| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 1999;100:II-287.)
© 1999 American Heart Association, Inc.
Aortic and Peripheral Vascular Surgery |
Predictors of Proximal Aortic Dissection at the Time of Aortic Valve Replacement
From the Department of Internal Medicine, Division of Cardiology (Y.v.K., C.A.N.), University Hospital Eppendorf, Hamburg; the Department of Cardiovascular Surgery (O.S., M.S., J.O.), St. Georg Hospital, Hamburg; the Department of Cardiovascular Surgery (C.D., A.H.), Hannover Medical School, Hannover; the Department of Cardiovascular Surgery (R.L.), Christian-Albrechts-University, Kiel, Germany; and the Department of Human Genetics (A.S.), MCP-Hahnemann School of Medicine, Pittsburgh, Pa.
Correspondence to Christoph A. Nienaber, MD, Department of Internal Medicine, Division of Cardiology, University Hospital Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany. E-mail nienaber{at}uke.uni-hamburg.de
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background—Type I aortic dissection develops in 0.6% of patients late after aortic valve replacement (AVR), and 13% of patients with type I aortic dissections have a history of AVR. Predictors of aortic dissection at AVR, however, have not been characterized.
Methods and Results—A study group of 33 patients with type I
aortic dissection had aortic surgery 49±55 months after routine AVR. A
group of 101 controls, who did not have morphological progression of
aortic diameters 
6 years after AVR, was used to identify predictors
of postsurgical dissection. Multivariate
analysis identified aortic regurgitation
(P<0.002) and fragility (P<0.001) or
thinning of the aortic wall (P<0.007) at AVR as
predictors, associated with a 14%, 22%, and 7% probability of late
aortic dissection, respectively. Clamping times, types of valve
prostheses, concomitant coronary artery bypass grafting, and
mean ascending aortic diameters of 43±10 mm at AVR did not
predict late dissection. A separate analysis of 29
nondissecting aneurysms of the ascending aorta developing
104±64 months after routine AVR revealed younger age at AVR
(P<0.003) and congenitally bicuspid aortic valves
(P<0.03) as predictors of late aneurysm
formation.
Conclusions—Aortic regurgitation combined with fragile and thinned aortic walls in patients with moderate aortic dilation may reflect aortic root disease, with a high risk for postsurgical aortic sequelae if it is treated incompletely by isolated valve replacement.
Key Words: aneurysm • aorta • risk factors • surgery • valves
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Type I aortic dissection occurs in 0.6% of patients late after aortic valve replacement (AVR).1 2 3 4 5 6 7 8 9 10 11 12 13 Moreover, 13% of patients with acute type I aortic dissections have a history of previous AVR.9 12 13 14 15 16 A lethal outcome is reported in 50% of such dissections occurring within 1 month to 16 years after AVR.1 2 7 9 12 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Combining AVR with prophylactic replacement of the ascending aorta is a treatment that may prevent aortic dissection; criteria for prophylactic aortic surgery at AVR, however, are established exclusively for Marfan patients.8 12 29 32
This retrospective study was performed to assess clinical and pathoanatomical characteristics of non-Marfan patients developing type I aortic dissection late after routine AVR. A control group with stable aortic diameters who were followed for a minimum of 6 years after AVR was used to identify predictors of late dissection. These predictors may be used as an aid to estimate the risk of subsequent dissection at the time of routine AVR. A similar analysis was performed to identify predictors for progressive nondissecting aneurysm of the ascending aorta at AVR.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Patients
Between January 1979 and December 1997, 666 consecutive patients had surgery for type I aortic dissection (210 men and 90 women with a mean age of 54±13 years)33 or a nondissecting aneurysm of the ascending aorta (231 men and 135 women with a mean age of 54±15 years). Aortic surgery was performed at the University Hospital of Eppendorf, Hamburg; St. Georg General Hospital, Hamburg; Hannover Medical School, Hannover; and Christian-Albrechts-University, Kiel; all in Germany. A total of 33 of the type I aortic dissections and 29 of the nondissecting aneurysms of the ascending aorta were associated with previous AVR (Tables 1
and 2). In the group of type I aortic
dissections, 28 cases were acute and 5 were chronic (15%; cases
4, 19, 24, 25, and 30), with onset of pain within or after 14 days of
admission, respectively.34 In the group of patients with
nondissecting aneurysms after AVR, aortic surgery was performed
electively for progression of an aortic diameter >50 mm in 24
cases and as an emergency repair for rapid expansion or rupture in 5
cases (17%; patients 38, 47, 52, 57, and 62).
