The French National Authority for Health reportson thoracic stent graftsEmmanuel Corbillon, MD,a Patrice Bergeron, MD,b Anne-Isabelle Poullié, MSc,a Celia Primus, MSc,a

Tiiu Ojasoo, PhD,a and Joel Gay, MSc,b Saint-Denis La Plaine and Marseille, France

Objective: This study was conducted to determine the efficacy and safety of stent grafts in the treatment of thoracic aorticaneurysms and dissections.Method: Our health technology assessment method combined a critical review of the literature with experts’ opinions.Several databases, useful Web sites, and the gray literature were searched from January 1995 to December 2004. Somemanually retrieved major articles published in 2005 were also included. The draft report was submitted to and discussedby a working group of 12 members nominated by relevant medical societies. The amended report was submitted to amultidisciplinary group of 12 peer reviewers for comment.Results: Endovascular stent grafting (ESG) repair for lesions of the thoracic aorta, including aneurysms, dissections, andaortic isthmus ruptures, is probably beneficial in terms of operative mortality and severe morbidity, with an incidence ofparaplegia of 2.1% (range 0%-7%) for ESG vs 5% (range, 3%-15%) for surgery, provided that there is a rigorousmedium-term assessment and that anatomic factors are favorable. A proximal neck length of at least 2 cm is needed toinsert the stent graft. Indications for ESG in thoracic aortic aneurysm and dissection are similar to those for surgery.Endovascular stent grafting is particularly appropriate in patients with multiple traumas to the thoracic aorta in whomconcomitant lesions are a contraindication to open surgery. Endovascular stent graft repair can only be done in public orprivate centers with expertise in both endovascular and surgical procedures and with adequate technical facilities. Patientsshould be informed of the advantages and drawbacks of both methods. A multidisciplinary discussion should addressrisks of converting to open surgery and possible need for a cardiopulmonary bypass. Patients should be monitoredannually by computed tomography scan or magnetic resonance imaging and plain radiographs because long-term resultsare uncertain (possible stent graft deterioration, onset of aortic disease). They should be told of the need for surveillanceand possible further treatment.Conclusions: A prospective registry of all thoracic aorta procedures (endovascular treatment, open surgery, thoracic ESGplus extra-anatomic bypass) needs to be set up. All practitioners and stent graft manufacturers should contribute to thisregistry. It should include information on patient monitoring in order to (1) select patients who could be treated by ESGrepair, (2) assess the feasibility of a randomized controlled study comparing ESG and surgery, (3) assess the medium-termoutcome of different devices, and (4) obtain a better understanding of the health economics aspects. ( J Vasc Surg 2008;

47:1099-1107.)

The two main degenerative diseases affecting the tho-racic aorta (TA) are aneurysms (TAAs) and dissections(TADs). If left untreated, they may be life threatening.Treatment is medical or surgical, or both. The aim ofmedical treatment is to control blood pressure, but itsefficacy is limited once symptoms have appeared (impend-ing aneurysms, back pain, malperfusion syndrome) or whenthe degree of true lumen constriction exceeds two-thirds.At this stage of the disease, open surgery is the goldstandard treatment; however, only patients with an accept-able surgical risk can benefit from open surgery because itmay involve cardiopulmonary bypass. Endovascular stentgrafting (ESG) might prove to be an acceptable alternativeto open surgery.

From the Haute Autorité de Santé, Saint-Denis La Plainea and HôpitalSaint-Joseph, Marseille.b

Competition of interest: Dr Bergeron has shares in Medtronic & LeMaitreVascular.

Reprint requests: Patrice Bergeron, MD, Department of Thoracic & Car-diovascular Surgery, Saint Joseph’s Hospital, 26 boulevard de Louvain,13285 Marseille Cedex 08, France (e-mail: pbergeron@hopital-saint-joseph.fr).

0741-5214/$34.00Copyright © 2008 by The Society for Vascular Surgery.

doi:10.1016/j.jvs.2007.11.021

One of the main activities of the Haute Autorité deSanté (HAS), or French National Authority for Health, isto assess drugs, medical devices, and procedures to encour-age their proper use by professionals and to provide healthauthorities with the information they require to make de-cisions on reimbursement by National Health Insurance.The French Society of Vascular Surgery therefore askedHAS to assess the efficacy and safety of ESG in the treat-ment of TAAs and TADs. This article is based on a HASreport published in February 2006.

METHOD

The assessment report on ESG for TAAs and TADs wasproduced using an established method combining a criticalappraisal of the literature and the opinion of a panel ofexperts recruited from several medical societies. We per-formed an electronic search of several bibliographic data-bases (MEDLINE, EMBASE, and PASCAL; articles inEnglish or French) from January 1995 to December 2004and consulted useful Web sites and the grey literature (eg,conference reports, fact sheets, patent applications, sympo-sia, technical reports, etc). A manual search was used to

retrieve major articles published in early 2005.

1099

JOURNAL OF VASCULAR SURGERYMay 20081100 Corbillon et al

The following types of studies were included in ouranalysis:

1. comparative diagnostic and therapeutic studies on TAdisease, regardless of sample size;

2. safety and efficacy studies on the treatment of athero-matous aneurysms, unruptured type B or traumaticdissections, intramural hematomas, and penetratingaortic ulcers (TA segments II and III), regardless of

Table I B. Results of the literature search

Articles, No.

Articles retrieved 529Articles analyzed 281Articles cited in report 58

Descending thoracic aorta treatment 34b

Emergency treatment (ruptures and traumas) 10Disease spread to the aortic arch. 8Reviews 6

bIncluding a double study.

