Histopathological Grading of Ascending Aortic Aneurysm: Comparison

© Copyright by ICR Publishers 2003

Histopathological Grading of Ascending AorticAneurysm: Comparison of Patients with Bicuspid versusTricuspid Aortic ValveJ. F. Matthias Bechtel1, Frank Noack2, Friedhelm Sayk2, Armin W. Erasmi1, Claus Bartels1, Hans-Hinrich Sievers1

1Department of Cardiac Surgery, University Hospital Lübeck, 2Institute of Pathology, Medical University of Lübeck, Lübeck,Germany

Bicuspid aortic valve is the most common congenitalcardiac defect, with a prevalence among the generalpopulation of 1-2% (1). Although the prognosis of mostpatients with bicuspid aortic valve is likely to be good,a significant number of these patients will eventuallysuffer from aortic valve stenosis, regurgitation orendocarditis, or will experience aortic dilatation, dis-section or rupture (2).

An association of bicuspid aortic valve with ascend-ing aortic aneurysm and dissection has long been rec-ognized (3,4). Patients with a bicuspid aortic valveusually have increased diameters of the aortic root,regardless of the functional status of the valve (5-8);therefore, the presence of a genetic defect has been pro-

posed in patients with bicuspid aortic valve that man-ifests both at the valve level and within the aortic wall(9).

A recent histological study in patients with non-dilated ascending aorta suggested that those withbicuspid aortic valve had more severe aortic wallabnormalities than those with tricuspid aortic valve(10). In the latter patients, there appears to be a posi-tive relationship between the degree of aortic dilata-tion and the severity of aortic wall abnormalities(11,12). Unfortunately, histological data on patientswith bicuspid aortic valve and aortic aneurysm are rel-atively sparse. Hence, a review was conducted of thepresent authors’ experience in ascending aortic sur-gery in order to determine the severity of associatedaortic wall abnormalities in patients with bicuspid aswell as tricuspid aortic valve.

Clinical material and methods

Patient populationBetween 1995 and 1999, a total of 167 operations on

Presented in part at the First Biennial Meeting of the Society forHeart Valve Disease, 15th-18th June 2001, Queen Elizabeth IIConference Centre, London, United Kingdom

Address for correspondence:Prof. Dr. H. H. Sievers, Klinik fuer Herzchirurgie,Universitaetsklinikum Lübeck, Ratzeburger Allee 160, D-23538Lübeck, Germany

Background and aims of the study: Bicuspid aorticvalve (BAV) is a common inherited condition that isoften accompanied by ascending aortic aneurysm. Ahigh level of histological wall abnormalities wasreported to be present in non-dilated aortas ofpatients with BAV. In patients with tricuspid aorticvalve, there appears to exist a direct relationshipbetween the diameter of the ascending aorta anddegree of histopathological aortic wall abnormali-ties. Whether this situation exists in patients withBAV has not yet been investigated.Methods: Surgical and medical records of all patientsundergoing surgery of the ascending aorta werereviewed. A total of 65 patients was identified inwhom an aortic wall specimen was obtained intraop-eratively. These specimens were systematically re-evaluated, and graded according to the severity ofseven histopathological conditions: fibrosis, athero-sclerosis, medionecrosis, cystic medial necrosis,

smooth muscle cell orientation, elastic fiber frag-mentation, and inflammation.Results: BAVs were present in 26 patients (40%).Patients with BAV had significantly less aortic wallalterations than patients with tricuspid aortic valves(p <0.001) in all variables examined. The severity ofaortic wall abnormalities was significantly depend-ent on aortic diameter in patients with BAV as well astricuspid aortic valve (p = 0.036 and 0.019), butdependent on age (p = 0.009) only in patients with tri-cuspid aortic valve.Conclusion: The study results provide evidence thatascending aortic aneurysm in patients with BAV dif-fers clinically and histologically from that in patientswith tricuspid aortic valve. Further studies are need-ed to elucidate the impact of inherited and acquiredaortic wall abnormalities on the development ofaneurysms.The Journal of Heart Valve Disease 2003;12:54-61

Aortic aneurysm histopathologyJ. F. Matthias Bechtel et al.

