Changes in False Lumen After Transluminal Stent-Graft Placement in Aortic Dissections
Six Years’ Experience
Background— Transluminal stent-graft placements (TSGPs) are a new, less invasive procedure now recognized as the choice for aortic disease repair. Treatment of aortic dissections with TSGPs has resulted in good early results, but the long-term results and changes in the false lumen have not been elucidated in detail.
Methods and Results— TSGPs were performed in 49 patients with primary tears in their descending aortas, and the follow-up period ranged from 4 months to 6 years. The patients were divided into 32 acute-onset and 17 chronic dissections; of the acute-onset cases, there were 15 Stanford type A retrograde dissections. Periodic enhanced spiral CT was conducted after TSGP. The false lumen in the ascending aorta in 14 (93%) of the Stanford type A cases was obliterated completely within 3 months. The CT study was continued for >2 years for 17 acute-onset dissection and 11 chronic dissection patients. The average false lumen diameters of the proximal, middle, and distal descending aorta before treatment were 15.9, 16.2, and 15.6 mm in the acute-onset dissection group and 28.1, 25.2, and 21.0 mm in the chronic dissection group, respectively. The false lumen diameters 2 years after treatment were 3.0, 3.7, and 3.1 mm in the acute-onset dissection group and 10.6, 10.5, and 11.9 mm in the chronic dissection group, respectively. Two years after TSGPs, the false lumen of the thoracic aorta totally disappeared in 76% of the acute-onset dissection group and 36% of the chronic dissection group. No cases showed rupture after TSGP.
Conclusions— Complete obliteration of the false lumen is more likely in acute-onset cases than in chronic cases.
Received June 3, 2004; revision received February 2, 2005; accepted March 2, 2005.
Lethal complications such as rupture, impending rupture, end-organ ischemia, and progressive aneurysmal changes often accompany aortic dissections. There are 2 subtypes of aortic dissection involving primary tears in the descending aorta: type B1–3 and type A retrograde dissections.4,5 The risk of open surgery in treating these dissections is high, and the preoperative state of patients, other than their aortas, is often poor. Associated mortality and morbidity are therefore high, although improvements have been observed.
Transluminal stent-graft placements (TSGPs) are a comparatively new procedure for sealing primary tears without open surgery. The authors have been performing TSGPs on aortic dissection patients with primary tears in descending aortas for the last 6 years.6 TSGP is less invasive than open surgery and has shown good early results with low mortality,7,8 but the long-term results remain unknown. Continuous surveillance of the entire aorta after TSGPs for aortic dissections is important to monitor complications. This article presents the results of an intermediate-term follow-up study of the false lumen after TSGP and evaluates the feasibility and limitations of TSGPs.
Between July 1997 and June 2003, 49 consecutive patients with primary tears in their descending aorta underwent TSGPs. There were 32 acute-onset and 17 chronic dissections; of the acute-onset cases, 15 were Stanford type A retrograde dissections. No Stanford type A cases were in the chronic group. In the same period, 173 aortic dissections were managed at our institution, and 44 acute type A dissections containing 4 cases of retrograde subtype and 20 chronic dissections, totally 64 dissections, underwent surgery in this period. In 49 TSGP cases, hypertension was coexistent, with 26 cases of 32 acute-onset cases and 11 cases of 17 chronic cases, but there was no Marfan patient and no case with bicuspid aortic valve. Only 3 patients in the chronic group had undergone ascending aortic replacement for acute type A dissection. The inclusion and exclusion criteria for TSGP were previously described in detail.7 Briefly, hemodynamically unstable patients with acute-onset dissections were excluded. All other patients underwent digital subtraction angiography to evaluate the distance of the landing zone, existence of multiple entries, blood supply to the branched vessels, existence of static obstructions, and access route feasibility. Spiral CT was performed in all patients to measure the diameters of the aorta and false lumen where the dissections had occurred and to evaluate the diameter and quality of the aorta in the landing zone. If the digital subtraction angiography and CT findings were deemed suitable, TSGP was conducted. Acute-onset patients with dissection-related complications, such as type A retrograde dissection, rupture, impending rupture, and end-organ ischemia, underwent emergent TSGP. Acute-onset patients without dissection-related complications at high risk for aneurysmal enlargement in the chronic phase, in whom the maximum aortic diameter was >40 mm on CT, underwent scheduled TSGP at least 2 weeks after the onset. In the treatment of chronic dissection, patients who had growing ulcerlike projections or whose maximal aortic diameter was >50 mm on CT were enrolled to TSGP. Patients whose critical branched vessels were supplied from false lumen were excluded from this treatment.
