Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Apr 9.
Published in final edited form as: Ultrasound Med Biol. 2018 Jul 6;44(10):2081–2088. doi: 10.1016/j.ultrasmedbio.2018.04.017

PHOTOACOUSTIC OXYGENATION QUANTIFICATION IN PATIENTS WITH RAYNAUD’S: FIRST-IN-HUMAN RESULTS

JOHN R EISENBREY *, MARIA STANCZAK *, FLEMMING FORSBERG *, FABIAN A MENDOZA-BALLESTEROS †,‡, ANDREJ LYSHCHIK *
PMCID: PMC8994565  NIHMSID: NIHMS1784929  PMID: 30207278

Abstract

The purpose of this study was to investigate the use of photoacoustic imaging for quantifying fingertip oxygenation as an approach to diagnosing and monitoring Raynaud’s phenomenon. After 30 min of acclimation to room temperature, 22 patients (7 patients with secondary Raynaud’s associated to Scleroderma and 15 healthy controls) provided informed consent to undergo fingertip Doppler imaging and high-frequency photoacoustic imaging before and 5, 15 and 30 min after cold stimulus (submerged hand in a 15 °C water bath for 1 min). High-frequency ultrasound and photoacoustic imaging was performed on the nail bed of each patient’s second through fifth finger on their dominant hand, using a Vevo 2100 LAZR system with an LZ-250 probe (Fujifilm VisualSonics, Toronto, ON, Canada) in oxy-hemoglobin quantification mode. During each exam, volumetric data across a 3-mm span of data was acquired to produce a volumetric image of percent oxygenation and hemoglobin concentration. Changes in fingertip oxygenation between Raynaud’s patients and healthy volunteers were compared, using receiver operator characteristic (ROC) analysis. Photoacoustic signal was detected in both the nail bed and nailfold in all study participants. Doppler ultrasound resulted in poor differentiation of Raynaud’s patients from healthy volunteers, with an area under the ROC curve (Az) of 0.51. Photoacoustic imaging demonstrated improved accuracy at baseline (Az = 0.72), which improved when quantifying normalized changes after cold stimulus (Az = 0.89 5-min post stimulus, Az = 0.91 15-min post stimulus, and Az = 0.85 after stimulus). Oxygenation levels derived using photoacoustic imaging are able to identify patients with Raynaud’s and safely evaluate their response to a cold stimulus over time.

Keywords: Photoacoustic imaging, Raynaud’s, Tissue oxygenation

INTRODUCTION

Raynaud’s phenomenon (RP) is a debilitating disorder that can lead to finger and toe amputations in severe cases. It is caused by an exaggerated peripheral vasospasm in response to cold or emotional stress. The subsequent peripheral blood flow reduction is clinically characterized by a triphasic color change (blanching, cyanosis, and erythema) and, although RP is generally inconsequential for patients with no additional pathologies (primary RP), it a disabling disorder for patients with connective tissue disease, such as scleroderma (Maricq et al. 1996). RP is almost a universal finding in patients with scleroderma and is driven by endothelial dysfunction, intimal fibroproliferation and vessel lumen occlusion (Maricq et al. 1996). It is present in more than 90% of patients with scleroderma (Herrick 2008; Maricq et al. 1996). RP can be seriously disabling, leading to digital ulcers, finger necrosis and amputation or associated complications, such as infection and osteomyelitis. Digital ulcerations can be present in up to 30% of the patients with systemic sclerosis-RP and is an expression of its clinical severity (Herrick 2008; Maricq et al. 1996).

