In addition, we recorded non–valve-related strokes, that is, major cerebral events proven to be thrombotic or proven to be hemorrhagic in patients who did not receive anticoagulant treatment. Primary tissue failure (n = 258) of the allograft was recorded in case of significant regurgitation (at least grade 3/4) or stenosis (gradient 35 mm Hg) at aortic root angiography (n = 60) or Doppler echocardiography (n = 99), discovered during redo AVR (n = 52) or postmortem examination (n = 21), or when the following clinical criteria were fulfilled: the presence of a diastolic murmur, a wide pulse pressure (>50 mm Hg), increased cardiothoracic index (>0.50), and increasing left ventricular hypertrophy and strain pattern in the electrocardiogram, ⁴ all in the absence of infectious endocarditis.

Statistical analysis
All analyses were computerized by means of the BMDP Dynamic version 7.0 software package (BMDP Software, Los Angeles, Calif). ¹⁹ Simple comparisons were done with the use of a nonpaired t test, a 1-way analysis of variance, or a Pearson ² test as appropriate. Cumulative survival and event-freedom curves were made by means of the Kaplan-Meier product-limit method, and differences between curves were tested with a log-rank and a Gehan test. Multivariable analysis of early mortality was done with a stepwise logistic regression analysis, whereas late mortality and valve-related complications were tested with a Cox proportional hazards regression analysis. All variables of Tables I to III were tested in the analyses. The multivariable tests were done by means of a comprehensive formalized analysis sequence, which has been previously described. ^11,18 The odds ratio was calculated ¹⁸ for each independent risk factor to indicate the increase of risk associated with the risk level of a dichotomous risk factor or the risk increase relative to 1 unit increase of a quantitative risk factor. Estimated event-freedom curves for subgroups of patients were made by means of the Cox model; this method re-uses the common underlying event-freedom function of all 618 patients for each subgroup estimate. ^18,19 Cumulative survival and event freedom, as well as linearized incidences, are given with ± 1 standard error (SE); 95% confidence limits (95% CL) were identified by the "± 2 SE" cutoff points. ²⁰ Quantitative data are given as mean ± standard deviation (SD).

	Results

Top Abstract Introduction Patients and methods Results Discussion Appendix: Discussion References

 
Early and late mortality
The causes of deaths are given in Table IV; cardiac causes dominated and only 15% of late deaths could be related to the allograft.

Early mortality
By univariate analysis, early mortality (5.0%) was unrelated to allograft procedure and associated coronary bypass grafting. It was higher for patients who had the ascending aorta replaced with a Dacron prosthesis, and it tended to be higher for those operated on because of active endocarditis (Table III). The independent risk factors for early mortality are shown in Table V.

View larger version (10K): [in this window] [in a new window]   Fig. 1. Cumulative survival of all 618 patients. Cumulative survivals ± 1 SE are given at 5-year intervals. Numbers of patients at risk are given at 5-year intervals above the abscissa.

View larger version (15K): [in this window] [in a new window]   Fig. 2. Cumulative survival in relation to type of allograft (see text). AB, Antibiotic sterilized (the group includes 12 patients who received a cryopreserved allograft). Cumulative survivals ± 1 SE (number of patients at risk) are given at 5-year intervals.

View larger version (21K): [in this window] [in a new window]   Fig. 3. Cumulative survival in relation to allograft procedure: Full root replacement with reimplantation of the coronary ostia; subcoronary implantation of a scalloped allograft with 2-line suturing technique; and subcoronary implantation preceded by tailoring of the aortic root in patients operated on for chronic valve disease (aortic stenosis, regurgitation, or a combination), excluding patients operated on primarily because of endocarditis or ascending aortic aneurysm or dissection. Cumulative survivals ± 1 SE (number of patients at risk) are given at 5-year intervals.

Bleeding and embolism
The incidence of anticoagulant-related bleeding episodes was negligible, whereas there were 82 embolic episodes (vascular events that could not be proven to be hemorrhagic or thrombotic) in 66 patients; 76% (n = 62) of the events were cerebral and 67% (n = 55) were classified as minor (Table VII). The independent determinants for embolism (Table VIII) included long harvest time (for a cadaver valve; see Table II).

