Primary Aortic Valve Replacement With Cryopreserved Aortic Allograft

The aortic allograft valve (AAV) has been in use for the surgical treatment of aortic valves since the late 1950s. Major advantages of the AAV in early studies^1–5 were related to good hemodynamics of AAV function, without the need for anticoagulation. Relatively recent studies have shown that the AAV should be considered the valve of choice in patients with active endocarditis due to the low risk of recurrent postoperative valve infection.^6–10 However, like other tissue valves, AAV durability is still suboptimal.^11–13 AAV degeneration usually accelerates 10 to 15 years after implantation, making a reoperation a likely scenario for younger patients.⁷ The causes of AAV degeneration are under intense investigation, but they seem to be complex and multifactorial.¹⁴ Immunological rejection, mechanical stress, ischemia, and/or chemical insults all may cause AAV degeneration, with the likely trigger mechanism being endothelial activation in response to tissue injury leading to smooth muscle cell proliferation and cell apoptosis. Thus, recipient and donor factors, including methods of AAV sterilization and preservation that may influence the severity of immunological reaction and factors such as implantation technique that can modify the mechanical function of the allograft, may all influence AAV longevity.¹⁵

The aims of this study were (1) to document, using transthoracic echocardiography, the characteristic features of AAV degeneration and (2) to investigate the potential factors influencing the development of specific forms of AAV dysfunction.

Methods

Subjects

From December 1969 to July 1998, 1022 patients (mean age, 48.4±16.9 years; range, 1.2 to 80.8 years; 669 were male) underwent primary AAV replacement at the Department of Surgery, the Prince Charles Hospital and St Andrew’s War Memorial Hospital, Brisbane, Australia. Until June 1996, when the current study commenced, 223 patients had died, 126 patients had undergone reoperation, and 7 patients were lost to follow-up. Thus, from the entire group of 1022 patients, 666 AAV patients were eligible for the inclusion into the study. However, 51 patients did not undergo echocardiographic assessment for various reasons, and 37 AAV patients with a follow-up <1-year and 8 patients with fresh (“nonviable”) AAV were excluded from the study. The final study group consisted of 570 patients. In our previous study based on the current cohort of patients, we assessed AAV durability and the associated risk of reoperation in view of AAV structural degeneration.¹⁶ All study patients had cryopreserved (in liquid nitrogen at −196°C) AAVs.⁷

Cryopreserved AAVs are considered to have at least 50% leaflet viability (interstitial cells) at the time of valve implantation.¹⁷ All AAVs came from the Queensland homograft valve bank at the Prince Charles Hospital, where all donor characteristics were kept for permanent record. Sizing of the annulus was performed in a meticulous order to minimize potential measurement-related errors. For the donor, this was done by distending the aorta to a pressure of ≈80 mm Hg and measuring the ventricular aspect of the actual diameter of the valve annulus with coapted valve leaflets. For the host, measurements were made using standard, commercially available valve size obturators (of the St Jude type).

AAV replacement was performed using the subcoronary technique in 278 patients, the aortic root replacement technique in 258 patients, and the inclusion cylinder technique in 34 patients (Table 1). After observing that the subcoronary technique is (1) more difficult to perform in an asymmetric/dilated aortic root and (2) related to immediate trivial/mild AAV regurgitation, the root replacement technique was used predominantly in our center to implant AAV (since 1985).¹⁶ All operations were performed with a single, 6-member surgical team, with two-thirds of the operations performed by 2 senior surgeons. Technical differences in the different surgical procedures of AAV replacement were previously described in detail.¹⁸

Study Protocol

Transthoracic echocardiographic studies were performed between June 1996 and December 1998 in all 570 patients in the following echocardiographic laboratories in Brisbane and other metropolitan centers: The Prince Charles Hospital, St Andrew’s War Memorial Hospital, Wesley Hospital, Princess Alexandra Hospital, Toowoomba Base Hospital, and Queensland X-ray service. Each patient underwent standard echocardiographic study according to the same protocol using commercially available ultrasound scanners. Informed consent was obtained from all subjects before the study. AAV stenosis was divided into 3 grades of severity, and AAV regurgitation was divided into 4 grades using previously validated criteria (Table 2).^19–23

Statistical Analysis

The data are expressed as mean±SD for continuous variables and as frequency [number (%)] for dichotomous variables. ANOVA with Scheffe’s F adjustment for multiple comparisons was used to assess the differences between AAV patients operated on using different surgical techniques. χ² tests were used when appropriate. Cox regression analysis was performed to assess the relationship between the patient/donor factors (age, sex, aortic annulus size, blood groups), patient clinical information (NYHA classification, type of native aortic valve disease, other underlying disorders) and the severity of AAV dysfunction. P<0.05 was considered significant.

