University of Pisa
Department of Translational Research and New
Technologies in Medicine and Surgery
Diagnostic and Interventional Radiology
Chairman: Prof. Davide Caramella
Vascular applications of Very-High Frequencies
Ultrasonography
Supervisor
Candidate
Prof. Davide Caramella Saverio Vitali
Academic Year 2015-2016
Abstract
The aim of this thesis is to evaluate vascular applications of very high frequencies ultrasound (VHFUS) through the analysis of preliminary results obtained from two different scientific studies.
VHFUS allows visualization of the vessel wall ultrastructure of several arterial districts with a correlation between the different ultrasound interfaces and the layers that histologically compose the vessel wall.
Compared to what previously seen with conventional ultrasound, arterial wall showed an additional echogenic interface, possibly corresponding to the external elastic lamina (media-adventitia interface), allowing the measure of many wall parameters: most important being intima-media thickness (IMT), adventitia thickness (AT) and global thickness (IMAT).
These new diagnostic possibilities have been applied in two different clinical scenarios, analyzing the preliminary results of two different scientific studies:
• The first work evaluates caliber and ultrastructure of the vessel wall of the interdigital arteries of hands in patients with secondary Raynaud's phenomenon, comparing the results with those obtained in healthy subjects population. Preliminary data shows significant differences in intimal thickness and "Media to lumen ratio" between the two study groups. Furthermore, considering only the cryoglobulinemic patients compared to the control group, significant differences were identified for the above-mentioned parameters as well as for the intima-media thickness and lumen diameter.
• The second work shows preliminary results obtained from the FUCHSIA study (Very high-Frequency Ultrasonography for arterial phenotyping in patients with Cervico-Cerebral Artery Dissection (CCeAD), Hypertension, Spontaneous Coronary Artery Dissection (SCAD) and FibroMuscular DysplasIA (FMD). This study is based on the identification of radial vascular wall abnormalities by very high frequency ultrasound in patients with fibromuscular dysplasia. This case-control study aim is to identify radial vascular wall abnormalities analyzing this disease with VHFUS and automated image analysis. Furthermore, the disarray level of the vessels echogenic interfaces was assessed calculating the root mean square error (RMSE) between 20 profiles crossing the two interfaces and the profile obtained averaging them. For each echogenic interface, the RMSE was normalized for the maximum value of the corresponding mean profile (RMSE/mean). Results show a similar radial internal diameter in the two groups, the maximum values of the mean profiles corresponding to the two interfaces tended to be lower in FMD patients. RMSE/mean was significantly higher in FMD than in control group both for 1st and 2nd echogenic interface. IMT, AT and IMAT were significantly higher in FMD: the difference in IMT and IMAT remained significant even when considering age and mean Blood Pressure (BP) as covariates. Wall/lumen ratio was similar and wall cross sectional area (WCSA)
increased in FMD, calculated with IMAT. In conclusion wall ultrastructure of radial arteries of hypertensive FMD patients was extensively altered: a peculiar “blurred” pattern, characterized of loss of echogenicity and inhomogeneity of the two echogenic layers, independent of age and mean BP. Increased wall thickness and WCSA were also found, indicating a eutrophic remodeling.
General Introduction
"Ultrasound Biomicroscopy" (UBM), "High Frequency Ultrasound" (HFUS) and “Very High Frequency Ultrasound” (VHFUS) are defined as exams obtained using ultrasound frequencies between 30 and 100 MHz [1]. With these devices it is possible to correlate the information deriving from the ultrasonographic image with tissue histology [2] (Fig. 1).
Fig. 1: cystic basal cell carcinoma; (A) Histological analysis with hematoxylin-eosin, (B) VHFUS using 70 MHZ transducer.
VHFUS is therefore an innovative tool [1, [3], [4, 5] that allows visualizing healthy and pathological tissues, providing information often similar to those obtained with bioptic sampling.
These devices were initially introduced in the pre-clinical setting and mainly used in the study of animal models. For example the first VHFUS system, used for imaging of small animal embryology, dates back to 2000 and is equipped with transducers with frequencies of 40-55 MHz [6].
VHFUS was later used to monitor tumor growth in animal models and their changes after antineoplastic therapy, to visualize the microvascularization of neoplastic mass and to study the progression of atherosclerotic lesions and the medium-intimate thickness of the wall Arterial (IMT) in murine models [7].
Since 2016 Vevo MD (Fig. 2), VisualSonics (Toronto, Canada) of the FUJIFILM group, is available. It’s one of the first VHFUS systems for clinical use, recently acquired by the Department of Translational Research and New Technologies in Medicine and Surgery of Pisa.