The TSGP procedure was conducted as previously described.7 The stent-grafts consisted of self-expanding, stainless steel Z-shaped Gianturco stents (Cook) covered with e-PTFE grafts (Bard). The stent-grafts were delivered and deployed through 18F to 22F sheaths (Cook) under general anesthesia.
Follow-Up of False Lumen and Whole Aorta Diameters
Follow-up CTs were performed 1 month, 6 months, and 1 year after TSGP and annually thereafter. Measurements of the whole aorta, false lumen, and true lumen diameters were conducted on the level of proximal (level P), middle (level M), and distal (level D) descending aorta and abdominal aorta (level A), as shown in Figure 1.
In this CT follow-up study, 17 cases of acute-onset dissections containing 7 Stanford type A retrograde dissections (acute-onset dissection group) and 11 cases of chronic dissections (chronic dissection group) were followed up for >2 years, and the results were analyzed. Preoperatively, all patients in the acute-onset dissection group had dissections on levels P, M, D, and A. One patient in the chronic dissection group had a localized dissection of the distal thoracic aorta and was followed up by level D CTs only, and 3 cases had no preoperative dissections in the abdominal aorta; therefore, level A CT measurements were not conducted.
Clinically important rates are reported with 95% CIs, and continuous variables are expressed as mean±SD. Only the diameter values in the figures of the periodic change study are expressed as mean±SE. Comparisons of lumen diameter before and at each period after TSGP were performed with a Wilcoxon signed rank test; probability values of <0.05 were considered statistically significant. Actuarial survival rates were calculated with the Kaplan-Meier method.
The patients’ characteristics are shown in Table 1. TSGP was performed in 32 acute-onset and 17 chronic cases, for a total of 49 cases. There were 15 retrograde type A aortic dissections only in the acute-onset group. Rupture occurred in 3 acute-onset type B dissections preoperatively. Ischemia of the kidneys and lower extremitas was observed in 2 acute-onset type B cases before TSGP. Of the 49 patients studied, 2 acute-onset type B patients (4.1%) died in the hospital after TSGP. One patient with rupture of the 2 who suffered from preoperative multiple organ failure died as a result of long-term poor general condition. The other patient died of bowel ischemia owing to SMA involvement before TSGP. One acute-onset patient with type B dissection had a persistent endoleak, and therefore the overall primary success rate was 93.9%. The primary success rates in acute-onset Stanford type A retrograde dissections, acute-onset type B dissections, and chronic type B dissections were 100%, 82.4%, and 100%, respectively.
The mean follow-up period was 3.6±1.5 years. The acute-onset patient with type B dissection died of pneumonia 1 year after open surgery for the persistent endoleak. In addition, 4 TSGPs and 2 open surgeries were conducted in 6 acute-onset patients as additional interventions. Of these, 2 TSGPs and 1 open surgery were performed because of intimal tears in the landing zone 1 to 4 months after the initial TSGP. The locations of these 3 tears were a proximal landing zone in a type B dissection treated by additional TSGP, a distal landing zone in a type A dissection treated by additional TSGP, and a proximal landing zone in a type B dissection treated by surgery. Moreover, 1 additional TSGP was performed because of a new intimal tear into the thrombosed false lumen in a type B dissection, and the final additional TSGP was performed because a secondary endoleak at a peripheral landing zone was observed 1 year after the initial TSGP in 1 type A dissection. The other 1 of 3 additional surgeries was performed in a Stanford type A patient because of a patent false lumen in the ascending aorta due to an intimal tear thought to be a reentry in the cervical branch. No additional interventions were performed in the chronic dissection cases. Treatment failure rate was 20.0% in acute-onset type A retrograde dissections, 23.5% in acute-onset type B dissections, and 0% in chronic type B dissections, totaling 14.3%. The 4-year actuarial survival rate was 89% in acute-onset cases and 100% in chronic cases. No ruptures were observed after the TSGPs.
