Catheterization and Cardiovascular Interventions 00:00–00 (2014)
Original Studies Coronary Artery Aneurysms Associated With Ascending Aortic Aneurysms and Abdominal Aortic Aneurysms: Pathophysiologic Implications Jessica R. Balderston,1 MD, Jay Giri,2 MD, Daniel M. Kolansky,2 MD, Joseph E. Bavaria,3 MD, and Zachary M. Gertz,1* MD Background: Coronary artery aneurysms (CAA) are seen in 1–5% of angiograms. Aneurysmal coronary disease has been thought to be a variant of atherosclerotic coronary artery disease (CAD) in most patients, but this has not been systematically studied. Methods: To better understand the pathophysiology of CAA, we reviewed the cardiac catheterization films of 403 patients with ascending thoracic aortic aneurysms and 74 patients with abdominal aortic aneurysms (AAA) who underwent surgery for their aortic aneurysms at our institution. Coronary aneurysms had diameters 1.5-fold that of a reference segment. Results: The incidences of CAA in patients with ascending aneurysms and AAA were 17% and 16% respectively (P 5 0.92). CAAs in the ascending group were larger (mean diameter 5.9 vs. 5.0 mm, P 5 0.12) with larger reference vessel size (3.1 vs. 2.6 mm, P 5 0.03). CAAs in the patients with ascending aneurysms were less likely to be CAD-associated within the same vessel (12% vs. 75%, P < 0.001). This difference remained significant after controlling for the presence of generalized CAD. No other differences in CAAs between the two groups were found. Within the ascending aneurysm group, the only clinical variable independently associated with CAA was bicuspid aortic valve (OR 0.47, 95% confidence interval 0.25–0.89, P 5 0.02). The majority of patients with CAA in the ascending aortic aneurysm population did not have CAD or any other previously identified cause of CAA. Conclusions: There is a high incidence of CAA in patients with aortic aneurysms. In patients with ascending aortic aneurysms there is likely a mechanism distinct from CAD that causes CAAs. VC 2014 Wiley Periodicals, Inc. Key words: coronary artery aneurysm; thoracic aortic aneurysm; abdominal aortic aneurysm; bicuspid aortic valve
INTRODUCTION
Coronary artery aneurysms (CAA) are seen in 1–5% of patients undergoing coronary angiography [1,2]. It has long been suggested that such aneurysms may be a 1
Division of Cardiology, Virginia Commonwealth University Medical Center, Richmond, Virginia 2 Division of Cardiology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania 3 Division of Cardiovascular Surgery, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania Conflict of interest: Nothing to report.
C 2014 Wiley Periodicals, Inc. V
variant of atherosclerotic coronary artery disease (CAD) given their frequent coincidence in certain populations, namely those undergoing cardiac catheterization for angina or positive stress test [1,3]. However, studies have *Correspondence to: Zachary M. Gertz, MD, 1200 E Broad St, West Hospital 5th Floor, West Wing, Room 529-B, Richmond, VA 23298. E-mail: [email protected] Received 15 October 2014; Revision accepted 3 November 2014 DOI: 10.1002/ccd.25726 Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com)
2
Balderston et al.