|
|
In an attempt to identify features at AVR associated with the risk of
either late postsurgical dissection or aneurysm, we selected a
group of 101 consecutive patients (74 men and 27 women; mean age,
63±17 years) who had routine AVR performed from January 1989 through
December 1991 at the University Hospital of Eppendorf. This control
group had 6 years of uneventful follow-up at our institution (mean,
94±9 months), with no morphological progression in the diameter of the
ascending aorta, as assessed by transesophageal
echocardiography, computed tomography, or magnetic
resonance imaging (MRI). In this group, the aortic diameter measured at
AVR was 41±10.1 mm. Stigmata of Marfan syndrome, acute
endocarditis, and concomitant mitral valve surgery at the time of AVR
excluded patients from this group, and complete documentation was
required for inclusion in a prediction model.
Study Variables
Sixteen variables were assessed in each patient. These
included (1) age at the time of AVR, (2) sex, (3) cardiac functional
class (New York Heart Association), (4) reduced left
ventricular ejection fraction (<25%), and (5) an
angiographic diagnosis of coronary heart disease, as derived
from preoperative medical records. The type of aortic valve disease
was categorized according to echocardiographic and/or
angiographic findings as (6) aortic stenosis, (7) aortic
regurgitation, or (8) a combination of both. Surgical
records at AVR were used to assess (9) a congenitally bicuspid
aortic valve, (10) the maximum diameter of the ascending aorta (mm),
and (11) thinning or (12) fragility of the aortic wall (according to
intraoperative inspection and palpation). (13) Concomitant
coronary artery bypass grafting and (14) time of aortic
cross-clamping (minutes) were also assessed in each patient. In 15
patients with no intraoperative measurements available at AVR (24%),
images from preoperative transesophageal
echocardiography, computed tomography, MRI, and/or
angiography were reevaluated for aortic diameters.34
Images were also screened for (15) aortic coarctation in all
patients. (16) Chronic systemic hypertension was considered
present with evidence of long-standing hypertension 2 years
before AVR.
Both location and distal extent of post-AVR dissection or aneurysm, maximum ascending aortic diameter, and leakage of the aortic valve prosthesis were assessed from findings on transesophageal echocardiography, computed tomography, MRI, and/or angiography performed before aortic surgery.34 Surgical records were used to assess the location of the entrance tear to the aortic dissection. Histological examination of the ascending aorta was done for evidence of cystic media necrosis (considered present in at least a moderate degree) in all study patients35 but not in controls. Follow-up after aortic surgery was obtained through repeat visits at the outpatient clinic and/or by communication with the patient’s primary physician; 9 patients (14%; patients 3, 26, 31, 45, 48, 51, 53, 54, and 55) were lost to follow-up.
Literature Review
The English, French, and German literature was screened for
reports on aortic dissection or aneurysm developing after AVR
using Medline (key words: aneurysm, aorta, risk factors,
surgery, valves) and literature lists provided in articles on this
subject. Detailed results of this survey are shown in Tables 3 and 4.
|
|
|
Statistical Analysis
The risk of developing aortic dissection or aneurysm
after AVR associated with the presence of each of the 16 variables
was evaluated separately for aortic dissection and aneurysm by
univariate, unconditional logistic regression
analysis with SAS.36 37 Age, aortic diameters, and
time of aortic cross-clamping were analyzed as continuous
variables, and they are given as mean±SD. Variables emerging
as significant predictors of risk at the 5% level were included in a
multivariate model. Estimates of risk (odd ratios) were
calculated based on the coefficients from the logistic models.
Predicted probabilities of disease (P[D]) given a particular
variable (Xi) were calculated from the
coefficients (bi) in the final
multivariate models using the equation
 |
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Findings at Aortic Surgery
Patients with type I aortic dissection and nondissecting ascending aortic aneurysm after previous AVR had similar age (55±13 versus 51±13 years) and sex distributions (79% men in both groups). Surgery for aortic dissection and nondissecting aneurysm was performed 49±55 and 104±64 months, respectively, after previous AVR (P<0.01). Diameters of the ascending aorta were similar after postsurgical dissection and aneurysm (67±21 versus 63±9 mm), whereas involvement of the descending aorta was found only in patients with aortic dissection (54%; patients 1 to 18). Paravalvular leakage was more frequent in the aneurysm group than in the dissection group (59% versus 12%; P<0.01). In all postsurgical dissections, the entry site was located in the ascending aorta; in 45% of these, the entry was found 1 to 3 cm above the site of previous aortotomy, in 17%, previous cross-clamping and cannulation sites were excluded as entry locations, and in 38%, the precise location in the ascending aorta was not described. In 3 dissections, vena cava obstruction syndrome was observed (9%; patients 15, 24, and 25). Thirty-day survival was similar in the postsurgical dissection and aneurysm groups (82% versus 90%; Tables 1
and 2).
