Chikara Ueki, MD, Genichi Sakaguchi, MD, PhD, Takeshi Shimamoto, MD, PhD, and Tatsuhiko Komiya, MD Department of Cardiovascular Surgery, Kurashiki Central Hospital, Kurashiki, Japan
Background. The benefit of thoracic endovascular aortic repair (TEVAR) for uncomplicated acute type B aortic dissection is unclear. Reliable prognostic factors are needed to identify candidates for prophylactic TEVAR. The aim of this study was to detect prognostic factors in patients with uncomplicated acute type B aortic dissection. Methods. From January 2003 to April 2012, a total of 228 patients with uncomplicated acute type B aortic dissection were admitted to our institute. Cox proportional hazards analysis was performed to identify risk factors for death, dissection-related death, and aortic events. Results. Independent predictors of mortality were age (hazard ratio [HR], 1.08; p < 0.001) and false lumen thickness (mm) (HR, 1.10; p [ 0.013), and the risk factor for dissection-related death was false lumen thickness
(mm) (HR, 1.14; p < 0.001). Independent risk factors for aortic events were diameter of the descending aorta (mm) (HR, 1.14; p < 0.001) and entry in a proximal site (HR, 2.90; p [ 0.02). The actuarial freedom from aortic events in patients with a descending aortic diameter of less than 40 mm and no entry in a proximal site at 1, 3, and 5 years was 96.6%, 90.8%, and 82.5%, respectively, whereas in patients with 1 of these 2 factors, it was 80.1%, 66.8%, and 53.5%, respectively. Conclusions. In uncomplicated acute type B aortic dissection, the diameter of the descending aorta and entry in a proximal site were independent prognostic factors for aortic events. Prophylactic TEVAR should be considered for patients with these risk factors.
S
tear [5–8]. However, most studies included not only uncomplicated ABAD but medically treated patients with complications, and few studies have focused on uncomplicated ABAD [9]. The aim of this study was to detect prognostic factors in patients with uncomplicated ABAD to identify the most appropriate indications for prophylactic TEVAR.
ince its initial introduction, thoracic endovascular aortic repair (TEVAR) has emerged as a new option for the treatment of acute type B aortic dissection (ABAD). Recent studies have shown satisfactory early outcomes and mid-term survival benefits of TEVAR for ABAD complicated with rupture and organ malperfusion [1–3]. Thoracic endovascular aortic repair has become a viable option in the treatment of complicated ABAD. On the other hand, the role of TEVAR in the treatment of uncomplicated ABAD remains unclear. In the first randomized comparison between elective TEVAR and best medical treatment, performed in the Investigation of Stent grafts in Aortic Dissection (INSTEAD) trial, TEVAR failed to improve 2-year survival and adverse event rates [4]. Reliable prognostic factors for adverse events are needed for the development of more effective prophylactic intervention in patients with uncomplicated ABAD. Some studies have identified prognostic factors in ABAD such as aortic diameter, partial false lumen thrombosis, false lumen thickness, and location of the primary entry
(Ann Thorac Surg 2014;97:767–73) Ó 2014 by The Society of Thoracic Surgeons
Patients and Methods The Institutional Review Board of Kurashiki Central Hospital approved this study and waived the individual consent because of the study’s retrospective design. Between January 2003 and April 2012, a total of 256 patients were admitted to our institution during the acute phase (within 14 days of onset) of ABAD. Of the 256 patients, 28 had complications of aortic rupture or organ ischemia on admission. These patients were excluded, and the remaining 228 patients with uncomplicated ABAD were entered into this study.
Diagnosis and Predictive Variables Accepted for publication Oct 11, 2013. Presented at the Forty-ninth Annual Meeting of The Society of Thoracic Surgeons, Los Angeles, CA, Jan 26–30, 2013. Address correspondence to Dr Ueki, Department of Cardiovascular Surgery, Kurashiki Central Hospital, Miwa, Kurashiki City, Okayama 7108602, Japan; e-mail: [email protected].
