Description of the condition
Aneurysms are usually defined as a permanent localised dilation of an artery where the diameter is at least 50% greater than the expected normal size of the artery. The majority of aortic aneurysms are localised in the abdominal aorta, but the thoracic part of the aorta is not immune to vascular degeneration. Thoracic aortic aneurysms (TAAs) are usually associated with atherosclerosis. Due to the uncommon nature of the disease the data on natural history, prevalence, and mortality are limited and sometimes differ. A study published in 1982 reported an incidence of TAAs of 5.9 per 100,000 in a stable community over 30 years of observation (Bickerstaff 1982), while another survey documented an incidence of 10.4 per 100,000 from 1980 to 1994 (Clouse 1998). The two-fold increase in the incidence of TAAs may be attributed to the advent of more sophisticated diagnostic instruments, such as computed tomography (CT), or the result of an aging population. However, according to some reports, TAAs occur mainly in older individuals and, contrary to abdominal aortic aneurysms which affect mainly males, the incidence of TAAs is evenly distributed among both sexes (Bickerstaff 1982; Clouse 1998).
Given the silent nature of TAAs it is difficult to identify signs and symptoms that warn of the presence of the disease. Most of the time such aneurysms are diagnosed by chance. When the diagnosis is made, identification of factors that might help us to recognise patients who are in most need of treatment, or who are at high risk of rupture, is necessary. While age, the presence of hypertension, chronic obstructive pulmonary disease, and renal failure have been identified as risk factors associated with aneurysmal disease and its rupture, the diameter of the aneurysm is the most important predictor to be considered when evaluating patients with TAAs (Griepp 1999). A number of retrospective studies have documented that the size of the aortic aneurysm is the major factor predicting rupture. Although small aneurysms may rupture, risk of rupture and mortality related to rupture become significant when the diameter of the affected aorta exceeds 5 cm (Cambria 1995; Coady 1997; Lobato 1998; Perko 1995). The five-year mortality rate of aneurysms exceeding 6 cm varies from 38% to 64% (Coady 1997; Joyce 1964). In addition, the rate of increase in the size of aortic aneurysms is part of the natural course of the disease and is an another indicator that predicts rupture. Indeed, TAAs may have an expansion rate of between 0.2 cm and 0.4 cm per year. Growth rate is directly related to the initial size of the diseased aorta (Cambria 1995; Coady 1997; Hirose 1993). Some authors estimate that each 1 cm increase in the size of the thoracic aneurysm is translated into a further increase in the probability of rupture by a factor of 1.9, compared with a factor of 1.5 observed for abdominal aneurysms (Juvonen 1997).
Data from several retrospective surveys indicate that for patients who did not receive surgery at the time of diagnosis, rupture was the most common cause of death, with death rates ranging from 42% to 74% (Bickerstaff 1982; Crawford 1986; Perko 1995). The five-year survival rate was 39% for patients with TAAs and 23% for those with thoracoabdominal aneurysm (Perko 1995). When rupture occurs, patients mostly die within six hours and, although 46% of patients arrive alive at the hospital, the overall mortality rate of ruptured TAAs reaches 97% (Johansson 1995).
Description of the intervention
The high mortality of aneurysmal disease has lead many to suggest surgical repair of the dilated aorta, even though the anatomical morphology of the thoracic aorta together with the characteristics of patients affected by TAAs pose a considerable challenge to surgeons. Open surgical repair using prosthetic graft interposition is the conventional treatment for TAAs mainly because of its feasibility and effectiveness in excluding the degenerated aorta from the systemic circulation. Open surgical repair of TAAs is associated with significant perioperative complications including 30-day mortality and paraplegia, with rates of 4.8% and 4.6% respectively. Stroke and renal failure are also important complications to be considered (Coselli 2000; Deeb 1995; LeMaire 2003). Despite noteworthy improvements in surgical procedures with extracorporeal circulation for peripheral organ preservation, the development of different techniques for spinal cord protection, together with the refinement of prosthetic grafts for aortic repair, morbidity and mortality rates remain high (Hamerlijnck 1989; Huynh 2002; Thurnher 2002).
How the intervention might work
In view of these complexities, and the fact that many high risk patients were regularly excluded from open surgical repair, a less invasive and potentially safer technique became imperative. Thus an alternative for open surgical repair for aortic aneurysm has become the performance of endovascular repair using stent grafts. The aim of this technique is to reach the target site by performing an access through a remote vessel to deliver the stent, secure endograft fixation, and allow the formation of a haemostatic seal between the graft and the vessel wall. The endovascular repair (EVAR) technique has been successfully used in the treatment of abdominal aortic aneurysms and with respect to conventional open surgical repair has been shown to reduce early post-operative complications and death (Greenhalgh 2004; Prinssen 2004). The success of the stent-graft placement performed in abdominal aortic aneurysms prompted experts to translate the technique to thoracic aneurysm repair. Dake was the first to report the use of stent graft in 13 patients with TAAs with positive outcomes over 11 months of observation (Dake 1994).