Unfortunately, there are very few direct methods to detect and quantify peripheral microangiopathy in RP (Andrade et al. 1990; Korn et al. 2004; Merkel et al. 2002). Widefield nailfold capillaroscopy (NFC) and video capillaroscopy (Andrade et al. 1990) are the most commonly utilized methods. In healthy individuals and those with primary Raynaud’s, the NFC pattern is characterized by capillary loops that are similar in size and shape and demonstrate uniform distribution (Cutolo et al. 2003). On the other hand, patients with active NFC abnormalities present with abnormally dilated capillary loops (wire loops) that progress to late vasculopathy with disorganized and missing capillary loops and varying degrees of microhemorrhages (Maricq 1981). However, NFC and video capillaroscopy as used in the clinical setting do not provide quantitative measurements, are intrinsically observer dependent and only measure structural capillary abnormalities of the subjacent vasculopathy, but not the severity of the blood flow impairment (Herrick 2008; Maricq 1981). Activity and severity of RP in clinical trials have been measured by the Raynaud’s Condition Score and other patient-reported outcomes, such as severity, duration and frequency of attacks. However, the intrinsic subjectivity of these measurements, based in patient-reported outcomes, stress the need for an objective, noninvasive measure of digital blood flow and tissue oxygenation.

Ultrasonography with Doppler evaluation of peripheral blood flow is commonly used for diagnostic evaluation in patients with a wide variety of connective tissue disorders. Unfortunately, relatively low resolution (~0.2 mm) of current clinical scanners limits its use in studying circulation in small body parts, such as fingers. In addition, quantification of these findings is often difficult and user dependent, resulting in similar drawbacks to those associated with NFC. Photoacoustic (PA) imaging is an emerging imaging modality and PA systems are now commercially available for preclinical research (Becker et al. 2018; Eisenbrey et al. 2015; Needles et al. 2013). Using an optical laser, light (generally in the 650–970 nm wavelength range) is directed through the tissue from a tunable laser. As this light is absorbed, tissue chromophores undergo localized thermal expansion, which in turn generate acoustic signatures that can be detected by an ultrasound transducer. One such measurement is the oxygenation of blood by comparing the ratio of absorption at multiple wavelengths (the absorption spectra of hemoglobin alters based on its oxygenation state). Such optical absorption measurements form the basis of clinically used pulsed-ox measurements (Yoshiya et al. 1980). However, PA benefits from the same specificity of optical imaging, along with the resolution and increased penetration depths of ultrasound, while providing tissue oxygenation information throughout the imaging plane. This modality has been applied to a variety of preclinical models (Eisenbrey et al. 2015; Mallidi et al. 2011; Needles et al. 2013) and has more recently shown some promise in early human trials of cancer detection and characterization (Mallidi et al. 2011). However, PA has yet to be applied as a tool for imaging and quantification of vascular dysfunction of connective tissue diseases in humans.

Quantification of microperfusion and oxygenation remains an important goal in systemic sclerosis-RP. Magnetic resonance angiography has been used to successfully quantify peripheral circulation in RP, but this technique requires an injection of contrast dye and extensive magnetic resonance imaging (MRI) workup, which may limit its applicability for screening (Park et al. 2014; Zhang et al. 2011). Other optical methods such as those using laser-speckle tracking have also been proposed to quantify microperfusion, but they do not quantify tissue oxygenation levels or provide anatomic registration (Leahy et al. 2007; Zhang et al. 2017). Hence, the purpose of this pilot study was to explore the feasibility of PA imaging in RP patients and to compare oxygen quantification from this patient population with healthy volunteers.

MATERIALS AND METHODS

Patient enrollment

All participants (both RP patients and healthy volunteers) provided informed consent to participate in this institutional review board-approved and HIPPA-compliant study. Inclusion criteria for both groups consisted of the subject being 18 y of age or older and being able to provide informed consent. Within the RP group, patients were required to have a documented history of severe scleroderma-related (secondary) RP history of finger ulceration. High-resolution video capillaroscopy was performed by an expert (F.A.M.-B.) in each RP patient (as part of their clinical standard of care) before enrolling to confirm severe scleroderma-related RP. Study exclusion criteria for all patients consisted of nail polish (other than clear), active smoking, occupational exposure to a cold environment, occupational exposure to strong mechanical vibrations, uncontrolled systemic arterial hypertension or diabetes mellitus and clinical evidence of proximal arterial disease. Healthy volunteers from Thomas Jefferson University Hospital (Philadelphia, PA, USA) were approached between November and December 2016. Patients with RP were consecutively approached at The Jefferson Scleroderma Center by Dr. Mendoza-Ballesteros between December 2016 and April 2017. Before participation, all patients were asked to rest for 30 min in the scanning room to ensure temperature acclimation (21°C).