55J Heart Valve DisVol. 12. No. 1January 2003

the ascending aorta was performed at the authors’department. Among patients, one aortic leaflet waspresent in one patient (0.6%), while two leaflets werepresent in 42 patients (25.1%) and three leaflets in 96(57.5%). An unknown number was present in 28patients (including 10 who had undergone prior aorticvalve replacement). Patients with bicuspid aortic valvewere significantly younger than those with tricuspidaortic valve (53 ± 14 versus 62 ± 13 years, respectively;p <0.001). Among patients with either a bicuspid or tri-cuspid aortic valve, an aortic wall specimen wasexcised from the anterior aspect of the convexity of theascending aorta at the time of surgery in 65 cases, andthese specimens form the basis of this report.

Histopathological evaluationThe resected material was fixed in 4.5% pH-buffered

formalin for approximately 24 h. Representative por-tions of the resected material were selected macroscop-ically for further processing. If necessary,decalcification of atherosclerotic lesions was undertak-en by use of Ossa fixona solution (Diagonal, Muenster,Germany). The tissue was processed for lightmicroscopy, embedded in paraffin blocks, and sections(4 µm thickness) were taken from each specimen.Sections were stained with hematoxylin and eosin, elas-tica-van Gieson, Alcian blue and Masson’s trichromestains. For the purpose of this study, all specimens werere-evaluated by two experienced histopathologists who

Figure 1: Elastica-van Gieson staining of the aortic wall(original magnification, ×100), showing grade 3 elastic

fragmentation with presence of foci of elastic fragmentationin more than ten neighboring elastic lamellae.

Figure 2: Alcian blue staining of two aortic wallspecimens. A) Grade 1 cystic medial necrosis with mucoid

material within lamellar units (original magnification,×200). B) Grade 3 cystic medial necrosis with mucoidmaterial surpassing more than one lamellar unit in the

presence of fragmented elastic fibers (originalmagnification, ×100).

Table I: Clinical characteristics of all patients.

Characteristic Aortic valve p-value__________________________________________

Bicuspid Tricuspid

Male/female (n) 18/8 25/14 0.79Age (years)* 53 ± 15 57 ± 14 0.23Hypertension (n) 10 (42) 24 (63) 0.12Diabetes mellitus (n) 0 1 (3) 1.0Marfan syndrome (n) 0 4 (10) 0.14Aneurysm diameter (mm)* 61 ± 11 58 ± 6 0.41Dissection (n) 2 (8) 27 (69) <0.001Need for aortic valve surgery (n) 22 (85) 28 (72) 0.37Type of aortic surgery (n) 0.40

Replacement 39 (100) 25 (96)Wrapping 0 1 (4)

Values in parentheses are percentages.*Values are mean ± SD.

were blinded to the clinical data. The following histo-logical alterations were analyzed semiquantitatively:(1) fibrosis (defined as an increase in interstitial colla-gen); (2) atherosclerosis (defined as the presence of inti-mal fibrous plaques and/or complex or complicatedatheromas); (3) medionecrosis (defined as a focal loss ofsmooth muscle cell nuclei in the media); (4) cysticmedial necrosis (defined as mucoid material accumula-tion); (5) changes in smooth muscle cell orientation; (6)elastic fragmentation (defined as focal fragmentation ofelastic lamellae in the media); and (7) periaortic inflam-mation (defined as the presence of inflammatory cells).Each variable was graded from 0 (no change) to 3 (mostsevere change) when examined at a magnification of×100 or ×200, using an Olympus microscope (OlympusBX 50, Japan). The grades were determined on the basisof the worst area observed. Examples are shown in

Figures 1 and 2. The criteria for histological gradingwere used as proposed by Schlatman and Becker (13),Klima et al. (14) and de Sa et al. (10) and are detailed inAppendix I. The sum of the results of all variables wascalculated for each individual patient, and this wasreferred to as the aortic wall score.