demonstrated subtle differences in the disease patterns and biochemical markers between these two processes, suggesting that coronary aneurysms and CAD may have distinct underlying pathophysiologic processes [4]. In order to learn more about the mechanisms of CAA formation, we compared two populations of patients who had either ascending thoracic aortic aneurysms or abdominal aortic aneurysms (AAA). Both disease groups are prone to pathologic arterial dilation of the aorta, but with different pathophysiologies. AAA are generally associated with atherosclerosis and, in part, driven by traditional risk factors of CAD [5]. Abdominal aortic aneurysm tissue is characterized by chronic local inflammation, with inflammatory infiltrates similar to those found in atherosclerotic coronary plaques [6]. In contrast, ascending aortic aneurysmal disease manifests itself without a clear association with atherosclerosis and is thought to have a strong genetic component [7]. Traditionally, its pathology is characterized by medial degeneration in the absence of atherosclerotic inflammation [8]. If such a mechanism were shown to take place in the coronary arteries, the translational and clinical implications would be significant. We sought to delineate the role that atherosclerosis plays in CAA by comparing patients with ascending aortic aneurysms to those with AAA. The aim of our study was to describe the incidence of CAA and to define risk factors for CAA in patients with ascending aortic aneurysms or AAA with the goal of better understanding the pathophysiology of CAA. METHODS
We performed a retrospective cohort analysis of patients with aortic aneurysms who presented for surgical repair and who also underwent cardiac catheterization at the Hospital of the University of Pennsylvania between January 1, 2007 and May 31, 2013. The University’s Institutional Review Board approved the study. Patient Selection Patients were identified using the Hospital of the University of Pennsylvania’s electronic clinical database. Patients who had undergone either ascending aortic aneurysm repair or AAA repair and who also had cardiac catheterization were selected. Patients undergoing either open or endovascular repairs were included. At our institution, pre-operative cardiac catheterization is performed prior to ascending aortic aneurysm repair in males 30 years of age or older and females 40 years of age or older. Younger patients are referred
for catheterization if they are considered to be at increased risk for coronary disease. Pre-operative cardiac catheterization is not routinely performed in patients undergoing repair of AAA. These patients are evaluated and are referred for catheterization at the discretion of treating clinicians based on clinical factors or abnormal stress test. Patients were included in the ascending aortic aneurysm group if they had a preoperative ascending aneurysm 4.0 cm and in the AAA group if they had a pre-operative AAA 3 cm. Patients who had both an ascending aneurysm 4.0 cm and an AAA 3 cm were included in the ascending aortic aneurysm group. Patients were excluded if they had a history of heart transplant, history of aortic surgery or aortic valve surgery, had predominately aortic arch or descending thoracic aortic disease, or were undergoing surgery for pseudoaneurysm. Coronary Angiography
All coronary angiograms were reviewed by a boardcertified interventional cardiologist (ZG or JG) who was blinded to the patients’ study group. Measurements of coronary vessels were made using quantitative coronary angiography in at least two views (McKesson, San Francisco, CA). Coronary artery aneurysm was defined as a luminal dilation exceeding 1.5-fold of the diameter of an adjacent segment of normal vessel. CAD was defined as a stenosis >70% in any epicardial coronary artery, >2 mm in diameter, or a stenosis >50% in the left main coronary artery. We used a stricter definition of whether a CAA was associated with CAD. A CAA was considered to be CADassociated if there was a stenosis of >50% within 2 mm of the CAA. Because our definition of CADassociated CAA was novel, we also tested whether patients with CAA had any CAD. In addition to standard clinical characteristics, patient charts were reviewed for the presence of predisposing conditions including connective tissue disease (Marfan disease, Ehlers-Danlos) and inflammatory disease (Rheumatoid arthritis, systemic lupus erythematosus, scleroderma, polyarteritis nodosa, giant cell arteritis, takayasu arteritis, Kawasaki disease, temporal arteritis, polymyalgia rheumatica, psoriatic arthritis, sarcoidosis), bicuspid aortic valve (BAV), and size of pre-operative aortic aneurysm. Statistical Analysis Continuous variables are presented as the mean value 6 1 standard deviation, and dichotomous variables as a percentage. Categorical variables were compared using Chi-squared test, and a student’s t-test was used for continuous variables with parametric distributions.