A total of 15 healthy volunteers and 7 RP patients (out of 13 approached patients, 6 declined to participate) were enrolled and successfully completed the study protocol. The sex breakdown of the healthy volunteer group consisted of 67% women (10/15). A total of 80% were Caucasian (12/15), 13% Asian (2/15), and 7% (1/15) African American. The average age in this group was 53.4 ± 5.6 y (range = 41–62 y). The RP group consisted of 86% women (6/7). A total of 71% (5/7) were Caucasian and the remaining 29% (2/7) were African American. The average age in this group was 47.2 ± 13.1 y (range = 32–63 y).

Baseline doppler ultrasound

Color Doppler ultrasound was performed by a registered sonographer (M.S.) with more than 20 y of experience in vascular ultrasound. Scanning was performed using an Aplio 500 Platinum scanner with an 18 L7 probe (Toshiba Medical, Tustin, CA, USA) imaging at 14 MHz. With the top of the hand resting on an exam table, three or four representative Doppler images were acquired from each finger of each participant’s dominant hand, using the small parts preset to visualize vascularity within the fingertips. Consistent with clinical standards, all imaging parameters were optimized by the sonographer to minimize Doppler artifacts, and still allowing detection of microvessel (estimated microvessel limit of detection = 150 μm; pulse repletion frequency range 13.2–19.5 kHz, color Doppler gain range of 5–40, dynamic range = 65–70 dB). Images were then stored for later off-line data analysis.

High-frequency ultrasound and photoacoustic imaging

High-frequency ultrasound and photoacoustic imaging was performed to quantify tissue oxygenation of all participants at both baseline and after cold stimulus. High-frequency ultrasound and photoacoustic imaging was performed using a Vevo 2100 LAZR system with an LZ250 photoacoustic probe (Fujifilm VisualSonics, Toronto, ON, Canada) with a center frequency of 18 MHz and optical wavelength range of 680–970 nm. In collaboration with and with approval from the university’s laser safety office, this preclinical imaging system was modified to enable imaging without the laser safety enclosure (as photoacoustic imaging systems are not yet approved for clinical use). Modification resulted in the system being classified as a Class 4 laser, requiring laser protective eyewear for personnel and participants in the room during scanning. Oxygenation measurements were performed using the system’s built-in oxy-hemoglobin package, which generates a PA signal at transmit wavelengths of 750 nm and 850 nm and computes a percent hemoglobin oxygenation by comparing these PA signal at these two wavelengths (Needles et al. 2013). After application of acoustic coupling gel, imaging was performed through sagittal scanning of the nail bed (through the fingernail), cuticle, and nailfold from each finger of the participant’s dominant hand, with their palm facing down on a custom brace. All PA imaging was acquired in 3-D at a step size of 0.18 mm, using a stepper motor to cover 3-mm across the nail bed. Photoacoustic imaging was obtained at baseline (immediately after Doppler imaging) and at 5, 15 and 30 min after cold stimulus. All volumetric data was then stored for later offline analysis.

Cold stimulus

Cold stimulus was performed by participants placing their dominant hand in 15°C water for 1 min under direct observation. During this period, participants were monitored for any signs of Raynaud’s crisis flare-up or general distress. After stimulus, each participant’s hand was quickly dried and returned to the imaging stage for analysis.

