| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2001;20:835-841
© 2001 Elsevier Science NL
Department of Cardiothoracic and Vascular Surgery, Deutsches Herzzentrum Berlin, Augustenburger Platz 1, D-13353 Berlin, Germany
Received 11 December 2000; received in revised form 22 June 2001; accepted 22 June 2001.
Corresponding author. Tel.: +49-30-4593-2021; fax: +49-30-4593-2021
e-mail: yankah{at}dhzb.de
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
Key Words: Aortic allograft • Geometric mismatch • Ventricular remodeling
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
The limited availability of aortic homografts might pose a situation at an emergency operation to implant a mismatched homograft valve. Early valve failure could result, and subsequently, the enthusiasm for the excellent clinical results of this relatively technical artistry achieved by others [3–6,9–13] might fade away. Other factors related to early homograft failure are immune response, especially at a young age, and the learning phase of homograft implantation [14,15].
The selection of a stentless aortic valve substitute and accurate sizing will have an impact on early and long-term valve performance and myocardial remodeling, as well as bringing about an improvement in the patient's clinical and cardiac performance. The valve performance will be disturbed by non-structural deterioration, such as cusp rupture, leaflet prolapse or distortion, commissural displacement, central and paravalvular leak (all by technical errors), aortic root dilatation from progressive dilatation which is not a technical error, and a progressive loss of leaflet extensibility by structural deterioration [5,6,10,13].
The purpose of this report is to analyze our 14-year clinical results of freehand subcoronary aortic valve replacement (AVR) and aortic root replacement (ARR) with cryopreserved homovital homografts with particular reference to sizing and homograft/native aortic root geometrical mismatch.
|
2. Materials and methods |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
There were 216 males and 76 females. Two-hundred and eighteen patients had aortic incompetence, 22 had aortic stenosis and 52 had combined lesions. One-hundred and seventy-seven patients had active infective aortic roots (104 native and 73 prosthetic endocarditis). Of these, 119 had ring abscesses. Of the 115 sterile roots, 15 had a history of a previous healed endocarditis, four had homograft degeneration, 12 had prosthetic valve dysfunction due to valve thrombosis and recurrent paravalvular leaks, 29 had congenital bicuspid valves and one quadricuspid valve, and there were 59 patients with rheumatic/degenerative valves. The preoperative mean left ventricular end-diastolic diameter (LVEDD) was between 60.0±9.9 and 63.7±12.5 mm in patients with aortic valve incompetence and the left ventricular (LV) mass was between 229.3±16.1 and 269.2±25 g in patients with aortic stenosis, respectively.
Matched homografts were implanted in 169 (70%) patients and were defined as an internally sized aortic homograft annulus of 2 mm less than or equal to the recipient aortic root annulus. Mismatched homografts were implanted in 74 (30%) patients and defined as internally sized aortic homografts 3 mm less (undersize, n=46) and 3 mm larger (oversized, n=28) than the recipient aortic root annulus. Among the 184 survivals, 125 matched and 59 mismatched (37 undersize, 22 oversize) homografts were selected for AVR, and 44 matched and 15 mismatched (nine undersize, six oversize) homografts for ARR.
All of the homografts were harvested from hearts of cardiac transplant recipients, or from hearts of multi-organ donors if they were deemed unsuitable for heart transplantation, under sterile conditions. The donor age ranged from 12 to 56 years, with a mean age of 40.8 years.
The implanted homograft valves were obtained from our institution-based bank in Berlin, Bio-Implant Service of Eurotransplant in Rotterdam, the European Homograft Bank in Brussels and the Homograft Bank in Barcelona. The size of homografts implanted varied between 19 and 29 mm (mean, 22.5±3 mm). Hegar dilators were used to measure the internal diameter of the homografts within 2 mm. For the selection of proper homograft size, the host annulus size was measured preoperatively by a two-dimensional echocardiographic measurement.
The intraoperative measured native annulus sizes were compared and correlated with the preoperative measured host annulus sizes and the selected homograft sizes.