Endocarditis
Twenty-year freedom from allograft endocarditis (Table VII) was nearly 90% (Fig 4). The independent risk factors are shown in Table IX. An allograft donor who was 10 years or more older than the patient was a risk factor and was related to a 20-year freedom from endocarditis of 83% ± 4% (n = 173) compared with 93% ± 1% for a lesser age difference (n = 445; P < .01). "Other" as a limiting symptom (see Table I) was more prevalent (29%) among the 52 patients primarily operated on for active endocarditis than in the remaining 566 patients (3%; P < .0001).

View larger version (22K): [in this window] [in a new window]   Fig. 4. Cumulative freedom from valve-related complications in all 618 patients; see Table VII and text. AC, Anticoagulant (warfarin, 22 patients; acetylsalicylic acid and dipyridamole, 31 patients; the remaining 565 patients did not receive any anticoagulant treatment). Cumulative freedoms ± 1 SE are given at 10, 15, 20, and 25 years. Numbers of patients at risk are given in parentheses for thromboembolism and tissue failure.

View larger version (20K): [in this window] [in a new window]   Fig. 5. Cumulative freedom from tissue failure in relation to allograft procedure; see legends to Fig 3 for identification of procedure groups. Cumulative freedoms ± 1 SE (number of patients at risk) are given at 10, 15, and 20 years.

View larger version (23K): [in this window] [in a new window]   Fig. 6. Cumulative freedom from tissue failure in relation to age difference between allograft donor and patient. D-P Age, Donor minus patient age. Cumulative freedoms ± 1 SE (number of patients at risk) are given at 10 and 20 years.

View larger version (20K): [in this window] [in a new window]   Fig. 7. Estimated freedoms from tissue failure using a Cox regression model; see text and Table X. A1, A2, A3, and A4 curves, A 70-year-old patient, and –A1, donor age of 30 years, viable allograft with implantation time of 48 hours, and no other risk factors (Table X); A2, similar to the preceding but with an antibiotic-sterilized allograft and an implantation time of 120 hours; A3, similar to the preceding but with a donor age of 67 years; and A4, similar to the preceding but with all other risk factors of Table X. B1, B2, B3, and B4 dotted curves, like A1, A2, A3, and A4 curves, but with a 30-year old patient. Estimated freedoms are given at 10 and 20 years.

	Discussion

Top Abstract Introduction Patients and methods Results Discussion Appendix: Discussion References

 
This study has served to document the "true" long-term pattern of performance of aortic valve allografts over a period of up to 27 years and has identified several patient-, allograft-, and procedure-related factors that could influence patient outcome parameters. Several previous reports from our center ^4,5,13,15 and others ^6,7 were based on maximum follow-up periods varying around 10 to 15 years for current models of allografts. We believe that this underestimates the incidence of complications and that the true results are better represented by longer follow-up times.

A major advantage of the allograft as an aortic valve substitute is the restoration of blood flow in the aortic root, sinuses, and coronary ostia to normal or near normal. The normal aortic valve is a highly complex and sophisticated structure that starts opening and closure preceding hemodynamic events. ²³ The hemodynamic advantage translates into pressure gradients across allografts, which are usually negligible and significantly lower than the gradients across stented xenografts and modern disc valves. ^24,25 The impact may be a more complete regression of left ventricular hypertrophy after a successful AVR with an allograft. ²⁴ Complete regression of hypertrophy is a main factor in keeping the left ventricle well functioning in the long term, ²⁶ which may be an important determinant of long-term survival. ¹⁸ The current interest in unstented xenografts in general and in patients with a small aortic root in particular is understandable, particularly given the limited availability of allografts.

The hemodynamic advantage of the allograft can be exploited only if other allograft-related factors do not impair long-term survival. The present method of antibiotic sterilization ^4,12 has significantly improved results compared with earlier methods of sterilization and preservation. ¹⁰ The present viable allograft ⁵ and the viable cryopreserved allograft, ⁷ both harvested from live donors, may have improved results further.

