Clinical Characteristics and CT Findings in Adult Patients With an Aberrant Right Subclavian Artery: A Single-Center Retrospective Cohort Study

Introduction: An aberrant right subclavian artery (ARSA) in an anomalous artery that arises from the descending aorta distal to the left subclavian. Although most ARSA are discovered incidentally given the absence of symptoms, they may be associated with significant morbidity and mortality. We sought to determine features associated with symptoms in adult patients diagnosed with an aberrant right subclavian artery. Methods: In this single-center, retrospective study, 386 adult patients were diagnosed with ARSA on chest CT scans performed between June 2016 and April 2021. Patients were grouped by the presence of symptoms, which included dysphagia, shortness of breath, cough, and upper airway wheezing. Four cardiothoracic radiologists reviewed the chest CT scans to measure features of ARSA. Agreement and multivariable logistic regression analyses were performed to determine interobserver variability and features associated with the presence of symptoms, respectively. Results: The prevalence of ARSA was 1.02% and 81.3% of patients were asymptomatic. Shortness of breath (74.6%) and dysphagia (18.6%) were the most common symptom. Interobserver agreement amongst the reading radiologists was acceptable with most variables having an interclass correlation coefficient or kappa > 0.80. A patient’s height > 158 cm (OR: 2.50, P=0.03), cross-sectional area > 60 mm2 of ARSA at the level of the esophagus (OR: 2.39, P=0.046), distance increase per 1 mm between ARSA and trachea (OR: 0.85, P=0.02), and angle > 108 degrees formed with the aortic arch (OR: 1.99, P=0.03) were associated with symptoms. (Table) (Figure) Conclusion: In our single center study, we found being taller, having a larger cross-sectional area of ARSA at the level of the esophagus, a greater angle at the junction with the aortic arch, and a shorter distance between the ARSA and trachea, were associated with the presence of symptoms, while having a dilated esophagus, and/or atherosclerosis was not. Importantly, the absence of dysphagia should not rule out an ARSA. These findings may help to predict which patients will develop symptoms and potentially become candidates for surgical consideration.Figure 1.: Panels A, B, and C show multiplanar reformats to obtain measurements of ARSA at the level of crossing the esophagus. Panel D shows measurement of the distance between ARSA and posterior wall of trachea Table 1. - Association with being symptomatic Median (IQR) or fraction (%) of patients Unadjusted analysis Multivariable analysis SymptomaticN = 59 AsymptomaticN = 257 OR (95% CI) p-value OR (95% CI) p-value Age at diagnosis, years (10 year increase) 60.0 (43.9-68.5) 57.3 (41.5-67.3) 1.08 (0.91-1.29) 0.41 1.00 (0.83-1.21) 0.83 Male 25/59 (42.4%) 85/257 (33.1%) 1.49 (0.83-2.65) 0.18 1.12 (0.60-2.11) 0.73 Weight, kg (10 kg increase) 78.9 (63.3-90.5) 76.5 (62.5-90.7) 0.98 (0.86-1.11) 0.77 0.98 (0.86-1.12) 0.78 Height, cm (10 cm increase) 167 (160.0-175.0) 167 (160.0-173.0) 1.14 (0.89-1.46) 0.29 1.13 (0.88-1.48) 0.37 Height > 158 cm 54/59 (91.5%) 205/254 (80.7%) 2.58 (1.07-7.70) 0.03 2.50 (1.01-7.54) 0.034 Body mass index, kg/m2 (5 kg/m2 increase) 25.8 (22.4-30.3) 26.7 (22.9-32.4) 0.93 (0.76-1.05) 0.46 0.98 (0.86-1.11) 0.74 Obesity 16/58 (27.6%) 83/254 (32.7%) 0.78 (0.42-1.48) 0.45 1.09 (0.56-2.15) 0.80 White race 53/59 (89.8%) 206/251 (82.1%) 2.19 (0.96-5.93) 0.09 1.95 (0.83-5.38) 0.15 Ethnicity, Hispanic/Latino 