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
Coronary Aneurysms and Aortic Aneurysms
3
TABLE I. Clinical Characteristics
Age (years) Gender (male) Diabetes mellitus Hypertension Hyperlipidemia History of smoking Coronary artery disease Bicuspid aortic valve Connective tissue/inflammatory disease
Ascending aortic aneurysm (n ¼ 403)
Abdominal aortic aneurysm (n ¼ 74)
P value
62.2 (613.0) 66% (n ¼ 265) 10% (n ¼ 41) 85% (n ¼ 342) 58% (n ¼ 235) 59% (n ¼ 236) 15% (n ¼ 60) 46% (n ¼ 184) 12% (n ¼ 47)
71.3 (68.1) 76% (n ¼ 56) 19% (n ¼ 14) 96% (n ¼ 71) 96% (n ¼ 71) 92% (n ¼ 68) 64% (n ¼ 47) 0% (n ¼ 0) 1% (n ¼ 1)
Original Studies Coronary Artery Aneurysms Associated With Ascending Aortic Aneurysms and Abdominal Aortic Aneurysms: Pathophysiologic Implications Jessica R. Balderston,1 MD, Jay Giri,2 MD, Daniel M. Kolansky,2 MD, Joseph E. Bavaria,3 MD, and Zachary M. Gertz,1* MD Background: Coronary artery aneurysms (CAA) are seen in 1–5% of angiograms. Aneurysmal coronary disease has been thought to be a variant of atherosclerotic coronary artery disease (CAD) in most patients, but this has not been systematically studied. Methods: To better understand the pathophysiology of CAA, we reviewed the cardiac catheterization films of 403 patients with ascending thoracic aortic aneurysms and 74 patients with abdominal aortic aneurysms (AAA) who underwent surgery for their aortic aneurysms at our institution. Coronary aneurysms had diameters 1.5-fold that of a reference segment. Results: The incidences of CAA in patients with ascending aneurysms and AAA were 17% and 16% respectively (P 5 0.92). CAAs in the ascending group were larger (mean diameter 5.9 vs. 5.0 mm, P 5 0.12) with larger reference vessel size (3.1 vs. 2.6 mm, P 5 0.03). CAAs in the patients with ascending aneurysms were less likely to be CAD-associated within the same vessel (12% vs. 75%, P < 0.001). This difference remained significant after controlling for the presence of generalized CAD. No other differences in CAAs between the two groups were found. Within the ascending aneurysm group, the only clinical variable independently associated with CAA was bicuspid aortic valve (OR 0.47, 95% confidence interval 0.25–0.89, P 5 0.02). The majority of patients with CAA in the ascending aortic aneurysm population did not have CAD or any other previously identified cause of CAA. Conclusions: There is a high incidence of CAA in patients with aortic aneurysms. In patients with ascending aortic aneurysms there is likely a mechanism distinct from CAD that causes CAAs. VC 2014 Wiley Periodicals, Inc. Key words: coronary artery aneurysm; thoracic aortic aneurysm; abdominal aortic aneurysm; bicuspid aortic valve
INTRODUCTION
Coronary artery aneurysms (CAA) are seen in 1–5% of patients undergoing coronary angiography [1,2]. It has long been suggested that such aneurysms may be a 1
Division of Cardiology, Virginia Commonwealth University Medical Center, Richmond, Virginia 2 Division of Cardiology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania 3 Division of Cardiovascular Surgery, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania Conflict of interest: Nothing to report.
C 2014 Wiley Periodicals, Inc. V
variant of atherosclerotic coronary artery disease (CAD) given their frequent coincidence in certain populations, namely those undergoing cardiac catheterization for angina or positive stress test [1,3]. However, studies have *Correspondence to: Zachary M. Gertz, MD, 1200 E Broad St, West Hospital 5th Floor, West Wing, Room 529-B, Richmond, VA 23298. E-mail: [email protected] Received 15 October 2014; Revision accepted 3 November 2014 DOI: 10.1002/ccd.25726 Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com)
2
Balderston et al.