A first concern regarding allograft AVR is the more complex implantation technique compared with those needed for stented xenografts or mechanical valves. However, the present early mortality compares favorably with results of numerous published AVR series, both those spanning the same time period as the present series and more recent studies. Furthermore, we identified widely known risk factors for early mortality, in terms of both associated surgical procedures and patient-related factors. Notably, a full root replacement with reimplantation of the coronary arteries did not entail increased risk.

During a more than 25-year follow-up with an average starting age of 51 years, normal epidemiologic risk factors and death rates are in effect. ¹⁸ Age as an independent risk factor needs no explanation, whereas the influence of male sex may be related to a generally shorter life span of men compared with women. Other independent risk factors for late mortality of the present study underscored the influence of advanced heart disease before the operation, in agreement with other long-term studies spanning the same time period. ^11,27 The present long-term survival was comparable with the survivals of a predominantly biologic ²⁷ and a predominantly mechanical AVR series, ¹⁸ both with more than 20 years of follow-up.

In the current series, valve characteristics including method of preparation and insertion appeared to have an important influence on survival. The viable allograft and root replacement were associated with better long-term survival than the antibiotic-sterilized allograft and subcoronary implantation. The frequency of root replacement increased during the 25 years of surgery. As a result, more than half were done with a viable valve, indicating that the 2 determinants were complimentary in their beneficial influence on survival. Our results regarding root replacement confirmed previous findings. ^5,14 The beneficial influence of a viable allograft on long-term survival was probably explained by its lesser tendency to failure (as was the case for root replacement) than the antibiotic-sterilized allograft. Equally important was the finding that donor age was related to late mortality; this was obviously a reflection of increased risk of both endocarditis and primary tissue failure of an allograft from elderly donors.

Allografts and other biologic valves are generally considered nonthrombogenic. However, in this series we have observed a definite incidence of thromboembolic complications. It should be recognized that there are normal rates of thromboembolic phenomena such as stroke in the general background population. ^21,22 The internationally accepted guidelines for reporting thromboembolism after heart valve replacement ¹⁶ state, in essence, that a systemic vascular event should be recorded as embolic unless there is proof of thrombosis (or hemorrhage), indicating that the rate of valve-related embolism will be overestimated. During the present very long follow-up, the rate of embolism was 1.4%/pt-y, which is lower than the average for mechanical valves, despite the absence of anticoagulant treatment in the vast majority of the present patients. In a recently published paper, a current stented pericardial xenograft aortic valve was associated with an embolic rate of 1.6%/pt-y. ²⁸ However, using published tabulated age- and sex-specific rates of stroke of a general United Kingdom population, ²¹ we were able to calculate ²² the stroke incidence of a background population with the same sex and age composition as the present patients: the background incidence did not differ from the observed (embolic, hemorrhagic, thrombotic) stroke rate of our patients. Furthermore, systemic hypertension and age, being the predominant risk factors for stroke in the general population, ²² were also independent determinants of embolism in the present patients. Our data thus confirm that the allograft in general is not thrombogenic. The exception may be the cadaver valve that is harvested too long after death of the donor.

The present 25-year freedom from endocarditis of 89% indicates that the allograft is resistant to infection. A 10-year freedom of 97% after allograft AVR in prosthetic valve endocarditis ²⁹ strongly supports that notion. The allograft, especially used as a full root replacement, is recommended in complex aortic valve endocarditis, especially with destruction of the aortic root. ²⁹ The present incidence of endocarditis was comparable with the average of reported figures for mechanical valves and somewhat lower than the incidence associated with stented xenograft aortic valves. The latter was directly confirmed by Agnihotri, McGiffin, Galbraith, and O'Brien. ³⁰ Interestingly, we observed that donor-patient age mismatch of 10 years or more independently increased the risk of endocarditis. This finding needs to be confirmed in future studies, and similarly the underlying mechanisms require further research.

