Autosomal dominant hyper-IgE (AD-HIES) is a primary immunodeficiency caused by mutations in STAT3. AD-HIES is characterized by elevated levels of IgE, an ineffective immune response, connective tissue and coronary arterial dilation and tortuosity. To date, coronary artery evaluation in AD-HIES patients has been limited to lumenography using conventional angiography², multi-detector computer tomography angiography (MD-CTA)^1,2,3 or magnetic resonance angiography (MRA)¹, revealing marked increases in ectasia, aneurysms and tortuosity relative to control subjects³. However, compared to controls, there was an absence of significant luminal stenosis, which appeared consistent with less atherosclerosis. Direct in vivo coronary vessel wall (VW) imaging may allow for better interrogation of coronary vessel abnormalities. The goal of this study was to evaluate the coronary VW of AD-HIES patients using MRI and compare to healthy subjects and subjects with known coronary artery disease (CAD). Additionally, histopathology was used to directly interrogate the vessel wall and associated structures in one coronary artery from a patient with AD-HIES.

A total of 28 subjects(10 with AD-HIES, 8 healthy, 10 with CAD) were prospectively included in the study. The 10 subjects with AD-HIES had confirmed dominant negative STAT3 mutations.The 10 patients with known CAD were confirmed by coronary Computer. Additionally, a post-mortem coronary artery from one AD-HIES patient was examined. Groups were age- and BMI-matched as shown in the table.

Scout Scanning: The beginning of the cardiac rest period was identified and anatomic slices perpendicular to the proximal part of the right coronary artery (RCA) during diastole were planned similar to a previously published methodology⁴. Patient-specific time delay between the R-wave of the electrocardiogram (ECG) and this rest period were used for the subsequent coronary MRA and wall imaging. This was followed by a fast 3D segmented k-space gradient-echo low resolution, navigator, and Vector ECG gated (VCG) whole-heart scan for localization of the coronaries.

Coronary Wall MRI: Free-breathing time-resolved dark-blood (TRAPD-MRI)⁵ proximal coronary vessel wall datasets were acquired with a fixed inversion time (TI=200ms) and phase-sensitive reconstruction. Data were acquired in the proximal RCA segment at a location without noticeable stenosis or atherosclerotic disease on the coronary MRA using a segmented k-space spiral acquisition with spectral spatial excitation, using a 32-channel phased array cardiac received coil and VCG triggering. Images were anonymized, wall thickness was measured as previously published.

Coronary MD-CTA: Coronary MD-CTA scans with ECG gating were performed in 10 CAD subjects. The MDCT protocol was similar to previously described techniques. Image analysis and interpretation of the axial and the multiplanar re-formatted images were performed using a commercial three-dimensional software tool.

Pathology: Autopsy specimens of coronary arteries were taken from one AD-HIES patient less than 24 hours after death and processed immediately. Staining was performed on all the samples to delineate the connective tissue structures.

Statistical Analysis: The mean ± standard deviation of age, body mass index (BMI), systolic and diastolic blood pressures, Framingham risk score (FRS) and coronary vessel wall thickness and luminal areas were calculated. Analysis of covariance (ANCOVA) was used to test for statistical difference between these variables. Additional ANCOVA test for coronary luminal area and vessel wall thickness was performed to adjust for the potential effect of age. A P value of < 0.05 was considered statistically significant.

Examples of the coronary vessel wall images in the three groups are shown in Figure 1a. There was no statistical difference between the age, BMI, blood pressures and FRS between the three groups (Table 1). However, MRI imaging of coronary vessel walls of AD-HIES patients showed thicker vessel walls than those of healthy controls (Table 1).

There was no statistically significant difference in vessel wall thickness between non-AD-HIES subjects with atherosclerosis and AD-HIES patients (Figure 1b). Six of the 10 AD-HIES subjects had either tortuosity and/or dilation of one of the coronary vessels. All CAD subjects had atherosclerosis in their coronary arteries as seen on MDCT.

The autopsy specimen showed vessel dilation at the bifurcation. The vessel wall was expanded and thickened by an atherosclerotic plaque, which occupied more than 60% of the vessel area. However, lumen stenosis was only 2.7%, relative to the reference section. Morphological evaluation of the plaque site showed thinning of both medial and adventitial layers and decrease in connective tissue relative to less affected areas of the vessel.

This is the first study to image the coronary vessel wall of patients with AD-HIES by MRI. Coronary wall thickness of AD-HIES patients was thicker than in healthy subjects but comparable to CAD patients. These findings suggest that coronary arteries in Job’s syndrome are affected with atherosclerosis, contrary to prior beliefs and study findings. Direct histologic evaluation confirms the presence of atherosclerosis with lack of needed supportive adventitial thickening and elastic components. These findings suggest mechanisms for weakened vessel wall that may lead to coronary dilation and aneurysm in AD-HIES.