Findings at AVR
At AVR, patients with late dissection and aneurysm had
similar maximum aortic diameters (43±10 versus 47±9 mm)
and New York Heart Association functional classes (2.5±0.8 versus
2.0±0.7); the 2 groups also had an equal distribution of patients with
systemic hypertension (33% versus 24%), cystic media necrosis (39%
versus 24%), congenitally bicuspid aortic valves (15% versus 24%),
and severely reduced left ventricular ejection fractions
(15% versus 14%). AVR was more frequently performed for aortic
regurgitation in late dissections than in post-AVR
aneurysms (73% versus 21%; P<0.01). Acute
endocarditis was not observed in any patient. At AVR, thinning (51%
versus 7%; P<0.01) and fragility (24% versus 0%;
P<0.05) of the aortic wall were noted frequently in
patients who developed aortic dissection, but rarely in those who
developed nondissecting aneurysms. Clamping times at AVR were
similar in both groups (72±27 versus 72±34 minutes). Biological valve
prostheses were equally distributed among patients with postsurgical
dissection and aneurysm (42% versus 38%). Monodisc devices
were used in postsurgical dissection and aneurysm (27% versus
34%) with similar frequency, as were bileaflet prostheses (30% versus
27%). Concomitant mitral valve operations were performed in no
patients, and coronary artery bypass grafting was performed in
6 of the study patients (10%; patients 2, 6, 16, 42, 47, and 57)
(Tables 1 and 2).
Prediction of Dissection and Aneurysm at AVR
Of the 16 variables assessed in this study,
multivariate analysis revealed aortic wall
fragility (P<0.001), aortic regurgitation
(P<0.002), and aortic wall thinning (P<0.007)
as independent predictors of late dissection. According to this model,
the probability of late dissection associated with each predictor is
22%, 14%, and 7%, respectively. Aortic wall fragility in combination
with aortic thinning or aortic regurgitation increases
the probability of late dissection to 64% and 79%, respectively;
presence of all 3 predictors results in a 96% probability of
dissection. Younger age at AVR (P<0.003) and a congenitally
bicuspid aortic valve (P<0.03) were predictors of late
nondissecting aortic aneurysm on multivariate
analysis (Table 5).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The 30-day survival in patients who had dissection after previous AVR was 82% in our study cohort, whereas analysis of the literature documents a lethal outcome in 50% of such patients (Tables 3
and 4). Moreover, in 14 442
patients who underwent AVR in 13 different centers, 0.6% developed
type I aortic dissection 1 month to 16 years after AVR (Tables 3 and 4). Similarly, 13% of 744 reported type I aortic
dissections revealed a history of AVR (Table 3). Thus, our
analysis confirms a high prevalence of previous AVR in type I
aortic dissection (11%) and documents a similar proportion of previous
AVR in nondissecting aneurysms of the ascending aorta
(8%).
Surgical Techniques
In our population, the intimal tear was located 1 to 3 cm above
the previous aortotomy in 45% of postsurgical dissections. However,
the site of the aortotomy itself was intact in all cases, and it was
not involved in the dissecting process in any case. Moreover, no
evidence exists that any of the dissections arose from the site of
previous aortic cross-clamping or cannulation. Similarly, logistic
regression analysis excluded any association of aortic clamping
time, concomitant coronary artery bypass grafting, or type of
aortic valve prosthesis with postsurgical dissection or
aneurysm. The literature review also failed to identify the
predominance of a specific aortic valve prosthesis in
postsurgical cases of dissection (Table 4).1 2 7 9 12 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Dissections at clamping or
cannulation sites are usually observed at the time or soon after
surgery.39 Thus, we think that dissections late after AVR
are usually not related to specific operative features or
techniques.
Risk for Late Dissection
Aortic diameters <50 mm usually carry a low risk of
dissection or rupture.32 40 Interestingly, the present
study reveals that postsurgical dissection developed in 33 patients who
had a mean aortic diameter of only 43±10 mm at the time of valve
replacement. Moreover, multivariate analysis
excludes the aortic diameter at AVR as an independent predictor of
dissection. Thus, in patients with moderate aortic dilation at AVR,
additional factors play a role in the subsequent development of late
aortic dissection. Multivariate analysis
identified fragility of the aortic wall, aortic
regurgitation, and aortic wall thinning as independent
predictors of dissection. Assessment of these independent risk
variables at AVR permits prediction of the risk of late dissection.