The main drawback of allografts and of any other biologic valves is their limited durability. It is important to emphasize that length of follow-up may have a profound influence on estimated freedoms from primary tissue failure. In the present more than 25-year follow-up series, with a significant number of patients at risk in the long term, the hazard of failure seemed to be especially high during 16 to 20 years after the operation before it seemed to return to a phase of low and constant risk. In a previous publication from the present center with a maximum follow-up of just above 15 years, the 10- and 15-year freedoms from failure of the antibiotic-sterilized allograft were 77% and 48%, respectively ⁴; in the present series the figures were 61% and 32%, respectively. Similarly, with 14 years of follow-up we previously reported an 89% 10-year freedom from failure for the viable allograft ⁵; in the present study, with 17 years of follow-up, the 10-year freedom was 71%. The present article may thus give a picture close to the real failure characteristics of allograft aortic valves, at least as regards the antibiotic-sterilized valve.

Primary tissue failure is not a trivial complication, and any measure that can reduce the incidence is directly warranted. The technical surgical procedure seemed to influence the risk of failure. A full root replacement was associated with a low incidence of failure, whereas aortic root tailoring to accommodate a subcoronary implantation seemed to be associated with a very high failure rate between 5 and 10 years after the operation. Root replacement and root tailoring had direct beneficial and adverse influences, respectively, in the multivariable model for redo AVR that was undertaken primarily because of failure. These findings could be explained by the fact that the essential feature of root replacement is that the anatomic integrity of the valve is perfectly maintained from root to sinotubular junction. At the other extreme stands the attempt of the surgeon to recreate the anatomy by first shaping and tailoring the native root and then stitching the scalloped allograft in place in the subcoronary position with a 2-line freehand technique. Others have also observed the increased failure tendency associated with root tailoring, ⁸ as well as the favorable influence of root replacement. ^5,31,32

Young patient age is a time-proven widely known risk factor for degenerative failure of porcine ³³ and pericardial ²⁸ xenografts, as well as of allografts. ^5,8,34 In the present study, patient age played a significant role but donor age was an even stronger determinant of failure. The 2 variables were interrelated: the donor-patient age match or mismatch appeared to be the dominant factor. Furthermore, the very old allograft (donor age > 65 years) in its own right significantly increased the risk of failure. It has previously been reported from the present institution that donor age of more than 65 years was related to early allograft failure. ³⁵ Time from allograft harvest to implantation was also directly related to risk of failure. Donor age, donor minus patient age, and implantation time were all significantly higher for the antibiotic-sterilized allograft than for the viable valve. The former factors may thus in part explain the better freedom from failure of the viable valves.

The risk factors for redo AVR reflected the multivariable models for endocarditis and primary tissue failure. Interestingly, diabetes mellitus adversely influenced primary tissue failure and consequently also redo AVR. With a relation to "general tissue weakness" of diabetes mellitus, this finding may pinpoint factors of the tissue in which the allograft rests. The remaining risk factors, high cardiothoracic index, previous myocardial infarction, and atrioventricular block, indicate a deleterious influence of preoperatively impaired heart status. It has previously been shown that impaired left ventricular function was related to increased intravascular hemolysis in patients with aortic ball valves ³⁶ and with a modern bileaflet disc valve. ²⁵ A regionally or globally malfunctioning pump may give rise to higher turbulent shear forces in the flow profiles of mechanical valves, which may traumatize erythrocytes ^18,25,36 as well as the leaflet edges of an allograft. A dilated left ventricle (and root) may also add to difficulties of restoring the anatomic integrity of the valve and left ventricular outflow tract, especially in a subcoronary implantation.

In conclusion, our study has defined the long-term behavior of the allograft both in terms of survival and complications, and it has identified several factors that could influence these parameters and therefore be of clinical relevance. During the very long follow-up of the present study, the hazard of primary tissue failure "re-entered" a phase of low and constant risk first after 20 years. The failure characteristics of the present allograft series may thus indicate the true behavior of the allograft, inasmuch as studies with a shorter follow-up significantly underestimated the failure rate. Our data suggest that allograft AVR can give acceptable results for up to 25 years. Furthermore, the late results can be improved by the use of a viable allograft, by matching patient and donor age, by avoiding the old donor, and by more liberal use of free root replacement with reimplantation of the coronary ostia rather than tailoring the root to accommodate a subcoronary implantation. It is hoped that our findings will help to optimize the long-term results of allograft AVR.