Results

In all 570 patients, the average time to an echocardiogram after the AAV replacement was 6.8±4.1 years (range, 1.0 to 22.9 years). In the subcoronary group, this time period was longer than in the root replacement or inclusion cylinder group (9.1±4.1 versus 4.2±2.4 and 7.1±1.3 years, respectively; P<0.001).

In the entire group, 27.9% patients had no signs of AAV dysfunction at the time of an echocardiogram (Figure 1). A total of 309 patients (54.2%) had grade 1 AAV regurgitation or stenosis (44.2% and 10.0%, respectively); 76 patients (13.3%) had grade 2 AAV regurgitation (10.9%) or grade 2 AAV stenosis (2.5%). Only 4.6% had moderately severe or severe AAV dysfunction (regurgitation of grades 3 or 4 or stenosis of grade 3). Grade 3 and 4 AAV regurgitation was present in 3.9% of patients, and grade 3 AAV stenosis existed in <1% patients. As expected, AAV dysfunction occurred more often at a later stage after AAV implantation. However, there were no differences in time between patients with grade 1 regurgitation or grade 3 AAV stenosis and patients with no signs of AAV dysfunction (6.2±3.3 and 6.7±2.7 versus 5.3±3.4 years, respectively; P=NS). Also, there was no difference between patients with grade 2 to 4 AAV regurgitation and those with grade 2 or 3 AAV stenosis (8.9±4.6 versus 10.3±6.2 years; P=NS).

• Download figure
• Download PowerPoint

In the subcoronary subgroup, only 15.5% of patients compared with 42.2% in the root replacement and 20.6% in the inclusion cylinder subgroup had no signs of AAV dysfunction. The root replacement group had the smallest number of patients with grade 2 to 4 AAV regurgitation (5.4%) compared with the subcoronary (23.7%) or the inclusion cylinder technique (14.7%). Grade 2 and 3 AAV stenosis was observed in a similar percentage of patients in all subgroups (4.3% in the subcoronary, 1.9% in the root replacement, and 2.9% in the inclusion cylinder subgroups). Also, hemodynamic indices of AAV function were lower in the subcoronary subgroup compared with the other subgroups (Table 3). After 5 to 10 years after AAV replacement, significant valve dysfunction (grade ≥2) was present in 21.5% of patients in the subcoronary subgroup (n=38) compared with 11.3% (n=7) in the root replacement subgroup (P<0.001). By 10 to 15 years after AAV replacement, grade ≥2 AAV dysfunction was present in 40% of patients (n=20) in the subcoronary subgroup, but no significant increase in dysfunction (P<0.001) was observed in patients in the root replacement subgroup (n=12; Figure 2).

• Download figure
• Download PowerPoint

Cox regression analysis of factors influencing AAV degeneration showed that in the subcoronary subgroup, AAV dysfunction (grade ≥2) was related to smaller host annulus size and the younger age of the recipient. In the root replacement subgroup, AAV degeneration was related to the older age of the donor, but not recipient, and smaller host annulus size (Table 4).

Discussion

The “ideal” replacement aortic valve, that is, one with perfect hemodynamics, devoid of early and late morbid events, readily available, and usable in all patients with any type of aortic valve disease, does not yet exist and probably never will. However, throughout the quest for this ideal valve, the AAV has always been appealing because human valve tissue is used to replace a diseased human valve. Over the last 3 decades, continuing research work on the cryopreservation technique of “viable” AAV has produced an improvement in their durability.^{7,12,13,15,24–26} However, the causes and the mechanisms of AAV degeneration are not clearly understood.¹⁴

A recently published analysis of our entire cohort of 1022 AAVs demonstrated clear advantages of the aortic allograft for the treatment of acute endocarditis and for use in patients aged ≥20 years.¹⁶ However, younger patients (≤20 years) experience only a 47% freedom from reoperation from structural degeneration. The current study shows an echocardiographic pattern of AAV dysfunction up to a period of 22 years. We have described the relation between the types and the severity of AAV dysfunction, including an analysis of different types of surgical replacement procedures. We have identified patient/donor-related factors influencing AAV degeneration. The data from the entire cohort of AAV patients operated on in our institution showed that the freedom from reoperation due to structural AAV degeneration was indeed similar after the subcoronary and the root replacement technique.¹⁶ However, the present study indicates that in patients who have aortic root replacement, AAV degeneration is less frequent (apparent) and does not accelerate after 10 years compared with the subcoronary technique. We think that our findings may have an important impact on establishing more strict selection criteria for surgical replacement procedure type and patient/donor factors for AAV replacement.

