Interact CardioVasc Thorac Surg 2008;7:407-411. doi:10.1510/icvts.2007.166835
© 2008 European Association of Cardio-Thoracic Surgery
Institutional report - Valves |
Aortic root motion remodeling after aortic valve replacement – implications for late aortic dissection
Carsten J. Bellera,*,
Michel R. Labrosseb,
Siegfried Hagla,
Martha M. Gebhardc and
Matthias Karcka
a Department of Cardiac Surgery, University of Heidelberg, INF 326, 69120 Heidelberg, Germany
b Department of Mechanical Engineering, University of Ottawa, Ottawa, Canada
c Department of Experimental Surgery, University of Heidelberg, Heidelberg, Germany
Received 5 September 2007;
received in revised form 18 February 2008;
accepted 19 February 2008
 Presented at the 21st Annual Meeting of the European Association for Cardio-thoracic Surgery, Geneva, Switzerland, September 16–19, 2007.
*Corresponding author. Tel.: +49/6221/566246; fax: +49/6221/564571.
E-mail address: Carsten.Beller{at}urz.uni-heidelberg.de (C.J. Beller).
 |
Abstract
|
|---|
Aortic root motion was previously identified as an additional risk factor for aortic dissection. This study analyzed if the magnitude of aortic root motion changed in patients after aortic valve replacement (AVR) and acute proximal aortic dissection. An institutional database (1984–2005) was used to measure the downward motion of the aortic root (perpendicular to the plane of the sinotubular junction) in contrast injections in 48 patients with aortic insufficiency (AI), aortic stenosis (AS) and proximal aortic dissection pre- and postoperatively, when available. Postoperative aortic root motion was significantly reduced after AVR for AI, while it was significantly increased after AVR for AS. By contrast, aortic root motion was unchanged when functional AI due to paravalvular leak was present post-AVR for AI. In patients with acute aortic dissection, both aortic root motion and aortic diameter were unchanged from pre-dissection. However, in patients who dissected again, aortic root motion was significantly smaller than pre-dissection, and the aortic diameter was significantly less than at first dissection. Removal of aortic stenosis was associated with increased aortic root motion, theoretically heightening the threat of dissection posed to the aortic wall by mechanical stress, although this was not confirmed by our study of dissection patients. Yet, mechanical principles command to include higher magnitude of aortic root motion during follow-up of patients after AVR as an additional risk factor for dissection.
Key Words: Aortic root motion; Aortic valve replacement; Aortic stenosis; Aortic insufficiency; Aortic dissection
|
1. Introduction
|
|---|
Most aortic dissections occur a few centimeters above the aortic valve, and exhibit a transverse intimal tear [1]. Although abnormalities in the aortic wall likely promote dissections, similar defects have been observed in normal aging without causing serious damage [2, 3]. Hypertension and aortic dilatation are other recognized risk factors for dissection, mechanical stress in the aortic wall being proportional to blood pressure and vessel diameter.
Displacement of the aortic root during the cardiac cycle is also a well-established phenomenon [4–6]. Aortic root motion was found to range from 0 to 22 mm in cardiac patients, and aortic insufficiency (AI) to be a predictor of higher values, while hypokinesis and hypertrophy of the left ventricle were associated with lower values [7]. By finite element analysis of the aortic root, aortic arch and supra-aortic vessels, we determined that aortic root motion significantly increased the mechanical stress present in the aortic wall, and could determine both the tear location and orientation observed in aortic dissections [8].
Aortic dissection may also be a late sequel of cardiac surgery. Dissections occurred in up to 0.16% of patients who underwent previous cardiovascular surgery [9] and specifically, in up to 0.6% of patients who had an aortic valve replacement (AVR) [10]. Previous studies suggested that sites of cross-clamping [11], suture lines [12] and cannulation [13] are locations for peri-operative injury of the aortic wall that can potentially lead to dissection.