This equipment has two transducers: one reaching a maximum frequency of 70 MHz (UH F 70) and the other a maximum frequency of 46 MHz (UH F 48).
UH F 70 UH F 48
Bandwidth 29-71 MHz 20-46 MHz
Axial resolution 30 µm 50 µm
Laterale resolution 65 µm 110 µm
Maximum depth 10.0 mm 23.5 mm
Image width (Max) 9.7 mm 15.4 mm
Image depth (Max) 10 mm 23.5 mm
Focal depth 5 mm 9 mm
Table 1: Transducers technical characteristics
Ultrasonic frequency is directly proportional to the image resolution and inversely proportional to the depth of penetration: using a maximum frequency of 70 MHz, a resolution of 30 µm/pixel may be obtained, allowing a maximum depth of 10.0 mm; so, to view deeper structures, it is necessary to reduce the frequency of the ultrasound beam, thus decreasing the resolution.
Compared to common ultrasound equipment, the VEVO MD allows extreme resolution of structures located within a maximum depth of 2.3 cm. These features make it suitable for evaluation in various areas:
• NEONATOLOGY AND PEDIATRICS: blood vessels, lymph nodes, abdomen, spine, joints;
• BLOOD VESSELS: radial artery, venous valves, IMT, peripheral vascularization, blood flow characteristics;
• SMALL PARTS: nerves, thyroid and other glands, lymph nodes, male reproductive system;
• DERMATOLOGY: skin layers, melanoma, lipoma, hair follicles, skin lesions and ecchymosis, identification of foreign bodies;
• MSK: superficial joints, tendons and pulleys, medial menisci, carpal and tarsal tunnel.
We focused on the Vevo MD vascular application reporting preliminary results of two ongoing work:
1) Evaluation of patients with secondary raynaud. 2) Evaluation of fibrodysplastic patients.
Interdigital artery analysis in patients with secondary raynaud
phenomenon
Introduction
Raynaud’s phenomenon is the clinical manifestation of vasospasm of digital blood vessels. It includes three phases: ischemia, with pallor of the digits due to vasoconstriction of the digital arteries, precapillary arteries, and cutaneous arteriovenous shunts; hyperemia with redness of the digits; and a return to normal. Whereas the ischemic phase is required for the diagnosis, the hyperemic phase may be lacking. The abnormalities usually spare the thumb but involve most of the other digits. The phenomenon resolves within an hour after the end of cold exposure [8].
Primary Raynaud’s phenomenon is related to functional alterations alone, secondary Raynaud’s phenomenon also reflects structural microvascular abnormalities [9]. Raynaud’s phenomenon is a common complication in patients with connective tissue disease. In systemic sclerosis, Raynaud’s phenomenon is uniformly a major problem and peripheral vascular complications are common. Digital ulcers are primarily due to a vasculopathy of the peripheral arteries in the fingers and toes, in which the intima of vessels becomes thickened and the lumen occluded [10].
Diagnostic imaging modalities in the past have highlighted the involvement of large and medium arteries in patients with extremities’ trophic lesions.
MRA was used to study the digital vascularization, pointing out substantial arterial and venous damage in patients with systemic sclerosis, emphasizing that both the microcirculation and also small caliber vessels are involved in this cluster of patients as also shown by previous x-ray angiography studies [11, 12].
CT angiography demonstrates luminal lesions like thrombi or vessel wall changes such as thickening seen in inflammatory vascular disease or secondary calcifications [12].
Doppler derived pressure measurements may detect hemodynamically significant large vessel disease and Color duplex ultrasound may be used to point out the affected vessels. This modality has good sensitivity and specificity detecting significant stenosis [13].
The resolution provided by common imaging methods (MR, CT and US) does not, however, allow an adequate assessment of the small distal arteries. The advent of high resolution ultrasound methodologies for clinical use allows the non-invasive ultrastructural evaluation of arteries with diameter less then millimetre, such as interdigital arteries . Through B-mode images, obtained with maximum frequency probes up to 70 MHz, it is possible to evaluate the different layers of the vessel wall, the caliber and the patency of the arteries, up to the distal phalanx.
The aim of this paper is to evaluate alteration of small arteries using the Ultrasound device VEVO MD, with transducer with an upper frequency of 70 MHZ, in patients with secondary Raynaud’s phenomenon without trophic lesions,
evaluating patency, diameter and parietal thickness. The results are compared to those obtained on healthy controls.