Ó 2014 by The Society of Thoracic Surgeons Published by Elsevier Inc
The diagnosis of type B aortic dissection was confirmed with the use of enhanced computed tomography (CT) immediately after emergent admission in all patients. Aortic geometries were measured using these CT images, including aortic diameter, false lumen thickness, false lumen status, presence of entry, and location. The aortic 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2013.10.038
ADULT CARDIAC
Prognostic Factors in Patients With Uncomplicated Acute Type B Aortic Dissection
768 ADULT CARDIAC
UEKI ET AL RISK FACTORS IN TYPE B AORTIC DISSECTION
Ann Thorac Surg 2014;97:767–73
diameter and false lumen thickness were measured in the descending thoracic aorta at the level of the tracheal bifurcation. The distance from the primary entry tear to the left subclavian artery was measured on parasagittal maximum intensity projections angulated across the aortic arch. The entry site was considered to be proximal when the entry tear was within 50 mm of the left subclavian artery. The status of the false lumen on imaging was classified as patent (group P) if the false lumen was patent in the absence of a thrombus, as partially thrombosed (group PT) if the false lumen was patent and a thrombus was present, and as completely thrombosed (group C) if the false lumen was closed. A completely thrombosed false lumen with only a small intimal tear (ulcer-like projection [ULP]) was categorized into group C. The ULP was defined as any focal, blood-filled pouch projecting into the thrombosed false lumen and was treated as an entry tear in this study.
Follow-Up and Definitions Patients who survived to hospital discharge underwent follow-up care in the outpatient clinic on a regular basis. Follow-up CT was performed every 3 to 24 months, depending on the aortic diameter and progression of aortic disease. An aortic event was defined as a dissection-related death, surgical intervention, aortic enlargement of greater than 60 mm, or occurrence of type A aortic dissection.
Statistical Analysis The Cox proportional hazards model was used to identify predominant predictors for aortic events throughout the entire follow-up period with the use of univariate and stepwise multivariate analyses (entry and removal thresholds, 0.05 and 0.1, respectively). Survival rate, dissection-related death-free rate, and aortic event-free rate were computed according to the Kaplan-Meier technique, and event-free curves were compared with the use of the log-rank test. Unless stated otherwise, all results are expressed as mean standard deviation. A value of p less than 0.05 was considered statistically significant. Data analysis was performed using SPSS for Windows, Advanced Statistics (SPSS Inc, Chicago, IL).
Results Basic and Clinical Characteristics Patients’ basic and clinical characteristics are shown in Table 1. The mean age was 70.4 11.8 years. The majority of patients (67.1%) were male, and 61.0% were 70 years of age or older. On CT imaging upon admission, the false lumen was found to be patent in 27 patients (11.8%), partially thrombosed in 47 (20.6%), and completely thrombosed in 154 (67.5%). Patients in group P were significantly younger than those in the other 2 groups, with a mean age of 59.2 15.4 years versus 70.1 11.1 years in patients of group PT and 72.5 10.1 years in patients of group C. The diameter of the descending aorta was 35.7 5.5 mm, and the false lumen thickness was 12.0 5.2 mm. Entry tears were detected in 170 patients (74.6%), and 72 of those tears were at a proximal site. Of the 154 patients in group C, ULP were detected by CT imaging on admission in 96 (62.3%) and proximal ULP were detected in 43 (27.9%).
