Aortic Compliance After Root Replacement With Decellularized Homografts Versus in Donor Age-Matched Healthy Controls

Abstract

Objectives

We sought to evaluate compliance and flow of the ascending aorta of patients with decellularized aortic homografts compared to donor age-matched healthy controls.

Methods

Male patients and donor age-matched male healthy controls were included. Cardiac function was evaluated by retrospectively electrocardiography-gated cine balanced steady-state free precession magnetic resonance imaging (MRI). Time-resolved 2- and 3-dimensional phase-contrast sequences were used to determine relative area change and pulse wave velocity as surrogate parameters for vessel compliance as well as maximum blood flow velocity.

Results

Thirteen patients were matched according to the age of their homograft donor (median 42 years, interquartile range [IQR] 32-50) to 7 healthy controls (median 40 years, IQR 36-48). Time to post-operative MRI was 3.33 (1.33-4.50) years. Relative area change in the proximal ascending aorta was significantly lower in the homograft group compared to healthy controls (26%, IQR 23-44 vs 38%, IQR 24-44, P < .001), with no significant difference observed in the distal ascending aorta (22%, IQR 22-33 vs 34%, IQR 22-41, P = .438). Maximum blood flow velocity in the proximal ascending aorta was significantly higher in the homograft group compared to healthy controls (168 cm s⁻¹, IQR 148–188 vs 115 cm s⁻¹, IQR 114-120, P = .009).

Conclusions

Decellularized aortic homograft patients seem to have a reduced compliance of the proximal ascending aorta compared to donor age-matched healthy controls. This may be attributable to the in vitro decellularization process or post-operative graft degeneration. These findings highlight the ultimate need for follow-up data to understand the long-term in vivo effects of decellularized human tissue. This study is a follow-up study of the patients included in the ARISE Study registered on ClinicalTrials.gov (NCT02527629). For the purposes of this manuscript, healthy individuals were subsequently recruited to serve as the control group.

Decellularized aortic homografts (DAHs) are a promising alternative in the surgical treatment of aortic valve and ascending aortic diseases, especially for younger patients.

GRAPHICAL ABSTRACT

INTRODUCTION

Decellularized aortic homografts (DAHs) are a promising alternative in the surgical treatment of aortic valve and ascending aortic diseases, especially for younger patients.^1,2 This approach eliminates the need for synthetic prosthetic materials and long-term anticoagulation, making DAH particularly attractive for patients who require aortic replacement but wish to avoid the risks of thromboembolism and early degeneration associated with mechanical or biological prostheses.^3,4 Recent studies have highlighted the potential of DAH to restore normal aortic haemodynamics and reduce the likelihood of complications such as thromboembolism and endocarditis.^5,6 However, there is still only limited data on the biomechanical properties of DAH, particularly regarding the compliance and elasticity of the implanted aortic wall tissue compared to healthy aortic tissue over time.

A critical aspect of aortic function is its compliance, which refers to the ability of the aorta to expand and contract in response to changes in blood pressure during the cardiac cycle. Compliance is essential for damping the pulsatile nature of blood flow and reducing the workload on the heart (ie, the Windkessel principle).^7,8 Aortic compliance is impaired by increased stiffness.

Distensibility as reflected by relative area change (RAC) and pulse wave velocity (PWV) are magnetic resonance imaging (MRI) surrogate parameters for vessel stiffness.⁹ Relative area change quantifies the relative change in cross-sectional area of the aorta between systole and diastole. A reduced RAC indicates increased local stiffness and impaired aortic function, which can lead to cardiovascular complications such as hypertension and left ventricular hypertrophy.^10,11 Pulse wave velocity, on the other hand, measures the speed at which pressure waves travel along the vessel. An increased PWV is associated with higher arterial stiffness, which in turn elevates systolic blood pressure and increases cardiac afterload.^12,13 Studies have shown that elevated PWV is a predictor of cardiovascular events, including myocardial infarction and stroke, and is an indicator of vascular ageing.^11,14

Given the importance of both RAC and PWV in assessing aortic function, we sought to evaluate whether long, valved DAHs exhibit a compliance similar to that of the native aorta of a donor age-matched healthy control group. To our knowledge, this is the first MRI-based analysis that matches the controls’ age to the age of the homograft donors.

PATIENTS AND METHODS

Study population

This study included 2 cohorts: patients who underwent DAH implantation, and healthy controls, who were matched to the homograft donor’s age.