Materials and methods
This analytical transverse study was performed on a sample of 15 patients with a secondary raynaud phenomenon (mean age: 61yy) versus a control group of 7 subjects (mean age: 42yy). In particular, the patients group consisted of 2 patients with dermatomyositis, 2 cryoglobulinemic patients, 6 patients with systemic sclerosis, 3 patient with ANCA vasculitis and 2 patients with systemic lupus erythematosus (Table 2).
Characteristics Statistics
Sperimental group Control group
Number 15 7 Measurements 180 84 Mean age 61 42 Gender M 33% F 67% M 71% F 29% Primary pathology Vasculitis (V) 3 (20%) absence Systemic sclerosis (SSc) 6 (40%) absence
LES 2 (13%) absence
Dermatomyositis (DM) 2 (13%) absence Cryoglobulinemia (CM) 2 (13%) absence Table 2: clinical characteristics of the population.
The ultrasound examination was performed at ambient room temperature of about 23°C using the Vevo MD with 70 MHz probe. Parameters such as depth and focus position were appropriately adjusted to optimally visualize the interdigital arteries.
Multiple scans were performed on the radial and ulnar interdigital arteries of each finger of both hands, evaluating the whole course from proximal to distal phalanx. Furthermore, at each phalanx level, 5-second (300 frames) clips of radial and ulnar interdigital arteries of the second, third and fourth finger were obtained. Using dedicated offline post-processing tools, we measured the thickness of the different ultrasound interfaces in the arterial wall. Such ultrasound interfaces in previous scientific work were related to the different histological composition of the arterial wall [14].
Based on these observations, the thickness of three distinct layers was measured: • The first layer, called I (intima), begins with the inner wall of the vessel and
• The second layer, called IMT (intima-media thickness), includes I and it extends to the beginning of the second hyperechogenic interface including the overlying hypoechoic layer.
• The third layer, called IMAT (intima-media-adventitia thickness), includes IMT and it extends to the end of the second hyperechogenic interface (Fig. 3).
Fig. 3: Performed measurements on interdigital arterial wall; I (red cross), IMT (blue cross), IMAT (green cross).
The thickness of these layers was manually measured using offline post-processing tools, Horos® [15]. Seven measurements a clip were performed on artery segment perpendicular to the direction of the ultrasonic beam.
The thickness of the intermediate hypoechoic layer, Media (M), and the second hyperechogenic interface, adventitia (A), derived from the differences of the three layers.
Likewise, the vessel diameter (L) was obtained measuring seven times the thickness from the inner surface of the first ultrasound interface to the same on the opposite side (Fig. 4).
Fig. 4: measurements on interdigital artery; L (green cross).
From the seven samples performed for each parameter, the mean value was obtained and a derived value was calculated: "media to lumen ratio" (M/L).
Same evaluations were performed in the control group and the results were compared with the patients group.
Statistical analysis
Categorical data were described by frequency, continuous data by mean, median, range and interquartile range (IQR).
To evaluate the normality of the quantitative data distributions, the Kolmogorov-Smirnov test was performed.
Mann-Whitney test (two tailed) was applied in order to compare the quantitative variables such as I, IMT, IMAT, M, A, L and M/L to the groups (experimental and control) and to the different primitive pathologies (dermatomyositis, cryoglobulinemia patients, systemic sclerosis, ANCA vasculitis and 2 systemic lupus erythematosus.
Differences were considered significant at p < 0.05.
All analyses, descriptive and inferential, were performed using the SPSS v.23 technology.
Results
The results obtained from the comparison between patients and the control group are shown in Table 3.
Parameters Median (IQR)
Sperimental group Control group p-value
I [µm] 58.60 (20) 48.83 (19) 0.001 IMT [µm] 124.8 (51) 115.65 (41) 0.027 IMAT [µm] 191.8 (69) 180.60 (63) 0.022 M [µm] 66.1 (46) 66.70 (36) 0.814 A [µm] 62.8 (29) 55.60 (29) 0.099 L [µm] 850 (404) 917.50 (326) 0.017 M/L 0.15 (0.10) 0.13 (0.06) 0.001
Table 3: Comparison between sperimental and control group of the vascular anatomic parameters in order to evaluate the modifications produced by secondary Raynaud. IQR: interquartile range.
The data shows significant difference between the two groups with regard to I, IMT and IMAT, with greater thickness in the patient group.
Further interesting considerations may be made evaluating the caliber of the vessel, which appears significantly smaller in the patient group. Considering these previous observations also a significant difference even in the case of M/L, with higher values in the patient group was found.
Due to the patients group’s different etiologies, a further subdivision was made to compare patients with the same pathology with the control group (Table 4).