Data analysis

Doppler ultrasound was evaluated by a board-certified radiologist (A.L.), with more than 15 y of experience in diagnostic ultrasound. The radiologist was blinded to the participant group (RP patient or healthy volunteer) and asked to count the number of detected microvessels in each image (sensitivity of detection approximately 150-μm diameter vessels). The number of detected microvessels from each image was then averaged for each participant as an indicator of vascular supply. Volumetric PA data was analyzed using stand-alone Vevo Labs 2.2.0 analysis software (FujFilm VisualSonics). Briefly, a region of interest was placed encompassing the entire nail bed of each fingertip up to a depth of 11 mm in each plane from the volume. The region of interest selection was based on the nail bed providing the most consistent source of PA signal, with scanning and a signal depth limitation occurring at approximately 11 mm. Percent tissue oxygenation was then calculated by the analysis software across the entire acquired volume and averaged. Quantitative data from each dataset (baseline Doppler; PA at baseline; PA 5 min post cold stimulus; PA 15 min post cold stimulus; and PA 30 min post cold stimulus) was averaged across all four fingers. The number of detected vessels in Doppler imaging, baseline PA oxygenation levels and normalized changes in oxygenation (computed as a percentage relative to baseline) were compared between healthy volunteers and Raynaud’s patients. Statistical analysis was performed in GraphPad Prism v. 5 (GraphPad Software, San Diego, CA, USA) with p < 0.05 considered statically significant and receiver operator characteristic (ROC) curves plotted in Stata v. 12.0 (StataCorp LLC, College Station, TX, USA).

RESULTS

No adverse events were observed during imaging or cold stimulus. Doppler images were acquired from all 22 enrolled participants and demonstrated at least some vascularity in all participants, albeit with a high degree of variability from plane to plane and finger to finger. On average, 2.6 ± 1.5 vessels were detected per finger, with no significant difference between healthy volunteers and Raynaud’s patients (p = 0.8). Two example Doppler images are presented in Figure 1, showing numerous (top image) and little (bottom image) microvessel detection from the same RP patient. Notice in the image with limited vessel detection the presence of motion artifacts, particularly at the fingertip (left edge of figure) and additional artifact outside the finger itself.

Fig. 1.

Fig. 1.

Open in a new tab

Doppler images of the distal and middle phalanx from the same patient, (a) identifying multiple microvessels, but (b) limited detection from an alternate plane in the same finger. Doppler flow within the region of interest was used to detect directional flow in microvessel in both the distal phalanx (red and blue scale). In (a), nine microvessels were counted by the blinded radiologist and, in (b), only one vessel was counted.

High-frequency ultrasound and PA demonstrated improved resolution relative to the clinical system (Fig. 1) of the finger anatomy, including the nail bed, cuticle, and nailfold of a healthy volunteer (Fig. 2a). Photoacoustic imaging showed strong PA signal in both the nailbed and nailfold at depths up to 11 mm (Fig. 2b). Notice the high levels of oxygenation within the nail bed (red) and limited or no detection and artifact within the nail itself. It was also observed that the nail bed provided more consistent and homogenous PA signal detection compared with the nailfold, and this area was therefore chosen for subsequent analysis.

Fig. 2.

Fig. 2.

Open in a new tab

(a) High-frequency B-mode image and (b) PA image of a healthy volunteer. High resolution was observed on B-mode imaging, enabling visualization of the nail bed, root and nailfold. Strong PA signal (bottom) was observed in both the nail bed and nailfold, but with greater homogeneity and filling in the nail bed (with the color scale in the region of interest, denoting percent hemoglobin oxygenation). A region of interest was mapped and used to quantify percent oxygenation in the nail bed.

After cold stimulus, limited-to-no changes in finger oxygenation were observed in the healthy volunteers. Raynaud’s patients demonstrated more pronounced changes in oxygenation levels in both the nailfold and nail bed, albeit with much greater variability within the group. An example of these changes over time is provided in Figure 3, showing a severe RP patient (left) column and a healthy volunteer (right). Each set of images was taken from the ring finger at baseline, 5-min post cold stimulus, 15-min post cold stimulation and 30-min post cold stimulation. While oxygenation levels (%) were lower at baseline in the patient with RP, this was highly variable among RP patients. More consistent was the drop in oxygenation following cold stimulus only in RP patients as shown in the series of images.

Fig. 3.

Fig. 3.

Open in a new tab

PA images from an RP patient (left) and a healthy volunteer (right) at baseline (top row) and at 5-min (second row), 15-min (third row) and 30-min (bottom row) post cold stimulus. Consistent with all healthy volunteers, no decreases in finger oxygenation were observed over time. However, RP patients demonstrated decreases in oxygenation in response to cold stimulus. Provided numbers indicate the percent oxygenation quantified in the nail bed, using photoacoustic imaging.