Clinical Outcomes ALL-CAUSE DEATH. The mean follow-up period was 3.2 2.6 years. Figure 1 shows the Kaplan-Meier survival curves stratified according to false-lumen status. Survival was lowest in group PT, with 1-, 3-, and 5-year survival rates of 85.5% 5.5%, 81.2% 6.7%, and 75.0% 8.6%, respectively versus 92.6% 5.0%, 86.8% 7.3%, and 86.8% 7.3% in group P, and 97.6% 1.4%, 91.2% 3.1%, and 85.9% 4.2% in group C. Separate log-rank testing revealed a significant increase in mortality in group PT compared with group C (p ¼ 0.013). There was no significant difference between groups P and C or between groups P and PT. DISSECTION-RELATED DEATH. During the follow-up period, 10 patients died of dissection-related causes (6 of rupture of the dissected aorta, 3 of acute respiratory distress syndrome, and 1 of acute type A aortic dissection). The dissection-related death-free rate was highest in group C, with 1-, 3-, and 5-year dissection-related death-free rates of 99.3% 0.7%, 99.3% 0.7%, and 97.4% 2.0%, respectively, versus 92.6% 5.0%, 92.6% 5.0%, and 92.6% 5.0% in group P, and 91.1% 4.3%, 91.1% 4.3%, and 79.7% 11.3% in group PT (Fig 2).
Table 1. Baseline Demographics and Clinical Characteristics Variable
All (n ¼ 228)
Patent (n ¼ 27)
Partially Thrombosed (n ¼ 47)
Completely Thrombosed (n ¼ 154)
Age (years) Male Marfan syndrome Diameter of descending aorta (mm) Thickness of false lumen (mm) Entry detected on CT imaging Entry in proximal site
70.4 11.8 153 (67.1%) 2 (0.9%) 35.7 5.5 12.0 5.2 170 (74.6%) 72 (31.6%)
59.2 15.4 17 (63.0%) 2 (7.4%) 34.8 8.0 17.6 5.9 27 (100%) 13 (48.1%)
70.1 11.1 33 (70.2%) 0 36.5 5.8 15.4 4.6 47 (100%) 16 (34.0%)
72.5 10.1 103 (66.9%) 0 35.6 4.9 10.0 3.8 96 (62.3%) 43 (27.9%)
Entry detected on CT upon admission includes ulcer-like projections in patients with thrombosed false lumens. CT ¼ computed tomography.
Fig 1. Cumulative survival in each group. (C ¼ completely thrombosed false lumen [FL]; P ¼ patent false lumen; PT ¼ partially thrombosed false lumen.)
Separate log-rank testing revealed a significant increase in dissection-related mortality in group PT compared with group C (p ¼ 0.006). Log-rank testing did not reveal significant differences between groups P and C (p ¼ 0.15) nor with group PT (p ¼ 0.65). AORTIC EVENTS. In this study cohort, 47 patients had aortic events during the follow-up period. In these 47 patients, the mean period from onset to aortic event was 38 31 months (range, 0 to 104 months). Of the 47 patients, 30 underwent surgical intervention (17 had open surgery, 13 had TEVAR), 10 had dissection-related death as mentioned above, 6 met the criterion of a maximum aortic diameter of greater than 60 mm, and 1 had retrograde type A aortic dissection in the course of medical treatment. In patients who underwent open surgery, there were 2 in-hospital deaths. On the other hand, there was no death and stroke in patients who underwent TEVAR. Actuarial freedom from aortic events in all cases at 1, 3, and 5 years was 89.2% 2.1%, 80.2% 3.3%, and 70.0% 4.5%, respectively. The aortic event-free rates in the 3 groups are shown in Figure 3. Actuarial freedom from aortic events in group C at 1, 3, and 5 years was 90.9% 2.4%, 81.7% 3.8%, and 72.6% 5.2%, respectively; these values were 83.8% 5.7%, 73.3% 8.5%, and 64.1% 11.4% in group PT and 88.7% 6.1%, 83.2% 7.9%, and 55.1% 14.5% in group P. Log-rank testing did not reveal significant differences between any 2 groups.