Parameters Median (IQR)
V SSc LES DM C I [µm] 57 (18) 56 (23) 61 (12) 55 (25) 58 (17) IMT [µm] 121 (31) 130 (58) 112 (37) 121 (68) 150 (74) IMAT [µm] 190 (34) 201 (76) 167 (55) 177 (68) 216 (90) Media [µm] 63 (35) 67 (52) 50 (30) 63 (47) 86 (39) Avventizia [µm] 63 (23) 62 (29) 58 (48) 52 (23) 74 (22) L [µm] 1082 (393) 766 (438) 995 (284) 957 (316) 709 (552) M/L 0.13 (0.04) 0.18 (0.14) 0.12 (0.04) 0.14 (0.11) 0.21 (0.26) Table 4: Median and IQR of the vascular anatomic parameters in the different primitive pathologies:
Vasculitis (V), Systemic Sclerosis (SSc), Systemic Lupus Erythematosus (LES), Dermatomyositis (DM) and Cryoglobulinemia (C).
Comparison p-value
Int IMT I/A Med Avv LV LIMT Ltot Ratio M/L
V vs Ctrl .014 .224 .322 .922 .264 .057 .066 .057 .859 .643
SCL vs Ctrl .077 .018 .012 .493 .235 .000 .000 .002 .000 .000
LES vs Ctrl .002 .306 .926 .004 .515 .309 .525 .572 .169 .112
DMM vs Ctrl .032 .282 .618 .953 .408 .471 .243 .225 .252 .387
C vs Ctrl .009 .004 .004 .024 .001 .000 .001 .022 .000 .000
Table 5: Multiple comparisons between medians associated with different primitive pathologies and control group. Control group (Ctrl), Vasculitis (V), Systemic Sclerosis (SSc), Systemic Lupus Erythematosus (LES), Dermatomyositis (DM) and Cryoglobulinemia (C).
These data show the absence of significant difference in vascular parameters compared to the control group for specific pathologies such as dermatomyositis and ANCA vasculitis.
Concerning LES significant differences were found in the intimal and media thickness.
Other significant differences were identified in the scleroderma group: the most interesting values do not refer to the wall structure, but to the size of the vessel, less than the control group, also affecting the M/L ratio which consequently has a higher value.
In the cryoglubilenic group, the identified values differ significantly from those of the control group, both for ultrastructural wall parameters and vascular diameter, with M/L ratio greater than the control group.
Finally, the obtained parameters from the different pathological groups were compared.
Comparison p-value
Int IMT I/A Med Avv LV LIMT Ltot Ratio M/L
V vs SCL .518 .356 .117 .584 .984 .000 .000 .000 .000 .000 V vs LES .386 .042 .305 .006 .874 .536 .455 .264 .291 .271 V vs DMM .868 1.000 .809 .964 .118 .032 .035 .027 .312 .373 V vs CRG .587 .043 .021 .049 .011 .000 .000 .003 .000 .000 SCL vs LES .186 .024 .067 .004 .980 .000 .006 .016 .000 .000 SCL vs DMM .403 .655 .209 .764 .087 .022 .111 .251 .051 .037
SCL vs CRG .338 .187 .202 .126 .011 .493 .684 .791 .430 .190
LES vs DMM .821 .135 .673 .035 .216 .131 .208 .248 .070 .066
LES vs CRG .680 .003 .016 .000 .080 .000 .005 .047 .000 .000
DMM vs CRG .861 .174 .050 .132 .001 .012 .056 .298 .020 .006
Table 6: Multiple comparisons between medians associated with different primitive pathologies. Vasculitis (V), Systemic Sclerosis (SSc), Systemic Lupus Erythematosus (LES), Dermatomyositis (DM) and Cryoglobulinemia (C).
Discussion
In secondary raynaud, vascular arterial abnormalities have been demonstrated both with invasive angiographic studies and non-invasive studies [11, 12]; ultrasonography typically allows performing an inexpensive, minimally invasive and easily repeatable examination. The use of 70MHz frequency probes results in a spatial resolution of 30 µm/pixel allowing a complete visualization of the distal arteries of the upper limb to the terminal branches of the third phalanx, with resolution that allows evaluation of the wall ultrastructure. The maximum scan depth of 1 cm does not determine a significative limit in the visualization of the arterial circulation of the hand as most vessel, especially digital ones, can be visualized at lower depths. The clinical value of assessing the different wall layers and their thickness with conventional ultrasound imaging have been well established [16]. For example, carotid IMT is often referred as a surrogate marker of atherosclerotic damage and a cardiovascular risk prediction as well as other markers such as arterial thickening and loss of elasticity, calcification of coronary blood vessels and volume of coronary plaque [17].