Receiver operator characteristic (ROC) curves were generated for each modality and time point to identify each set’s accuracy and optimal threshold point in identifying patients with RP. These results are presented in Figure 4. Quantification of the number of microvessels did a poor job in differentiating RP patients from healthy controls (consistent with the results in Fig. 1) with an area under the ROC curve (Az) of 0.51 and p = 0.94. Photoacoustic imaging before cold stimulus resulted in improved accuracy with Az = 0.72 (p = 0.10). Normalized changes in oxygenation 5-min post cold stimulus resulted in a statistically significant differentiation of RP patients from healthy controls with Az = 0.89 (p = 0.004). Using the ROC curve, the optimal threshold point, defined as the point closest to 100% sensitivity and specificity, was a net change of −0.001% (essentially any drop in oxygenation) with a sensitivity of 85.7% (95% confidence interval [CI] of 42.1%–99.6%) and specificity of 80% (CI = 51.9%–95.7%). Normalized changes in oxygenation 15-min post cold stimulus also demonstrated significant differentiation with Az = 0.91 (p = 0.003). This curve resulted in an identical optimal threshold value with a sensitivity of 85.7% (CI = 42.1%–99.6%) and specificity of 93.3% (CI = 68.1%–99.9%). Finally, changes in oxygen levels 30-min post cold stimulus also showed significant differentiation among patient groups with Az = 0.85 (p = 0.01). At this time point, a threshold of −0.003% was determined to be optimal, with a sensitivity of 71.4% (CI = 29.4%–96.3%) and specificity of 93.3% (CI = 68.0%–99.8%).

Fig. 4.

Fig. 4.

Open in a new tab

Receiver operator characteristic curves, differentiating RP patients from healthy volunteers. Doppler provided an Az of 0.51, PA at baseline resulted in Az = 0.72, PA 5-min post cold stimulus resulted in Az = 0.89, PA 15-min post cold stimulus resulted in Az = 0.91 and PA 30-min post cold stimulus provided an Az = 0.85.

DISCUSSION

In this pilot study, we demonstrate the usefulness of PA imaging for both identifying RP patients and tracking the response of fingertip oxygenation to cold stimulus. Figure 4 demonstrates that, although PA imaging alone may be somewhat useful for identifying RP in patients (Az = 0.72), the true benefit comes from tracking oxygenation response from cold stimulus (with Az ranging 0.85–0.91). The limited value of the modality with no cold stimulus is attributed to the higher variability among the RP group, potentially because of previous RP complications or rate of acclimation to room temperature (for the 30-min waiting period). Normalization to base-line levels and tracking response to cold stimulus appears to limit this variability and improve on the technique’s sensitivity and specificity. Of note, the optimal thresholding point based on the three ROC curves was close to 0% change, indicating that RP patients undergo drops in oxygenation after cold stimulus, while healthy volunteers show no change or slight increases. Also of note, Doppler imaging at the baseline provided no value in differentiating RP patients from healthy volunteers. This is not surprising given the high degree of variability within each individual (demonstrated in Fig. 1 and by a standard deviation of roughly 75% of the mean), and an inability to differentiate actual low flow vessels from artifact in many cases. As a result of this variability and the time constraints during scanning, multiple follow-up time points after cold stimulus were not investigated with Doppler imaging.

Although an encouraging proof of concept, this study has several limitations that should be addressed. Sample sizes were relatively small, particularly the number of RP patients (n = 7), and findings should be replicated in a larger population moving forward. Severe RP patients with a history of ulceration were selected as an experimental group to best demonstrate variations with healthy volunteers. Now that clinical validation has occurred, less severe RP cases should be investigated because the clinical benefit of PA diagnosis will be more significant in patients before severe RP onset. Doppler imaging was performed with a separate probe because our experience has shown this system (Toshiba 500) to be more sensitive to lower flow detection than the Vevo 2100 system. In addition, a separate plane was used for Doppler acquisition to avoid transmission through the fingernail. In the future, more sensitive Doppler approaches could be used to increase the limits of microvessel detection within the same plane as photoacoustic acquisition. For example, Superb Micro-Vascular Imaging (Toshiba) is an advanced Doppler approach (Machado et al. 2016) and has demonstrated improved detection of synovial vascularity compared with power Doppler in patients with rheumatoid arthritis (Lim et al. 2018; Orlandi et al. 2017). Additionally, changes in Doppler signal and temperature should be monitored post cold stimulus to better determine the influence of tissue perfusion, but this was logistically difficult in this pilot study, which focused on feasibility and optimal image-acquisition time after cold stimulus.