In addition, bicuspid aortic valves, systemic hypertension, and cystic
media necrosis were found in 15%, 33%, and 39% of postsurgical
dissections, respectively; statistical analysis, however,
failed to establish their association with aortic dissection.
Analysis of 57 dissections reported late after AVR confirms a
high prevalence of aortic regurgitation (92%),
arterial hypertension (57%), cystic media necrosis (39%),
and bicuspid aortic valves (24%) (Table 4).
It had been shown that—in contrast to previous AVR—previous isolated coronary artery bypass grafting is not an independent risk factor for type I aortic dissection.13 22 Intact native aortic valves probably exclude the presence of marked aortic root disease (comprising aortic regurgitation with aortic wall abnormalities) and may thus explain a lower risk for postsurgical aortic dissection in isolated coronary artery bypass grafting procedures.
Risk for Late Nondissecting Aneurysm
Younger age at AVR and presence of a congenitally bicuspid aortic
valve in moderately dilated aortas predict nondissecting aortic
aneurysms after AVR but not late dissection. Predominance of
aortic valve stenosis, lower prevalence of aortic wall
thinning, absence of wall fragility, and a longer time interval between
AVR and aortic surgery in nondissecting aneurysms suggest that
different mechanisms exist in the development of nondissecting
aneurysms and dissection after AVR. Moreover, the high
prevalence of paravalvular leakage after AVR indicates a
potentially causative role in the pathogenesis of nondissecting
progressive aneurysms.
Study Limitations
The present study was conducted in a retrospective fashion and
includes some subjective data, such as thinness and fragility of the
aortic wall. However, aortic wall thinness and/or fragility were
described independently by 20 different surgeons from 4 surgical
centers. In addition, reports from 4 centers confirm aortic wall
fragility as a key finding in 5 cases of AVR later complicated by
aortic dissection (Table 4).2 7 19 23 Future
high-resolution tomography and aortic distensibility measurements may
be used for noninvasive and objective assessment of potential wall
abnormalities.41
Conclusions
Type I aortic dissection after AVR seems unrelated to surgical
techniques. Patients with aortic diameters 43 mm at AVR should
have intraoperative assessment of risk factors for potential late
aortic dissection. Aortic wall fragility in combination with aortic
thinning or aortic regurgitation carries a 64% or 79%
probability, respectively, of late dissection, which increases to a
96% probability of dissection in the presence of all three predictors.
Thus, any patient at AVR with an aortic diameter 43 mm and the
presence of at least 2 predictors of late dissection will likely
benefit from prophylactic aortic surgery. The combined
presence of these predictors identifies a disease process of the entire
aortic root rather than isolated valve disease. We thus suggest that in
such patients, surgical treatment should encompass the entire aortic
root. Prospective studies, however, may be required for defining
optimal prophylactic surgical techniques.8
|
Acknowledgments |
|---|
The authors thank Prof Reed E. Pyeritz and Dr Walter J. Rogers from the MCP-Hahnemann School of Medicine, Pittsburgh, Pa, for their thoughtful reviews. Dr von Kodolitsch was supported by grants from the Deutsche Forschungsgemeinschaft (KO 1828/1 and KO 1828/1-2).
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Muna WF, Spray TL, Morrow AG, Roberts WC. Aortic dissection after aortic valve replacement in patients with valvular aortic stenosis. J Thorac Cardiovasc Surg. 1977;74:65–69.[Abstract]
2. Bachet J, Mesnildrey P, Goudot B, Tawil N, Dubois C, Brodaty D, Schlumberger S, Guilmet D. Dissection de l’aorte ascendante apres remplacement valvulaire aortique. Presse Med. 1984;13:2253–2256.
3. Burckhardt D, Striebel D, Vogt S, Hoffmann A, Roth J, Weiss P, Kiowski W, Pfisterer M, Burkart F, Althaus U, Goy JJ, Sadeghi H, Grädel E. Heart valve replacement with St. Jude Medical valve prothesis: long-term experience in 743 patients in Switzerland. Circulation 1988;78(suppl I):I-18–I-24.