UEKI ET AL RISK FACTORS IN TYPE B AORTIC DISSECTION
Fig 2. Cumulative dissection-related death-free rate in each group. (C ¼ completely thrombosed false lumen [FL]; P ¼ patent false lumen; PT ¼ partially thrombosed false lumen.) DISSECTION-RELATED DEATH. The results of univariate and multivariate analyses for predictors of dissection-related death are shown in Table 3. On multivariate analysis, false lumen thickness (HR, 1.14; 95% CI, 1.03 to 1.27; p ¼ 0.013) was confirmed to be the significant predictor of dissection-related death. Patients with a false lumen of greater than15 mm had significantly lower dissectionrelated death-free rates than the other patients (p < 0.001) (Fig 4). AORTIC EVENTS. Table 4 shows the results of univariate and multivariate analyses for predictors of aortic events. Multivariate Cox regression analysis showed a descending aortic diameter (HR, 1.14; 95% CI, 1.09 to 1.19; p < 0.001) and the presence of a proximal entry tear (HR, 2.90; 95% CI, 1.47 to 5.75; p ¼ 0.002) to be significantly predictive of aortic events.
Predictors of Adverse Events MORTALITY. Cox regression multivariate analysis after univariate analysis of all-cause mortality showed age (hazard ratio [HR], 1.08; 95% confidence interval [CI], 1.03 to 1.14; p < 0.001) and false lumen thickness (HR, 1.10; 95% CI, 1.02 to 1.19; p ¼ 0.01) to be independent risk factors for mortality (Table 2).
769
Fig 3. Cumulative aortic event-free rate in each group. (C ¼ completely thrombosed false lumen [FL]; P ¼ patent false lumen; PT ¼ partially thrombosed false lumen.)
ADULT CARDIAC
Ann Thorac Surg 2014;97:767–73
770 ADULT CARDIAC
UEKI ET AL RISK FACTORS IN TYPE B AORTIC DISSECTION
Ann Thorac Surg 2014;97:767–73
Table 2. Univariate and Multivariate Predictor Analyses of All-Cause Death Univariate Analysis Variable a
Age (years) Female Diameter of descending aorta (mm)b Thickness of false lumen (mm)b Status of false lumen Patent Partially thrombosed Completely thrombosed Entry tear Proximal entry tear a
b
Per year.
Multivariate Analysis
HR
95% CI
p Value
HR
95% CI
p Value
1.07 0.62 1.08 1.09
1.02–1.11
0.002 0.285 0.028 0.014
1.08
1.03–1.14
Background. The benefit of thoracic endovascular aortic repair (TEVAR) for uncomplicated acute type B aortic dissection is unclear. Reliable prognostic factors are needed to identify candidates for prophylactic TEVAR. The aim of this study was to detect prognostic factors in patients with uncomplicated acute type B aortic dissection. Methods. From January 2003 to April 2012, a total of 228 patients with uncomplicated acute type B aortic dissection were admitted to our institute. Cox proportional hazards analysis was performed to identify risk factors for death, dissection-related death, and aortic events. Results. Independent predictors of mortality were age (hazard ratio [HR], 1.08; p < 0.001) and false lumen thickness (mm) (HR, 1.10; p [ 0.013), and the risk factor for dissection-related death was false lumen thickness
(mm) (HR, 1.14; p < 0.001). Independent risk factors for aortic events were diameter of the descending aorta (mm) (HR, 1.14; p < 0.001) and entry in a proximal site (HR, 2.90; p [ 0.02). The actuarial freedom from aortic events in patients with a descending aortic diameter of less than 40 mm and no entry in a proximal site at 1, 3, and 5 years was 96.6%, 90.8%, and 82.5%, respectively, whereas in patients with 1 of these 2 factors, it was 80.1%, 66.8%, and 53.5%, respectively. Conclusions. In uncomplicated acute type B aortic dissection, the diameter of the descending aorta and entry in a proximal site were independent prognostic factors for aortic events. Prophylactic TEVAR should be considered for patients with these risk factors.