Some literature studies had also focused on the diameter of distal arteries [18] showing a 10-15% reduction in radial arteries in patients with primitive and secondary raynaud compared to the general population, not dependent on blood pressure or parietal thickening, but rather related to low blood flow.
Our study is one of the first performed using VHFUS device on interdigital arteries in patients with secondary raynaud.
The results, contrary to what was shown in literature for the radial artery [18], revealed a significantly increased thickness of the interdigital artery wall layers in the patient group compared to the control group; indicating a different feature of the interdigital arteries, which tend to have thicker walls. Mourad et al [18] showed that radial arteries maintain a normal parietals thickness; thus indicating that, in secondary raynaud’s phenomenon patients, the arteries with a greater muscle component (diameter of about 1mm) tend to have more marked wall modifications compared to arteries with greater elastic component (diameter approximately 1 cm).
Regarding vascular diameter, the values indicate a significant difference between patients with secondary raynaud’s phenomenon and the control group, determining questions and requiring further study. In particular, it would be
useful to understand if these modifications are evolutive and tend to get worse over time. In such case, further evaluations on patients with different stage of disease may clarify if there is a role of these vascular wall alterations in the genesis of trophic lesions in secondary Raynaud's phenomenon. Furthermore, evaluations based on different degrees of injury may establish a M/L value that acts as a cut-off for the onset of trophic lesions, distinguishing between non-risk patients and patients requiring a specific therapeutic approach to avoid such complications.
These preliminary observations are supported by the analysis of values extrapolated by dividing patients according to pathology:
• Concerning patients with scleroderma, our results agree with that obtained by Mourad et al [18] in radial arteries, indicating a significant increase in M/L compared to the control group, mostly due to the reduction of L. Further studies should be performed to evaluate the vessel diameter, also in more distal segments. Even in these patients it’s necessary to understand if these alterations are evolutive and a cut-off value for the development of trophic lesions may be found.
• In the patients with ANCA vasculitis, the only significant identified difference is the thickness of I. In this group we consider patients with different type of vasculitis, so it would be desiderable to complete the evaluation with a larger and more homogeneous sample.
• The only identified differences in dermatomiosytis patients are the intimal thickness, increased compared to the control group. Significant differences were not found for the other parameters, particularly as regards M/L. Although a larger patients sample is needed for further evaluation, probably to suggest that the wall of distal arteries are not involved in this type of lesion. In patients suffering from LES we can make similar considerations. Due to the multiple clinical manifestations of the disease (antiphospholipid antibody syndrome, etc ...) it becomes crucial to have a larger sample with different stages of disease.
• Finally, the cryoglobulinemic patients show significant differences involving all the analysed parameters, suggesting that cryoglobulinemic patients would experience an increase wall thickness, a reduction in artery diameter, with a signigivant increased M/L. Given the numerous differences identified in these types of patients, it is crucial to increase the sample to recognize the presence of cut-off values for trophic lesions.
Unfortunately, our study is subject to some limitations that must be taken into account:
• The first is linked to patient selection: although all patients have a secondary raynaud phenomenon, the group has multiple pathologies with different pathogenetic mechanisms; thus enabling us to perform a comparison between those pathologies identifying disease that most
significantly determine a modification of the wall ultrastructure or the arterial diameter.
• Another limit is the inadequate matching of mean age and gender of patients compared to the control group, which creates a measurement BIAS, as wall structure alterations can be emphasized by aging processes or by gender-related factors. This depends on the difficulty encountered in the recruitment of elderly subjects without any cadiovascular disease. Nevertheless, in the future it is indispensable to expand the samples so the different groups are more comparable by age and gender.
• Another limit is determined by the measuring method. Manual detection, although repeated with a mean end value, may cause BIAS due to the non-complete repeatability of the measurements. This limit can be overcome using a software that automatically analyzes and segmentates images and automatically identifying the required values.
• Another critical issue is the lack of correlation between interdigital results with data obtained at a more proximal and more distal level; it would be interesting to be able to compare this data to that obtained from the radial artery or from the middle phalanx interdigital artery, analyzing greater and smaller diameter vessel with different muscular component.
• One last criticism is the absence of specific histologic models. Our study is based on the experience of Sarkola et al [14], who compared the observations made in "in vivo", “in vitro” and in animal models with 55MHz maximum frequency probes. Due to the substantial correspondence of our images and our findings with them, we used their findings as basis for our observations.
Despite these limitations it’s necessary to understand what the application of these results may be, in particular to be able to analyze patients with different degrees of disease, to establish a direct correlation between ultrastructural wall vessel modification and vascular involvement of the upper limb to identify parameters that describe early and rapidly trend of disease.