The preoperative false lumens in the ascending aortas of the 15 acute type A retrograde dissection cases were thrombosed in 7 cases, open in 4 cases, and thrombosed at onset but recanalized within 4 weeks in 4 patients. Within 1 to 3 months after TSGP, the false lumen in the ascending aorta was obliterated, and the diameter was reduced to <5 mm in 14 patients (93%).
Changes in maximum false lumen diameter in the descending thoracic aorta of the acute-onset and chronic cases before and after TSGP are shown in Figure 2. Terms were defined as follows: obliterated, diameter became <5 mm with thrombosis of the false lumen; decreased, diameter decreased by >5 mm; and enlarged, diameter increased by >5 mm. Among the acute-onset cases, the rates of false lumen obliteration 1 month, 6 months, and 1, 2, 3, and 4 years after TSGP were 13%, 56%, 67%, 76%, 82%, and 71%, respectively, whereas the rates of decreased false lumen 1 month, 6 months, and 1, 2, 3, and 4 years after TSGP were 67%, 84%, 88%, 94%, 100%, and 100%, respectively. Among the chronic cases, the rates of false lumen obliteration and rates of decrease were 13%, 29%, 31%, 36%, 40%, and 40%, and 44%, 64%, 69%, 73%, 80%, and 80%, respectively. The false lumen was expected to be more obliterated or reduced by TSGP in the acute-onset than the chronic cases. Although 10 acute-onset cases first showed false lumen obliteration between 1 and 6 months after TSGP, the rate of obliteration dramatically increased from 13% at 1 month to 56% at 6 months after TSGP. False lumen enlargement was observed 1 month after TSGP in the acute-onset case who had a persistent endoleak and underwent open surgery.
The characteristics of the patients who underwent TSGPs and a CT follow-up for >2 years (acute-onset and chronic dissection groups) are shown in Table 2. Fifteen cases were excluded from the 32 acute-onset cases in the size measurement study: 2 hospital deaths, 5 cases followed up for <2 years, 3 cases who needed additional surgery, 3 failed CT follow-up cases, and 2 cases who underwent additional TSGPs because of a new dissection in the long segment of the aorta. The last 2 cases were treated as new chronic cases after the second procedure. One acute-onset case who had a minor endoleak and underwent an additional TSGP because of a local dissection in the landing zone was continuously followed up. Moreover, 6 cases (3 cases followed up for <2 years and 3 failed CT follow-up cases) were excluded from the 17 chronic cases.
The periodic changes in the false lumen, whole aortic, and true lumen diameters of the 17 patients in the acute dissection group and 11 patients in the chronic dissection group who were followed up for >2 years after TSGP were measured. The average diameters of the false lumen at each level for each group are shown in Figure 3. In the acute dissection group, the average diameters before TSGP and 1 month, 6 months, 1 year, and 2 years after TSGP were 15.9, 9.1, 2.8, 2.2, and 3.0 mm on level P; 16.2, 9.5, 2.8, 2.2, and 3.6 mm on level M; 15.6, 8.9, 3.7, 3.2, and 3.1 mm on level D; and 9.0, 9.5, 6.6, 5.3, and 4.8 mm on level A, respectively. In the chronic dissection group, the average diameters before TSGP and 1 month, 6 months, 1 year, and 2 years after TSGP were 28.1, 19.4, 13.7, 11.9, and 10.6 mm on level P; 25.2, 20.7, 14.9, 12.3, and 10.5 mm on level M; 21.0, 16.9, 14.2, 12.9, and 11.9 mm on level D; and 23.3, 23.4, 18.9, 19.1, and 18.9 mm on level A, respectively. The absolute diameter value was significantly larger in the chronic dissection group than in the acute-onset dissection group during the whole follow-up period. The average diameter of the false lumen in the descending thoracic aorta decreased significantly 1 month and 6 months after TSGP in the acute-onset dissection group, after which no significant reductions were seen. In the chronic dissection group, the average diameter of the false lumen in the descending thoracic aorta decreased significantly 6 months and 1 year after TSGP on levels P and M compared with 6 months after TSGP. On level A, the false lumen diameters of the acute-onset dissection and chronic dissection groups did not decrease significantly, although the absolute diameter values of the acute-onset dissection group were small.