	Appendix: Discussion

Top Abstract Introduction Patients and methods Results Discussion Appendix: Discussion References

 
Dr Mark F. O'Brien (Brisbane, Australia). Dr Lund, I congratulate you and your colleagues, specifically Sir Magdi Yacoub, for the valuable analysis of your allograft experience. You highlight the importance of donor age, reasonably short donor death–to–implantation time (although you talk of harvest-to-implantation time), and a better valve matching with the technical implantation of the root replacement.

Eleven years ago in 1987 at this very meeting we presented and published from our Brisbane experience, which is now over 1000 allografts, the incremental risk factors of reoperation for degeneration—younger age of patient at AVR and older age of donor. You have gone further, to focus on the additional factors that influence the quality of the valve, factors that affect viability and preservation of the tissue matrix. We have been saying for some decades now that a valve should be implanted early or cryopreserved as early as possible. Our mean time to preservation is 15 hours from donor death. We have shown clinically a marked difference in the durability between early cryopreserved valves and the antibiotic-sterilized valves that stayed in the refrigerator for days or weeks. The freedom from structural deterioration and reoperation at 20 years in our experience is 60%.

I read your manuscript as carefully as I could. You would be surprised to know that we would not use 51% of your homovital valves and 78%, at least, of your antibiotic-sterilized valves because the time from death to implantation is too long. This may explain why our freedom from primary tissue failure in the whole cohort, without subdivision of the cohort, was 60% at 20 years, whereas your very best results are in the 70-year-old patient with a 30-year-old donor, giving 10- and 20-year freedoms from tissue failure of 91% and 64%.

I have a number of questions. A homograft is an excellent valve at 10 years, it is a very good valve at 15 years, but it is far from optimal at 20 years. Yet you make the statement that the homograft AVR can give acceptable results for up to 25 years. I would not accept those results out to 25 years. We do not accept our own results, which really are better than yours. There are many avenues for improvement, on some of which you have focused.

My first question is this: Do you think the homograft is an acceptable device to 25 years?

What was the time from donor death to cryopreservation?

Could you expand on the influence of patient age on reoperation, particularly in that younger age group? We are beginning to learn of the poor results in infants and teenagers, but what are your thoughts on the reasons why the results are not as good in these younger recipients?

Why do you think endocarditis was more prevalent with the older donor valve? Is it because degeneration occurs more quickly?

What modifications have you made for your collection and harvesting protocol, particularly in regard to the cutoff age of the donor?

Last, what happens if you cannot implant the valve within that magic homovital time period of 72 hours? Do you cryopreserve it, which we would regard as late cryopreservation?

Mr Lund. Thank you, Dr O'Brien. With regard to the comparison with your own results, I think there are two important things that should be said. First, I noticed from a previous publication from your center that less than 30% of your patients receive a homograft valve, so yours is a selected series whereas ours is not. The homograft was the valve of choice at Harefield Hospital, and that may significantly influence results.

Second, as your follow-up has just passed 20 years, at least with the results you have published, we might suspect that with a follow-up closer to 30 years, your 20-year results may not look as good as they do today. We have noticed with previous publications from Harefield, with the antibiotic-sterilized homograft, that the estimated 15-year freedom reported in 1985 and in 1990 has been successively dropping down to current figures with an almost 30-year follow-up. Selection of patients most certainly will influence long-term results and so will length of follow-up.

I do not know the answer to your question about the time from harvest to cryopreservation of the 12 valves.

You asked about the possible mechanism of an increased tendency for endocarditis for the old donor valve. In our analyses, the donor-patient age match or mismatch was the important factor. The worst subset seemed to be old donor in young patient. We can only speculate about the mechanism. Further research certainly is needed. Maybe the donor valve from an elderly patient is more prone to degeneration and weakness of tissues and therefore more susceptible to endocarditis, but that is only speculative.