S
tear [5–8]. However, most studies included not only uncomplicated ABAD but medically treated patients with complications, and few studies have focused on uncomplicated ABAD [9]. The aim of this study was to detect prognostic factors in patients with uncomplicated ABAD to identify the most appropriate indications for prophylactic TEVAR.
ince its initial introduction, thoracic endovascular aortic repair (TEVAR) has emerged as a new option for the treatment of acute type B aortic dissection (ABAD). Recent studies have shown satisfactory early outcomes and mid-term survival benefits of TEVAR for ABAD complicated with rupture and organ malperfusion [1–3]. Thoracic endovascular aortic repair has become a viable option in the treatment of complicated ABAD. On the other hand, the role of TEVAR in the treatment of uncomplicated ABAD remains unclear. In the first randomized comparison between elective TEVAR and best medical treatment, performed in the Investigation of Stent grafts in Aortic Dissection (INSTEAD) trial, TEVAR failed to improve 2-year survival and adverse event rates [4]. Reliable prognostic factors for adverse events are needed for the development of more effective prophylactic intervention in patients with uncomplicated ABAD. Some studies have identified prognostic factors in ABAD such as aortic diameter, partial false lumen thrombosis, false lumen thickness, and location of the primary entry
(Ann Thorac Surg 2014;97:767–73) Ó 2014 by The Society of Thoracic Surgeons
Patients and Methods The Institutional Review Board of Kurashiki Central Hospital approved this study and waived the individual consent because of the study’s retrospective design. Between January 2003 and April 2012, a total of 256 patients were admitted to our institution during the acute phase (within 14 days of onset) of ABAD. Of the 256 patients, 28 had complications of aortic rupture or organ ischemia on admission. These patients were excluded, and the remaining 228 patients with uncomplicated ABAD were entered into this study.
Diagnosis and Predictive Variables Accepted for publication Oct 11, 2013. Presented at the Forty-ninth Annual Meeting of The Society of Thoracic Surgeons, Los Angeles, CA, Jan 26–30, 2013. Address correspondence to Dr Ueki, Department of Cardiovascular Surgery, Kurashiki Central Hospital, Miwa, Kurashiki City, Okayama 7108602, Japan; e-mail: [email protected].
Ó 2014 by The Society of Thoracic Surgeons Published by Elsevier Inc
The diagnosis of type B aortic dissection was confirmed with the use of enhanced computed tomography (CT) immediately after emergent admission in all patients. Aortic geometries were measured using these CT images, including aortic diameter, false lumen thickness, false lumen status, presence of entry, and location. The aortic 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2013.10.038
ADULT CARDIAC
Prognostic Factors in Patients With Uncomplicated Acute Type B Aortic Dissection
768 ADULT CARDIAC
UEKI ET AL RISK FACTORS IN TYPE B AORTIC DISSECTION
Ann Thorac Surg 2014;97:767–73
diameter and false lumen thickness were measured in the descending thoracic aorta at the level of the tracheal bifurcation. The distance from the primary entry tear to the left subclavian artery was measured on parasagittal maximum intensity projections angulated across the aortic arch. The entry site was considered to be proximal when the entry tear was within 50 mm of the left subclavian artery. The status of the false lumen on imaging was classified as patent (group P) if the false lumen was patent in the absence of a thrombus, as partially thrombosed (group PT) if the false lumen was patent and a thrombus was present, and as completely thrombosed (group C) if the false lumen was closed. A completely thrombosed false lumen with only a small intimal tear (ulcer-like projection [ULP]) was categorized into group C. The ULP was defined as any focal, blood-filled pouch projecting into the thrombosed false lumen and was treated as an entry tear in this study.
Follow-Up and Definitions Patients who survived to hospital discharge underwent follow-up care in the outpatient clinic on a regular basis. Follow-up CT was performed every 3 to 24 months, depending on the aortic diameter and progression of aortic disease. An aortic event was defined as a dissection-related death, surgical intervention, aortic enlargement of greater than 60 mm, or occurrence of type A aortic dissection.