The average diameters of the whole aorta at each level for each group are shown in Figure 4. In the acute-onset dissection group, the average diameters before TSGP and 1 month, 6 months, 1 year, and 2 years after TSGP were 40.2, 40.9, 36.2, 35.2, and 35.1 mm on level P; 38.9, 37.8, 33.4, 33.2, and 32.7 mm on level M; 35.1, 35.9, 31.9, 31.7, and 31.5 mm on level D; and 29.8, 30.1, 29.2, 28.5, and 28.5 mm on level A, respectively. In the chronic dissection group, the average diameters before TSGP and 1 month, 6 months, 1 year, and 2 years after TSGP were 46.4, 45.7, 42.5, 40.9, and 40.0 mm on level P; 44.2, 44.4, 41.4, 39.1, and 38.4 mm on level M; 38.7, 38.6, 37.3, 36.8, and 35.7 mm on level D; and 38.6, 39.3, 35.4, 35.4, and 37.4 mm on level A, respectively. The aortic diameter in the descending thoracic aorta of both groups gradually decreased during the follow-up period. This decrease became significant >6 months after TSGP in the acute-onset dissection group and >1 year after TSGP on level P and >6 months after TSGP on level M in the chronic dissection group.
The average diameters of the true lumen on each level for each group are shown in Figure 5. In the acute-onset dissection group, the average diameters before TSGP and 1 month, 6 months, 1 year, and 2 years after TSGP were 24.3, 31.8, 33.4, 33.0, and 32.0 mm on level P; 22.8, 28.4, 30.7, 30.8, and 29.1 mm on level M; 19.2, 25.1, 28.2, 28.5, and 27.7 mm on level D; and 20.8, 20.9, 22.8, 23.2, and 23.8 mm on level A, respectively. In the chronic dissection group, the average diameters before TSGP and 1 month, 6 months, 1 year, and 2 years after TSGP were 20.9, 26.3, 28.8, 29.0, and 29.4 mm on level P; 21.2, 23.7, 26.6, 26.8, and 28.0 mm on level M; 19.6, 21.6, 22.9, 23.9, and 23.9 mm on level D; and 15.4, 15.9, 16.6, 16.3, and 18.5 mm on level A, respectively. The true lumen diameter in the descending thoracic aorta increased 1 month and 6 months after TSGP in the acute-onset dissection group and significantly >6 months after TSGP in the chronic dissection group, but on level P it increased significantly 1 month after TSGP in the chronic dissection group.
In 1999, along with the Stanford University Group, the authors initially proposed the possibility of TSGPs as the choice of repair for acute aortic dissections originating in the descending aorta.8 TSGPs were shown to be technically successful in all 19 patients studied, and complete and partial thoracic aortic thromboses were achieved in 79% and 21%, respectively. Several other facilities are now trying this procedure and have shown good early results even among poor-risk patients.9,10 The authors have also reported the early to midterm outcomes of TSGP for aortic dissections and concluded that it is a reasonable treatment.7 However, in this series false lumen size did not decrease in some chronic cases, irrespective of whether it was open or thrombosed.
The long-term outcomes of TSGP are a major concern for physicians because this procedure is relatively new, and long-term outcomes have yet to be sufficiently discussed. Continuous surveillance of the entire aorta after TSGPs for aortic dissections is important to monitor complications. Changes in the size of the false lumen after TSGP seem to be related to long-term outcomes. The authors have been performing TSGPs for aortic dissections for the last 6 years, and in this series the sizes of the false lumen, whole aorta, and true lumen have been periodically examined at different levels by CT.