demonstrated subtle differences in the disease patterns and biochemical markers between these two processes, suggesting that coronary aneurysms and CAD may have distinct underlying pathophysiologic processes [4]. In order to learn more about the mechanisms of CAA formation, we compared two populations of patients who had either ascending thoracic aortic aneurysms or abdominal aortic aneurysms (AAA). Both disease groups are prone to pathologic arterial dilation of the aorta, but with different pathophysiologies. AAA are generally associated with atherosclerosis and, in part, driven by traditional risk factors of CAD [5]. Abdominal aortic aneurysm tissue is characterized by chronic local inflammation, with inflammatory infiltrates similar to those found in atherosclerotic coronary plaques [6]. In contrast, ascending aortic aneurysmal disease manifests itself without a clear association with atherosclerosis and is thought to have a strong genetic component [7]. Traditionally, its pathology is characterized by medial degeneration in the absence of atherosclerotic inflammation [8]. If such a mechanism were shown to take place in the coronary arteries, the translational and clinical implications would be significant. We sought to delineate the role that atherosclerosis plays in CAA by comparing patients with ascending aortic aneurysms to those with AAA. The aim of our study was to describe the incidence of CAA and to define risk factors for CAA in patients with ascending aortic aneurysms or AAA with the goal of better understanding the pathophysiology of CAA. METHODS
We performed a retrospective cohort analysis of patients with aortic aneurysms who presented for surgical repair and who also underwent cardiac catheterization at the Hospital of the University of Pennsylvania between January 1, 2007 and May 31, 2013. The University’s Institutional Review Board approved the study. Patient Selection Patients were identified using the Hospital of the University of Pennsylvania’s electronic clinical database. Patients who had undergone either ascending aortic aneurysm repair or AAA repair and who also had cardiac catheterization were selected. Patients undergoing either open or endovascular repairs were included. At our institution, pre-operative cardiac catheterization is performed prior to ascending aortic aneurysm repair in males 30 years of age or older and females 40 years of age or older. Younger patients are referred
for catheterization if they are considered to be at increased risk for coronary disease. Pre-operative cardiac catheterization is not routinely performed in patients undergoing repair of AAA. These patients are evaluated and are referred for catheterization at the discretion of treating clinicians based on clinical factors or abnormal stress test. Patients were included in the ascending aortic aneurysm group if they had a preoperative ascending aneurysm 4.0 cm and in the AAA group if they had a pre-operative AAA 3 cm. Patients who had both an ascending aneurysm 4.0 cm and an AAA 3 cm were included in the ascending aortic aneurysm group. Patients were excluded if they had a history of heart transplant, history of aortic surgery or aortic valve surgery, had predominately aortic arch or descending thoracic aortic disease, or were undergoing surgery for pseudoaneurysm. Coronary Angiography
All coronary angiograms were reviewed by a boardcertified interventional cardiologist (ZG or JG) who was blinded to the patients’ study group. Measurements of coronary vessels were made using quantitative coronary angiography in at least two views (McKesson, San Francisco, CA). Coronary artery aneurysm was defined as a luminal dilation exceeding 1.5-fold of the diameter of an adjacent segment of normal vessel. CAD was defined as a stenosis >70% in any epicardial coronary artery, >2 mm in diameter, or a stenosis >50% in the left main coronary artery. We used a stricter definition of whether a CAA was associated with CAD. A CAA was considered to be CADassociated if there was a stenosis of >50% within 2 mm of the CAA. Because our definition of CADassociated CAA was novel, we also tested whether patients with CAA had any CAD. In addition to standard clinical characteristics, patient charts were reviewed for the presence of predisposing conditions including connective tissue disease (Marfan disease, Ehlers-Danlos) and inflammatory disease (Rheumatoid arthritis, systemic lupus erythematosus, scleroderma, polyarteritis nodosa, giant cell arteritis, takayasu arteritis, Kawasaki disease, temporal arteritis, polymyalgia rheumatica, psoriatic arthritis, sarcoidosis), bicuspid aortic valve (BAV), and size of pre-operative aortic aneurysm. Statistical Analysis Continuous variables are presented as the mean value 6 1 standard deviation, and dichotomous variables as a percentage. Categorical variables were compared using Chi-squared test, and a student’s t-test was used for continuous variables with parametric distributions.
Catheterization and Cardiovascular Interventions DOI 10.1002/ccd. Published on behalf of The Society for Cardiovascular Angiography and Interventions (SCAI).
Coronary Aneurysms and Aortic Aneurysms
3
TABLE I. Clinical Characteristics
Age (years) Gender (male) Diabetes mellitus Hypertension Hyperlipidemia History of smoking Coronary artery disease Bicuspid aortic valve Connective tissue/inflammatory disease
Ascending aortic aneurysm (n ¼ 403)
Abdominal aortic aneurysm (n ¼ 74)
P value
62.2 (613.0) 66% (n ¼ 265) 10% (n ¼ 41) 85% (n ¼ 342) 58% (n ¼ 235) 59% (n ¼ 236) 15% (n ¼ 60) 46% (n ¼ 184) 12% (n ¼ 47)
71.3 (68.1) 76% (n ¼ 56) 19% (n ¼ 14) 96% (n ¼ 71) 96% (n ¼ 71) 92% (n ¼ 68) 64% (n ¼ 47) 0% (n ¼ 0) 1% (n ¼ 1)