Statistical analysisData were presented as absolute numbers and rela-

tive percentages or mean (± SD), except where other-wise stated. Relative frequencies were compared usingFisher’s exact test; continuous data were comparedusing the Mann-Whitney U-test. Linear regressionanalysis was performed with the aortic wall score asdependent variable, and age and aortic diameter asindependent variables. All analyses were performedusing SPSS for Windows (SPSS Inc., Chicago, IL, USA).

56 Aortic aneurysm histopathologyJ. F. Matthias Bechtel et al.

J Heart Valve DisVol. 12. No. 1January 2003

Table II: Results of histopathological evaluation.

Variable Grade Aortic valve p-value___________________________________________

Bicuspid (n) Tricuspid (n)

Fibrosis None 19 (73) 11 (28)I 3 (12) 15 (39) 0.002II 3 (12) 11 (28)III 1 (4) 2 (5)

Atherosclerosis* None 12 (46) 11 (29)I 10 (39) 9 (24) 0.018II 2 (8) 5 (13)III 2 (8) 13 (34)

Medionecrosis None 18 (69) 9 (23)I 5 (19) 10 (26) <0.001II 1 (4) 11 (28)III 2 (8) 9 (23)

Cystic medial necrosis None 16 (62) 10 (26)I 5 (19) 13 (33) 0.008II 3 (12) 10 (26)III 2 (8) 6 (15)

SMC orientation+ Normal 24 (92) 22 (63)I 1 (4) 7 (20) 0.010II 1 (4) 6 (17)III 0 0

Elastic fragmentation* None 12 (46) 5 (13)I 6 (23) 6 (16) <0.001II 6 (23) 6 (16)III 2 (8) 21 (55)

Inflammation None 19 (73) 11 (28)I 5 (19) 15 (39) <0.001II 1 (4) 6 (15)III 1 (4) 7 (18)

Aortic wall score‡ 3.8 ± 3.8 9.2 ± 4.6 <0.001

Values in parentheses are percentages.*Variable not evaluated sufficiently in one patient with tricuspid aortic valve.+Variable not evaluated sufficiently in four patients with tricuspid aortic valve.‡Values are mean ± SD.SMC: Smooth muscle cell.

Results

The demographic and surgical data of patients arelisted in Table I. A total of 26 patients (40%) had bicus-pid aortic valve.

Five patients had no dectectable histopathologicalchanges; four of these had bicuspid aortic valve (p =0.15). The mean aortic wall score was 7.1 ± 5.1. Inpatients with tricuspid aortic valve, regression analysisrevealed a significant association between age at oper-ation, ascending aortic diameter, and aortic wall score(aortic wall score = 0.42 × age + 0.33 × diameter; p =0.009 and 0.036, respectively). In contrast, patients witha bicuspid aortic valve showed a significant associa-tion between aortic diameter (but not age) and aorticwall score (aortic wall score = 0.49 × diameter; p =0.019). The association between aortic wall score andaortic diameter is shown in Figure 3.

Among patients with hypertension or diabetes mel-litus, all of the examined histopathological variableswere of similar severity. Patients with aortic dissectionhad a more severe extent of cystic medionecrosis (p =0.017) as compared with patients with aneurysm, butall other variables examined were of similar severity.The results of the histopathological evaluation forpatients with bicuspid versus tricuspid aortic valve arelisted in Table II. Patients with bicuspid aortic valvehad a significantly lower mean aortic wall score(3.8 ± 3.8 versus 9.2 ± 4.6; p <0.001) due to significant-ly less severe histopathological changes in all variablesexamined.

Discussion

The results of this retrospective study suggest that -at the time of aortic surgery - patients with a bicuspid

aortic valve have less severe aortic wall abnormalitiesaccording to histological standard criteria than thosepatients with a tricuspid aortic valve, despite the pres-ence of similar degrees of aortic dilatation. In addition,these results confirm that bicuspid aortic valve is fre-quent among patients undergoing surgery for diseasesof the ascending aorta, and that a wide variety of aor-tic wall abnormalities of the ascending aorta can beobserved in patients with tricuspid as well as bicuspidaortic valve.