In the present study, we explored possible changes in aortic root motion after AVR in relation to aortic dissection.
|
2. Materials and methods
|
|---|
The magnitude of aortic root motion was analyzed in two groups of patients: 1) patients who underwent AVR, and 2) patients whose aorta dissected one or multiple times. Note that the aortic root motion of the patients before dissection could only be assessed when usable records of a previous cardiac surgery existed.
2.1. Patients who underwent AVR (Table 1)
Group 1 of 32 patients was further divided into: 1) eight patients with AVR due to aortic insufficiency (AI subgroup), 2) fourteen patients with AVR due to AI who experienced post-AVR functional AI due to paravalvular leak from endocarditis (AI PL subgroup), and 3) ten patients with AVR due to aortic stenosis (AS subgroup). The mean age for the three subgroups at the time of the first operation was similar, at 65, 66 and 70 years for AI, AI PL and AS subgroups, respectively. A predominance of males, and four bicuspid aortic valves were noted in the AI subgroups. In all subgroups, the grade of aortic valve pathology was high. When available, the post-AVR degree of hypokinesis or hypertrophy of the left ventricle was recorded.
View this table:
[in this window]
[in a new window]
|
Table 1 Pre- and post-operative aortic displacements in patients who underwent aortic valve replacement (AVR) due to 1) AI, aortic insufficiency; 2) AI PL, AI, followed by paravalvular leak after AVR; 3) AS, aortic stenosis
|
|
2.2. Dissection patients (Table 2)
Group 2 consisted of 17 patients who were first hospitalized at 60 years on average. Both genders were present, with only few more males. Most of the patients showed signs of aortic medial degeneration at the time of aortic dissection; only three patients presented with connective tissue disorders (two with Marfan syndrome, one with Ehlers-Danlos-syndrome). Concomitant AI and sinotubular junction (STJ) diameter were recorded.
View this table:
[in this window]
[in a new window]
|
Table 2 Values of aortic displacement and sinotubular junction (STJ) diameter in patients who experienced one or multiple (consecutive) dissections. In the initial surgery, patient D8 received additional supracoronary ascending aortic replacement, and patient D16 received additional reduction aortoplasty with wrapping of the aorta at the time of the first operation. Patient D12 dissected three times, with similar data at dissections 2 and 3.
|
|
Nine (53%) patients with acute aortic dissection had previous cardiac surgery: coronary artery bypass grafting (CABG) in four cases, AVR in four cases (three due to AI, one due to AS), and mitral valve replacement for mitral insufficiency in one case. Among them, three (33%) patients (with AVR as initial operation) dissected twice, one of them three times; two of these patients suffered from the Marfan syndrome. The remaining eight (47%) patients with acute aortic dissection without previous cardiac surgery dissected again as well (had Ehlers-Danlos syndrome).
AI was present in six cases of (first) acute dissection (including two cases with previous cardiac surgery) and in three cases of second dissection at a later time (none with cardiac surgery before initial dissection).
2.3. Measurement of aortic root motion
Videos of aortic root contrast injections recorded on compact discs were analyzed frame by frame, and the aortic root outlines in the most upward and downward positions were traced on a transparency (Fig. 1), following a procedure described elsewhere [8]. The outline of the 5F-angiocatheter present in the field was also traced for distance calibration. The distances between the marked points were determined using image analysis software (Matlab, The MathWorks). The actual-size downward motion (axial displacement) of the aortic root perpendicular to the plane of the sinotubular junction was measured in millimeters (Fig. 1).
View larger version (7K):
[in this window]
[in a new window]
|
Fig. 1. For measurement of aortic root displacement, outlines from angiograms of the most upward and downward positions of the aortic root in a cardiac cycle are overlaid. Landmarks are noted with crosses. In schematic (a), for patients with native aortic roots, the crosses represent the base of the sinuses of Valsalva and the sinotubular junction. In schematic (b), for patients who underwent aortic valve replacement, the crosses represent the prosthetic valve bottom edge and the sinotubular junction. The aortic root movement is represented by the arrow between middle points M1 and M2. The projected length of this arrow in the direction perpendicular to the plane of the STJ was measured to obtain the axial displacement of the aortic root.