Identification of the best vascular access through which to deliver the stent is important, as well as performing imaging studies that help assess the best site for the insertion of the stent-graft delivery system. Non-surgical repair may be an attractive therapeutic option for patients with TAAs. According to some authors the morbidity and mortality associated with EVAR of thoracic aneurysm (TEVAR) appear to be less than with conventional treatment. In addition, while mid-term survival seems satisfactory in patients receiving endovascular treatment the long-term outcomes for high risk patients are disappointing, probably due to co-existing disease (Demers 2004; Ehrlich 1998; Greenberg 2000; Marcheix 2006; Mitchell 1999).
Why it is important to do this review
Thus, convincing evidence that TEVAR is better than open surgical treatment for TAAs is needed. Potential advantages of TEVAR include reduction in the rate of paraplegia; reduced time under anaesthesia; avoidance of arterial cross-clamping, renal failure, and cardiovascular complications; reduced length of hospital stay, and particularly a reduction in the length of stay in an intensive care unit.
This review aimed to assess the efficacy of TEVAR versus conventional open surgery in patients with thoracic aortic aneurysms.
Criteria for considering studies for this review
Types of studies
Randomised controlled trials (RCTs) evaluating TEVAR compared with conventional open surgical repair.
Types of participants
All asymptomatic patients in whom a thoracic aortic aneurysm was diagnosed by computed tomography (CT), magnetic resonance angiography (MRA), or conventional angiography. Only studies reporting the size of TAAs were eligible for consideration.
Mycotic aneurysms (aneurysms caused by an infection of the arterial wall or when a pre-existing aneurysm has become secondarily infected), acute or chronic dissections, and all patients with connective tissue disorders were excluded.
Types of interventions
All types of endovascular devices were considered when compared with conventional open surgical treatment.
Types of outcome measures
The following outcome measures were considered:
- short-term mortality rates (30 days, or in hospital, i.e. procedure related);
- aneurysm exclusion (no flow in the aneurysmal sac or extravasations (discharge of blood from a vessel to the tissues) beyond the sac) on follow-up imaging 30 days after the procedure;
- major complications (haemorrhage, open conversion, myocardial infarction, stroke, renal failure, respiratory failure, spinal chord ischaemia, bowel ischaemia, lower limb ischaemia);
- minor complications;
- long-term complications and mortality (re-intervention rates for problems related to the TAA or its treatment were sought where possible, as was cause of death with or without re-intervention, i.e. device related);
- quality of life (based on validated questionnaires);
- economic analysis (based on analysis of costs).
Search methods for identification of studies
For this update the Cochrane Peripheral Vascular Diseases Group Trials Search Co-ordinator (TSC) searched the Specialised Register (last searched March 2013) and the Cochrane Central Register of Controlled Trials (CENTRAL) 2013, Issue 2, part of The Cochrane Library, (www.thecochranelibrary.com). See Appendix 1 for details of the search strategy used to search CENTRAL. The Specialised Register is maintained by the TSC and is constructed from weekly electronic searches of MEDLINE, EMBASE, CINAHL, AMED, and through handsearching relevant journals. The full list of the databases, journals and conference proceedings which have been searched, as well as the search strategies used are described in the Specialised Register section of the Cochrane Peripheral Vascular Diseases Group module in The Cochrane Library (www.thecochranelibrary.com).
For the previous version of the review the authors searched MEDLINE (1966 onwards) and EMBASE (1980 onwards) using the search strategy developed for the Cochrane PVD Group. The MEDLINE strategy via Ovid (see Appendix 2) was modified when applied to other databases.
The following electronic databases were searched:
- Best Evidence;
- Biological Abstracts;
- HMIC (Health Information Management Consortium - comprising DH-Data, the King's Fund Database, and Helmis);
- NHS DARE (Database of Assessment of Reviews of Effects);
- NHS EED (Economic Evaluations Database);
- NHS HTA (Health Technology Assessment);
- Science Citation Index;
Searching other resources
The reference lists of all relevant studies found were screened. Authors of relevant papers, pharmaceutical and stent device manufacturers were contacted to identify further published and unpublished studies. We also handsearched relevant surgical and radiology journals for the 2008 version of the review.
Data collection and analysis
Two authors (IA and RC) assessed all trials identified from the described literature search. Disagreements were resolved through discussion with the review team and the agreed arbitrator (AM).
Given that there were no included studies in this review, the following methods will be used for future updating of the review.