Finally, we should understand what wall structure modification are not related to systemic changes due to chronic inflammatory status, as it happens for other chronic nosological entities.
Identification of radial vascular wall abnormalities by very-high
frequency ultrasound in patients with fibromuscular dysplasia
IntroductionFibromuscular dysplasia (FMD) is an idiopathic, systemic, non-atherosclerotic, non-inflammatory vascular disease leading to stenosis, aneurysms, dissections and occlusion of small and medium-sized arteries. Recent data suggest that FD is not so rare as previously thought, showing a prevalence up to 2.6%, In addition, since renal and carotid arteries are more often involved, FMD may lead to cerebral hemorrhage or ischemia, accelerated and renovascular hypertension [19, 20]. Renal FMD can be classified histologically in medial, intimal or perimedial, or angiographically in multifocal or unifocal, but the significance of these classifications for clinical outcome is still controversial. Therefore, there is a strong need of non-invasive biomarkers able to predict clinical evolution of FMD. In a preliminary study by radiofrequency-based US, common carotid wall abnormalities, evaluated by a visual-based phenotypic score, were able to discriminate FMD patients from healthy volunteers and were associated with different histological patterns [21]. The study of medium and small-size arteries might be even more informative for predicting clinical outcomes in FMD, but to date they have not been extensively explored non-invasively due to limited spatial resolution of standard US machines. Furthermore, though FMD shows a certain degree of heritability, with 7.3-11% of FMD patients having an affected family member, no screening strategies have been defined yet in relatives.
This case-control study is aimed at identifying, by means of a novel non-invasive approach based on very-high frequency ultrasound and automated image analysis, radial vascular wall abnormalities characterizing this disease.
Materials and methods:
High-frequency US scans of the radial arteries and of FMD patients and healthy controls were obtained by Vevo MD (70 MHz probe).
Inclusion criteria for all groups: • age >= 18 years;
• informed consent;
• full possession of one’s faculties. Inclusion criteria for FMD group:
• diagnosis of fibromuscular dysplasia of the renal artery, of the cervicocephalic arteries or of other vascular territories according to diagnostic criteria of the European consensus on the diagnosis and management of FMD [20], confirmed by Angio-MRI / Angio-CT and/or arteriography.
Inclusion criteria for Control group: • apparent good health;
• blood pressure < 140/90 mmHg and under no blood pressure-lowering drugs;
• exclusion of FMD, CCeAD and SCAD by medical history and carotid and renal duplex ultrasound.
Exclusion criteria for all groups:
• Serious comorbidity reducing life expectancy to < 1 year; • Pregnancy.
The measurements were performed with subjects in supine position in a quiet air-conditioned room (24°C). Two 5”-US-clips were obtained from the left radial artery. Radial wall showed in all individuals an additive echogenic interface, possibly corresponding to the external elastic lamina (media-adventitia interface). Thus, intima-media thickness (IMT), adventitia thickness (AT), and global thickness (IMAT) were measured and wall cross-sectional area calculated. Furthermore, the disarray level of the two echogenic interfaces was assessed calculating the root mean square error (RMSE) between 20 profiles crossing the two interfaces and the profile obtained averaging them (Fig. 5). For each echogenic interface, the RMSE was normalized for the maximum value of the corresponding mean profile (RMSE/mean)
For each echogenic interface, the RMSE was normalized for the maximum value
of the corresponding mean profile (RMSE/mean).
Results
Eleven treated hypertensive female FMD patients and 8 healthy control women (C) were enrolled (age 52±5 vs 45±13 years, p=0.51; BMI 26±3 vs 23±3, p=0.12; mean BP 97±7 vs 85±10, p=0.01).
Radial internal diameter was similar (1.938±0.432 vs 1.701±0.532 mm, p=0.21). The maximum values of the mean profiles corresponding to the two interfaces tended to be lower in FMD patients (1st: 121±43 vs 157±22, p=0.09; 2nd vs 109 ±44 133±18, p=0.09). RMSE/mean was significantly higher in FMD than in C both for 1st and 2nd echogenic interface (1st 0.83±0.32 vs 1.31±0.24, p=0.006; 2nd 0.94±0.32 vs 1.37±0.38, p=0.03). The difference was attenuated for the 1st but Fig. 5: The homogeneity of the two echogenic interfaces was assessed calculating the root mean square error (RMSE) between 20 gray-level profiles crossing the two interfaces and the profile obtained averaging them. For each echogenic interface, the RMSE was normalized for the maximum value of the corresponding mean profile (RMSE/mean).
not for the 2nd interface when considering age and mean BP as covariates (p=0.054 and p=0.016 respectively) (Table 7).