Table I A. Method of the critical analysis of the scientific

Documentary search on French and English articles Documen● MEDL● Cochra● Web sit● Evaluat● Web sit

(AFSSA● Industr

Subjects Study typ● Thoracic surgery medical treatment of aortic

aneurysms and dissections● Imaging and implantation of thoracic stent graft● Anesthesia, intensive care, and surgery of

thoracic aorta

● Recom● Meta an● Literatu● Contro● Cohort● Econom

Period Jan 1995Key words (various combinations have been used) Aneurysm

AnévrisdissectiThoracaneurysCohortstatemeanalysisCrossovDissectDoubleEconomrésumé[Titre,GuideliHospitareview,MagnetMinimaPracticstudy,a

aFrench Health Ministry.

study design and evidence level;

3. studies on type B retrograde dissections of TA segmentI (non-A, non-B TADs); and

4. recent good-quality reviews of TA surgery (includinganesthesia), medical treatment of aneurysms and aorticdissections, TA imaging, and ESG.

We excluded studies on segment I TAAs, thoracoab-dominal aneurysms, or patients presenting both thoracicand abdominal aneurysms because ESG may not be indi-cated in these cases. We also excluded cohort studies with�30 patients for combined TAA and TAD studies, �20 forTAA or TAD studies, �10 for traumatic aortic disease,intramural hematomas, or penetrating aortic ulcers. Finally,we excluded studies of homemade stent grafts because ofdevice variability and absence of assessments during vigi-lance monitoring.

The results of the literature search and analysis are givenin Table I. For each study, we determined internal validity(bias due to selection, nonblinding, and confoundingvariables) and external validity (population characteris-tics, follow-up, relevance of results). Validity was classified

economic literature

ourcesEMBASE, PASCAL databasesbrarym scientific societies for endovascular or thoracic surgerygencieshe French Agency of Sanitary Security of Health Products

ationsviewnd randomized trialsies, French literatureudiesc 2004, and early 2005

urysm rupture, aneurysm surgery, aneurysm thoracic,angiography, antilipemic agent, aorta, aorta aneurysm, Aortaeurysm, Aorta surgery, Aorta, thoracic/surgery, AorteAortic aneurysm, Aortic aneurysm, thoracic, Aortic ruptureardiovascular agent, Cardiovascular surgery, Cohort analysis,es, Comparative study, Consensus development/conference/tle], Controlled clinical trial,a Cost, Cost allocation/benefittrol, Cost effectiveness analysis, Cost of illness, Cost saving,ocedure, Cross-over studies, Diagnostic imaging, Dissecting,rta, thoracic Aortic aneurysm thoracic, Dissection aortique,procedure, Double-blind method, Drug therapy, Economic,

pect, Economic, Economics, Endoprosth, Endoprosth [Titre,oprosthesis, Endoprothèse,a Endovasculaire, Endovascularé], Endovascular stents, Financial management, Follow up,ealth care, Health economic, Hospital, Hospital billing,

nce, Hospital purchasing, Hospital running cost, Literaturerature francophone, Longitudinal studies, Magnetic resonance,onance angiography, Meta analysis, Minimally invasive,vasive surgery, Nuclear magnetic resonance, Pharmaeconomics,eline, Health planning guideline Recommendation, Prospective

hesis and implants, Vascular surgery

and

tary sINE,ne Lies froion ae of tPS)ialesmendalysire relled astudic st

to De, aneme,ang anique,m, Cstudint [ti/coner pr

ing aoblindic as

], Endrésumne,a Hl finaLittéic reslly in

e guidProst

as “poor,” “average” or “satisfactory.”

sesop

JOURNAL OF VASCULAR SURGERYVolume 47, Number 5 Corbillon et al 1101

The HAS project manager produced a draft criticalreview of the literature that was submitted for discussion toa multidisciplinary working group of 15 members consist-ing of cardiologists, cardiac surgeons, vascular surgeons,vascular physicians, radiologists, engineers, anesthesiolo-gists, and an account manager. Working group memberswere nominated by the relevant medical societies. An anal-ysis of French ESG practice based on data from the Pro-gramme of Medicalization of Information Systems (PMSI)and from the Common Classification of Medical Proce-dures (CCAM), and on a description of care at Saint JosephHospital in Marseilles was included in the report. Theamended report was then submitted to a multidisciplinarygroup of 12 peer reviewers for comment.

RESULTS

Study quality. Seven nonrandomized, comparativestudies (evidence level 2) were included in the 58 articlesanalyzed (Table II).1-7 Five of these studies comparedsurgical and endovascular treatments, one compared med-ical and endovascular treatments, and one concerned imag-ing techniques. Only the studies by Brandt et al5 andNienaber et al2 compared ESG with conventional surgeryin two groups of statistically comparable patients, either byidentification of similar subpopulations or prospectively.

In the three other studies of ESG vs surgery, thedemographic and clinical characteristics of the two groupsof patients differed, with the ESG patients presenting a highcomorbidity rate, such as high blood pressure, heart dis-ease, or chronic pulmonary obstructive disease.1,3,4 In thestudy by Ehrlich et al,1 all the patients undergoing surgerycould have been eligible for ESG; however, whereas all ESGpatients had either coronary artery disease or a history ofmyocardial infarction, only 53% of the surgical patients did.In the study by Najibi et al,3 the proportion of men andwomen was not the same in both groups. None of thesethree studies accounted for confounding factors.1,3,4 Thestudy by Bortone et al6 compares ESG with medical treat-ment but makes no mention of a comparison between the

Table II. Nonrandomized comparatives studies design

First author Treatment Patient

Ehrlich,1 1998 First commercial ESG vs surgery 10Nienaber,2 1999 ESG vs surgery 12Najibi,3 2002 ESG vs surgery 46

Doss,4 2003 ESG vs surgery with cardio-pulmonary assistance

26

Brandt,5 2004 ESG vs surgery 22

Bortone,6 2004 ESG vs medical treatment 110

Rocchi,7 2004 Diagnostic role of intrasurgicalTEE and pre-op angiographyin ESG patients

42

ESG, Endovascular stent grafting; TAD, thoracic aorta dissection; TEE, tran

groups under study.

