Abstract
Objective
We reviewed data pertaining to fenestrated endograft technique and chimney stent repair of complex aortic aneurysm for comparative analysis of the outcomes.
Methods
A comprehensive search of relevant databases was conducted to identify articles in English, related to the treatment of complex aortic aneurysm with fenestrated endovascular aneurysm repair and chimney stent repair, published until January 2015.
Results
A total of 42 relevant studies and 2264 patients with aortic aneurysm undergoing fenestrated endovascular aneurysm repair and chimney stent repair were included in our review. A total of 4413 vessels were involved in these processes. The cumulative 30-day mortality was 2.4% and 3.2% (p = 0.459). The follow-up aneurysm-related mortality was 1.4% and 3.2% (p = 0.018), and target organ dysfunction was 5.0% and 4.0% in fenestrated endovascular aneurysm repair and chimney stent repair, respectively (p = 0.27). A total of 156 vessels showed restenosis or occlusion after primary intervention (3.6% and 3.4% in fenestrated endovascular aneurysm repair and chimney stent repair, respectively, p = 0.792). The cumulative type I endoleak was 2.0% (38/1884) after fenestrated endovascular aneurysm repair compared with 3.4% (13/380) after chimney stent repair (p = 0.092), and the type II endoleak was 5.4% (102/1884) and 5.3% (20/380), respectively (p = 0.905). Approximately, 1.1% and 1.6% increase in aneurysm was observed following fenestrated endovascular aneurysm repair and chimney stent repair, respectively (p = 0.437). The re-intervention frequency was 205 and 19 cases after fenestrated endovascular aneurysm repair and chimney stent repair, respectively (11.7%, 5.6%, p = 0.001).
Conclusions
Fenestrated endovascular aneurysm repair and chimney stent repair are safe and effective in treating patients with complex aortic aneurysm. A higher aneurysm-related mortality was observed in chimney stent repair while fenestrated endovascular aneurysm repair was associated with a higher re-intervention rate.
Introduction
Endovascular aneurysm repair (EVAR) was first published in 1994 by Parodi 1 and has since been recognized as a standard method for treatment of infra-renal abdominal aortic aneurysms (AAA). Complex aortic aneurysm (CAA) is a type of aortic aneurysm affecting visceral arteries, including the renal, the celiac trunk and the superior and inferior mesenteric arteries. Newer techniques such as fenestrated EVAR (F-EVAR) and chimney stent repair (Ch-EVAR) raise the possibility of successful treatment of juxta-, para- and supra-renal aneurysms.2,3 We reviewed the literature reporting F-EVAR and Ch-EVAR of CAA for comparative analysis of the outcomes.
Methods
Eligibility criteria
A feasible protocol was established for the meta-analysis in advance, and the review was developed, accordingly. We formulated the inclusion criteria on the basis of participants, interventions, comparisons, outcomes, study design and time principle 4 : (1) trials with patients afflicted with CAA involve renal, celiac trunk, superior or inferior mesenteric arteries; (2) interventions included F-EVAR or Ch-EVAR; (3) trials with outcomes of technical success, death, patency, complications and re-intervention and (4) published in English before January 2015.
Trials containing one of the following conditions were excluded: (1) concomitant aneurysms of other arteries; (2) non-endovascular intervention; (3) less than five patients or average follow-up less than six months; (4) inability to extract intact data and (5) trials involving the brachiocephalic artery.
Search strategy and study selection
We performed a systematic search according to the PRISMA statement, 5 including PubMed, Embase, CENTRAL, Ovid and other pertinent databases. We also searched the relevant systematic reviews published previously as supplements. We used aortic aneurysm and chimney or aortic aneurysm and fenestration as keywords, searching in English and Chinese, until January 2015. A single author screened the studies by title and abstract for inclusion. Identified articles were assessed independently by another author to confirm their eligibility. Studies eligible for full-text review were reviewed by two independent reviewers for inclusion in the analysis.
Data abstract
Data extraction form was designed according to the Cochrane handbook, 6 and tested in one trial randomly, for improvement. Two authors extracted data independently, revised the data entry using EpiData software (ver. 3.1), checked for consistency and negotiated with the third party in case of ambiguity. The major data extracted included: (1) article information: first author, study types, year of publication, country, follow-up time, capacity and vessels involved; (2) participant profile: gender, age, comorbidities and classification of vessels and (3) outcomes: technical success, 30-day mortality, aneurysm-related mortality, target organ dysfunction, vascular restenosis or occlusion, endoleak, aneurysm growth and re-intervention.
Quality assessment and statistical analysis
Quality of the study was assessed with the Cochrane collaboration’s tool for assessing risk of bias. 7 Seven aspects including the random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting and other bias were assessed and documented by two reviewers, independently. The data were negotiated with the third party in case of ambiguity. RevMan software (ver. 5.3) was used for risk of bias summary.