In patients with tricuspid aortic valve, there appearsto be a direct positive correlation between the degreeof ascending aortic dilatation and the degree of histo-logical aortic wall abnormalities (11,12), but whetherthe same situation exists among patients with bicuspidaortic valve has not yet been investigated. In the pres-ent study, less severe histopathological changes werefound in patients with bicuspid aortic valve despite asimilar degree of aortic dilatation; thus, these resultsmay be interpreted in such a way that the ‘true’ lesionin the aorta of patients with bicuspid aortic valve is notidentified using standard light microscopy criteria andthat the mechanism of dilatation may differ from thatin patients with tricuspid aortic valve. In accordancewith the first hypothesis, Parai et al. (15) have shownthere to be subtle (but significant) differences regard-ing the amount of elastic tissue between the aorta ofpatients with bicuspid and tricuspid aortic valvewhich could only be identified using morphometry.Recently, Bauer et al. (16) confirmed that patients withbicuspid aortic valve have less elastic tissue in theirascending aorta, while Nistri et al. (17) found evidencethat the aorta of patients with bicuspid aortic valveappears to be stiffer when compared with that inpatients with tricuspid aortic valve. Bonderman et al.(18) found evidence of a generally increased rate ofapoptosis in patients with bicuspid aortic valve,whereas in patients with tricuspid aortic valve the rateof apoptosis was elevated only in dilated aortas.

In studying patients with non-dilated aortas, de Sa etal. (10) reported that patients with bicuspid aorticvalve had more severe aortic wall abnormalities thanpatients with tricuspid aortic valve. These authors alsofound that patients with bicuspid aortic valve fre-quently have severe wall abnormalities in the mainpulmonary artery, which develops from the sameembryological structures as the ascending aorta (19).This apparent contrast to the results of the presentstudy cannot be explained easily. Indeed, some evi-dence was found of a direct positive correlationbetween the degree of aortic wall abnormalities andaortic diameter in patients with bicuspid as well as tri-cuspid aortic valve, but this occurred on a generallylower level in patients with bicuspid aortic valve. Aspatients with non-dilated aortas were excluded from



Figure 3: Scatterplot of ascending aortic diameter versusaortic wall score. Patients with bicuspid aortic valve

(triangles) had significantly (p <0.001) lower aortic wallscores than those with tricuspid aortic valve (circles).

the present study, the possibility of a two-phase modelcannot be excluded: a constant first phase (a constantdegree of aortic wall abnormalities in the range of nor-mal aortic diameters) followed by a steady increase ofaortic wall abnormalities after a certain degree ofdilatation has been exceeded. Further studies shouldbe conducted in order to determine the relationshipbetween aortic diameter and associated aortic wallabnormalities.

Most patients with bicuspid aortic valve will neverexperience aortic dilatation, dissection or rupture,though some will require ascending aorta replacementearly in life (2). Besides genetic differences amongpatients with bicuspid aortic valve, other factors suchas postvalvular flow may also contribute to this highlyvariable prognosis.

More recent studies have indicated that the aorticroot is an asymmetric, highly complex structure(20,21). Flow in the ascending aorta is usually eccen-tric, and studies on prosthetic aortic valves haveshown that orientation of the leaflets has a majorimpact on blood velocity and the presence of turbu-lence in the ascending aorta (22,23). The precise orien-tation and morphology of the bicuspid aortic valvevaries widely (24-26), and thus the morphology of abicuspid valve may cause abnormal blood flow in theascending aorta, even in the absence of a significantdegree of valvular disease. Whether these proposedflow disturbances are a cofactor in the development ofascending aortic dilatation/dissection has not yet beeninvestigated, but experimental studies have providedsome evidence that flow disturbances or hemodynam-ic stress can cause dilatation of vessels (27,28). Theimpact of flow disturbances on histological aortic wallalterations is, at present, unknown.