The above mentioned authors will independently extract the following information for each included trial: outcome measures, blinding of outcome evaluators, and balance of prognostic factors (age, sex, size of the aneurysm). If such information is not available from the published paper, we will contact the authors of trials. We will evaluate concealment of allocation, blinding in outcome evaluation, intention-to-treat analysis, and balance of baseline prognostic factors to grade them as present, absent, or unclear and develop an overall assessment of validity. For this purpose three categories will be used: A (low risk of bias, all of the criteria present); B (moderate risk of bias, one or more criteria unclear); C (high risk of bias, one or more criteria absent).
We will independently extract data for outcome assessment according to the measures defined in the previous section and using proformas designed by the Cochrane PVD Group. We will perform intention-to-treat analysis extracting the number of patients originally allocated to each treatment group, irrespective of compliance. If numbers extracted by the two authors are different, differences will be resolved by a third author (AM). We will contact trial investigators if additional information is required.
Agreement between investigators for study selection will be assessed using a weighted kappa statistic. Statistical analysis will be performed according to the statistical guidelines provided by the Cochrane PVD Group. Relative risk will be used as the measure of effect for each dichotomous outcome. A summary statistic will be calculated for each dichotomous outcome using both a fixed-effect model and a random-effects model provided that sufficient data are gathered. Heterogeneity will be explored using threshold analysis based on quality assessment and characteristics of the studies. A Chi
If outcomes were measured with continuous scales, we will analyse data on treatment effects with the weighted mean difference. Where different trials use different scales, the results will be standardised and then combined (standardised mean difference).
We will perform subgroup analysis, if possible and as necessary, and this may consider the size of the aneurysm, its anatomical localisation (ascending, descending, or thoracoabdominal) and condition of patients (high risk versus low risk).
Description of studies
Results of the search
No studies were identified that met any of the inclusion criteria. A number of case series, cohort studies, and controlled studies with historic controls were found. These were excluded because of lack of randomisation (see Characteristics of excluded studies).
Risk of bias in included studies
It was not possible to review methodological quality in the absence of studies eligible for inclusion in the review.
Effects of interventions
No published or unpublished randomised controlled trials were found comparing TEVAR with conventional open surgical repair for the treatment of TAAs.
Thoracic surgery for the treatment of aneurysms is performed in a complex setting and entails general anaesthesia, thoracotomy with extensive surgical resection, left lung collapse, and aortic cross-clamping. As a result, important postoperative complications such as haemorrhage, spinal cord injury, stroke, and renal insufficiency must be considered. Distal aortic perfusion, cerebrospinal drainage, and intercostal artery implantation are some of the advances made to avoid complications associated with thoracic surgery but morbidity remains high. Moreover, factors such as the age of the patient and the presence of serious co-morbidities such as coronary heart disease, chronic lung disease, diabetes, and hypertension may exclude a number of patients from receiving surgical treatment.
The advent of endovascular stent interposition, first described by Parodi (Parodi 1991) for abdominal aneurysm and later by Dake (Dake 1994) for the thoracic tract, has generated great interest among thoracic surgeons. This is mainly because an endovascular stent is considered a minimally invasive device and therefore, a good alternative for patients affected by TAAs who are not suitable for open surgical treatment.
This review documents that there are no published or pending randomised controlled trials that compare TEVAR with open surgical treatment for patients with TAAs, assessing early and late mortality and major complications. Most of the current information relating to endovascular intervention comes from reports of case series, observational studies, or registry data with differing presentations. However, at least two controlled clinical trials can be mentioned. The first study, published in 2002, compared 19 patients who received endovascular stents with an historic cohort of 10 patients who received open surgical repair. Endovascular deployment was successful in all but one patient and early outcomes such as mean length of intensive care unit stay and hospital stay were better in the TEVAR group. The study, however, was not randomised and follow up was limited (Najibi 2002).
The second trial, published in 2005, started as a prospective non-randomised multicentre phase II study. Promising results in terms of technical success and positive early outcomes prompted the FDA to approve the Gore TAG thoracic endoprosthesis (Makaroun 2005). At a later stage the study became a controlled trial in which the data of the 142 patients were compared with 94 controls (Bavaria 2007). According to the final analysis of the trial, mortality at 30 days was better in the endograft group, as well as other outcomes such as paraplegia and the length of hospital stay. However, the two-year survival rate was similar in the two groups. The significant limitations of the trial were that it was not a randomised trial; at least half of the control group was historical, which may augment further the selection bias; and symptomatic aneurysms were significantly higher in the open surgical cohort (38% versus 21%). In addition, data on aortic characteristics were unavailable in many of the control patients and the follow up was limited to two years, within which time 17% of the patients had exclusion of their aneurysm of a size greater than 0.5 cm, and at least 17% of the patients were lost to follow up (Bavaria 2007).