4. Cohn LH, Allred EN, DiSesa VJ, Sawtelle K, Shemin RJ, Collins JJ. Early and late risk of aortic valve replacement: a 12-year concomitant comparison of the porcine bioprosthetic and tilting disc prosthetic aortic valves. J Thorac Cardiovasc Surg. 1984;88:695–705.[Abstract]
5. Douglas PS, Hirshfeld JW, Edie RN, Harken AH, Stephenson LW, Edmunds LH. Clinical comparison of St. Jude and porcine aortic valve prostheses. Circulation. 1985;72(suppl II):II-135–II-139.
6. Lytle BW, Cosgrove DM, Taylor PC, Goormastic M, Stewart RW, Golding LAR, Gill CC, Loop FD. Primary isolated aortic valve replacement: early and late results. J Thorac Cardiovasc Surg. 1989;97:675–694.[Abstract]
7. Hedoire F, Grollier G, Maragnes P, Valette B, Lognone T, Breut C, Scanu P, Bonnefoy L, Khayat A, Potier JC. Dissections de l‘aorte après remplacement valvulaire aortique. Arch Mal Coeur Vaiss. 1991;84:967–974.[Medline] [Order article via Infotrieve]
8. Carrel T, von Segasser L, Jenni R, Gallino A, Egloff L, Laske A, Turina M. Dealing with dilated ascending aorta during aortic valve replacement: advantages of conservative surgical approach. Eur J Cardiothorac Surg. 1991;5:137–143.[Abstract]
9. Presbitero P, Aruta E, Rabajoli F, Agaccio G, Donegani E, Orzan F, Malara D, Di Summa M, Morea M. Aortic dissection after aortic valve replacement: clinical and anatomic features. G Ital Cardiol. 1992;22:941–947.[Medline] [Order article via Infotrieve]
10. Aris A, Padró JM, Cámara ML, Lapiedra O, Caralps JM, Borrás X, Carreras F, Pons-Lladó G. The monostrut Björk-Shiley valve: Seven years’ experience. J Thorac Cardiovasc Surg. 1992;103:1074–1082.[Abstract]
11. David TE, Armstrong S, Sun Z. Clinical and hemodynamic assessment of the Hancock II bioprosthesis. Ann Thorac Surg. 1992;54:661–668.[Abstract]
12. Prenger K, Pieters F, Cheriex E. Aortic dissection after aortic valve replacement: incidence and consequences for strategy. J Card Surg. 1994;9:495–499.[Medline] [Order article via Infotrieve]
13. von Kodolitsch Y, Simic O, Dresler C, Haverich A, Nienaber CA. Aortic valve replacement: an independent predisposing factor for late aortic dissection? Circulation 1997;96(suppl I):I-507. Abstract.
14. Epperlein S, Mohr-Kahaly S, Erbel R, Kearney P, Meyer J. Aorta and aortic valve morphologies predisposing to aortic dissection: an in vivo assessment with transesophageal echocardiography. Eur J Cardiol. 1994;15:1520–1527.
15. Miller DC, Mitchell RS, Oyer PE, Stinson EB, Jamieson SW, Shumway NE. Independent determinations of operative mortality for patients with aortic dissectios. Circulation. 1984;70(suppl I):I-153–I-164.
16. Crawford ES, Svensson LG, Coselli JS, Safi HJ, Hess KR. Aortic dissection and dissecting aortic aneurysms. Ann Surg. 1988;208:254–272.[Medline] [Order article via Infotrieve]
17.
Derkac W, Laks H, Cohn LH, Collins JJ. Dissecting
aneurysm after aortic valve replacement. Arch Surg. 1974;109:388–390.
18.
Spitzer S, Blanco G, Adam A, Spyrou PG, Mason D.
Superior vena cava obstruction and dissecting aortic aneurysm.
JAMA. 1975;233:164–165.
19.
Fukuda T, Tadavarthy SM, Edwards JE. Dissecting
aneurysm of aorta complicating aortic valvular
stenosis. Circulation. 1976;53:169–175.
20. Charnsangarvej C. Occlusion of right pulmonary artery by acute dissecting aortic aneurysm. AJR Am J Roentgenol.. 1979;132:274–276.[Medline] [Order article via Infotrieve]
21. Gooch AS. Downward displacement of competent aortic valve prosthesis without left ventricular enlargement: a new sign of ascending aortic root dissection in absence of valvular regurgitation. Am Heart J. 1982;103:920–922.[Medline] [Order article via Infotrieve]
22. Orszulak TA, Pluth JR, Schaff HV, Piehler JM, Smith HC, McGoon DC. Results of surgical treatment of ascending aortic dissections occuring late after cardiac operation. J Thorac Cardiovasc Surg. 1982;83:538–545.[Abstract]
23. Le Heuzey JY, Guize L, Guerinon J, Forman J, Sellier P, Maurice P. Dissection de l‘aorte ascendante après replacement valvulaire aortique. Arch Mal Coeur Vaiss. 1983;11:1349–1356.