The inclusion criteria required participants to be male and at least 18 years old, with no contraindications for MRI, such as incompatible implants. All patients had undergone surgery at our institution between 2014 and 2020 and had been classified as New York Heart Association Class I post-operatively, ensuring exclusion of any cases with impaired left ventricular function or suboptimal surgical results that could potentially bias the outcomes. During the study period, we implanted a total of 152 decellularized aortic homografts in both adults and children, including 67 adults over 18 years of age. However, for this study, we included only those patients who received long decellularized homografts with concomitant replacement of the proximal ascending aorta. These were cases in which the ascending aorta was dilated—from the supracoronary region up to the brachiocephalic trunk—and therefore required replacement together with the aortic valve. Patients who received shorter homografts were excluded to ensure accurate parameter measurements and to minimize the risk of unintentionally assessing regions affected by anastomotic sutures or native tissue. All participants in the control group were specifically age-matched, recruited for this research, and provided informed consent prior to inclusion.

Surgical procedure

Surgical access was achieved via full median sternotomy. Cold blood cardioplegia was employed for myocardial protection during cardiopulmonary bypass. The aortic homografts were obtained from 2 tissue banks (Deutsche Gesellschaft für Gewebetransplantation, Hannover, and the European Homograft Bank, Brussels) and were decellularized at Corlife oHG, Hannover, according to previously published protocols.¹⁵

After successful processing, long-valved DAHs were implanted using a full root technique, with distal anastomoses typically placed 1-2 cm proximal to the innominate artery. Surgeries were conducted using standard cardiopulmonary bypass, hypothermic circulatory arrest was not employed. No additional prosthetic material was used during DAH implantation.

Magnetic resonance imaging and data analysis

Magnetic resonance imaging was conducted using a 1.5-T scanner (MAGNETOM Avanto, Siemens Healthineers GmbH, Erlangen, Germany), equipped with a 8-channel torso phased-array coil. Cardiac function was evaluated by retrospectively electrocardiography-gated cine balanced steady-state free precession sequences. Short-axis views covering the whole heart were acquired during short inspiratory breath-holds. Blood flow was evaluated with phase-contrast imaging. Time-resolved 2-dimensional (2D) phase-contrast images (ie, 2D flow) were acquired in 2 locations perpendicular to the vessel: (1) proximal ascending aorta and (2) distal ascending aorta. Time-resolved 3-dimensional (3D) phase-contrast imaging (ie, 4-dimensional [4D] flow) was performed within a sagittal oblique volume covering the thoracic aorta. Two-dimensional and 4D flow imaging was prospectively gated by electrocardiography and was conducted during free breathing. Four-dimensional flow imaging used a respiratory navigator placed at the lung–liver interface. Magnetic resonance imaging data were analysed using the software cvi42 version 5.17 (Circle Cardiovascular Imaging Inc., Calgary, Canada). Semi-automatic segmentation was applied with manual adjustment where necessary. Left ventricular cardiac index and ejections fraction were derived from functional cardiac images. Relative area change—used as robust and easily interpretable surrogate marker of vessel compliance which does not depend on complex modelling or assumptions^16,17—was derived from 2D flow images, measuring differences in cross-sectional area between systole and diastole as follows: (A_max−A_min)/A_min. Regarding 4D flow images, post-processing involved the segmentation of the ascending aorta and the definition of 3 distinct locations at which cross-sectional planes were drawn: (1) aortic valve, (2) proximal ascending aorta, and (3) the distal ascending aorta. Maximum mean blood flow velocity (V_max) was measured at all 3 locations, whereas 4D flow PWV was calculated from the aortic valve to the distal ascending aorta. The main variable of interest was the difference in ascending aortic RAC between both groups, the other analysed variables were mainly looked at for hypotheses generation.

Statistical analysis

Metric data are reported as median with interquartile range (IQR). The Mann–Whitney U-test was used for comparisons between 2 groups as most data did not follow a normal distribution. The significance level was set to 5%. The Hodges–Lehmann estimator was used to calculate the median difference between 2 groups, and a 95% CI was reported for this difference. For post hoc power analysis, we used ‘PASS 2025 Power Analysis and Sample Size Software (2025), NCSS, LLC., Kaysville, Utah, USA, ncss.com/software/pass’. In order to visualize possible developments over time in this study, we plotted several examined parameters, such as RAC, in relation to the time since implantation and fitted a regression line (see Supplementary Material).

Study setting and ethical approval

This study is a follow-up of the patients included in the ARISE Study. The ARISE Study was registered on ClinicalTrials.gov (NCT02527629) and recognized as a Post Authorization Safety Study (EU PAS 10201) by ENCePP. It was also registered with the German Federal Institute for Vaccines and Biomedicines (Ref. NIS322). The ARISE Study was originally designed without a control group. For the purposes of this manuscript, healthy individuals were subsequently recruited to serve as the control group.