2.1. Preoperative echocardiographic prediction of annulus size of the homograft recipient for homograft selection
The homograft sizes selected on the basis of preoperative echocardiographic prediction were plotted against the intraoperative measured annulus diameter. By linear regression analysis, the matched homograft sizes were equal to or 2 mm less than the native annulus (r2=0.73; P<0.001). The estimations of the preoperative echocardiographic non-calcified annular measurements for emergency operations correlated weakly with the intraoperative direct measurements (r2=0.45; P<0.001). The distensibility (percentage systolic/diastolic diameter change) of the annulus of non-calcified aortic lesions was 2.5–5.8% (mean, 3.8%) as compared with 4.5–12% (mean, 7.5%) of a normal aortic root. Patients beyond the age of 50 years demonstrated a distensibility of 2.5–4.5%.
ABO blood group and Human Leucocyte Antigen (HLA) matching between homograft donors and recipients was not practiced in all patients because of logistic reasons. Homograft recipients did not receive immunosuppressive therapy.
The performance of the homografts was analyzed in 243 patients using the homograft/native aortic root geometric annulus mismatch as a risk factor for reoperation.
2.2. Surgical technique
All of the homograft valve implantations were performed through a median sternotomy with the aid of conventional cardiopulmonary bypass with moderate hypothermic perfusion at 30–32°C, and myocardial protection with cold crystalloid cardioplegic arrest was achieved by maintaining the myocardial temperature at 10–12°C.
A preselected and thawed aortic homograft valved conduit was prepared by trimming the proximal muscle to 5 mm below the valve and removing the attached anterior of the mitral valve if it was not needed for annulus enlargement.
2.2.1. The freehand subcoronary implantation (n=207)
The homograft/native aortic root sizing was 2 mm less than or equal to the recipient annulus internal diameter. The implantation was carried out by two suture lines: the proximal line with continuous interrupted or mattress 4-0 Prolene while the homograft conduit was held outside the aortic root. The upper line was accomplished with each sinus of Valsalva scalloped or leaving the non-coronary sinus in place. The depth of the sinus scallop was determined by the distance between the coronary ostia and the proximal line. The commissural posts of a properly sized homograft valve were fixed 1 cm above the recipient's commissures at the sino-tubular junction to facilitate 40–50% leaflet coaptation. The homograft was inserted in the natural position to restore the anatomic units of the aortic root. In seven patients, a homograft with its anterior mitral leaflet was used to enlarge the aortic annulus by extending the incision into the native anterior mitral leaflet as described by Manouguian [16] (Fig. 1) .
|
Additional cardiovascular surgical procedures were: replacement of the ascending aorta (n=12), coronary artery bypass surgery (n=13), mitral valve replacement (n=8) or reconstruction (29), and tricuspid valve reconstruction (n=4).
2.3. Follow-up patients and protocol
Follow-up patients were reviewed periodically either by DHZB cardiologists or by the patient's own cardiologist, initially at 1, 3 and 6 months after homograft valve replacement and at yearly intervals thereafter. At each visit, in addition to the clinical assessment, chest X-rays, electrocardiograms and echocardiograms were obtained and analyzed.
Homograft valve regurgitation was described by color Doppler echocardiography and categorized as follows: grade 0 as no aortic insufficiency, grade I as mild, grade II as moderate and grade III as severe [17].
Follow-up was complete at reoperation and removal of the homograft valve, death with the homograft valve in place or at the last follow-up in survivors with their homograft valves in place.
Follow-up information was available for evaluation in 94% of patients. The maximum follow-up was 14 years and 2 months, and the mean was 52 months. At the latest postoperative evaluation, 81% of the patients were in New York Heart Association (NYHA) class I and 19% in class II clinically.
2.4. Data analysis
Patient survival and valve failure were analyzed with Kaplan–Meier curves, and the probability and standard error were calculated. Continuous data were presented as means±standard deviation (SD). Variables relating to the recipient, including age, sex, underlying aortic root pathology, homograft/patient aortic root annulus size and implantation technique, were analyzed using the Cox proportional hazards model. Linear regression techniques were applied to establish the relationship between host and homograft annular size.