In the meta-analysis, we used the RevMan software (ver. 5.3) to complete heterogeneity test in pair-wise comparisons, and I2 was selected as the quantitative test index. 8 When I2 was greater than 50%, a random effects model 9 was used for data merging; otherwise, a fixed effect model was used. The odds risk ratios (OR) and 95% confidence intervals (CI) represented the effect magnitude. In the systematic review, characteristics and outcomes were pooled, respectively, and compared using the Pearson’s or Fisher’s exact test. Statistical significance was defined at the level of p < 0.05; all P values were two sided. Calculations were performed using SPSS software (version 18.0).
Results
Study selection
A total of 1474 studies were searched with matching keywords, duplicates were removed, records screened and full-text articles assessed. Finally, 42 studies matched the criteria and included in the review (Figure 1). The earliest study was published in 2001 and the latest in 2014.
Study selecting flow diagram.
Study data
A total of 25 studies of F-EAVR, 13 studies of Ch-EVAR and 4 studies containing both were included in our review. Ten of these studies represent prospective trials (eight in F-EVAR studies, two in Ch-EVAR) and the rest were retrospective trials. Thirty-three single arm studies (F-EVAR 21, Ch-EVAR 12) with 2264 patients (F-EVAR 1884, Ch-EVAR 380) and 4413 vessels (F-EVAR 1884, Ch-EVAR 380) were included. The mean follow-up time was 18.07 months in F-EVAR and 16.12 months in Ch-EVAR. A meta-analysis of four trials comparing F-EVAR with Ch-EVAR was performed. The bias of these trials is shown in Figure 2. The basic data of the 42 studies are shown in Tables 1 and 2. Most of the trials included patients at high risk for open repair and contraindicated for conventional endovascular repair.
Risk of bias summary. Overview of the C-EVAR studies. Overview of the C-EVAR studies.
Characteristics
F-EVAR
The mean age was 73.56 years, including 707 males (37.5%, 1884). Of the 29 studies, 27 provided data pertaining to comorbidities including 52.17% (946/1813) with coronary artery disease (CAD), 15.89% (288/1813) with diabetes mellitus (DM), 69.50% (1260/1813) with hypertension, 18.59% (337/1813) with dyslipidemia, 24.16% (438/1813) with renal insufficiency (RI) and 31.50% (571/1813) with chronic obstructive pulmonary disease (COPD). Twenty-three studies provided details of vessels treated: Of the total 3583 vessels, 2248 renal arteries were treated, accounting for 62.7%. The celiac trunk artery (CTA) accounted for the proportion of 7.3%, and the superior mesenteric artery (SMA) accounted for 26.2% of the total vessels treated (3583), while no inferior mesenteric artery was treated in F-EVAR.
Ch-EVAR
The mean age was 74.34 years, and 80.3% (305/380) were males. Only one study failed to provide the data pertaining to comorbidities. Among the 374 patients included in this study, 64.97% (243) afflicted with CAD, 8.29% (31) with DM, 52.40% (196) with hypertension, 35.29% (132) with dyslipidemia, 21.12% (79) with RI and 39.30% (147) with COPD. All studies contained intact data of the treated vessel. The renal artery accounted for 81.6% (511/626), the CTA accounted for 4.8% (30/626) and the SMA for 13.4% (84/626). A single inferior mesenteric artery was involved in Ch-EVAR.
Characteristics of patients in each groups.
F-EVAR: fenestrated EVAR; Ch-EVAR: chimney stent repair; CAD: coronary artery disease; DM: diabetes mellitus; RI: renal insufficiency; COPD: chronic obstructive pulmonary disease; RA: renal artery; CTA: celiac trunk artery; SMA: superior mesenteric artery; IMA: inferior mesenteric artery.
Outcomes
F-EVAR
The technical success rate was 98.2% for target vessels, with 68 vessels resulting in failure. The cumulative 30-day mortality was 2.4%, with 47 death during this period. Twenty-nine deaths included seven multi-organ failures, nine cases of myocardial infarction, six cases of intestinal ischemia, three cases of septic shock, two cases of arteriorrhexis (renal and iliac artery) and two major strokes. During the follow-up, 27 patients (1.4%) died from aneurysm, including four from ruptured aneurysm, eight from intestinal ischemia, one from aortoduodenal fistula, one from iliac in stent thrombosis, one from infection of the prosthesis and one from mycotic aneurysm. There were 38 cases (2.0%) of type I and 102 cases (5.4%) of type II endoleak. Five percent (185/3658) of patients showed target organ dysfunction including kidney, spleen and intestine; 3.6% (135/3787) of vascular restenosis or occlusion and 1.1% (20/1811) of patients with aneurysms showed increased vessel diameter. With these complications, 205 (11.7%) patients underwent secondary intervention. Twenty studies reported common complications including 63 cardiac, 16 cerebral and 33 respiratory.
Ch-EVAR
Outcomes of patients in each groups.