|
|
2.4. Statistical analysis
Assessment of differences between subgroup medians of patients' aortic root motion magnitude before and after AVR was performed using box plots with notches (Matlab, The MathWorks). The analysis was done on the presented mixed data as well as on paired (within patients) data, to investigate the influence of correlation within patients on the statistical findings. Subgroup differences in aortic root motion as a function of time after AVR were studied by one-way analysis of covariance with separate means. In patients whose aortas dissected one or multiple times, aortic root motion and STJ diameter before dissection and at first and second dissections were represented with notched box plots for comparison. A value of P<0.05 was considered statistically significant.
|
3. Results
|
|---|
3.1. Patients who underwent AVR
Detailed results are listed in Table 1, and Fig. 2 shows the box plots allowing comparison between pre- and post-AVR aortic root motion between subgroups AI, AI PL and AS. In the AI subgroup, the median aortic root motion was significantly lower after AVR than before (4.5 vs. 15 mm). Conversely, in the AS subgroup, the median aortic root motion was significantly higher after AVR than before (13 vs. 6 mm). In the AI PL subgroup, no significant difference in median aortic root motion was observed (9 vs. 7 mm). Similar findings were obtained when only paired data were analyzed. In addition, the median aortic root motion before AVR was significantly higher in AI than in AS subgroups (15 vs. 6 mm), while the inverse outcome was observed after AVR (4.5 vs. 13 mm for AI and AS subgroups, respectively). These findings are further illustrated in Fig. 3 representing the measurements of aortic root motion vs. time post-AVR. The AI and AS subgroups had significantly different means from the whole group, with 4.25 and 11.80 mm, respectively, vs. 8.47 mm.
View larger version (7K):
[in this window]
[in a new window]
|
Fig. 2. Box plots of pre- and post-AVR aortic root motions between patients with AVR due to aortic insufficiency (AI subgroup), patients with AVR due to AI and who experienced post-AVR AI due to paravalvular leak (AI PL subgroup), and patients with AVR due to aortic stenosis (AS subgroup). Notches display the variability of the subgroup medians, indicated by thick bars. Box plots whose notches do not overlap have significantly different medians. See comments in text.
|
|
View larger version (9K):
[in this window]
[in a new window]
|
Fig. 3. Graph of aortic root motion vs. time post-AVR, annotated with results from analysis of covariance with separate means. Notations as in Fig. 1; CI: 95% confidence interval. Patients with AVR for AI (patients with AVR for AS, respectively) had a significantly smaller (larger, respectively) aortic root motion than the whole group mean.
|
|
3.2. Dissection patients (Table 2)
Detailed results are listed in Table 2, and Fig. 4 presents the notched box plots for aortic root motion and STJ diameter before dissection and at first and second dissections. The median aortic root motion was significantly higher before dissection than at second dissection (10 vs. 4 mm). The median aortic root motion at first dissection was 5 mm. On the other hand, the median STJ diameter before dissection was 42 mm, and that at first dissection was significantly larger than at second dissection (75 vs. 46 mm, the latter value being consistent with the prosthetic supra-coronary aortic replacement performed upon first dissection). The time period before or between dissections varied considerably, from 0.1 to 12 years.
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 4. Box plots for aortic root motions and sinotubular junction (STJ) diameters before dissection and at first and second dissections. Box plots whose notches do not overlap have significantly different medians. See comments in text.
|
|
|
4. Discussion
|
|---|
4.1. Aortic root motion remodeling after AVR
To our knowledge, this is the first study examining the magnitude of aortic root motion before and after AVR. In cases with initial AI, aortic root motion was notably reduced after AVR. This is consistent with AI being a predictor of higher values of aortic root motion [7] in the absence of AVR. Further, when functional AI after AVR was present due to a paravalvular leak, the magnitude of aortic root motion was in a similar range as in the situation with AI before AVR. Conversely, we found a significant increase of mean aortic root motion after AVR in cases of initial AS. Since we previously showed that larger aortic root motion induces higher longitudinal mechanical stress in the wall of the ascending aorta [8], our new findings suggest that the remodeling of aortic root motion after AVR for AS may increase mechanical stress in the aortic wall. The magnitude of the observed aortic root motion after AVR for AS was in the same range as in healthy volunteers [5]. However, the possibility of a post-stenotic dilated and/or thinned aorta with more fragile wall properties exposed to increased longitudinal stress following AVR should not be taken lightly.