This MRI study of patients and healthy controls was approved by the Institutional Ethics Committee of Hannover Medical School (Hannover, Germany; No. 7960_BOS_2018c), which includes permission for data storage and further analysis. All participants provided written informed consent prior to inclusion in the study.

RESULTS

We included 13 male patients, who had undergone DAH implantation between 2014 and 2020 at our institution, with a median age of 27 years (IQR 27-39), and 7 healthy male controls with a median age of 40 years (IQR 36-48 years). These healthy controls were matched to the median donor age of 42 years (IQR 32-50; P = 0.362).

Among the 13 DAH patients, the indication for surgery was aortic regurgitation in 5 patients, aortic stenosis in 3 patients, and combined valvular pathology in 5 patients. These patients had previously undergone 10 cardiac procedures in total (see Table 1). Nine of them had a bicuspid aortic valve, while all healthy controls had tricuspid aortic valves. None of the DAH patients needed any additional left ventricular outflow tract procedures, such as sub-annular resection, Konno operation, or patch plasty, during their DAH implantation. The median follow-up period for MRI after DAH surgery was 3.33 (IQR 1.33-4.50) years. Detailed characteristics, including heart function data for both groups, are presented in Table 1.

Table 1.

Demographic Data and Heart Function in DAH Patients and Healthy Controls

Characteristics	DAH (n=13)	Control (n=7)	Hodges–Lehmann median difference with the 95% CI	P-value
Homograft age/control group age (years)	42 (32-50)	40 (36-48)	1.26 (−7.89 to 10.41)	.362
Time interval between surgery and MRI (years)	3.33 (1.33-4.50)	NA
Number of previous surgeries on cardiopulmonary bypass involving the left ventricular outflow tract	10	NA
Diameter of the implanted DAH (mm)	27 (26-27)	NA
Diameter of the proximal ascending aorta (mm)	32 (28-36)	33 (28-36)	0 (−3 to 5)	.817
BSA (Mosteller formula, m2)	2.04 (1.97-2.16)	2.06 (1.97-2.28)	0.08 (−0.14 to 0.26)	.485
Systolic blood pressure (mmHg)	120 (112-120)	117 (115-120)	0 (−5 to 7)	.938
Diastolic blood pressure (mmHg)	70 (63-76)	70 (70-75)	0 (−5 to 10)	.757
Heart rate (beats min−1)	67 (60-76)	60 (59.5-68.5)	−3 (−15.67 to 7.95)	.699
Height (cm)	182 (177-187)	178 (175-181)	−4 (−13 to 5)	.393
Weight (kg)	80 (78-92)	89 (81-104.6)	9 (−7 to 24)	.275
LVCI (L min−1 m−2)	3.9 (3.4-4.1)	3.8 (3.5-3.9)	−0.56 (−0.89 to 0.14)	.067
LVEF (%)	66 (60-69)	61 (59-65)	0.25 (−7.46 to 7.23)	.877

Data are reported as median with interquartile range (IQR).

Abbreviations: BSA, body surface area; DAH, decellularized aortic homograft; IQR, interquartile range; LVCI, left ventricular cardiac index; LVEF, left ventricular ejection fraction; MRI, magnetic resonance imaging; n.a., not applicable; SD, standard deviation.

The results of the 2D and 4D flow analysis are shown in Table 2. Relative area change in the proximal ascending aorta was significantly reduced in the DAH group (26%, IQR 23-44) compared to the control group (36%, IQR 24-44, P < .001). However, as shown in Figure 1, no significant difference was observed in the distal ascending aorta between DAH patients (22%, IQR 22-33) and controls (34%, IQR 22-41, P = .588).

Table 2.

Two-Dimensional and 4D Flow Magnetic Resonance Imaging Data of DAH Patients and Healthy Controls

Variable	Location	DAH (n=13)	Control (n=7)	Hodges–Lehmann median difference with the 95% CI	P-value
V max (cm s−1)	AV	159 (130-209)	136 (130-151)	−13.85 (−74.64 to 15.27)	.261
V max (cm s−1)	PAA	168 (148-188)	115 (114-120)	−38.78 (−73.07 to 25.08)	.010
V max (cm s−1)	DAA	156 (145-190)	118 (116-174)	−27 (−53.96 to 21.63)	.384
RAC (%)	PAA	26 (23-44)	36 (24-44)	28.29 (7.46 to 36.67)	<.001
RAC (%)	DAA	22 (20-33)	34 (22- 41)	6.94 (−5.01 to 21.73	.438
PWV (m s−1)	AA	5.11 (4.17-8.05)	4.53 (3.79-6.99)	−0.14 (−3.52 to 2.52)	.536

Data are reported as median with interquartile range (IQR). Values shown in bold denote statistical significance, whereas values in italics are used to facilitate easier interpretation.