The periodic CT studies conducted in the present study suggest that in acute-onset dissections, the false lumen is completely obliterated within 6 months after TSGP if no endoleaks or intimal tears occur. Four of the 17 acute-onset cases examined had open false lumens in some segments of the aorta 2 years after TSGP. Of these, 2 had open false lumens in the thoracic and abdominal aortas. One had intimal tears in the thoracic and abdominal aorta and a peripheral endoleak, and therefore the false lumen diameter showed no changes in the descending thoracic aorta and slight enlargement in the abdominal aorta. The other case had intimal tears in the abdominal aorta only and showed shrinkage of the false lumen in both the descending thoracic and abdominal aorta. The last 2 cases with open false lumen only in the abdominal aorta showed shrinkage of the false lumen in the descending thoracic aorta, although the false lumen in the abdominal aorta became slightly enlarged in 1 case and showed no change in 1 case. In the acute-onset cases, the remaining intimal tears in the descending thoracic aorta were shown to be a critical factor that hampered shrinkage of the thoracic false lumen, but the remaining intimal tears in the abdominal aorta were not. The management of retrograde type A acute dissection by TSGP is still controversial, although our follow-up result was favorable in the present study. More longer-term data are needed to show the stability of proximal aorta.
The periodic CT study results of the chronic cases suggested that the size of the false lumen in the descending aorta before TSGP, which decreased in the first 6 months after TSGP and subsequently decreased gradually, was larger than that of the acute-onset cases. Therefore, the false lumen diameters of the chronic cases 2 years after TSGP were larger than those of the acute-onset cases. Only 3 of the 11 chronic cases had a false lumen <5 mm in diameter in all segments of the aorta 2 years after TSGP and localized thick thromboses in the false lumen that did not shrink.
In 1 case, a thick localized thrombosis in the false lumen of the proximal descending aorta did not shrink despite the absence of intimal tears and endoleaks. This situation is similar to the phenomenon reported for endovascular stent-grafting of true aneurysms. During a long-term follow-up CT study, Sakai et al11 reported that for true aneurysms the maximum aneurysmal diameters increased slightly after stent-graft placement in 30% of the patients without perigraft leaks.
In the chronic cases, the false lumens at the level with intimal tears did not shrink. Only 1 case had intimal tears in the distal descending and abdominal aorta, and therefore the false lumen diameters on levels P and M decreased gradually, but on levels D and A they did not. Four cases had intimal tears only in the abdominal aorta, and therefore the false lumen diameters on levels P, M, and D decreased gradually, but on level A they did not. No cases of false lumen enlargement were seen among the chronic cases. Buffolo et al12 described distal reentry tears after the closure of primary entry tears by TSGP in chronic type B dissections. They suggested that although the fate of this incomplete correction is unknown, it is possible that by treating the proximal tears, false lumen pressure could be reduced. Although in this study no false lumen enlargements were observed in the chronic dissections and no ruptures were observed after TSGP, there is a risk of rupture in thrombosed thick false lumen, and these cases should be followed up.
The timing of TSGPs in acute-onset dissections remains controversial. TSGP treatments of acute aortic dissections after their onset seem to have the potential risks of new intimal tears and aneurysmal degeneration due to the fragility of the dissected intima.13 Therefore, we perform TSGPs for acute-onset dissections >2 weeks after onset if there are no complications caused by the dissections. However, we perform TSGPs immediately if complications, such as type A retrograde dissection, rupture, impending rupture, and end-organ ischemia, are detected. The clinical results of the emergent TSGPs seem to be relevant because of their low mortality rates and high risks of conventional open surgeries.
This study also suggests that the treatment of acute-onset dissections with TSGP shows higher midterm obliteration rates of the false lumen compared with chronic dissections. TSGP for acute-onset dissections seems to be a good alternative to conventional open surgery, although the fragility of the dissected intima remains problematic. TSGP in aortic dissections, particularly in acute aortic dissections, is not without risks. Close follow-up is mandatory to evaluate the necessity of adjunctive procedures and possibly open surgery.