Despite the reported early advantages, stent devices for thoracic aneurysm have late complications that are uncommon with open surgery. Such complications include endoleaks, graft migration, stent fractures, and aneurysm-related death. Type I or attachment site endoleak is the most frequent complication and its incidence may reach 30% of cases (Neuhauser 2005). Usually Type I endoleaks require additional stent-graft deployment for sealing or spot coiling, but sometimes such endoleaks may close spontaneously due to the formation of spontaneous thrombosis of the aortic false lumen and not require any further treatment. While resolution may occur within one week to eight months, endoleak may progress with dilatation of the false lumen necessitating close follow up, supplementary procedures, or even open surgical treatment.
The available controlled clinical trials on TEVAR efficacy had promising results in terms of early mortality and reduction in postoperative complications but they failed to give generalisable conclusions, not only because of their internal low methodological quality but also in that high risk patients were not the target population. This reduces the original expectation that stent grafts could be an alternative for patients excluded from open surgical treatment because of their high risk status. The recent report of the Society of Thoracic Surgeons in fact states that the indication for stent grafting of a thoracic aortic aneurysm should be based on a predicted operative risk that is clearly lower than the risk of either conventional open repair or optimal medical management (Svensson 2008). To this it must be added that patients who receive TEVAR will be in need of frequent surveillance CT scans after the intervention is performed and at a later stage they will probably need surgical re-intervention.
Although graft stenting for TAAs is technically feasible and apparently reduces the number of severity of early outcomes, high quality randomised controlled trials assessing all clinically relevant outcomes, including early and late mortality, open conversion, aneurysm exclusion, endoleaks, are needed.
Implications for practice
Technically, endovascular repair of thoracic aneurysms could be a good alternative to open surgical repair. However, its benefit cannot be established as there are no published randomised controlled trials. The available information comes from non-randomised studies, which show benefit in terms of early mortality and complications such as paraplegia. Although such evidence may suggest that endovascular repair can be appropriate in selected patients, high quality studies are needed to produce generalisable conclusions.
Implications for research
High quality randomised controlled trials evaluating thoracic endovascular aneurysm repair (TEVAR) for thoracic aortic aneurysms are needed. These prospective trials should have adequate follow up, enough to evaluate the durability of endovascular treatment in terms of endoleak rate, re-intervention rate, open-conversion rate, and rupture-free survival. In addition, clinically relevant outcomes including early and late mortality, major complications, and hospital and intensive care unit stay must be considered. However, it would probably be extremely difficult to perform a randomised trial given the current stage of surgical endovascular practice.
Data and analyses
This review has no analyses.
Appendix 1. CENTRAL search strategy
Appendix 2. 2008 MEDLINE search strategy via Ovid
1. RANDOMIZED CONTROLLED TRIAL.pt.
2. CONTROLLED CLINICAL TRIAL.pt.
3. RANDOMIZED CONTROLLED TRIALS/
4. RANDOM ALLOCATION/
5. DOUBLE BLIND METHOD/
6. SINGLE-BLIND METHOD/
8. (ANIMALS not HUMANS).sh.
9. 7 not 8
10. CLINICAL TRIAL.pt.
11. exp CLINICAL TRIALS/
12. (clin$ adj25 trial$).ti,ab.
13. ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab.
17. RESEARCH DESIGN/
19. 18 not 8
20. 19 not 9
21. 9 or 20
22. AORTIC ANEURYSM,THORACIC/
23. exp AORTIC ANEURYSM/
24. AORTIC RUPTURE/
26. AORTIC ANEURYSM,ABDOMINAL/
27. (abdominal adj5 dilation).ti,ab.
28. (thoracic adj5 dilation).ti,ab.
29. (aort$ adj5 aneurysm).ti,ab.
30. (abdominal adj5 aneurysm).ti,ab.
31. (thoracic adj5 aneurysm).ti,ab.
32. (thoracoabdominal adj5 aneurysm).ti,ab.
33. ANEURYSM,FALSE/ and AORTA,ABDOMINAL/
34. ANEURYSM,FALSE/ and AORTA,THORACIC/
35. SURGERY/ and AORTA, ABDOMINAL/
36. SURGERY/ and AORTA,THORACIC/
38. 37 and 21
Last assessed as up-to-date: 13 March 2013.
Protocol first published: Issue 4, 2007
Review first published: Issue 1, 2009
Contributions of authors
Dr Iosief Abraha wrote the majority of the text of this review, which was checked and discussed with the other authors who approved the final version. Dr Iosief Abraha and Dr Roberto Cirocchi independently selected and assessed trials for possible inclusion. Dr Alessandro Montedori resolved any differences in the selection of trials and helped in writing the 'Description of studies' and wrote the 'Plain language summary'. Dr Carlo Romagnoli prepared the abstract.
Declarations of interest
Sources of support
- No sources of support supplied
- Chief Scientist Office, Scottish Government Health Directorates, The Scottish Government, UK.The PVD Group editorial base is supported by the Chief Scientist Office.
Medical Subject Headings (MeSH)
MeSH check words