4.2. Implications of our study for aortic dissection after AVR
The time period between AVR and aortic dissection varied greatly (Table 2), which is in agreement with observations from other studies [10, 15]. A mildly dilated ascending aorta was a common characteristic at the time of first cardiac operation (not exclusively AVR) in our series of dissection patients. This is consistent with findings from others: in a meta-analysis of 24 studies of aortic dissection after AVR [15], a dilated ascending aorta was observed at the time of AVR in 88% of patients. Extensive thinning and/or fragility of the aortic wall in the presence of a mildly dilated aorta (45±5 mm) at the time of AVR were associated with a high risk of late dissection [10]. In our patients, although the prosthetic supra-coronary aortic replacement performed upon first dissection was successful in significantly reducing the aortic diameter, and although their aortic root motion was comparatively small, 65% dissected again.
Surprisingly, while the magnitude of aortic root motion clearly increased in (Group 1) patients who underwent AVR for AS, none of the dissection patients in our series (Group 2) had especially large motion due to AVR for AS when acute dissection occurred. Furthermore, three out of nine (33%) patients who had acute dissection after previous cardiac surgery had received AVR for AI, though such patients were expected to experience reduced aortic root motion, and therefore enjoy a reduced threat from dissection. Overall, while four out of nine (44%) patients who dissected following cardiac surgery had received AVR (and three dissected twice or more), eight patients without any previous cardiac surgery dissected twice.
It appears from these somewhat conflicting observations that an otherwise sensible correlation between AVR and dissection may prove elusive to quantify. An important fact is that not only proximal, but also distal aortic dissections have been observed after AVR, suggesting that risk factors other than aortic root motion are involved [14, 15]. Incidentally, our observations also point to the very difficult task of preventing repeated dissections.
4.3. Study limitations
In spite of our institutional database covering 22 years, our patient series were of modest sizes, which limited the statistical power of our results. Multi-center trials with long-term follow-up, e.g. using the International Registry of Aortic Dissection, could provide more insight into aortic root motion remodeling after AVR, and help identify predictors of late aortic dissection.
|
5. Conclusions
|
|---|
Restored aortic valve competence is associated with reduced aortic root motion in cases of AI. By contrast, removal of aortic stenosis is associated with increased aortic root motion, theoretically heightening the threat of dissection posed to the aortic wall by mechanical stress, although this was not confirmed by our study of dissection patients. Yet, mechanical principles command to include higher magnitude of aortic root motion during follow-up of patients after AVR as an additional risk factor for dissection.
|
Conference discussion
|
|---|
Dr. C. Yankah (Berlin, Germany): You discussed the distensibility of the aortic root, if I am correct, which is dependent on the elasticity of the wall. You didn't mention the role of collagen and elastin fibres in your methodology and results especially the genetics aspect of collagen diseases. You might have patients with defects in the fibrillin genes of the aortic collagen fibers which might trigger dissection after the operation. Do you have some histological findings from your patients, because this relates actually to this study?
Dr. Beller: I agree that wall alterations are a possibility, but the paper from IRAD recently published in Circulation showed that in patients with acute type-A dissection their aortic size initially was not a good predictor for later dissection. So I think despite that fact, and even in Marfan patients, IRAD showed that they dissect at a larger diameter compared to the patients without tissue disorders. Very likely, not only the known risk factors such as aortic size or wall abnormalities play a role. In my opinion other factors, I want to call them biomechanical factors, such as aortic root motion play a role.
Dr. M. Poullis (Liverpool, UK): If aortic root motion is so important, why in the setting of non-aneurysmal aortic disease is aortic stenosis the predominant valvular lesion, not regurgitation for dissection, because that has reduced root motion compared with regurgitation in the non-redo setting?