Of the 58 studies analyzed, 34 (59%) were noncom-parative cohort studies. The study of Ramaiah et al,8 inwhich patients were split into two distinct series accordingto year of ESG treatment, was the only exception even if itwas not a genuine comparative study. Whether the studywas prospective or retrospective was specified in 22 of 34(65%) studies (8 prospective, 14 retrospective), and theprimary end point was clearly stated in 32 (94%). Thenumber of patients to be included was rarely calculated:only three studies (9%) reported intention-to-treat results,and only nine (26.5%) performed an intention-to-treatanalysis. Even if the methodologic quality of these cohortstudies was rather poor, 31 (91%) reported results thatdirectly concerned the primary end point. All of the studiesaddressed the clinical relevance of ESG and its use as aroutine procedure. Patients formed representative samplesof the populations usually treated for these diseases and notof the general population.

Efficacy of endovascular treatment of descendingthoracic aorta disease regardless of cause. The technicalsuccess rate was mentioned in 31 studies and ranged from77% to 100% according to study. The overall success ratewas 94.6% (Table III).1-3,7,9-31 Five studies defined tech-nical success at the outset.3-7,28,32 According to Czernyet al12 and Taylor et al,29 technical success is “perfect ESGdeployment without primary leak.” Leurs et al20 completedthis definition as follows: “successful ESG deployment inthe absence of primary endoleak and intra-surgical compli-cation.” However, the latest definitions are given by Erbelet al,23 “insertion and successful deployment of ESG with-out surgical conversion necessity nor per-surgical mortality,in the absence of type 1 or 3 endoleak, till the 24th postsurgical hour,” and by Najibi et al,3 who reported thetechnical success rate at 30 days using the definition givenby the Society of Vascular Surgery/International Societyfor Cardiovascular Surgery. The mean technical success ratein these five studies defining success rate was 88.8%(Table IV).

Seven studies reported blood loss. Among them, 6

Description of patient population

Two historically successive groups of patientsProspective study in TAD patients (2 centers)Historical controls (patients treated by surgery who were

retrospectively considered to be good candidates for ESG)Nonrandomized, based on surgical risk

No significant difference in the demographic and clinicalcharacteristics of both groups of patients

Retrospective study of 4 groups: 1, ESG for atheroscleroticaneurysms; 2, ESG for traumatic lesions; 3, ESG forcomplicated type B dissections; 4, medication

Long-term study in patients with a chronic type B dissectionon whether intra-op TEE modifies the planned ESGintervention

hageal echocardiography.

s, No.

studies have quantified a total of 372 patients with a mean

17

tic dis

JOURNAL OF VASCULAR SURGERYMay 20081102 Corbillon et al

blood loss of 358.5 mL (range, 0-3000 mL).3,9,17,18,21,24

The mean surgical conversion rate, reported in 26 studies,was 1.4% (range, 0%-7.4%). Seven studies reported the ESGprocedure lasted an average of 118 minutes (range, 37-580minutes). The mean length of hospital stay in 17 of 19 studiesdocumenting this variable was 6.6 days (range, 2.8-10 days)for a total of 542 patients. One study reported a mean lengthof stay of just 4 days.9 In contrast, eight studies mentioned amean length of stay in the intensive care unit of 2.9 days

Table III. Technical success rate

First author

Patients, No.T

Total Elective Emergency

Bell,9 2003 67 42 25Bergeron,10 2003 33 32 1Bortone,6 2004 110Criado,11 2002 47Czerny,12 2004 54Dessl,13 2004 84 66 18Destrieux-Garnier,14 2004 32 21 11Doss,4 2003 26Ehrlich,1 1998 10Fattori,15 2003 70 56 14Hansen,16 2004 60 47 13Heijmen,17 2002 27 27Lepore,18 2002 43 20 23Leprince,19 2003 26Leurs,20 2004 443 266 155Lonn,21 2003 20Makaroun,22 2004 142Erbel,23 2003 94Najibi,3 2002 46Neuhauser,24 2004 31 18 13

Nienaber,2 1999 12Orend,25 2003 74 48 26Ouriel,26 2003 31 26 5Ramaiah,8 2003 46Ramaiah,8 2003 37Rocchi,7 2004 42Scharrer-Pamler,27 2003 45 34 11Schoder,28 2003 28Taylor,29 2001 37 19 18Totaro,30 2002 32Tse,31 2004 36Total 1885

Table IV. Technical success rate in studies defining techn

Indication

First author Patients, No. TAA TAD Trauma

Czerny,12 2004 54 54 0 0Leurs,20 2004 443 249 131 50Erbel,23 2003 94 NA NA NANajibi,3 2002 19 NA NA NATaylor,29 2001 37 18 6 5Total 647

NA, Not available; TAA, thoracic abdominal aneurysm; TAD, thoracic aor

(range, 1.5-4.3 days) for a total of 384 interventions.