Study limitationsThe main limitation of the present study was its ret-

rospective design. Some selection bias was clear: ahistopathological examination was more likely to beordered in younger patients, elective settings, or whenno etiology of a dissection (such as bicuspid aorticvalve or prior valve replacement) was apparent. Thisbias might explain the low incidence of aortic dissec-tion in patients with bicuspid aortic valve: 18 of the 28patients with an unknown number of aortic valveleaflets had acute type A aortic dissection. However, theresults remain virtually unchanged if all patients withaortic dissection are excluded from the analysis.Furthermore, aortic wall abnormalities are not sym-metrically distributed among the ascending aortic cir-cumference (29), although the total aortic circumferencewas not examined. Nonetheless, any sampling error islikely to occur at random and it is likely that the mainfinding was unaffected by this limitation.

In conclusion, the present study provides some evi-dence that ascending aortic aneurysm in patients withbicuspid aortic valve differs histologically from that inpatients with tricuspid aortic valve. Further studiesshould be conducted in order to elucidate the impactof inherited and acquired (for example, by age or flowdisturbances) aortic wall abnormalities on the devel-opment of aneurysms.

References1. Roberts WC. The congenitally bicuspid aortic

valve. A study of 85 autopsy cases. Am J Cardiol1970;26:72-82

2. Ward C. Clinical significance of the bicuspid aorticvalve. Heart 2000;83:81-85

3. Gore I. Dissecting aneurysms of the aorta in per-sons under 40 years of age. Arch Pathol 1953;55:1-13

4. Edwards WE, Leaf DS, Edwards JE. Dissecting aor-tic aneurysm associated with congenital bicuspidaortic valve. Circulation 1978;57:1022-1025

5. Hahn RT, Roman MJ, Mogtader AH, Devereux RB.Association of aortic dilation with regurgitant,stenotic and functionally normal bicuspid aorticvalves. J Am Coll Cardiol 1992;19:283-288

6. Nistri S, Sorbo MD, Marin M, et al. Aortic rootdilatation in young men with normally functioningbicuspid aortic valves. Heart 2000;82:19-22

7. Keane MG, Wiegers SE, Plappert T, et al. Bicuspidaortic valves are associated with aortic dilatationout of proportion to coexistent valvular lesions.Circulation 2000;102(Suppl.III):III-35-III-39

8. Pachulski RT, Weinberg AL, Chan KL. Aorticaneurysm in patients with functionally normal orminimally stenotic bicuspid aortic valve. Am JCardiol 1991;67:781-782

9. McKusick VA. Association of congenital bicuspidaortic valve and Erdheim’s cystic medial necrosis(letter). Lancet 1972;1:1026-1027

10. de Sa M, Moshkovitz Y, Butany J, David TE.Histologic abnormalities of the ascending aorta andpulmonary trunk in patients with bicuspid aorticvalve disease: Clinical relevance to the Ross proce-dure. J Thorac Cardiovasc Surg 1999;118:588-596

11. Agozzino L, de Vivo F, Falco A, de Luca TupputiSchinosa L, Cotrufo M. Non-inflammatory aorticroot disease and floppy aortic valve as cause of iso-lated regurgitation: A clinico-morphologic study.Int J Cardiol 1994;45:129-134

12. Bellitti R, Caruso A, Festa M, et al. Prolapse of the‘floppy’ aortic valve as a cause of aortic regurgita-tion. A clinico-morphologic study. Int J Cardiol1985;9:399-410

13. Schlatman TJM, Becker AE. Histologic changes inthe normal aging aorta: Implications for dissecting





aortic aneurysm. Am J Cardiol 1977;39:13-2014. Klima T, Spjut HJ, Coelho A, et al. The morphology

of ascending aortic aneurysms. Hum Pathol2001;14:810-817

15. Parai JL, Masters RG, Walley VM, Stinson WA,Veinot JP. Aortic medial changes associated withbicuspid aortic valve: Myth or reality? Can JCardiol 1999;15:1233-1238

16. Bauer M, Pasic M, Meyer R, et al. Morphometricanalysis of aortic media in patients with bicuspidand tricuspid aortic valve. Ann Thorac Surg2002;74:58-62

17. Nistri S, Sorbo MD, Basso C, Thiene G. Bicuspidaortic valve: Abnormal aortic elastic properties. JHeart Valve Dis 2002;11:369-374