Regarding implantation of homovital valves, if the valve is still in the refrigerator after 72 hours, we will use it. Obviously we prefer to implant the valve as soon as possible after harvest from the viable donor, and preferably within 48 hours, but that is something that needs to be tightened up in our center.

Matching of patient and donor age is going to be relevant in the future, but we have not done it so far. However, this analysis is quite new.

Dr Charles A. Yankah (Berlin, Germany). I rise to congratulate Dr Lund and Sir Magdi Yacoub for this excellent clinical study. Twenty-five years of homograft valve function is impressive, and it also provides important information for us surgeons with regard to how to speak to the relatives and patients.

We have implanted more than 250 homografts in the aortic position, subcoronary and also root replacement. We define our homovital valves as those homografts harvested during our transplant program and preserved within 24 hours, maximum 35 hours. We made prostacyclin studies, which indicated cellular viability in about 20% of these valves at the time of implantation. The rest of them, however, possessed no endothelial cells. The viability of the homografts was defined in relation to structural viability, that is, those without endothelial cells but with some living fibroblasts that will not survive long after implantation.

My question is this: What is the cellular viability definition in your homovital valves? Can you derive your experience from the results whether the antibiotic treatment in the homovital group had an impact on the results of the implantation technique such as sizing, root replacement, and the subcoronary technique?

Mr Lund. Regarding viability of the valve: Will viability influence long-term results? In this respect, we would all think in terms of major histocompatibility antigens. Results on research in that field are conflicting. Results from Harefield and also from Holland seem to indicate that major antigenicity is not a factor in the viable homograft, whereas results from Dr O'Brien's group tend to indicate the opposite.

With regard to technical procedure, we do not distinguish between a homovital or an antibiotic-sterilized homograft.

	References

Top Abstract Introduction Patients and methods Results Discussion Appendix: Discussion References

 

1. Ross DN. Homograft replacement of the aortic valve. Lancet 1962;2:487. [Medline]
   
2. Barratt-Boyes BG, Lowe JB, Cole DS, Kelly DT. Homograft replacement for aortic valve disease. Thorax 1969;20:489.
   
3. Ross D, Yacoub MH. Homograft replacement of the aortic valve: a critical review. Prog Cardiovasc Dis 1969;11:275-93. [Medline]
   
4. Khaghani A, Dhalla N, Penta S, Qureshi S, Theodoropoulos S, Esposito G, et al. Patient status 10 years or more after aortic valve replacement using antibiotic sterilized homografts. In: Bodnar E, Yacoub M, editors. Biologic and bioprosthetic valves. New York: Yorke Medical Books; 1986. p. 38-46.
   
5. Yacoub M, Rasmi N, Sundt T, Lund O, Boyland E, Radley-Smith R, et al. Fourteen-year experience with homovital homografts for aortic valve replacement. J Thorac Cardiovasc Surg 1995;110:186-94. [Abstract/Free Full Text]
   
6. Barratt-Boyes BG, Roche AH, Subramanyan R, Pemberton JR, Whitlock RM. Long-term follow-up of patients with the antibiotic sterilized aortic homograft valve inserted freehand in the aortic position. Circulation 1987;75:768-77. [Abstract/Free Full Text]
   
7. O'Brien MF, McGiffin DC, Stafford EG, et al. Allograft aortic valve replacement: long-term comparative clinical analysis of viable cryopreserved and antibiotic stored 4^oC stored valves. J Card Surg 1991;6:534-43. [Medline]
   
8. Langley SM, Livesey SA, Tsang VT, et al. Long-term results of valve replacement using antibiotic-sterilized homografts in the aortic position. Eur J Cardiothorac Surg 1996;10:1097-106. [Abstract]
   
9. Albertucci M, Wong K, Petrou M, Mitchell A, Somerville J, Theodoropoulos S, et al. The use of unstented homograft valves for aortic valve reoperations: review of a twenty-three–year experience. J Thorac Cardiovasc Surg 1994;107:152-61. [Abstract/Free Full Text]
   