As a modality, PA imaging continues to move toward clinical trials and potential clinical usage. To date, the majority of clinical trials using PA have focused on cancer imaging (primarily of the breast) (Mallidi et al. 2011). A drawback to the widespread clinical adoption has been the limited penetration of light reducing (or eliminating) the generated PA signal, which in turn limits imaging to relatively superficial structures. Improvements in optics and equipment design, however, have shown some promise to overcoming these limitations in breast imaging, with encouraging, albeit preliminary, results (Kita et al. 2014; Kruger et al. 2010; Monohar et al. 2007). More recently, PA imaging has begun to be explored in dermatology, showing promising results for quantifying oxygenation in skin ulcers (Petri et al. 2018), visualization of skin cancer (Zeitouni et al. 2015) and the quantification of port wine stain severity (Viator et al. 2002). PA has also recently been used to study the curvature in the palmar arteries (Matsumoto et al. 2018), but this has only been performed in healthy volunteers. To our knowledge, the work presented here is the first to use PA imaging for quantification of tissue oxygenation in RP. Earlier studies have been reported using PA for arthritis imaging, but these have only involved imaging in healthy volunteers to establish feasibility (Liu et al. 2016; van Es et al. 2014). Hence, this is the first report to demonstrate clinical value of PA for identifying patients with rheumatological disease.

Raynaud’s is a debilitating disease caused by peripheral vasospasm in response to cold or emotional stress and often secondary to systemic fibrotic disease such as scleroderma. Clinically, diagnosis and management is performed via NFC or video analysis, which can be user dependent and requires highly specialized expertise. For example, although some studies have shown high accuracy and repeatability of both NFC and video capillaroscopy (Sekiyama et al. 2013), others have shown that quantification of this data often results in limits of agreement up to only 25%–50% (Bukhari et al. 2000). Quantitative imaging approaches such as PA may be well-positioned to overcome these limitations, enable visualization of response (be it cold stimulation or treatment) in real time.

CONCLUSIONS

RP is a debilitating disorder that can rapidly progress if not properly diagnosed and treated. However, current diagnostic methods are highly variable and require specialized expertise for qualitative diagnosis. In this pilot study, PA imaging was shown to successfully differentiate RP patients from healthy volunteers. This work justifies future studies into the use of PA for RP screening in larger clinical trials as well as for the monitoring of treatment response.

Acknowledgments

This work was supported in part by NIH S10 OD010408. Equipment was provided by Toshiba Medical but authors have sole control of the data.