18. Bonderman D, Gharehbaghi-Schnell E, Wollenek G,et al. Mechanisms underlying aortic dilatation incongenital aortic valve malformation. Circulation1999;99:2138-2143

19. Maron BJ, Hutchins GM. The development of thesemilunar valves in the human heart. Am J Pathol1974;74:331-344

20. Choo SJ, McRae G, Olomon JP, et al. Aortic rootgeometry: Pattern of differences between leafletsand sinuses of Valsalva. J Heart Valve Dis1999;8:407-415

21. Dagum P, Green GR, Nistal FJ, et al. Deformationaldynamics of the aortic root. Modes and physiologicdeterminants. Circulation 1999;100(Suppl.II):II-54-II-62

22. Paulsen PK, Nygaard H, Hasenkam JM, et al.Analysis of velocity in the ascending aorta inhumans. A comparative study among normal aorticvalves, St. Jude Medical and Starr-Edwards silasticball valves. Int J Artif Organs 1988;11:293-302

23. Laas J, Kleine P, Hasenkam MJ, Nygaard H.Orientation of tilting disc and bileaflet aortic valvesubstitutes for optimal hemodynamics. Ann ThoracSurg 1999;68:1096-1099

24. Moore GW, Hutchins GM, Brito JC, Kang H.Congenital malformations of the semilunar valves.Hum Pathol 1980;11:367-372

25. Duran AC, Frescura C, Sans-Coma V, et al. Bicuspidaortic valves in hearts with other congenital heartdisease. J Heart Valve Dis 1995;4:581-590

26. Sabet HY, Edwards WE, Tazelaar HD, Daly RC.Congenitally bicuspid aortic valves: A surgicalpathology study of 542 cases (1991 through 1996)and a literature review of 2715 additional cases.Mayo Clin Proc 1999;74:14-26

27. Holman E. The obscure physiology of poststenoticdilatation: Its relation to the development ofaneurysms. J Thorac Surg 1954;28:109-133

28. Stehbens WE. Structural and architectural changesduring arterial development and the role of hemo-

dynamics. Acta Anat 1996;157:261-27429. Agozzino L, Ferraraccio F, Esposito S, et al. Medial

degeneration does not involve uniformly the wholeascending aorta: Morphological, biochemical andclinical correlations. Eur J Cardiothorac Surg2002;21:675-682

Meeting discussion

DR. CRISTINA BASSO (Padova, Italy): It seems thatthe histological samples were not re-evaluated by thepathologist. Which staining do you normally use toinvestigate aortic wall pathology?DR. J. F. MATTHIAS BECHTEL (Lübeck, Germany):The specimens were stained with hematoxylin andeosin, and elastica-van Gieson; they were paraffin-buffered and sectioned at 10 µm thickness.DR. BASSO: When you specified the aortic wallabnormalities you mentioned only mucoid degenera-tion and cystic necrosis - not the elastic fibers. Whatwas the elastic fiber architecture of the aortic wall?DR. BECHTEL: These studies were not carried out inall patients, so we only focused on the standard proto-col.DR. BASSO: But this is an issue that we must look forin aortic wall pathology in aortic aneurysm.DR. BECHTEL: I agree with that, but in our opinionthe absence of any evident histologic abnormality inthese patients doesn’t exclude the presence of aorticwall abnormalities - but it does suggest that other fac-tors are operative. In my opinion, these factors are syn-ergistic to aortic wall abnormalities.DR. JAGDISH BUTANY (Toronto, Canada): Thankyou for raising such a provocative subject. Do youknow how much of the aorta was excised at surgery?DR. BECHTEL: The aortic wall specimen was usuallytaken from the anterolateral aspect of the convexity ofthe aneurysm. Among the patients with bicuspid aor-tic valve there were no statistical imbalances withregards to valve surgery or the type of ascending aor-tic aneurysm surgery performed.DR. BUTANY: There have been many reports showingthat as we grow older there is a sequential set ofchanges in the aortic wall - none of us will escape that.So those changes will be present regardless of anyother changes. I trust you are aware of that. Anotherpoint is, did you obtain a ring of aortic tissue to exam-ine for histopathology? The exaggerated changes thatyou can see will be localized - they won’t be presentcircumferentially, or on the medial side of the innermargin. You have to examine the entire circumferenceto detect these changes. A second point is that in theaortic valve, the changes that you showed - theshelf-like change - is more than likely a change as aconsequence of aortic incompetence rather than a pri-