Findings

Like published data from others,¹² we have shown that AAV replacement can give a good result for up to 22 years, although normal AAV function on echocardiogram is rare after 15 years. However, there are a number of important issues that differentiate our study from others. At our institution, from mid-1975 on, all aortic valves taken from donor hearts were cryopreserved immediately (within 24 hours) after sterilization with antibiotics. The purpose of the cryopreservation method of AAV storage was to retain the viability of the fibroblasts within the leaflets, with the aim of markedly enhancing subsequent valve durability.²⁰ In the present study, the whole group received cryopreserved AAVs. Thus, our group of AAVs was more homogenous than those used in other studies,¹² where only 20.6% patients had cryopreserved valves. In our institution, since 1985, most surgeons performed the root replacement technique, but other centers still preferred subcoronary implantation. In this study, both subgroups (subcoronary and root replacement) were represented in a similar percentage; however, the time between surgery and echocardiographic follow-up was shorter in the latter subgroup. To our knowledge, this is also the largest AAV group studied, with the longest follow-up by echocardiography.

We have found that only 17.9% of patients had at least moderate AAV dysfunction and, among them, only 3.2% had significant (grade ≥2) AAV stenosis. This is in agreement with previously published data, which showed that the major mode of AAV failure has been aortic regurgitation.²⁷ As expected, AAV dysfunction occurred more often at a later stage after AAV replacement. Our results are also in agreement with previously published data that showed that AAV primary degeneration rapidly accelerates 10 years after the operation when using the subcoronary approach.^7,11,28 Although our study suggests that in the root replacement technique AAV dysfunction does not accelerate after this time period, a follow-up of >15 years is required to strengthen our findings. In addition, we have shown that there is a different echocardiographic pattern of AAV degeneration in different types of surgical valve replacement procedures. Significant AAV stenosis was observed in a similarly small percentage in all the subgroups, but the root replacement subgroup had the smallest number of patients with significant AAV regurgitation. We are not sure how to explain these differences. Perhaps they are related to the relatively short time of observation in the root replacement subgroup. However, our previous results and the results of others have also shown the advantage of the root replacement technique over the subcoronary technique.^12,24,29 In cases when the complete aortic root is implanted instead of the aortic valve in the subcoronary position, the risk of valve degeneration seems to be smaller. This is due to a lesser mechanical and immunological insult to the valve itself and to the preservation of the “short channel” connecting the left ventricle to the aorta, which can interact well with the surrounding structures.³⁰

In the subcoronary subgroup, AAV dysfunction was more frequent in younger recipients. In contrast, in the root replacement technique, older donors (not recipients) had a negative effect on AAV longevity. Although it is already known that a younger recipient carries a higher risk of AAV degeneration,^12,27,28 we have found that this risk occurs predominantly with the subcoronary technique but not in the root replacement surgical procedure. A recent report¹² showed that an older donor (age >65 years) has an inverse effect on AAV durability. Although in our study the oldest donor was 58 years old, we have confirmed these findings. However, this relationship was predominantly seen in the root replacement rather than the subcoronary subgroup.

Previously published data emphasize the importance of accurate measurement of the recipient aortic annulus size, with the aim of more appropriate selection of the type of surgical AAV replacement.^12,27–30 It has been suggested that patients with a dilated aortic annulus should be considered for the root replacement rather than the subcoronary technique of AAV implantation. In the present study, we have shown that in patients who underwent either subcoronary or the root AAV replacement technique, a small recipient aortic annulus size had a negative effect on valve durability.

Limitations

In this study, all subgroups divided according to the surgical AAV replacement technique were not perfectly matched. Patients who had AAVs inserted using the root replacement technique were younger than those who had the subcoronary or the inclusion cylinder technique. The smallest aortic annulus diameter size was observed in the subcoronary subgroup, and the biggest diameter was present in those patients who had inclusion cylinder technique. Also, the patients who underwent subcoronary AAV replacement compared with the root and/or the inclusion cylinder AAV replacement had a lower incidence of native aortic valve regurgitation and the highest incidence of degenerative native aortic valve disease. Although patients who died or had a reoperation due to structural AAV degeneration were not included in the current study, they were all discussed in detail in our previous report.¹⁶ The present study was a “snapshot” of AAV function at a relatively small point in time, resulting in variable follow-up periods in studied subgroups. However, we think that our echocardiographic study group was representative of the entire cohort of AAV patients and, thus, our results, which include data from all AAV replacement techniques, are of important clinical value. We did not assess the degree of aortic calcification because echocardiography enables only a subjective assessment of calcification and, because our studies were performed in different centers, the risk of nonhomogenous classification was high. Whenever possible, AAV area measurements were used to minimize possible factors that influence Doppler velocity measurements.

Footnotes