REFERENCES

  1. Andrade LEC, Gabriel Júnior A, Assad RL, Ferrari AJ, Atra E. Panoramic nailfold capillaroscopy: A new reading method and normal range. Semin Arthritis Rheum 1990;20:21–31. [DOI] [PubMed] [Google Scholar]
  2. Becker A, Masthoff M, Clasussen J, Ford ST, Roll W, Burg M, Barth PJ, Heindel W, Schafers M, Eisenblatter M, Wildrubger M. Multispectral optoacoustic tomography of the human breast: Characterisation of healthy tissue and malignant lesions using a hybrid ultrasound-optoacoustic approach. Eur Radiol 2018;28:602–609. [DOI] [PubMed] [Google Scholar]
  3. Bukhari M, Hollis S, Moore T, Jarson MI, Herrick AL. Quantitation of microcirculatory abnormalities in patients with primary Raynaud’s phenomenon and systemic sclerosis by video capillaroscopy. Rheumatology 2000;39:506–512. [DOI] [PubMed] [Google Scholar]
  4. Cutolo M, Grassi W, Matucci Cerinic M. Raynaud’s phenomenon and the role of capillaroscopy. Arthritis Rheum 2003;48:3023–3030. [DOI] [PubMed] [Google Scholar]
  5. Eisenbrey JR, Merton DA, Marshall A, Liu JB, Fox TB, Sridharan A, Forsberg F. Comparison of photoacoustically derived hemoglobin and oxygenation measurements with contrast-enhanced ultrasound estimated vascularity and immunohistochemical staining in a breast cancer model. Ultrason Imaging 2015;37:42–52. [DOI] [PubMed] [Google Scholar]
  6. Herrick AL. Diagnosis and management of scleroderma peripheral vascular disease. Rheum Dis Clin North Am 2008;34:89–114. [DOI] [PubMed] [Google Scholar]
  7. Kita T, Torrii M, Sugie T, Kanao S, Mikami Y, Shiina T, Toi M. Photoacoustic mammography: Initial clinical results. Breast Cancer 2014;21:146–153. [DOI] [PubMed] [Google Scholar]
  8. Korn JH, Mayes M, Matucci Cerinic M, Rainisio M, Pope J, Hachilla E, Rich E, Carpentier P, Molitor J, Seibold JR, Hsu V, Guillevin L, Chatterjee S, Peter HH, Coppock J, Herrick A, Merkel PA, Simms R, Denton CP, Furst D, Nguyen N, Gaitonde M, Black C. Digital ulcers in systemic sclerosis: Prevention by treatment with bosentan, an oral endothelin receptor antagonist. Arthritis Rheum 2004;50:3985–3993. [DOI] [PubMed] [Google Scholar]
  9. Kruger RA, Lam RB, Reinecke DR, Del Rio SP, Doyle RP. Photoacoustic angiography of the breast. Med Phys 2010;37:6069–6100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Leahy M, Enfield JG, Clancy NT, O’Doherty J, McNamara P, Nilsson GE. Biophotonic methods in microcirculation imaging. Med Laser Appl 2007;22:105–126. [Google Scholar]
  11. Lim AKP, Satchithananda K, Dick EA, Abraham S, Cosgrove DO. Microflow imaging: New Doppler technology to detect low-grade inflammation in patients with arthritis. Eur Radiol 2018;28:1046–1053. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Liu Y, Wang Y, Yuan Z. Dual-modality imaging of the human finger joint systems by using combined multispectral photoacoustic computed tomography and ultrasound computed tomography. Biomed Res Int 2016;2016 1453272. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Machado P, Segal S, Lyshchik A, Forsberg F. A novel microvascular flow technique: Initial results in thyroids. Ultrasound Q 2016;32:67–74. [DOI] [PubMed] [Google Scholar]
  14. Mallidi S, Luke GP, Emelianov E. Photoacoustic imaging in cancer detection, diagnosis, and treatment guidance. Trends Biotechnol 2011;29:213–221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Maricq HR. Wide-field capillary microscopy. Arthritis Rheum 1981;24:1159–1165. [DOI] [PubMed] [Google Scholar]
  16. Maricq HR, Weinrich MC, Valter I, Palesch YY, Maricq JG. Digital vascular responses to cooling in patients with cold sensitivity, primary Raynaud’s phenomenon, or scleroderma spectrum disorders. J Rheumatol 1996;23:2068–2078. [PubMed] [Google Scholar]