mary change in the aortic valve cusp. So changes in theaortic wall are less likely, if not unlikely, to be relatedto that shelf-like change in the aortic valve cusp. Thechanges in the aortic wall probably occurred earlierthan the development of that shelf in the aortic cusps.DR. BECHTEL: We can’t exclude that, but as I men-tioned we did not have circumferential specimens.They were usually from the anterolateral aspect of theconvexity.DR. BUTANY: In a study that we published inToronto, we showed that even if you didn’t see signif-icant morphological changes, you find many fibrillinchanges - but you have to perform much more exten-sive studies.DR. BECHTEL: I am aware of those studies, but this isa retrospective study that is sort of provocative. Wewere not able to re-evaluate the specimens, or to carryout more extensive studies until now.DR. KARYN KUNZELMAN (Madison, Wisconsin,USA): In the follow up to that, you have acknowl-edged that your study is somewhat limited by the lackof histological data due to its retrospective nature. Areyou going to continue this in a prospective manner toanswer some of these questions that are being raised?DR. BECHTEL: Yes - these results only included dataup to the end of 1999 when the case patient was oper-ated on.DR. KUNZELMAN: Have you considered thatmechanical stress alterations rather than just hemody-namic changes might alter the properties of the root orthe ascending aorta?DR. BECHTEL: I won’t speculate too much on that. Iused the term ‘hemodynamics’ because I wasn’t quitesure what term should be used. I am not sure exactlywhat caused these aneurysms. We speculate that tur-bulence is developing behind the valve, and this caus-es the aneurysms to develop, but it is unclear whetherthis should be called mechanical stress or whether it ishemodynamic inasmuch as you can influence it bydecreasing dP/dt.DR. DANIEL LOISANCE (Creteil, France): Thankyou for raising a very difficult issue - I am veryimpressed by your provocative conclusions. However,they don’t fit in with what we observe clinically, orwhat we see when we examine the specimen very care-fully. Clinically, how can you explain the appearance ofaortic dilatation following aortic valve replacement inthese patients who had initially a bicuspid valve? Itcould be the hemodynamic parameters that explainthis secondary dilatation. A second observation is thatwhen we examine these aortic tissues they appearextremely abnormal - not only microscopically but alsoin their subcellular structures. For instance, gly-cosaminoglycan production is extremely different,TIMP (tissue metalloproteinase) is extremely different,

TIMP expression is very different, and MMP-1 (matrixmetalloproteinase-1) and MMP-3 are also extremelydifferent. So there is more and more convincing evi-dence that there is some kind of gene function behindthat, and that the hemodynamic parameters may offera secondary explanation. What is you opinion aboutthat?DR. BECHTEL: To answer your last point, many excel-lent reports have been published recently providingevidence that there is a genetically determined aorticwall abnormality in patients with bicuspid aorticvalve. A recent paper from De Sa and colleagues inToronto showed pulmonary wall abnormalities alsooccur very frequently, but I don’t want to challenge allthese findings. Nevertheless, we thought it possiblethat other factors might be operative in some patients.For example, hemodynamics may have a major impacton the question of whether or not these aortic wallabnormalities will develop into an aneurysm, or sim-ply be present. To return to your first point, there arestudies on disk orientation in bileaflet prosthetic heartvalves that show that the exact orientation of the valvehas a major impact on turbulence behind the valve. Itis possible that this could be why some, but not all,patients with bicuspid aortic valve develop ascendingaortic dissections after aortic valve replacement.DR. PENNY THOMAS (London, UK): Have you anyidea from your specimens whether your bicuspidvalves arose through the fusion of two leaflets early onin development, or whether there were always onlytwo leaflets?DR. BECHTEL: Most of them were noted in the surgi-cal records to be congenitally bicuspid, which meansthat there are not always only two sinuses, but most ofthem were not fused as a result of a pathologicalprocess in adulthood.DR. THOMAS: So the valves were like it when thepatients were born?DR. BECHTEL: That was the opinion of the operatingsurgeons. In this retrospective study I tried to includeonly those valves that appeared to be congenitallybicuspid.DR. THOMAS: I am just trying to see if there is anylink between any abnormality in the wall which mightbe neural crest-related or developmental with thenumber of leaflets, because neural crest cells doapproach the leaflets.DR. BECHTEL: I tried to include only those that werecongenitally bicuspid.DR. GAETANO THIENE (Padova, Italy): Did youhave cases of aortic dissection with bicuspid valve? Ifso, was the aortic wall normal or abnormal?DR. BECHTEL: We have cases with bicuspid aorticvalves and dissection, but I can’t tell you at presentwhat their pathology was like.