Dr. Beller: I didn't get completely your question.
Dr. Poullis: In non-aneurysmal disease, when you have a dissection without previous cardiac surgery, stenosis is more common than regurgitation, but this has reduced root motion. So there is something else going on.
Dr. Beller: Some series indicate that you are correct. Probably, it is because the concomitant valvular stenosis is associated with plaque formation in the aortic wall, and we know that sometimes we have a plaque rupture leading to dissection. So I think it is not a contradiction per se.
Dr. Poullis: At the risk of being done for self-publicity, aortic curvature might be one of the additional factors.
|
References
|
|---|
- Hirst AE, Johns VJ, Kime SW. Dissecting aneurysms of the aorta: a review of 505 cases. Medicine 1958;37:217–279.[Medline]
- Leonard JC, Hasleton PS. Dissecting aortic aneurysms: a clinicopathological study. Q J Med 1979;XLVIII:55–76.
- Schlatmann TJM, Becker AE. Pathogenesis of dissecting aneurysm of aorta. Comparative histopathologic study of significance of medial changes. Am J Cardiol 1977;39:21–26.[CrossRef][Medline]
- Mercer JL. Movement of the aortic annulus. Br J Radiol 1969;42:623–626.[Abstract/Free Full Text]
- Kozerke S, Scheidegger MB, Pedersen EM, Boesiger P. Heart motion adapted cine phase-contrast flow measurements through the aortic valve. Magn Reson Med 1999;42:970–978.[CrossRef][Medline]
- Stuber M, Scheidegger MB, Fischer SE, Nagel E, Steinemann F, Hess OM, Boesiger P. Alterations in the local myocardial motion pattern in patients suffering from pressure overload due to aortic stenosis. Circulation 1999;100:361–368.[Abstract/Free Full Text]
- Beller CJ, Labrosse MR, Thubrikar MJ, Szabo G, Robicsek F, Hagl S. Increased aortic wall stress in aortic insufficiency: clinical data and computer model. Eur J Cardiothor Surg 2005;27:270–275.[Abstract/Free Full Text]
- Beller CJ, Labrosse MR, Thubrikar MJ, Robicsek F. Role of aortic root motion in the pathogenesis of aortic dissection. Circulation 2004;109:763–769.[Abstract/Free Full Text]
- Still RJ, Hilgenberg AD, Akins CW, Daggett WM, Buckley MJ. Intraoperative aortic dissection. Ann Thorac Surg 1992;53:374–380.[Abstract]
- Von Kodolitsch Y, Loose R, Ostermeyer J, Aydin A, Koschyk DH, Haverich A, Nienaber CA. Proximal aortic dissection late after aortic valve surgery: 119 cases of a distinct clinical entity. Thorac Cardiovasc Surg 2000;48:342–346.[CrossRef][Medline]
- Litchford B, Okies JE, Sugimura S, Starr A. Acute aortic dissection from cross-clamp injury. J Thorac Cardiovasc Surg 1976;72:709–713.[Abstract]
- Archer AG, Choyke PL, Zeman RK, Green CE, Zuckerman M. Aortic dissection following coronary artery bypass surgery: diagnosis by CT. Cardiovasc Intervent Radiol 1986;9:142–145.[Medline]
- Williams CD, Suwansirikul S, Engelman MR. Thoracic aortic dissection following cannulation for perfusion. Ann Thorac Surg 1974;18:300–304.[Medline]
- Loisance D, Hirose H, Invited commentary for, Svensson LG, Bruce WL, Blackstone EH, Rajeswaran J, Cosgrove DM. Aortic dissection after previous cardiovascular surgery. Ann Thorac Surg 2004;78:2099–2105.[Abstract/Free Full Text]
- Pieters FAA, Widdershoven JW, Gerardy AC, Geskes G, Cheriex EC, Wellens HJ. Risk of aortic dissection after aortic valve replacement. Am J Cardiol 1993;72:1043–1047.[CrossRef][Medline]
|
|
Top
Abstract
1. Introduction