3/59 (5.1%) 21/257 (8.2%) 0.60 (0.17-2.09) 0.42 0.93 (0.25-3.50) 0.92 Family History, negative 57/59 (96.6%) 230/257 (89.5%) 3.35 (0.77-14.48) 0.10 2.29 (0.63-14.78) 0.28 Type of CT scan performed o Chest CT Angiogramo Contrast Chest CT 31/59 (52.5%)28/59 (47.5%) 122/257 (47.8%)133/257 (52.2%) 1.00 (reference)0.83 (0.47-1.46) 0.52 1.00 (reference)0.61 (0.33-1.12) 0.11 Diameter at origin max, mm (10 mm increase) 17 (14.0-19.0) 16 (13.0-18.0) 1.21 (0.63-2.34) 0.57 0.93 (0.42-2.07) 0.86 Diameter at origin min, mm (10 mm increase) 14 (11.5-17.0) 13 (11.0-16.0) 1.32 (0.65-2.67) 0.45 0.98 (0.43-2.27) 0.97 Cross-sectional area at origin, mm2 (50 mm2 increase) 166 (122.5-240.5) 150 (109.0-215.0) 1.03 (0.90-1.18) 0.67 0.98 (0.84-1.15) 0.81 Diameter at crossing of esophagus max, mm (10 mm increase) 12 (10.0-13.5) 11 (10.0-13.0) 1.03 (0.50-2.10) 0.94 0.76 (0.25-2.34) 0.63 Diameter crossing of esophagus min, mm (10 mm increase) 10 (9.0-12.0) 10 (8.0-12.0) 1.23 (0.56-2.69) 0.60 0.76 (0.25-2.34) 0.45 Cross-sectional area at crossing of esophagus, mm2 (50 mm2increase) 88 (68.0-120.0) 85 (59.0-120.0) 1.06 (0.83-1.36) 0.63 1.03 (0.76-1.35) 0.85 Area > 60 mm 52/59 (88.1%) 189/257 (73.5%) 2.67 (1.23-6.70) 0.02 2.39 (1.08-6.06) 0.046 Distance between ARSA and trachea, mm (1 mm increase) 5 (2.5-6.0) 5 (3.0-6.0) 0.91 (0.81-1.02) 0.11 0.85 (0.75-0.97) 0.02 Distance 7 mm 53/59 (89.8%) 213/257 (82.9%) 1.83 (0.79-4.97) 0.19 2.24 (0.88-5.73) 0.09 Angle of Proximate ARSA with the Arch, degrees (10 degree increase) 110 (90.0-121.0) 101 (73.0-118.0) 1.09 (0.99-1.21) 0.08 1.08 (0.98-1.20) 0.14 Angle > 108 degrees 32/59 (54.2%) 96/257 (37.4%) 1.99 (1.12-3.54) 0.02 1.90 (1.06-3.42) 0.032 Presence of Kommerell aneurysm 2/59 (3.4%) 4/257 (1.6%) 2.22 (0.40-12.41) 0.36 2.26 (0.39-13.07) 0.36 Presence of atherosclerotic plaque within ARSA 23/59 (39.0%) 77/257 (30.0%) 1.49 (0.83-2.69) 0.18 1.31 (0.69-2.47) 0.41 Severity of atherosclerotic plaque o Mildo Moderateo Severe 14/23 (60.9%)6/23 (26.1%)3/23 (13.0%) 49/77 (63.6%)23/77 (29.9%)5/77 (6.5%) 1.00 (reference)0.91 (0.31-2.68)2.10 (0.45-9.89) 0.870.35 1.00 (reference)0.73 (0.23-2.32)1.82 (0.35-9.38) 0.590.47 Presence of thrombus within ARSA 0/59 (0.0%) 3/257 (1.2%) NA 1.00 NA NA Presence of proximal esophagus dilation 0/59 (0.0%) 23/257 (8.9%) NA 0.01 NA NA IQR, interquartile range; OR, odds ratio; 95% confidence interval, CT, computer tomography; ARSA, aberrant right subclavian artery ORs, 95% CIs, and p-values result from logistic regression models. ORs correspond to the increase given in parenthesis (continuous variables) or presence of the given characteristic (categorical variables). Logistic regression was not possible for presence of thrombus within ARSA and presence of proximal esophagus dilation due to zero cell counts; p-values result from Fisher’s exact test. Family history included aortic aneurysm, congenital heart disease, or aberrant vessels Continuous variables were examined on the continuous scale but were also dichotomized based on the value with the highest Youden Index on the Receiver Operating Characteristic Curve in unadjusted logistic regression. Multivariable logistic regression models were adjusted for all variables that were associated with being symptomatic with a p-value ≤ 0.10 in unadjusted logistic regression analysis (where such analysis was possible). These variables included height > 158cm, cross-sectional area at crossing with esophagus > 60mm2, angle of proximate ARSA with aortic arch >108 degrees, negative family history, and race.