The log-rank test was used to identify statistical significance as an endpoint analysis. Implantation of mismatched homografts and reoperation for homograft explantation, and patient survival in relation to a variable period of homograft operation (1987–1993, and 1994–2000) were included in the analysis.
At the echocardiographic studies, both M-mode recordings and off-line measurements were guided by the two-dimensional image. The ejection fraction was calculated according to Teichholz et al. [18] and the LV mass by means of the cube formula. Blood flow velocity in the LV outflow tract was estimated by pulsed wave Doppler from an apical four-chamber view (sample size of 5 mm). Pressure gradients were calculated according to the simplified Bernoulli equation (Doppler pressure gradient=4x(peak velocity)2). The mean gradients were calculated from off-line planimetry of the continuous wave Doppler recordings [17–19].
Analysis of variance (ANOVA) was used to compare the differences between the three different valve sizes (match, undersize and oversize) and the pre- and postoperative values of LV mass and LVEDD. The results are presented with 95% confidence intervals (CI).
Homograft valve failure was diagnosed when there was an onset of postoperative evidence of clinically mild, moderate or severe valve stenosis or incompetence with or without conduit calcification. The time of censoring for significant valve stenosis or incompetence, or dysfunction due to structural or non-structural deterioration was the date of reoperation and homograft explantation, late death from the causes above, or the last follow-up for the remaining survivors.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
On survival analysis, 92% of the hospital survivors are alive at 10 and 14 years after homograft valve replacement, respectively (Fig. 2) . Mismatched homograft was not a risk factor for death in both periods of operation (risk ratio, 0.35; 95% CI, 0.10–1.23; P=0.17).
|
|
|
|
3.4. Matched homografts
The valve performance was excellent in 155 (91.7%) patients. Fourteen (8.3%) patients, including patients with acute aortic root endocarditis and abscesses and a Marfan patient, developed valve dysfunction due to paravalvular (n=2) and central leaks (n=12, including five with the mini-root technique) because of technical reasons. One patient developed homograft endocarditis with valve incompetence. The freedom from valve explantation for all causes was 77±4% at 10 and 14 years (Fig. 4).
3.5. Correlation between homograft size and patient body surface area
Hemodynamically, there was no disparity between the patient's body surface area and the cardiac output for the selected homograft size. For the 22 mm homograft valves, the mean cardiac index measured postoperatively was 4.09±0.5 (95% CI of 3.62 and 4.55 l/m2), and for a 24 mm valve, the mean cardiac index was 3.8±0.4 (95% CI of 3.40 and 4.2 l/m2). Two patients with a mini-root technique and an undersized homograft demonstrated cardiac indices of 2.7 l/m2.
3.6. Left ventricular remodeling
The postoperative LV remodeling was achieved by reduction of the LV mass (g) for aortic stenosis and the LVEDD (mm) for aortic incompetence. Over a period of 6–60 months, the LV mass reduced by 12–15% and the LVEDD by 10–12%. The aortic annulus diameter of patients, particularly with annuli greater than 27 mm, remained stable without calcification and progressive dilatation.
3.7. Thromboembolism
Our patients were treated with aspirin for 3 months after surgery and have not been complaining of symptoms related to thromboembolic complications, either in the early or late postoperative follow-up periods.
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
Accurate matching of a homograft to the host aortic root either equal to or 2 mm less than the host annulus at subcoronary implantation will therefore provide over 50% leaflet coaptation and excellent long-term clinical and hemodynamic performance which is identical to that of ARR [4–6,9,11,21].
The less distensible stentless xenografts, which appear to be complimentary, rather than a competitive valve substitute to the homograft, require oversizing by 2–3 mm to allow proper matching to a distensible native aortic root in order to maintain over 50% leaflet coaptation.