Statistical Analysis The Cox proportional hazards model was used to identify predominant predictors for aortic events throughout the entire follow-up period with the use of univariate and stepwise multivariate analyses (entry and removal thresholds, 0.05 and 0.1, respectively). Survival rate, dissection-related death-free rate, and aortic event-free rate were computed according to the Kaplan-Meier technique, and event-free curves were compared with the use of the log-rank test. Unless stated otherwise, all results are expressed as mean standard deviation. A value of p less than 0.05 was considered statistically significant. Data analysis was performed using SPSS for Windows, Advanced Statistics (SPSS Inc, Chicago, IL).
Results Basic and Clinical Characteristics Patients’ basic and clinical characteristics are shown in Table 1. The mean age was 70.4 11.8 years. The majority of patients (67.1%) were male, and 61.0% were 70 years of age or older. On CT imaging upon admission, the false lumen was found to be patent in 27 patients (11.8%), partially thrombosed in 47 (20.6%), and completely thrombosed in 154 (67.5%). Patients in group P were significantly younger than those in the other 2 groups, with a mean age of 59.2 15.4 years versus 70.1 11.1 years in patients of group PT and 72.5 10.1 years in patients of group C. The diameter of the descending aorta was 35.7 5.5 mm, and the false lumen thickness was 12.0 5.2 mm. Entry tears were detected in 170 patients (74.6%), and 72 of those tears were at a proximal site. Of the 154 patients in group C, ULP were detected by CT imaging on admission in 96 (62.3%) and proximal ULP were detected in 43 (27.9%).
Clinical Outcomes ALL-CAUSE DEATH. The mean follow-up period was 3.2 2.6 years. Figure 1 shows the Kaplan-Meier survival curves stratified according to false-lumen status. Survival was lowest in group PT, with 1-, 3-, and 5-year survival rates of 85.5% 5.5%, 81.2% 6.7%, and 75.0% 8.6%, respectively versus 92.6% 5.0%, 86.8% 7.3%, and 86.8% 7.3% in group P, and 97.6% 1.4%, 91.2% 3.1%, and 85.9% 4.2% in group C. Separate log-rank testing revealed a significant increase in mortality in group PT compared with group C (p ¼ 0.013). There was no significant difference between groups P and C or between groups P and PT. DISSECTION-RELATED DEATH. During the follow-up period, 10 patients died of dissection-related causes (6 of rupture of the dissected aorta, 3 of acute respiratory distress syndrome, and 1 of acute type A aortic dissection). The dissection-related death-free rate was highest in group C, with 1-, 3-, and 5-year dissection-related death-free rates of 99.3% 0.7%, 99.3% 0.7%, and 97.4% 2.0%, respectively, versus 92.6% 5.0%, 92.6% 5.0%, and 92.6% 5.0% in group P, and 91.1% 4.3%, 91.1% 4.3%, and 79.7% 11.3% in group PT (Fig 2).
Table 1. Baseline Demographics and Clinical Characteristics Variable
All (n ¼ 228)
Patent (n ¼ 27)
Partially Thrombosed (n ¼ 47)
Completely Thrombosed (n ¼ 154)
Age (years) Male Marfan syndrome Diameter of descending aorta (mm) Thickness of false lumen (mm) Entry detected on CT imaging Entry in proximal site
70.4 11.8 153 (67.1%) 2 (0.9%) 35.7 5.5 12.0 5.2 170 (74.6%) 72 (31.6%)
59.2 15.4 17 (63.0%) 2 (7.4%) 34.8 8.0 17.6 5.9 27 (100%) 13 (48.1%)
70.1 11.1 33 (70.2%) 0 36.5 5.8 15.4 4.6 47 (100%) 16 (34.0%)
72.5 10.1 103 (66.9%) 0 35.6 4.9 10.0 3.8 96 (62.3%) 43 (27.9%)
Entry detected on CT upon admission includes ulcer-like projections in patients with thrombosed false lumens. CT ¼ computed tomography.
Fig 1. Cumulative survival in each group. (C ¼ completely thrombosed false lumen [FL]; P ¼ patent false lumen; PT ¼ partially thrombosed false lumen.)