Death after endovascular treatment. The mean 30-day mortality rate in 1756 patients from 31 studies was6% (range, 0%-15%). Overall mortality, defined as deaths�30 days and during follow-up, was derived from 21studies and was 0% to 28%. Of these 21 studies, 17 gave amean follow-up that was used to calculate an overall mor-tality rate of 11.7% over an average of 18.4 months offollow-up. Seven studies provided actuarial survival rates,one of which reported a 2-year survival rate of �50%.23 The

cal success rate,No. (%)

Conversions,No. (%)

Early death,No. (%)

Endoleaks,No. (%)

67 (100) 5 (8) 5 (8)32 (97) 3 (9) 006 (96.4) 1 (0.9) 4 (3.6)46 (98) 1 (2.1)51 (94.4) 0 2 (3.7) 15 (28.9)73 (87.2) 1 (2.5) 1 (2.5) 4 (11)32 (100) 0 3 (9) 9 (28)20 (77)10 (100) 0 1 (10) 2 (20)68 (97) 5 (7.4) 1 (1.5) 5 (7.4)59 (98.3) 2 (3.4) 9 (15) 14 (24)26 (96)43 (100) 0 3 (7) 9 (21)26 (100) 1 (3.8) 2 (7.7)90 (87-89) 0 41 (9.3)20 (100) 0 3 (15) 2 (10)40 (98) 1 (1) 2 (1.5) 6/68 at 2 y (9)80 (85)46 (100)30 (95) 6 (19) 13 (45); type I,

7 (23)12 (100) 0 0 074 (100) 5 (6.7) 7 (9.5)31 (100) 1 (3.2) 4 (12.9) 14 (45)46 (100) 0 2 (4) 4 (8)37 (100) 0 4 (10.8) 2 (5.4)42 (100) 045 (100) 0 3 (6.7) 8 (17.8)28 (100) 0 0 9 (32)36 (97.3) 0 3 (8)32 (100) 1 (3.1) 0 3 (10)36 (100) 2 (5.6) 0 3 (8.3)84 (94.6)

uccess (type I)

Circumstances

ther Elective Emergency Technical success rate, No. (%)

0 NA NA 51 (94.4)13 266 155 393 (88.7)NA NA NA 80 (85)NA NA NA 17 (89)

8 19 18 36 (97.3)577 (89.2)

section.

echni

1

3

1

ical s

O

survival rate in the other six studies was �80% at 2, 3, or 5

JOURNAL OF VASCULAR SURGERYVolume 47, Number 5 Corbillon et al 1103

years15,20,24,26-28 and even reached 95% at 5 years (meanfollow-up, 25 � 15 months) in the study by Fattori et al.15

The 1-year survival rates were 80% and 90% in twostudies20,25 and 47% and 84% in two others.24,26

Emergency patients were at much higher risk of deaththan those receiving elective treatment, whether in the caseof ESG or open surgery. Of the 27 cohort studies thatincluded surgical and endovascular repair, 11 distinguisheddeath after elective and emergency surgery. Surgery was notreviewed in other articles. Emergency patients were a het-erogeneous group, including traumatic ruptures with lowmortality and TAA or TAD ruptures in patients with mul-tiple comorbidities. A single study defined emergency treat-ment as treatment �48 hours of admission, symptomatic orfissured aneurysms, complicated Stanford type B aorticdissections, and traumatic aortic sections.9 The mean 30-day mortality rate in these 11 studies was 5.9% (range,0%-16%) in electively treated patients and 16% (range,2.5%-23%) in emergency cases. The overall mortality rate inseven studies was 13.3% (range, 4%-45%) in electivelytreated patients and 28.2% (range, 5%-38.5%) in emer-gency cases for a median follow-up of 20.5 months.

Endoleaks after endovascular treatment. Endoleaksare a specific complication of endovascular treatment. TheHAS report used the endoleak classification given by Whiteet al33 for TAAs. Type I endoleaks are defined as anincomplete or ineffective seal at the proximal or distalend(s), or both, of the graft, and type II endoleaks as aretrograde blood flow from arterial branches. In TADs, thefalse lumen can also be perfused from distal re-entries andblood flow can circulate in a retrograde way; therefore, weprefer to use “false channel perfusion” instead of “en-doleaks” in TADs.

The rate of early endoleaks was reported in 18 studiescovering all types of TA disease. The mean rate was 10.5%(range, 0%-45%) for a total of 796 patients. No endoleakswere observed in three studies totalling 68 patients. Thetype of endoleak was specified in eight of 18 stud-ies.9,15,18,23,26-28,31 In patients with an endoleak (13.4% of371 patients), 76% of the endoleaks were type 1 leaks

Table V. Endoleaks according to indication

First author Patients, No.

Ind

TAA TAD

Bavaria,34 2004 50 48 2Bell,9 2003 67 44 14Bortone,6 2004 110 43 43Destrieux-Garnier,14 2004 32 3 17Lepore,18 2002 43 14 16Leurs,20 2004 443 249 131Lonn,21 2003 20 0 20Nienaber,2 1999 12 0 12Taylor,29 2001 37 18 6Total 814 419 261

TAA, Thoracic abdominal aneurysm; TAD, thoracic aortic dissection.

(remaining endoleaks at the end of the procedure), 20%

were type 2, and 4% were type 3. No early type 4 endoleakswere observed.

Late endoleaks were reported in 18 further studiestotalling 907 patients. The mean rate was 7.5% for a medianfollow-up of 17.9 months. Three studies totalling 93 pa-tients reported a complete absence of endoleaks for a meanfollow-up of at least 12 months.2,3,10 Only six studiesspecified endoleak type: 61.8% were type 1 leaks, 23.5%were type 2 leaks, and 14.7% were type 3 leaks. No studymentioned a type 4 endoleak caused by porosity of the graftmaterial.