FMD
(N=11) (N=8) C P value
1st interface: maximum value of
the mean profile 121±43 157±22 0.09
1st interface: RMSE/mean 1.31±0.24 0.83±0.32 0.006
2nd interface: maximum value
of the mean profile 109 ±44 133±18 0.09
2nd interface: RMSE/mean 1.37±0.38 0.94±0.32 0.03
Table 7: Results gray-level analysis of left radial arteries.
IMT (0.166±0.037 vs 0.128±0.022 mm, p=0.03), AT (0.114±0.029 vs 0.083±0.019 mm, p=0.008) and IMAT (0.281±0.042 vs 0.211±0.027 mm, p=0.003) were significantly higher in FMD: the difference in IMT and IMAT remained significant even when considering age and mean BP as covariates (p=0.04, p=0.17, p=0.009 respectively). Wall/lumen ratio was similar and WCSA increased in FMD, calculated with IMAT (Table 8).
FMD (N=11) (N=8) C P value Age (years) 52 (±5) 45 (±13) 0.52 BMI (kg/mq) 26 (±3) 23 (±3) 0.12 Mean BP 97 (±7) 85 (±10) 0.01 rIMT (µm) 166 (±37) 128 (±22) 0.03 rAT (µm) 114 (±29) 83 (±19) 0.008 rIMAT (µm) 281 (±42) 211 (±27) 0.003 r lumen (µm) 1938 (±432) 1701 (±532) 0.21 r M/L 0.15 (±0.04) 0.13 (±0.04) 0.14 r WCSA (mm2) 2.09 (±0.61) 1.20 (±0.34) <0.001
Table 8: Mean and Standard deviation of different parameters in FMD group and Control group with p-value of multiple comparison between the medians.
Discussion
This study combines a strong physiopathological background with the use of novel biomedical technologies to find innovative solutions to a relevant clinical problem. To our knowledge this is the first study aimed at identifying non-invasive, operator-independent tools for diagnosis, risk stratification and family surveillance in FMD, a disease with possible life-threatening consequences but whose pathogenesis and clinical management has not being clearly defined yet. The use of very high-resolution US, coupled with automated analysis (Fig. 6) of wall texture and with the study of mechanical properties of medium and small-size arteries, will allow building a tool that will support the clinician in the management and follow-up of FMD.
Fig. 6: Two example of 20 gray-level profiles crossing the two interfaces and the profile obtained averaging them: FMD patient (a) and normal subject (b).
With the help of VHFUS we were able to find the extensively altered wall ultrastructure of radial arteries in hypertensive FMD patients: a peculiar “blurred” pattern occurred, characterized by loss of echogenicity and inhomogeneity of the two echogenic layers, independent of age and mean Blood Pressure (Fig. 7).
Fig. 7: Radial arteries VHFUS images; a peculiar “blurred” pattern occurred, characterized by loss of echogenicity and inhomogeneity of the two echogenic layers in fibrodisplasic patient (a) compare to the normal subject.
This aspect is well related with the histological features of the disease in which the normal arterial wall pattern is altered. Furthermore our work shows an increased wall thickness and WCSA, indicating an eutrophic remodeling. These
features may be specific of FMD and, if confirmed in a wider sample, can therefore be considered as FMD diagnostic patterns in the radial district.
This study has some limitations:
• The first is the low number of subjects enrolled. The numerical increase in the sample will offer observations with greater significance.
• An additional limit is that only radial arteries have been examined at the moment. Further comparisons made in other districts may add important information regarding the evaluation of the arterial wall in FMD.
• Another limit is the absence of a comparison with hypertensive subjects. Patients with fibrodysplasia often suffering from secondary hypertension that complicates the underlying disease. It would be useful to provide a comparison with patients with essential hypertension to see if similar alterations can be detected in such patients or if FMD play an essential role in the development of these parietal alterations.
• One last criticism is the absence of specific histologic models. Our study is based on the experience of Sarkola et al [14], who compared the observations made in "in vivo", “in vitro” and in animal models with 55MHz maximum frequency probes. Due to the substantial correspondence of our images and our findings with them, we used their findings as basis for our observations.
In conclusion, despite these limitations, the preliminary results of our study show significant changes in the radial artery wall detected with VHFUS. It should be understood whether these alterations are specific of fibrodysplasia or can be detected also in other alterations of the circulatory system. This feature is crucial because the absence of such alterations in other types of vascular pathology could provide a new non-invasive and repeatable ultrasonic diagnostic marker of fibrodysplasia. It is also important to understand whether such marker correlates with the severity of the underlying disease by evaluating subjects with different degrees of disease and re-evaluating them in time.