10. Grunkemeier GL, Bodnar E. Comparison of structural valve failure among different "models" of homograft valves. J Heart Valve Dis 1994;3:556-60. [Medline]
    
11. Lund O. Preoperative risk evaluation and stratification of long-term survival after valve replacement for aortic stenosis: reasons for earlier operative intervention. Circulation 1990;82:124-39 [Abstract/Free Full Text]
    
12. Yacoub M, Kittle F. Sterilization of homografts by antibiotic solutions. Circulation 1970;41/42(Suppl):II29-31.
    
13. Gula G, Ahmed M, Thompson R, Radley-Smith R, Yacoub M. Combined homograft replacement of the aortic valve and aortic root with reimplantation of the coronary arteries. Circulation 1976;54(Suppl):II150.
    
14. Yacoub MH. Allograft aortic root replacement. In: Yankah AC, Hetzer R, Miller DC, Ross DN, Somerville J, Yacoub MH, editors. Cardiac valve allografts 1962-1987. 1st ed. New York: Springer-Verlag; 1988. p. 149-55.
    
15. Dhalla N, Khaghani A, Radley-Smith R, Yacoub M. Early and long-term performance of aortic homograft root replacement. New York: Yorke Medical Books; 1986. p. 65-72.
    
16. Edmunds LH Jr, Clark RE, Cohn LH, Miller DC, Weisel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. J Thorac Cardiovasc Surg 1988;96:351-3. [Medline]
    
17. Lund O, Pilegaard HK, Magnussen K, Knudsen MA, Nielsen TT, Albrechtsen OK. Long-term prosthesis-related and cardiac-related complications after valve replacement for aortic stenosis. Ann Thorac Surg 1990;50:396-406. [Abstract]
    
18. Lund O. Valve replacement for aortic stenosis: the curative potential of early operation. Scand J Thorac Cardiovasc Surg 1993;27(suppl 40):1-137.
    
19. Dixon NJ, editor. BMDP statistical software manual. Version 7.0. Vols 1 and 2. Berkeley: University of California Press; 1992. p. 1-1500.
    
20. Grunkemeier GL, Thomas DR, Starr A. Statistical considerations in the analysis and reporting of time-related events. Am J Cardiol 1977;39:257-8. [Medline]
    
21. Oxfordshire Community Stroke Project. Incidence of stroke in Oxfordshire: first year's experience of a community stroke register. BMJ 1983;287:713-7.
    
22. Lund O, Magnussen K, Knudsen M, Pilegaard H, Nielsen TT, Albrechtsen OK. The potential for normal long term survival and morbidity rates after valve replacement for aortic stenosis. J Heart Valve Dis 1996;5:258-67. [Medline]
    
23. Yacoub MH, Gehle P, Chandrasekaran V, Birks EJ, Child A, Radley-Smith R. Late results of a valve-preserving operation in patients with aneurysms of the ascending aorta and root. J Thorac Cardiovasc Surg 1998;115:1080-90. [Abstract/Free Full Text]
    
24. Jin XY, Zhang Z-M, Gibson DG, Yacoub MH, Pepper JR. Effects of valve substitute on changes in left ventricular function and hypertrophy after aortic valve replacement. Ann Thorac Surg 1996;62:683-90. [Abstract/Free Full Text]
    
25. Lund O, Emmertsen K, Nielsen TT, Jensen FT, Flø C, Pilegaard HK, et al. Impact of size mismatch and left ventricular function on performance of the St. Jude disc valve after aortic valve replacement. Ann Thorac Surg 1997;63:1227-34. [Abstract/Free Full Text]
    
26. Lund O, Jensen FT. Functional status and left ventricular performance late after valve replacement for aortic stenosis: relation to preoperative data. Eur Heart J 1988;9:1234-43. [Abstract/Free Full Text]
    
27. McGiffin DC, O'Brien MF, Galbraith AJ, McLachlan GJ, Stafford EG, Gardner MAH, et al. An analysis of risk factors for death and mode-specific death after aortic valve replacement with allograft, xenograft, and mechanical valves. J Thorac Cardiovasc Surg 1993;106:895-911. [Abstract]
    
28. Frater RWM, Furlong P, Cosgrove DM, et al. Long-term durability and patient functional status of the Carpentier-Edwards Perimount pericardial bioprosthesis in the aortic position. J Heart Valve Dis 1998;7:48-53. [Medline]
    
29. Petrou M, Wong K, Albertucci M, Brecker SJ, Yacoub MH. Evaluation of unstented aortic homografts for the treatment of prosthetic aortic valve endocarditis. Circulation 1994;90(part I2):II198-204.
    