24.
Stone DL, Petch MC, Verney GI, Dixon AK. Superior
mediastinal obstruction with aortic dissection after aortic valve
replacement: diagnosis by computed tomography. Br Heart
J. 1984;51:565–567.
25. Morin D, Sadeghi H. Dissection de l‘aorte après chirurgie cardiaque. Schweiz Med Wochenschr. 1984;114:1530–1434.[Medline] [Order article via Infotrieve]
26.
Nancarrow PA, Higgins CB. Progressive aortic dilatation
after aortic valve replacement. AJR Am J Roentgenol.. 1984;142:669–672.
27.
Schofield PM, Bray CL, Brooks N. Dissecting
aneurysm of the thoracic aorta presenting as right atrial
obstruction. Br Heart J. 1986;55:302–304.
28. Hirao D, Ishikura Y, Odagiri S, Shimazu A, Yano K, Shirakusa T. A case report of type II dissecting aneurysm after aortic valve replacement, successful repair by the modified Cabrol’s method. Nippon Kyobu Geka Gakkai Zasshi. 1991;39:1940–1944.[Medline] [Order article via Infotrieve]
29. Albat B, Thevenet A. Dissecting aneurysms of the ascending aorta occurring late after aortic valve replacement. J Cardiovasc Surg. 1992;33:272–275.[Medline] [Order article via Infotrieve]
30. Osada T, Nagae T, Hirayama T, Ishimura S, Furukawa K. A case of ascending aortic dissection after aortic valve replacement in congenital bicuspid aortic valve. Nippon Kyobu Geka Gakkai Zasshi. 1992;40:1928–1932.[Medline] [Order article via Infotrieve]
31. Tayama K, Aoyagi S, Akashi H, Oryoji A, Higa Y, Hiromatsu S, Yamana K, Kosuga K, Ohishi K. Aortic dissection after aortic valve replacement: report of a case with an aortocameral fistula. Thorac Cardiovasc Surg.. 1995;43:299–301.[Medline] [Order article via Infotrieve]
32. Pieters FAA, Widdershoven JW, Gerardy A-C, Geskes G, Cheriex EC, Wellens HJ. Risk of aortic dissection after aortic valve replacement. Am J Cardiol. 1993;72:1043–1047.[Medline] [Order article via Infotrieve]
33. Daily PO, Trueblood HW, Stinson EB, Wuerflein RD, Shumway NE. Management of acute thoracic dissections. Ann Thorac Surg. 1970;10:237–247.[Medline] [Order article via Infotrieve]
34.
Nienaber CA, von Kodolitsch Y, Nicolas V, Siglow V,
Piepho A, Brockhoff C, Koschyk D, Spielmann R. The diagnosis of
thoracic aortic dissection by noninvasive imaging procedures.
N Engl J Med. 1993;328:1–9.
35. 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.[Medline] [Order article via Infotrieve]
36. Selvin S. Statistical Analysis of Epidemiologic Data. New York: Oxford University Press; 1991:214–240.
37. SAS Institute Inc. SAS/STAT User’s Guide. Version 6. 4th ed. Vol 1–2. Cary, NC: SAS Insttute Inc.; 1990:1071–1126.
38. Bishop YMM. Models for measuring changes. In: Bishop YMM, Fineberg SE, Holland PW, eds. Discrete Multivariate Analysis: Theory and Practice. Cambridge, Mass: MIT Press; 1975:257–279.
39. Litchford B, Okies JE, Sugimura S, Starr A. Acute aortic dissection from cross-clamp injury. J Thorac Cardiovasc Surg. 1976;72:709–713.[Abstract]
40. Platis IE, Qader MA, Coady MA, Shaw RK, Kopf GS, Elefteriades JA. Should a moderately dilated aorta be resected at the time of routine AVR? Circulation. 1998;98(suppl I):I-266. Abstract.
41. von Kodolitsch Y, Raghunath M, Nienaber CA. The Marfan syndrome: diagnosis of cardiovascular manifestations. Z Kardiol. 1998;87:161–172.[Medline] [Order article via Infotrieve]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||