Possible future developments of this work may be:
• The evaluation of further arterial districts (carotid, brachial and interdigital) • The investigation of the possible evolution over time of vascular alterations
in patients with FMD, comparing them with normotensive healthy individuals and with essential hypertensive patients, in order to use vascular damage accrual as a predictor of clinical outcome in FMD and to study the possible different route of vascular aging in FMD in comparison to essential hypertension. Association between vascular variables and clinical outcome variables will be also tested. Endothelial dysfunction in medium- and small-sized arteries will be explored as a predictor of disease evolution • The identification among first-degree relatives of those with asymptomatic
disease and at risk to develop clinically manifest disease by the use of parameters derived from very high-resolution US scans in comparison to standard work-up by carotid and renal duplex US.
References
1. Eklund, C., P. Friberg, and L.M. Gan, High-resolution radial artery intima-media thickness
and cardiovascular risk factors in patients with suspected coronary artery disease--comparison with common carotid artery intima-media thickness. Atherosclerosis, 2012. 221(1): p. 118-23.
2. Osika, W., et al., Increasing peripheral artery intima thickness from childhood to seniority.
Arterioscler Thromb Vasc Biol, 2007. 27(3): p. 671-6.
3. Foster, F.S., et al., Advances in ultrasound biomicroscopy. Ultrasound Med Biol, 2000.
26(1): p. 1-27.
4. Shung, K.K., High Frequency Ultrasonic Imaging. J Med Ultrasound, 2009. 17(1): p. 25-30.
5. Foster, F.S., J. Hossack, and S.L. Adamson, Micro-ultrasound for preclinical imaging.
Interface Focus, 2011. 1(4): p. 576-601.
6. Foster, F.S., et al., A new ultrasound instrument for in vivo microimaging of mice.
Ultrasound Med Biol, 2002. 28(9): p. 1165-72.
7. Gan, L.M., et al., Non-invasive real-time imaging of atherosclerosis in mice using ultrasound
biomicroscopy. Atherosclerosis, 2007. 190(2): p. 313-20.
8. Wigley, F.M., Clinical practice. Raynaud's Phenomenon. N Engl J Med, 2002. 347(13): p.
1001-8.
9. Gayraud, M., Raynaud's phenomenon. Joint Bone Spine, 2007. 74(1): p. e1-8.
10. Steen, V., et al., Digital ulcers: overt vascular disease in systemic sclerosis. Rheumatology
(Oxford), 2009. 48 Suppl 3: p. iii19-24.
11. Allanore, Y., et al., Hand vascular involvement assessed by magnetic resonance
angiography in systemic sclerosis. Arthritis Rheum, 2007. 56(8): p. 2747-54.
12. Kim, Y.H., et al., Classification of Raynaud's disease based on angiographic features. J Plast
Reconstr Aesthet Surg, 2011. 64(11): p. 1503-11.
13. McMahan, Z.H. and F.M. Wigley, Raynaud’s phenomenon and digital ischemia: a practical
approach to risk stratification, diagnosis and management. Int J Clin Rheumtol, 2010. 5(3): p. 355-70.
14. Sarkola, T., et al., Transcutaneous very-high-resolution ultrasound to quantify arterial wall
layers of muscular and elastic arteries: validation of a method. Atherosclerosis, 2010. 212(2): p. 516-23.
15. Horos Project (2015). . 2015: http://www.horosproject.org/ p. DICOM image viewing and measuring. .
16. O'Leary, D.H., et al., Carotid-artery intima and media thickness as a risk factor for
myocardial infarction and stroke in older adults. Cardiovascular Health Study Collaborative Research Group. N Engl J Med, 1999. 340(1): p. 14-22.
17. Bots, M.L., et al., Common carotid intima-media thickness and risk of stroke and
myocardial infarction: the Rotterdam Study. Circulation, 1997. 96(5): p. 1432-7.
18. Mourad, J.J., et al., The wall to lumen ratio of the radial artery in patients with Raynaud's
phenomenon. J Vasc Res, 1997. 34(4): p. 298-305.
19. Persu, A., et al., Revisiting Fibromuscular Dysplasia: Rationale of the European
Fibromuscular Dysplasia Initiative. Hypertension, 2016. 68(4): p. 832-9.
20. Bozzani, A. and V. Arici, Diagnosis and management of fibromuscular dysplasia, in J
Hypertens. 2014: England. p. 2098.
21. Boutouyrie, P., et al., Evidence for carotid and radial artery wall subclinical lesions in renal
fibromuscular dysplasia. J Hypertens, 2003. 21(12): p. 2287-95.