Of the 21 selected studies specifying the disease beingtreated and distinguishing TAAs and TADs, nine reportedan endoleak rate but without specifying the disease, threedid not distinguish early and late endoleaks,7,24,32 and onlynine distinguished endoleaks in TAAs and circulating falselumens in TADs. The 814 procedures in these nine studiesconcerned 419 TAAs, 261 TADs, 87 aortic traumas (eitherTAA or TAD), and 47 procedures for other causes (TableV).2,6,9,14,18,20,21,29,34 No endoleaks occurred in two stud-ies of TAAs6,18 and two studies of TADs.2,31 The mean rateof endoleaks was highly similar for TAAs at 10.3% (range,4%-30%) and TADs at 11.1% (range, 7%-47%). With a totalof 32 procedures, the total rate of endoleaks for TAD was6.25%.

Other complications after endovascular treatment.Technical, neurologic, and other complications are sum-marized in Table VI. Other types of complications werementioned in 17 studies totalling 900 patients, the threemain being renal insufficiency, respiratory complications,and scar complications.

DISCUSSION

Methodologic limitations. Our assessment revealedthe many methodologic limitations of studies on endovas-cular treatment of diseases of the TA. These limitationsconcerned:

● Study design. Most studies were cohort studies, includ-ing 27 retrospective (evidence level 5) and 11 prospec-

nFollow-up, meanmonths (range)

Endoleaks, No. (%)

uma Other TAA TAD

12 14 (28) 03 6 17 (2-64) 3 (4) 2 (14.3)4 0 20.8 � 10 (1-55) 0 3 (7)5 7 13.5 (2-28) 1 (30) 8 (47)0 13 19 (0-34) 0 2 (12.5)0 13 (1-60) 23 (9.1) 11 (8)0 0 13 (2-28) — 2 (10)0 0 12 — 05 8 17.5 (6-45) 2 (9.1) 1 (8)7 47 15.6 43 (10.3) 29 (11.1)

icatio

Tra

2

5

8

tive (level 3); six studies were reviews.

ost-tr

JOURNAL OF VASCULAR SURGERYMay 20081104 Corbillon et al

● Patient and disease heterogeneity. In the six studies thatincluded an open surgery group (level 2), a compari-son with ESG was not possible because of patientheterogeneity in demographic characteristics and dis-ease severity. Studies often did not specify patientselection criteria and disease severity, distinguish aorticaneurysms and dissections from different causes, de-scribe marginal patients, or provide information onloss to follow-up. Results for TAAs and acute andchronic TADs and for emergency and elective treat-ment were seldom distinguished.

● Outcomes. Events and technical success were rarelydefined: only six of 31 studies on technical success pro-vided a definition. Few studies reported long-term resultsfor ESG. Only three studies had a mean follow-up of �2years. The durability of stent grafts is thus uncertain

Table VI. Complications

ComplicationStudies,

No.Patients,

No.Comp

%

Technical complications otherthan endoleaks

Vascular access complications 14 803 Mean,

ESG fracture 3 208 5.3 in

ESG migration 8 525 Mean,

Neurologic complicationsParaplegia 24 1480 Mean,

Stroke 18 836 Mean,

Other neurologic complications 3 228 Mean,

Nontechnical and non-neurologiccomplications

Severe renal failure or requiringdialysis

11 501

Respiratory complications 7 448 Mean,

Scar complications 6 220 Total,

Myocardial infarction 2 121

Post-transplantation syndrome 2 115Retrograde dissection 3 180 Mean,

ESG, Endovascular stent grafting; MI, myocardial infarction.aIn the study by Orend et al,25 4 patients (5%) required mechanical ventilat(1%) patients, respectively, required assisted ventilation for �48 hours.bIn the study by Fattori et al,15 81% of 68 patients presented a transient pprotein, and moderate hyperthermia.

over the long-term.

Our assessment of clinical outcomes was based on 42studies, most of which were cohort studies. Seven of thesestudies compared endovascular and conventional surgicaltreatment but nearly always in patient groups with dissim-ilar characteristics. The outcomes investigated were deathat 30 days, paraplegia rate, operative time, intraoperativeblood loss, and average length of stay in hospital.

Outcomes for endovascular treatment of dissec-tions of the thoracic aorta. During the last 30 years, theprognosis of patients with aortic dissections has improvedsubstantially. The European Echocardiographic Coopera-tive Study Group has reported 1-year post-ESG survivalrates of 52%, 69%, and 70% and 2-year survival rates of 48%,50%, and 60% for DeBakey types I, II (Stanford type A) andIII (Stanford type B) dissections, respectively.23 One-yearsurvival rates of 34% and 85% for Stanford type A and B

on rate,ge) Comment

1.4-14) 47 complications; lesions: iliac artery dissections,4,14

primitive iliac artery perforations,17 femoral arterydissections and/or rupture32 and iliac arteryruptures.10,15,25

atients No fractures in 2 studies24,34 (mean follow-up, 14.25mon); 11 fractures (13%) in a study of 84patients32 (mean follow-up, 15 mon).