DR. THIENE: The problem is that it is impossible todissect the aortic wall in the setting of a bicuspid valvewithout aortic wall abnormalities, because there is nohypertension there.

DR. BECHTEL: There were no dissections among thelast eight patients I presented - they only hadaneurysms.

Appendix I: Criteria for histological grading*

FibrosisGrade 1: an increase in collagen content in an area comprising less than one-third of the total width of the media.Grade 2: an increase in collagen in an area comprising between one- and two-thirds of the total width of the media.Grade 3: an increase in collagen in an area comprising more than two-thirds of the total width of the media.

AtherosclerosisGrade 1: intimal fibrous plaques, the thickness of which was less than one-fourth of the thickness of the media.Grade 2: intimal fibrous plaques thicker than one-fourth of the media, or intimal plaques with minimal calcification and/oratheroma.Grade 3: complex or complicated lesions of severe atheroma with thrombosis, calcifications and ulcerations.

MedionecrosisGrade 1: focal loss of nuclei in an area comprising less than one-third of the total width.Grade 2: focal loss of nuclei in an area comprising between one- and two-thirds of the medial thickness.Grade 3: focal loss of nuclei in an area comprising more than two-thirds of the total medial thickness.

Cystic medial necrosisGrade 1: minute foci of mucoid material (‘cysts’) were present within a single lamellar unit.Grade 2: the amount of mucoid material had increased, so that accumulation of ‘cysts’ covered the total width of one lamellarunit.Grade 3: the extent of mucoid material surpassed more than one lamellar unit, either because of focal accumulation of small‘cysts’ within intact elastin lamellae or because of large ‘cysts’ in an area with fragmented elastin fibers.

Smooth muscle cell orientationGrade 1: small foci with change in the orientation of the smooth muscle cells, which could be spread in different areas.Grade 2: area with change in the orientation of the smooth muscle cell orientation or several areas that together representbetween one-third and one-half of the thickness of the media.Grade 3: large area of changes in the smooth muscle cell orientation, consisting more than one-half of the media thickness.

Elastic fragmentationGrade 1: fewer than five foci with elastin fragmentation in one microscopic field (magnification ×200), each focus comprisingtwo to four neighboring elastin lamellae. The orientation of smooth muscle cells was preserved. Interruption of one elastinfiber alone was not interpreted as fragmentation.Grade 2: five or more foci with elastin fragmentation in one microscopic field, each focus comprising two to four neighboringelastin lamellae. The foci could be confluent or scattered throughout the media. The orientation of smooth muscle cells waspreserved.Grade 3: presence of foci with elastin fragmentation in five or more neighboring elastin lamellae, irrespective of the number offoci per microscopic field. The smooth muscle cells showed alterations in orientation.

InflammationGrade 1: sparse scattered chronic inflammatory cells or an occasional small focus of inflammatory cells.Grade 2: multiple small foci of inflammatory cells.Grade 3: multiple large foci of inflammatory cells or a diffuse, heavy inflammatory cellular infiltrate.

*Based on data according to Schlatmann and Becker (13), Klima et al. (14) and de Sa et al. (10).

Health & Medicine

Histopathological Grading of Ascending Aortic Aneurysm: Comparison