Abbreviations: AA, ascending aorta; AV, aortic valve; DAA, distal ascending aorta 1 cm proximal to the origin of the innominate artery; DAH, decellularized aortic homograft; IQR, interquartile range; PAA, proximal ascending aorta 1 cm above the sinotubular junction; PWV, pulse wave velocity; V_max, maximum mean velocity.

Figure 1.

Relative Area Change in Decellularized Aortic Homograft (DAH) Group and Healthy Control Group at 2 Distinct Locations at Which the Aorta was Segmented in a Cross-Sectional Plane. Abbreviations: DAA, distal ascending aorta 1 cm proximal to the origin of the innominate artery; PAA, proximal ascending aorta 1 cm above the sinotubular junction

V _max in the proximal ascending aorta was significantly higher in the DAH group, with a median of 168 cm s⁻¹ (IQR 148-188), compared to 115 cm s⁻¹ (IQR 114-120) in healthy controls (P = .009). No statistically significant differences were found for V_max at the aortic valve level (159 cm s⁻¹, IQR 130-209 vs 136 cm s⁻¹, IQR 130-151, P = .261) or in the distal ascending aorta (156 cm s⁻¹, IQR 145-190 vs 118 cm s⁻¹, IQR 116-174, P = .384), as can be seen in Figure 2.

Figure 2.

Maximum (V_max) Velocities in Decellularized Aortic Homograft (DAH) Group and Healthy Control Group at 3 Distinct Locations at Which the Aorta was Segmented in a Cross-Sectional Plane. Upper values show decellularized aortic homograft patients and lower values show the results in healthy control group. Bold numbers delineate results of DAH patients. Abbreviations: AV, aortic valve; DAA, distal ascending aorta 1 cm proximal to the origin of the innominate artery; IQR, interquartile range; PAA, proximal ascending aorta 1 cm above the sinotubular junction

Furthermore, PWV did not differ significantly between the 2 groups, with DAH patients showing a median PWV of 5.11 m s⁻¹ (IQR 4.17-8.05) compared to 4.53 m s⁻¹ (IQR 3.79-6.99) in healthy controls (P = .536).

DISCUSSION

In this study, we analysed compliance of long DAHs in comparison to the native aorta of donor age-matched healthy controls using MRI-derived surrogate parameters. Primarily, we observed a reduced RAC of the proximal ascending aorta of DAH patients in comparison to the healthy controls.

Donated homografts in the majority derive from elderly donors. Higher age has been clearly shown to increase stiffness of the ascending aorta.^11,14,18 However, due to the design of our study, we eliminated this confounding factor as we matched for donor age. Therefore, a reduced RAC, as seen in the DAH group, may be rather indicative of localized radial stiffening, either due to the decellularization process itself¹⁹ or due to subsequent degenerative changes within the graft. Recipient cell matrix interactions or immune-mediated responses could also impact its mechanical behaviour.^20,21

Less compliant vessels with impaired Windkessel functionality eventually lead to altered haemodynamics, such as higher flow velocities (ie, Bernoulli’s principle), which in turn increase the risk of complications such as graft degeneration or aortic dilation over time.^22–24

In our study, V_max of the DAH group was higher in comparison to the healthy controls, though not significantly, and with 1.59 m s⁻¹ in a range, that is clinically considered as almost normal.²⁵ Interestingly, PWV as a measure of arterial stiffness over a longer segment of the aorta did not differ significantly between DAH and control group compared to the results of our previous study.⁶ This suggests that the longitudinal stiffness of the ascending aorta may not be as profoundly affected 3 years post-operatively as the radial compliance. Our longitudinal analysis shows that changes in the relative area change of the distal ascending aorta and the increase in V_max over time may indicate a gradual reduction in aortic elasticity with the time after implantation. However, our study group is too small to draw definitive conclusions (see Supplementary Material).