Besides patients’ special wishes and a contraindication for long-term anticoagulation, the principal indication for ARR in our series was acute endocarditis with burrowing ring abscess. Sepsis and NYHA functional classes III and IV were major risk factors for early death in this group of patients.
Twenty-seven (56.3%) homograft early and late failures in our series were associated with geometric mismatching, and of the matched homografts, 14 (29.2%) failures were technical and seven (14.5%) were due to infection. In our series, cryopreserved homovital homograft valve explantation occurred frequently in the first 7 postoperative years due to structural deterioration in most patients with mismatched homografts. However, the valve performance of matched and oversized homografts remained stable after 2 and 5 years until 14 years postoperatively, respectively (Fig. 4; P=0.001). Similar clinical results have been reported by others [4,6,11].
In small aortic roots or in subvalvular obstructive lesions, it seems prudent to enlarge the aortic root to accommodate a larger aortic homograft. The aortotomy is made through the commissure between the left and non-coronary sinus into the posterior interleaflet triangle. The incision may end at the edge of the anterior leaflet of the mitral valve. When there is no adequate native annulus size or there is a subvalvular obstructive lesion, the incision is extended into the anterior mitral leaflet and onto the roof of the left atrium [16,21]. The attached anterior mitral leaflet of the homograft is used to fill the opening of the native anterior mitral leaflet, thereby enlarging the LV outflow (Fig. 1). The homograft may be used in certain cases of double valve replacement and repair (AVR and anterior mitral valve repair) in order to avoid the need for anticoagulation therapy.
Annulus enlargement for a root replacement using an oversized homograft with the anterior mitral leaflet poses no risk for postoperative valve incompetence, unless there is homograft ring distortion or a geometric mismatch between the native annulus and the homograft.
Although a root replacement technique offers 100% valve competence, homograft ring distortion at implantation in a necrotic region may cause valve incompetence as well [5,14].
Tailoring or reduction annuloplasty for root replacement might be beneficial, but not for the freehand subcoronary implantation technique because of the limited technical possibility to tailor all the three anatomic units of the aortic root.
Homograft valve dysfunction is associated with multilevel geometric disparity at the subannular, annular, sinus and sino-tubular level. Subsequently, non-structural valve failure, such as cusp rupture, leaflet prolapse or distortion, paravalvular leak pseudoaneurysmal formation and commissural displacement, resulting in a central leak, occurred. An ongoing loss of leaflet flexibility and expansibility, such as structural deterioration, may be associated with an undersized valve, as could be demonstrated, particularly in adult patients.
In most patients, despite homograft/native root geometric mismatch, there was no mismatch between the cardiac output and the patient's surface area as demonstrated in the postoperative measurements of the cardiac indices. Like other reports, a regression of LV hypertrophy and LVEDD could be achieved over a period of 6–60 months [12,13,22–24]. Reduction of LV hypertrophy, LVEDD and improvement of the ventricular performance underscore the hemodynamic superiority of a non-obstructive stentless bioprostheses over a stented bioprostheses of the same size [25]. It is possible therefore to have a relatively small homograft valve (up to 3 mm smaller) without a demonstrable gradient. The gradient to valve area was not studied in this series. Rahimtoola, however, underlines that the relationship of the gradient to valve area is curvilinear, and once the effective orifice size of the aortic valve is critically reduced to less than 35% of normal, the gradient rises precipitously [24,25]. We encountered this in two patients with mini-root and 13 patients with subcoronary implantations. The multivariate analysis identified undersized homografts as a risk factor for developing aortic valve incompetence, early and late recurrent endocarditis in the freehand subcoronary implantation technique. Due to standardization of the homograft sizing technique at the beginning of homograft implantation at our institution, the use of mismatched homografts was not specifically associated with surgical inexperience or a particular period of implantation, but rather with availability. Mismatched homografts were rather frequently used at emergency operations for complicated acute infective endocarditis, and the number of implanted matched and mismatched homografts was evenly distributed in the two operation periods (1987–1993 and 1994–2000). The P value of reoperation for explantation of mismatched homografts for the log-rank test adjusted for both implantation periods was 0.003. The use of mismatched homografts, therefore, was not associated with surgical inexperience, but rather, with the availability of proper homograft sizes at the time of emergency operations in both operation periods. Although the risk for reoperation was twice as high in the second operation period (risk ratio, 2.05; 95% CI, 1.07–3.90; P=0.028), it was demonstrated that mismatched homografts did not have an adverse effect on patient survival, and therefore were not a risk factor for death for both operation periods (risk ratio, 0.35; 95% CI, 0.10–1.23; P=0.17).