No migration in 5 studies3,8,11,32,34; 11 instances ofmigration in 543 patients in 4 studies

0-6.5) No paraplegia in 13 studies1,2,4,6,8,10,13,14, 26-30; 31cases in 10 studies; similar rates of paraplegia forelective and emergency surgery (5% and 4%, resp)in 1 study9

1.4-20) No strokes in one study4; 38 strokes in 836 patientsin 17 studies

6 cases of neurologic complications, incl 1 eachtransitional monoparesis15 and transitionalparaparesis24

26 cases of renal failure

44 cases of complications requiring mechanicalrespiratory assistance4,18,24 or pneumonia-typecomplications9a

Healing complications, incl 1 each groin hematoma,9femoral arterial access thrombosis,11 groininfection,4 groin hematoma or a superficialinfection,17 an inguinal lymphocele3

Isolated cases of acute coronary syndrome after 1 and22 mon in these 2 studies: a MI without Q-wave11

and a nontransmural MI25

56 cases in a mean follow-up of 25 mon11,15b

6 cases

r �24 hours. In the studies by Doss et al4 and Lepore et al,18 2 (8%) and 6

ansplantation syndrome characterized by leukocytosis, elevated C-reactive

licati(ran

5.8 (

208 p

2.0

2.1 (

4.5 (

2.6

5.2

9.8

8.2

3.3

ion fo

dissections, respectively, had been reported earlier.35 Be-

JOURNAL OF VASCULAR SURGERYVolume 47, Number 5 Corbillon et al 1105

cause survival seems to depend on the degree of communi-cation between the true and false lumen (ie, the pressureexercised over the wall within the false lumen), the prog-nosis of noncommunicating and retrograde Stanford typeB dissections restricted to the descending aorta was better:the 2-year survival rates were 80% and 86%, respectively.

Endovascular treatment provides better outcomes thanmedical treatment. The International Registry of AcuteAortic Dissection (IRAD) study group reported mortalityrates of 53% and 9% for Stanford types A and B (DeBakeyI-III) dissections, respectively, after medical treatmentcompared with 27% and 29% after surgical treatment in astudy of 464 patients.36

The morbidity rates of endovascular treated TAAs andTADs were similar. Vascular access seemed to cause fewercomplications because of flexible arteries, provided thestent graft was placed in the true lumen. Strokes after TADswere less frequent than after TAAs owing to the absence ofatheromatous lesions of the aortic arch. However, supra-aortic vessels could be the seat of dissection-induced malp-erfusion. The paraplegia rate was no higher for TADs thanTAAs. Reports of aortic arch and ascending aorta retro-grade dissection were rare.37-39 Lethal mechanical compli-cations have been reported, including rupture and fistulasbetween the aorta and esophagus or bronchus, related tothe uncovered part of the stent graft. The use of this type ofstent should be avoided.

Organization of endovascular treatment and surveil-lance based on the expert panel’s opinion. Endovasculartreatment of TAAs and TADs should be performed inexpert centers where catheterization and endovascular pro-cedures are routine. Multiple skills are required, includingknowledge of medical, surgical, and endovascular treat-ments and their indications; clinical experience; and abilityto interpret the results of imaging examinations. High-quality intensive care requires close cooperation betweencardiologists and radiologists. Nursing staff should be ex-perienced in the surveillance of patients with instability orcomplications arising from a surgical conversion or from aneed for respiratory or renal assistance or both. All sur-geons should have a university qualification in the field andhave attended relevant continuous medical training ses-sions. Their practice should have been certified in referencecenters.

Patient selection depends on anatomic and clinicalcriteria to be discussed during multidisciplinary staff meet-ings in order to assess the benefit–risk ratios of the treat-ments offered.

Patients should be informed of the advantages anddrawbacks of each procedure and of the conditions oflong-term surveillance in order to be able to reflect on theoptions and make an informed choice. Provision of infor-mation to the patient will be put down in writing but doesnot imply that the patient has understood the information.The patient, or his or her family if appropriate, must giveinformed consent for the chosen treatment.

Postoperative surveillance will depend on the patient’s

clinical condition. On discharge from the hospital, the

patient will be given a schedule of hospital visits and radio-logic examinations, including computed tomography (CT)scanning or magnetic resonance imaging (MRI) plus tho-racic radiograph. The French Agency for the Safety ofHealthcare Products (AFSSAPS) has not yet issued guide-lines for surveillance after endovascular treatment of TAAsand TADs. The recommended schedule for the abdominalaorta ESG should be applied in the meantime; that is,examinations on discharge from hospital or after 1 month,at 3 months if an early endoleak occurred, at 6, 12, 18, and24 months, and yearly thereafter. Surveillance should betailored to the patient and depends on endoleak occurrence,available equipment, and risks from radiation overexposure.The working group recommended that for reproducibilityand reliability, radiologic examinations should be done when-ever possible in the same center. Imaging reports shouldrecord the maximum aneurysm diameter, permeability, stentgraft migration, periprosthetic leaks and, if appropriate, theperfusion of the false lumen at both thoracic and abdominallevels.

CONCLUSIONS

Endovascular treatment of thoracic aorta disease, in-cluding aneurysms, dissections, and traumatic disruption, isprobably beneficial in terms of operative mortality andsevere morbidity (incidence of paraplegia, 2.1% [range,0%-7%] for ESG vs 5% [range, 3%-15%] for surgery), pro-vided there is a rigorous medium-term assessment andprovided that anatomic factors are favorable. In particular,a proximal neck length of at least 2 cm is needed to insertthe stent graft. The indications for endovascular treatmentare the same as those for open surgery. Endovascular stentgrafting seems particularly appropriate in patients withmultiple thoracic traumas in whom concomitant lesions area contraindication to open surgery.

Treatment should be done in public or private centerswith expertise in both endovascular and surgical proceduresand with adequate technical facilities. Patients should beinformed of the advantages and drawbacks of both meth-ods. Endovascular stent graft treatment should be subjectto prior multidisciplinary discussion, particularly with re-gard to risks of conversion and need for a cardiopulmonarybypass. Patients should be monitored annually by CT scanor MRI plus plain radiograph because long-term results areuncertain and include possible stent graft deterioration andonset of aortic disease. They should be told of the need forsurveillance and possible further treatment.