30. Agnihotri AK, McGiffin DC, Galbraith AJ, O'Brien MF. The prevalence of infective endocarditis after aortic valve replacement. J Thorac Cardiovasc Surg 1995;110:1708-24. [Abstract/Free Full Text]
    
31. Dearani JA, Orszulak TA, Daly RC, et al. Comparison of techniques for implantation of aortic valve allografts. Ann Thorac Surg 1996;62:1069-75. [Abstract/Free Full Text]
    
32. Prager RL, Fischer CR, Kong B, et al. The aortic homograft: evolution of indications, techniques, and results in 107 patients. Ann Thorac Surg 1997;64:659-64. [Abstract/Free Full Text]
    
33. Fann JI, Miller DC, Moore KA, et al. Twenty-year experience with porcine bioprostheses. Ann Thorac Surg 1996;62:1301-12. [Abstract/Free Full Text]
    
34. McGiffin DC, Galbraith AJ, O'Brien MF, et al. An analysis of valve re-replacement after aortic valve replacement with biologic devices. J Thorac Cardiovasc Surg 1997;113:311-8.[Abstract/Free Full Text]
    
35. Thompson R, Yacoub M, Ahmed M, Somerville W, Towers M. The use of "fresh" unstented homograft valves for replacement of the aortic valve: analysis of eight years' experience. J Thorac Cardiovasc Surg 1980;79:896-903. [Abstract]
    
36. Lund O. Late chronic hemolysis after valve replacement for aortic stenosis: relation to residual hypertrophy and impaired left ventricular function. Angiology 1990;41:836-47.
    

This article has been cited by other articles:

				2006 WRITING COMMITTEE MEMBERS, R. O. Bonow, B. A. Carabello, K. Chatterjee, A. C. de Leon Jr, D. P. Faxon, M. D. Freed, W. H. Gaasch, B. W. Lytle, R. A. Nishimura, et al. 2008 Focused Update Incorporated Into the ACC/AHA 2006 Guidelines for the Management of Patients With Valvular Heart Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1998 Guidelines for the Management of Patients With Valvular Heart Disease): Endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons Circulation, October 7, 2008; 118(15): e523 - e661. [Full Text] [PDF]

				R. O. Bonow, B. A. Carabello, K. Chatterjee, A. C. de Leon Jr, D. P. Faxon, M. D. Freed, W. H. Gaasch, B. W. Lytle, R. A. Nishimura, P. T. O'Gara, et al. 2008 Focused Update Incorporated Into the ACC/AHA 2006 Guidelines for the Management of Patients With Valvular Heart Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1998 Guidelines for the Management of Patients With Valvular Heart Disease) Endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons J. Am. Coll. Cardiol., September 23, 2008; 52(13): e1 - e142. [Full Text] [PDF]

				E. R. Nowicki, G. B. Pettersson, N. G. Smedira, E. E. Roselli, E. H. Blackstone, and B. W. Lytle Aortic Allograft Valve Reoperation: Surgical Challenges and Patient Risks Ann. Thorac. Surg., September 1, 2008; 86(3): 761 - 768. [Abstract] [Full Text] [PDF]

				T. M. Joudinaud, F. Baron, R. Raffoul, B. Pagis, M. Vergnat, C. Parisot, U. Hvass, and P. R. Nataf Redo aortic root surgery for failure of an aortic homograft is a major technical challenge Eur. J. Cardiothorac. Surg., June 1, 2008; 33(6): 989 - 994. [Abstract] [Full Text] [PDF]