The durability of the homografts achieved with both implantation techniques therefore was influenced by sizing and the implantation technique as demonstrated in the valve survival analysis in Fig. 3 and by others [6,9,11,12,14,21]. However, a mini-root technique should not be used as a standard root replacement technique because of its potential to develop valve incompetence from crimping the homograft following wrapping with the native aortic root. Furthermore, it can complicate root fibrosis with subsequent late calcification as well as coronary ostial stenosis leading to reoperation. The technique can, however, be beneficial for local root hemostasis, and therefore can be used in selected groups of patients and as a rescue procedure to save patients from catastrophic bleeding and prolonged operation times.
|
5. Conclusions |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
Geometric mismatch between the native and homograft root was not a risk factor for postoperative death, but it was a risk factor for postoperative aortic incompetence, reinfection with pseudoaneurysmal formation and reoperation in patients with a freehand subcoronary implantation technique and a common hazard for the mini-root implantation technique in the first 7 years after operation. The use of mismatched homografts was not associated with surgical inexperience, but rather, was associated with the availability of the proper homograft size at the time of emergency operations. In a situation where there is only a mismatched homograft available, it seems therefore prudent to choose a stentless xenograft as an alternative biological valve substitute, which is readily available in all sizes, to restore the anatomic and functional units of the patient's aortic root and to achieve a competent valve.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. Discussion5. ConclusionsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. A. Kumar, T. McGrath, B. P. Griffin, G. V. Gonzalez-Stawinski, A. Alfirevic, H. Arora, and L. K. Wallace Aortic Valve Homograft: A Rare Cause of Left Ventricular Outflow Tract Obstruction Anesth. Analg., August 1, 2009; 109(2): 361 - 363. [Full Text] [PDF]
|
|
|
|
 |
|
 C. R. Hampton and E. D. Verrier Stentless Aortic Valve Replacement: Autograft/Homograft Card. Surg. Adult, January 1, 2008; 3(2008): 895 - 914. [Full Text]
|
|
|
|
 |
|
 E. Hickey, S. M. Langley, O. Allemby-Smith, S. A. Livesey, and J. L. Monro Subcoronary Allograft Aortic Valve Replacement: Parametric Risk-Hazard Outcome Analysis to a Minimum of 20 Years Ann. Thorac. Surg., November 1, 2007; 84(5): 1564 - 1570. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Yankah, M. Pasic, H. Klose, H. Siniawski, Y. Weng, and R. Hetzer Homograft reconstruction of the aortic root for endocarditis with periannular abscess: a 17-year study Eur. J. Cardiothorac. Surg., July 1, 2005; 28(1): 69 - 75. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Siniawski, H. Lehmkuhl, Y. Weng, M. Pasic, C. Yankah, M. Hoffmann, I. Behnke, and R. Hetzer Stentless aortic valves as an alternative to homografts for valve replacement in active infective endocarditis complicated by ring abscess Ann. Thorac. Surg., March 1, 2003; 75(3): 803 - 808. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. R. Hampton, A. J. Chong, and E. D. Verrier Stentless Aortic Valve Replacement: Homograft/Autograft Card. Surg. Adult, January 1, 2003; 2(2003): 867 - 888. [Full Text]
|
|
|
|
 |
|
 A C. Yankah Forty Years of Homograft Surgery Asian Cardiovasc Thorac Ann, June 1, 2002; 10(2): 97 - 100. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