  17. Matsumoto Y, Asao Y, Yoshikawa A, Sekiguchi H, Takad M, Furu M, Saito S, Kataoka M, Abe H, Yagi T, Togashi K, Toi M. Label-free photoacoustic imaging of human palmar vessels: A structural morphologic analysis. Sci Rep 2018;8:786. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Merkel PA, Herlyn K, Martin RW, Anderson JJ, Mayes MD, Bell P, Korn JH, Simms RW, Csuka ME, Medsger TA, Rothfield NF, Ellman MH, Collier DH, Weinstein A, Furst DE, Jimenez SA, White B, Seibold JR, Wigley FM. Measuring disease activity and functional status in patients with scleroderma and Raynaud’s phenomenon. Arthritis Rheum 2002;46:2410–2420. [DOI] [PubMed] [Google Scholar]
  19. Monohar S, Vaartjes SE, van Hespen JC, Klasse JM, van den Engh FM, Steenbergen W, van Leeuwen TG. Initial results of in vivo non-invasive cancer imaging in the human breast using near-infrared photoacoustics. Opt Express 2007;15:12277–12285. [DOI] [PubMed] [Google Scholar]
  20. Needles A, Heinmiller A, Sun J, Theodororopoulos C, Bates D, Hirson D, Yin M, Foster FS. Development and initial application of a fully integrated photoacoustic micro-ultrasound system. IEEE Trans Ultrason Ferroelectr Freq Control 2013;60:888–897. [DOI] [PubMed] [Google Scholar]
  21. Orlandi D, Gitto S, Perugin Bernardi S, Corazza A, De Flaviis L, Silvestri E, Cimmino MA, Sconfienza LM. Advanced power Doppler technique increases synovial vascularity detection in patients with rheumatoid arthritis. Ultrasound Med Biol 2017; 43:1880–1887. [DOI] [PubMed] [Google Scholar]
  22. Park JK, Park EA, Lee W, Kim YK, Lee EY, Song YW, Lee EB. Base-line MRA predicts the treatment response to vasodilator udenafil in patients with secondary Raynaud’s phenomenon. Clin Exp Rheumatol 2014;32(Suppl 86):S167–S170. [PubMed] [Google Scholar]
  23. Petri M, Stoffels I, Griewank K, Jose J, Engels P, Schulz A, Potzschke H, Jansen P, Schadendorf D, Dissemond J, Klode J. Oxygenation status in chronic leg ulcer after topical hemoglobin application may act as a surrogate marker to find the best treatment strategy and to avoid ineffective conservative long-term therapy. Mol Imaging Biol 2018;20:124–130. [DOI] [PubMed] [Google Scholar]
  24. Sekiyama JY, Camargo CZ, Eduardo L, Andrade C, Kayser C. Reliability of widefield nailfold capillaroscopy and video capillaroscopy in the assessment of patients with Raynaud’s phenomenon. Arthritis Care Res (Hoboken) 2013;65:1853–1861. [DOI] [PubMed] [Google Scholar]
  25. van Es P, Biswas SK, Bernelot Moens HJ, Steenbergen W, Manohar S. Initial results of finer imaging using photoacoustic computed tomography. J Biomed Opt 2014;19 060501. [DOI] [PubMed] [Google Scholar]
  26. Viator JA, Au G, Paltauf G, Jacques SL, Prahl SA, Ren H, Chen Z, Nelson JS. Clinical testing of a photoacoustic probe for port wine stain depth determination. Lasers Surg Med 2002;30:141–148. [DOI] [PubMed] [Google Scholar]
  27. Yoshiya I, Shimada Y, Tanaka K. Spectrophotometric monitoring of arterial oxygen saturation in the fingertip. Med Biol Eng Comput 1980;18:27–32. [DOI] [PubMed] [Google Scholar]
  28. Zeitouni NC, Rohrbach DJ, Aksahin M, Sunar U. Preoperative ultrasound and photoacoustic imaging of nonmelanoma skin cancers. Dermatol Surg 2015;41:525–528. [DOI] [PubMed] [Google Scholar]
  29. Zhang L, Ding L, Li M, Zhang X, Su D, Jia J, Miao P. Dual-wavelength laser speckle contrast imaging (dwLSCI) improves chronic measurement of superficial blood flow in hands. Sensors (Basel) 2017;17(12):E2811. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Zhang W, Xu JR, Lu Q, Ye S, Liu XS. High-resolution magnetic resonance angiography of digital arteries in SSc patients on 3 Tesla: Preliminary studies. Rheumatology (Oxford) 2011;50:1712–1719. [DOI] [PubMed] [Google Scholar]

RESOURCES