Scientific evidence is currently of a low level. Random-ized studies would be useful but are difficult to set upbecause patient numbers are small and catchment areas arevery large. Data should be entered into a registry and bepart of a medical device vigilance study within the establish-ment or be included prospectively in study currently com-paring ESG with conventional surgery and medical treat-ment, or both, similar to the design of the Investigationof Stentgrafts in Aortic Dissection (INSTEAD) trial. Aregistry of all thoracic aorta interventions, including

endovascular treatment, open surgery, thoracic ESG plus

JOURNAL OF VASCULAR SURGERYMay 20081106 Corbillon et al

extra-anatomic bypass, and of follow-up data, accessibleto all practitioners and device manufacturers, would havea number of advantages. It would (1) help identify theprofiles of patients suitable for ESG treatment, (2) assessthe practicability of a long-term randomized study com-paring treatment modalities, (3) allow medium-termassessment of different devices, and (4) provide informa-tion for health economics studies.

AUTHOR CONTRIBUTIONS

Conception and design: ECAnalysis and interpretation: PB, EC, CP, TO, JGData collection: EC, PB, AP, CP, JGWriting the article: PB, TO, JGCritical revision of the article: EC, PBFinal approval of the article: EC, PB, TOStatistical analysis: EC, JGObtained funding: Not applicableOverall responsibility: EC, PB

REFERENCES

1. Ehrlich M, Grabenwoeger M, Cartes-Zumelzu F, Grimm M, Petzl D,Lammer J, et al. Endovascular stent-graft repair for aneurysms on thedescending thoracic aorta. Ann Thorac Surg 1998;66:19-25.

2. Nienaber CA, Fattori R, Lund G, Dieckmann C, Wolf W, von Kodo-litsch Y, et al. Nonsurgical reconstruction of thoracic aortic dissection bystent-graft placement. N Engl J Med 1999;340:1539-45.

3. Najibi S, Terramani TT, Weiss VJ, Mac Donald MJ, Lin PH, Redd DC,et al. Endoluminal versus open treatment of descending thoracic aorticaneurysms. J Vasc Surg 2002;36:732-7.

4. Doss M, Balzer J, Martens S, Wood JP, Wimmer-Greinecker G, FieguthHG, et al. Surgical versus endovascular treatment of acute thoracicaortic rupture: a single-center experience. Ann Thorac Surg 2003;76:1465-70.

5. Brandt M, Hussel K, Walluscheck KP, Müller-Hülsbeck S, Jahnke T,Rahimi A, et al. Stent-graft repair versus open surgery for the descend-ing aorta: a case-control study. J Endovasc Ther 2004;11:535-8.

6. Bortone AS, De Cillis E, D’Agostino D, Tupputi Schinosa L. Endovas-cular treatment of thoracic aortic disease: four years of experience.Circulation 2004;110(11 suppl):II262-7.

7. Rocchi G, Lofiego C, Biagini E, Piva T, Bracchetti G, Lovato L, et al.Transesophageal echocardiography-guided algorithm for stent-graftimplantation in aortic dissection. J Vasc Surg 2004;40:880-5.

8. Ramaiah V, Rodriguez-Lopez J, Diethrich EB. Endografting of thethoracic aorta: single-center experience with technical considerations.J Card Surg 2003;18:444-54.

9. Bell RE, Taylor PR, Aukett M, Sabharwal T, Reidy JF. Mid-termresults for second-generation thoracic stent grafts. Br J Surg 2003;90:811-7.

10. Bergeron P, De Chaumaray T, Gay J, Douillez V. Endovasculartreatment of thoracic aneurysms. J Cardiovasc Surg 2003;44:349-61.

11. Criado FJ, Clark NS, Barnatan MF. Stent-graft repair in the aortic archand descending thoracic aorta: a 4-year experience. J Vasc Surg 2002;36:1121-8.

12. Czerny M, Cejna M, Hutschala D, Fleck T, Holzenbein T, Schoder M,et al. Stent-graft placement in atherosclerotic descending thoracic aorticaneurysms: midterm results. J Endovasc Ther 2004;11:26-32.

13. Dessl A, Chemelli A, Steingruber I, Jaschke W, Waldenberger P.Stenting of the thoracic aorta - the radiologist’s view. Eur Surg ActaChir Austriaca 2004;36:145-51.

14. Destrieux-Garnier L, Haulon S, Willoteaux S, Decoene C, Mounier-Vehier C, Halna P, et al. Midterm results of endoluminal stent grafting

of the thoracic aorta. Vascular 2004;12:179-85.

15. Fattori R, Napoli G, Lovato L, Grazia C, Piva T, Rocchi G, et al.Descending thoracic aortic diseases: stent-graft repair. Radiology 2003;229:176-83.

16. Hansen CJ, Bui H, Donayre CE, Aziz I, Kim B, Kopchok G, et al.Complications of endovascular repair of high-risk and emergent de-scending thoracic aortic aneurysms and dissections. J Vasc Surg 2004;40:228-34.

17. Heijmen RH, Deblier IG, Moll FL, Dossche KM, van den Berg JC,Overtoom TT, et al. Endovascular stent-grafting for descendingthoracic aortic aneurysms. Eur J Cardiothorac Surg 2002;21:5-9.

18. Lepore V, Lönn L, Delle M, Bugge M, Jeppsson A, Kjellman U, et al.Endograft therapy for diseases of the descending thoracic aorta: resultsin 43 high-risk patients. J Endovasc Ther 2002;9:829-37.