While the clinical implications of these findings require further investigation, it is important to combine those findings with the clinical results of DAH. Even though we observed reduced proximal aortic compliance in comparison to controls, DAH may be the only option, which at all offers compliance for patients needing an aortic root replacement procedure in combination with a replacement procedure for a dilated ascending aorta.⁶ In comparison, a synthetic Gore-Tex vascular prosthesis has almost no compliance.²⁶ Moreover, an in vitro comparison of decellularized and non-decellularized human aortic and pulmonary homografts demonstrated that the hydrodynamic and expansion properties of pulmonary and aortic roots were not significantly affected by the decellularization process.²⁷ Yet, in vivo studies have reported that autograft roots become stiffer than native aortic roots 1 year after the Ross procedure.²⁸ Additionally, increased stiffness in the ascending aorta has been observed in patients following aortic valve replacement surgery, which both aligns with our findings.²⁹

Specific research is needed to explore whether specific modifications in the decellularization process could influence the observed compliance reduction and enhance the long-term performance of DAH implants. This also includes the assessment of decellularized xenografts for aortic root replacement.³⁰

Our findings highlight the potential role of advanced cardiovascular imaging techniques in monitoring haemodynamic properties of DAH over time. Non-invasive flow MRI, that does not even need the use of contrast agents, allows for detailed regional assessment of blood flow dynamics and offers surrogate parameters for the estimation of vessel compliance as well as stiffness, which may aid in early detection of abnormal remodelling processes. As DAH implants are increasingly utilized in clinical practice, understanding the impact of decellularization, degeneration, graft biomechanics, immunogenicity, and patient outcome is becoming more and more important and ultimately helps improving the durability and function of these grafts.

Limitations

Indeed, the cohort size is small and therefore a limitation. However, post hoc power assessment indicated a value of 84.4%. Hence, it is safe to say that small cohort size is no major limitation for the actual post-operative period we examined. Longitudinal studies have been initiated to assess longer term results. Furthermore, all study members were male. This was intentional to prevent potential gender-induced confounding effects in this small cohort, but of course limits applicability of the results to female patients. Additionally, we obtained 2D and 4D flow images, as the former offered a better temporospatial resolution. This is why we derived RAC from 2D flow data. Further development of 4D flow sequences will however offer better 4D flow resolution at a reasonable scan time. Moreover, it is important to note that we cannot definitively determine whether the measurements at the level of distal ascending aorta correspond to normal tissue, anastomotic sutures, or homograft tissue. Consequently, the non-existing differences between the 2 groups in terms of relative area change at the distal level of the ascending aorta should be interpreted with caution. Also, previous cardiac surgery may lead to scar tissue formation in the mediastinum, which could affect vessel compliance (ie, RAC) and consequently influence velocity parameters. As we did not have any previously operated participants in our control group, this could represent another potential limitation of our study. Lastly, DAHs from young donors were not analysed within this study. Data regarding the performance of DAH from these young donors are lacking, although it could potentially alter current allocation algorithms.

CONCLUSION

In conclusion, the results of this study suggest a reduced local aortic compliance in the proximal ascending aorta of DAH patients compared to age-matched healthy controls due to increased vessel stiffness. The decellularization process itself or subsequent degenerative effects may contribute to these localized graft changes. This highlights the need for long-term studies as, ultimately, a better understanding of DAH long-term behaviour is crucial for optimizing patient outcomes.

AUTHOR CONTRIBUTIONS

Tomislav Cvitkovic: conceptualization, data curation, formal analysis, investigation, methodology, writing—original draft. Alexander Horke: data curation, methodology, supervision, writing—review and editing. Dmitry Bobylev: data curation, investigation, methodology, writing—original draft. Murat Avsar: investigation, methodology, supervision, writing—review and editing. Theresa Holst: data curation, resources, supervision, writing—review and editing. Philipp Beerbaum: investigation, methodology, supervision, writing—review and editing. Dietmar Boethig: formal analysis, software, visualization, writing—review and editing. Elena Petena: data curation, resources, supervision, writing—review and editing. Valery Tsimashok: data curation, resources, supervision, writing—review and editing. Mechthild Westhoff-Bleck: conceptualization, data curation, resources, supervision. Marcel Gutberlet: conceptualization, data curation, resources, supervision. Frerk Hinnerk Beyer: conceptualization, data curation, resources, supervision. Frank Wacker: data curation, resources, supervision, writing—review and editing. Arjang Ruhparwar: data curation, resources, supervision, writing—review and editing. Jens Vogel-Claussen: conceptualization, project administration, resources, supervision, writing—review and editing. Samir Sarikouch: conceptualization, funding acquisition, project administration, supervision, writing—review and editing. Christoph Czerner: conceptualization, formal analysis, methodology, project administration, resources, supervision, writing—original draft

SUPPLEMENTARY MATERIAL

Supplementary material is available at ICVTS online.

FUNDING

This study was supported by a grant from the European Union’s HORIZON 2020 Programme under grant agreement no. 643597 (www.arise-clinicaltrial.eu).

CONFLICTS OF INTEREST

None declared.

DATA AVAILABILITY

The data underlying this article will be shared on reasonable request to the corresponding author.