Separate log-rank testing revealed a significant increase in dissection-related mortality in group PT compared with group C (p ¼ 0.006). Log-rank testing did not reveal significant differences between groups P and C (p ¼ 0.15) nor with group PT (p ¼ 0.65). AORTIC EVENTS. In this study cohort, 47 patients had aortic events during the follow-up period. In these 47 patients, the mean period from onset to aortic event was 38 31 months (range, 0 to 104 months). Of the 47 patients, 30 underwent surgical intervention (17 had open surgery, 13 had TEVAR), 10 had dissection-related death as mentioned above, 6 met the criterion of a maximum aortic diameter of greater than 60 mm, and 1 had retrograde type A aortic dissection in the course of medical treatment. In patients who underwent open surgery, there were 2 in-hospital deaths. On the other hand, there was no death and stroke in patients who underwent TEVAR. Actuarial freedom from aortic events in all cases at 1, 3, and 5 years was 89.2% 2.1%, 80.2% 3.3%, and 70.0% 4.5%, respectively. The aortic event-free rates in the 3 groups are shown in Figure 3. Actuarial freedom from aortic events in group C at 1, 3, and 5 years was 90.9% 2.4%, 81.7% 3.8%, and 72.6% 5.2%, respectively; these values were 83.8% 5.7%, 73.3% 8.5%, and 64.1% 11.4% in group PT and 88.7% 6.1%, 83.2% 7.9%, and 55.1% 14.5% in group P. Log-rank testing did not reveal significant differences between any 2 groups.
UEKI ET AL RISK FACTORS IN TYPE B AORTIC DISSECTION
Fig 2. Cumulative dissection-related death-free rate in each group. (C ¼ completely thrombosed false lumen [FL]; P ¼ patent false lumen; PT ¼ partially thrombosed false lumen.) DISSECTION-RELATED DEATH. The results of univariate and multivariate analyses for predictors of dissection-related death are shown in Table 3. On multivariate analysis, false lumen thickness (HR, 1.14; 95% CI, 1.03 to 1.27; p ¼ 0.013) was confirmed to be the significant predictor of dissection-related death. Patients with a false lumen of greater than15 mm had significantly lower dissectionrelated death-free rates than the other patients (p < 0.001) (Fig 4). AORTIC EVENTS. Table 4 shows the results of univariate and multivariate analyses for predictors of aortic events. Multivariate Cox regression analysis showed a descending aortic diameter (HR, 1.14; 95% CI, 1.09 to 1.19; p < 0.001) and the presence of a proximal entry tear (HR, 2.90; 95% CI, 1.47 to 5.75; p ¼ 0.002) to be significantly predictive of aortic events.
Predictors of Adverse Events MORTALITY. Cox regression multivariate analysis after univariate analysis of all-cause mortality showed age (hazard ratio [HR], 1.08; 95% confidence interval [CI], 1.03 to 1.14; p < 0.001) and false lumen thickness (HR, 1.10; 95% CI, 1.02 to 1.19; p ¼ 0.01) to be independent risk factors for mortality (Table 2).
769
Fig 3. Cumulative aortic event-free rate in each group. (C ¼ completely thrombosed false lumen [FL]; P ¼ patent false lumen; PT ¼ partially thrombosed false lumen.)
ADULT CARDIAC
Ann Thorac Surg 2014;97:767–73
770 ADULT CARDIAC
UEKI ET AL RISK FACTORS IN TYPE B AORTIC DISSECTION
Ann Thorac Surg 2014;97:767–73
Table 2. Univariate and Multivariate Predictor Analyses of All-Cause Death Univariate Analysis Variable a
Age (years) Female Diameter of descending aorta (mm)b Thickness of false lumen (mm)b Status of false lumen Patent Partially thrombosed Completely thrombosed Entry tear Proximal entry tear a
b
Per year.
Multivariate Analysis
HR
95% CI
p Value
HR
95% CI
p Value
1.07 0.62 1.08 1.09
1.02–1.11
0.002 0.285 0.028 0.014
1.08
1.03–1.14