19. Leprince P, Pavie A. Les endoprothèses de l’aorte thoracique descen-dante. Lett Cardiol 2003:33-40.

20. Leurs LJ, Bell R, Degrieck Y, Thomas S, Hobo R, Lundbom J.Endovascular treatment of thoracic aortic diseases: combined experi-ence from the EUROSTAR and United Kingdom Thoracic Endograftregistries. J Vasc Surg 2004;40:670-80.

21. Lonn L, Delle M, Falkenberg M, Lepore V, Klingenstierna H, RadbergG, et al. Endovascular treatment of type B thoracic aortic dissections.J Card Surg 2003;18:539-44.

22. Makaroun MS, Dillavou ED, Kee ST, Sicard G, Chaikof E, Bavaria J,et al. Endovascular treatment of thoracic aortic aneurysms: results of thephase II multicenter trial of the GORE TAG thoracic endoprosthesis. JVasc Surg 2005;41:1-9.

23. Erbel R, Oelert H, Meyer J, Puth M, Mohr-Katoly S, Hausmann D,et al. Effect of medical and surgical therapy on aortic dissection evalu-ated by transesophageal echocardiography. Implications for prognosisand therapy. The European Cooperative Study Group on Echocardi-ography. Circulation 1993;87:1604-15.

24. Neuhauser B, Perkmann R, Greiner A, Steingruber I, Tauscher T,Jaschke W, et al. Mid-term results after endovascular repair of theatherosclerotic descending thoracic aortic aneurysm. Eur J Vasc Endo-vasc Surg 2004;28:146-53.

25. Orend KH, Scharrer-Pamler R, Kapfer X, Kotsis T, Gorich J, Sunder-Plassmann L. Endovascular treatment in diseases of the descendingthoracic aorta: 6-year results of a single center. J Vasc Surg 2003;37:91-9.

26. Ouriel K, Greenberg RK. Endovascular treatment of thoracic aorticaneurysms. J Card Surg 2003;18:455-63.

27. Scharrer-Pamler R, Kotsis T, Kapfer X, Görich J, Orend K-H, Sunder-Plassmann L. Complications after endovascular treatment of thoracicaortic aneurysms. J Endovasc Ther 2003;10:711-8.

28. Schoder M, Cartes-Zumelzu F, Grabenwöger M, Cejna M, FunovicsM, Krenn CG, et al. Elective endovascular stent-graft repair ofatherosclerotic thoracic aortic aneurysms: clinical results and mid-term follow-up. Am J Roentgenol 2003;180:709-15.

29. Taylor PR, Gaines PA, McGuinness CL, Cleveland TJ, Beard JD,Cooper G, et al. Thoracic aortic stent grafts–early experience from twocentres using commercially available devices. Eur J Vasc Endovasc Surg2001;22:70-6.

30. Totaro M, Mazzesi G, Marullo AGM, Neri E, Fanelli F, Miraldi F.Endoluminal stent grafting of the descending thoracic aorta. ItalHeart J 2002;3:366-9.

31. Tse LW, MacKenzie KS, Montreuil B, Obrand DI, Steinmetz OK. Theproximal landing zone in endovascular repair of the thoracic aorta. AnnVasc Surg 2004;18:178-85.

32. Herold U, Piotrowski J, Baumgart D, Eggebrecht H, Erbel R, Jakob H.Endoluminal stent-graft repair for acute and chronic type B aorticdissection and atherosclerotic aneurysm of the thoracic aorta: an inter-disciplinary task. Eur J Cardiothorac Surg 2002;22:891-7.

33. White GH, Yu W, May J, Chaufour X, Stephen MS. Endoleak as acomplication of endoluminal grafting of abdominal aortic aneurysms:classification, incidence, diagnosis, and management. J Endovasc Surg1997;4:152-68.

34. Bavaria JE, Moser W, McBride A, Gleason T, Pochettino A, CarpenterJP, et al. Endovascular repair of thoracic aortic aneurysm. Results fromphase I and II study with Medtronic “talent” stent-graft [poster].

Philadelphia: University of Pennsylvania Health Services; 2005.

JOURNAL OF VASCULAR SURGERYVolume 47, Number 5 Corbillon et al 1107

35. Glower DD, Speier RH, White WD, Smith LR, Rankin JS, Wolfe WG.Management and long-term outcome of aortic dissection. Ann Surg1991;214:31-41.

36. Hagan PG, Nienaber CA, Isselbacher EM, Bruckman D, Karavite DJ,Russman PL, et al. The International Registry of Acute Aortic Dissection(IRAD): new insights into an old disease. JAMA 2000;283:897-903.

37. Herold U, Piotrowski J, Baumgart D, Eggebrecht H, Erbel R, Jakob H.Endoluminal stent graft repair for acute and chronic type B aorticdissection and atherosclerotic aneurysm of the thoracic aorta: an inter-

disciplinary task. Eur J Cardiothorac Surg 2002;22:891-7.

38. Doss M, Balzer J, Martens S, Wood JP,Wimmer-Greinecker G, FieguthHG, et al. Surgical versus endovascular treatment ofacute thoracicaortic rupture: a single-center experience. Ann Thorac Surg 2003;76:1465-70.

39. Bortone AS, De Cillis E, D’Agostino D, Tupputi Schinosa LDL.Endovascular treatment of thoracic aortic disease: four years of experi-ence. Circulation 2004;110(11 suppl):II262-7.

Submitted Jul 18, 2007; accepted Nov 10, 2007.