Meta-analysis
Four trials18,27,33,38 were included in the meta-analysis. The outcomes of technical success, 30-day mortality, types I and II endoleaks, target organ dysfunction, vascular restenosis/occlusion and re-intervention were analyzed (Figure 3). The results showed no significant differences between F-EVAR and Ch-EVAR (P > 0.05). The heterogeneity test indicated no discordance between the trials except for target organ dysfunction. The results support F-EVAR in 30-day mortality and type I endoleak, with an OR of 0.94 [0.25, 3.55] and 0.62 [0.10, 3.93], respectively. The results favored Ch-EVAR in technical success, type II endoleak, target organ dysfunction, vascular restenosis/occlusion and re-intervention, with an OR of 0.39 [0.07, 2.26], 1.25 [0.48, 3.28], 2.21 [0.11, 44.40], 1.02 [0.29, 3.52] and 1.51 [0.74, 3.10], respectively.
Forest plot of F-EVAR vs. Ch-EVAR. (a) Forest plot for technical success; (b) forest plot for 30-day mortality; (c) forest plot for type I endoleak; (d) forest plot for type II endoleak; (e) forest plot for target organ function impaired; (f) forest plot for vessels restenosis/occluded and (g) forest plot for re-intervention.
Discussion
Standard EVAR represented an alternative to AAA for years, 52 with proven success. 53 However, it was only appropriate for infra-renal AAA with a long neck. The option for para-renal, thoracic or juxta-renal AAA was unclear. For patients with complex AAA contraindicated for open surgery, F-EVAR and Ch-EVAR offer fewer invasive options. 54 F-EVAR uses fenestrations (holes) in the graft to access visceral arteries to enable extension of the graft sealing zone to an adequate, more proximal, landing zone incorporating the visceral vessels. Fenestrations are sealed with a covered stent into the target vessels. Newer techniques include the use of preloaded cannulating wires, double reducing ties and an enlarged proximal scallop. Ch-EVAR uses a covered stent parallel to the proximal graft to preserve the flow to visceral vessels without specially designed holes in the graft, which are stented to obtain an adequate seal. Studies have already compared the outcome of these two types of technique. Our review updated the comparative CAA data.
No randomized controlled trials (RCTs) compared F-EVAR and Ch-EVAR. We included 15 studies published in 2014, and 82.6% articles published in the last five years. Several prospective studies were reviewed, and population-based studies conducted. The characteristics of two groups differ from each other, with the mean vessels number treated by F-EVAR and Ch-EVAR at 1.9 and 1.6. The F-EVAR group treated more vessels per patient, and more number of patients accepted CTA and SMA fenestration (7.3% vs. 4.8% and 26.2% vs. 13.4%). The results suggest that the condition of aneurysm in F-EVAR group is more complex than in Ch-EVAR group, which may increase the difficulty in stent placement and impact the technical success rate.
The Ch-EVAR showed a better technical success rate than F-EVAR (99.5%, 98.2%, p = 0.028), with the difference probably related to technological challenges, operator skill, condition of the aneurysm and the target vessel. Based on the success rate, the outcome of re-intervention rate was 11.7% and 5.6%, respectively (p = 0.001), with F-EVAR associated with a higher re-intervention rate due to the more complex condition of aneurysms and difficulties with the fenestration techniques. Further, Ch-EVAR showed a higher aneurysm-related mortality than F-EVAR (3.2%, 1.4%, p = 0.018), with a similar mortality reported in previous single systematic reviews.17,55,56 Ch-EVAR is indicated for emergent intervention involving ruptured AAA, due to its higher technical success and lower re-intervention rate.
In F-EVAR, customized fenestrated grafts were inappropriate due to the long wait times. Home-made grafts were characterized by their off-the-shelf properties. However, no trials or analyses compared the advantages and disadvantages. In the F-EVAR trials included, we found that 26 out of 29 trials used the customized graft and three trials used off-the-shelf graft. Most of the trials used company-manufactured grafts, such as Zenith Cook device, and few home-made grafts were reported. For comparison of these two grafts, more trials with larger sample capacity were needed to highlight any statistically significant differences in outcomes.
The results of meta-analysis support Ch-EVAR in terms of technical success and re-intervention rates, consistent with the systematic review, although without any significant differences.
The limitations: (1) the outcome acquired under inconsistent base line; (2) no subgroup analysis; (3) failed to estimate the increased of bias even though the techniques used in the trials were not compared or analyzed and (4) higher levels of evidence such as RCTs are needed to corroborate our findings.
Conclusions
F-EVAR and Ch-EVAR are safe and effective in treating patients with CAA. No significant differences were seen between F-EVAR and Ch-EVAR in 30-day mortality, target organ dysfunction, target vessel occlusion or development of aneurysms. A higher aneurysm-related mortality in Ch-EVAR was observed. The F-EVAR was associated with a higher re-intervention rate than Ch-EVAR. Further evidence based on controlled clinical trials with a longer follow-up is needed to analyze the different outcomes.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Our review is supported by National Nature Science Foundation of China (81070256, 81100226).
