A case description of non-invasive plantar foot perfusion assessment pre-and post-angioplasty in patients with peripheral arterial disease using spatial frequency domain imaging

Introduction

Peripheral arterial disease (PAD) is a progressive atherosclerotic condition characterized by arterial narrowing or blockage to the lower extremities, typically seen in diabetic patients. When severe, PAD can lead to chronic limb-threatening ischemia, which causes pain and non-healing ulcers, which significantly increases the risk of limb amputation (1). Percutaneous transluminal angioplasty (PTA) is a standard revascularisation strategy to restore blood flow and relieve ischemic symptoms. For those with wounds, improved perfusion is essential to achieve wound closure (2). Technical success for PTA procedure is defined as less than 30% residual stenosis in the treated vessel, which is assessed via angiography with the use of contrast or carbon dioxide (3). While blood flow to foot can be evaluated, angiography cannot be used to quantify perfusion. Hence, accurately assessing the tissue oxygenation status following the procedure remains a challenge.

Current assessments primarily rely on ankle-brachial index (ABI), toe-brachial index (TBI), transcutaneous oxygen pressure (TcPO₂), continuous wave doppler, and clinical observation. However, these methods are not without limitations. In PAD patients with diabetes, arterial wall calcification causes blood vessels to be incompressible, which in turn results in unreliable ABI readings (4). Other contact-based procedures such as TBI and near-infrared spectroscopy (NIRS) are not suitable for patients with pain, severe ulcerations, or prior amputations, and have poor reproducibility. While NIRS can provide real-time perfusion data, it is a point-based method that does not generate spatial maps of tissue perfusion (5). Near-infrared (NIR) fluorescence imaging can offer real-time perfusion assessment, but requires injection of indocyanine green, which although non-nephrotoxic, is contraindicated in patients with liver dysfunction (6). These limitations highlight the need for a non-invasive, objective, and quantitative method to assess tissue oxygenation pre- and post-angioplasty.

Being able to assess intraprocedural perfusion has the potential to guide revascularisation strategies, which can be targeted to improve blood flow to angiosome-related artery. The impact of collateralization between angiosomes can also be assessed in real-time. Treating the source artery of targeted angiosome has the potential to improve wound healing and limb salvage rates (7,8).

Spatial frequency domain imaging (SFDI) is an emerging non-invasive optical imaging technique that quantifies tissue oxygen saturation (StO₂) by analyzing tissue absorption and scattering properties (9,10). It provides spatial maps of tissue oxygenation over a large area (15 cm × 20 cm) up to a depth of 3 mm, making it a promising tool for assessing microvascular perfusion changes following revascularisation (9). Unlike NIR fluorescence imaging, SFDI does not require contrast agents, making it safer and more accessible for real-time clinical applications. While SFDI has been used to study the wound healing progression before and after angioplasty in two subjects (11), this case report presents pre- and post-angioplasty StO₂ values in three PAD patients of varied Asian ethnicity, and compared with clinical angiogram images, demonstrating the potential of SFDI as an objective tool for evaluating procedural improvement and guide post-intervention management.

Case presentation

Ethics approval was obtained from SingHealth Centralised Institutional Review Board (CIRB Ref 2021/2648). All procedures performed in this study were in accordance with the ethical standards of SingHealth Centralised Institutional Review Board, and with the Helsinki Declaration and its subsequent amendments. Written informed consent was obtained from the patients for publication of this article and accompanying images. A copy of the written consent is available for review by the editorial office of this journal. In this single-centre prospective observational case series, we report the results of SFDI imaging performed for three patients presenting with symptomatic PAD at a tertiary hospital in Singapore. All patients planned for elective lower limb angioplasties, except for those with heel ulcers, forefoot amputations or contralateral major amputation, were eligible for enrolment. StO₂ measurements were obtained before and after the procedure, for both the control leg and the angioplasty leg, to assess changes in tissue oxygenation and to account for potential confounding factors that could influence the SFDI images during angioplasty. Pre-angioplasty measurements were conducted in the patient preparation room, while post-angioplasty measurements were performed in the operating theatre immediately after the procedure, following wound debridement and cleaning. Tissue oxygenation was assessed using the SFDI system (Reflect RS^®, Modulim Inc., Irvine, California, USA) as shown in Figure 1A,1B. The device employs LED illumination at 470, 525, 625, and 971 nm with sinusoidal patterns to project single spatial frequencies. Multiple scattering and absorption within the tissue reduces the amplitude of the projected wave, and the resulting diffused pattern is captured by a digital camera to generate reflectance images (10). The system is user-friendly, requiring only an initial calibration with a white reflectance standard to correct for instrument response and extract demodulated signals from the sample. Imaging was performed separately on the heel and pad regions of the plantar foot, which is commonly analysed in clinical studies (12,13), with a total imaging time of approximately five minutes or less per foot. For the analysis, heel and pad regions were processed separately, and background thresholding was applied using a contrast-based mask to eliminate background artifacts, as reported before (10). For better visualization, the heel and forefoot images are presented as merged in figure panels. Results are shown only for the leg undergoing angioplasty, as the control leg exhibited a decrease or a negligible change post-angioplasty in Cases 1 (pre: 0.14 & post: 0.11), 2 (pre: 0.69 & post: 0.73, and 3 (pre: 0.106 & post: 0.07, high melanin artifacts). These findings confirm that any increase in StO₂ observed in the SFDI images after angioplasty can be directly attributed to the procedural success. Some images exhibited blackened areas due to ambient light interference in the operation theatre, which were accounted for during post-processing.

Figure 1.

SFDI configuration. (A) Portable system equipped with a power supply and trolley for easy manoeuvrability. (B) Close-up view of the imaging head, featuring LED illumination and an integrated camera for capturing reflectance images. SFDI, spatial frequency domain imaging.

StO₂ increase in Case 1

Case 1 is a 61-year-old Indian male patient with Rutherford 5 disease for a left big toe wound. He was treated with PTA to popliteal, anterior tibial artery (ATA), peroneal, and dorsalis pedis artery (DPA). The pre-procedural StO₂ as measured by SFDI was critically low at 0.21, indicating severe ischemia. Following angioplasty, there was a remarkable change in StO₂ to 0.42 (+100%), suggesting improved perfusion (Figure 2A,2B). At the end of the procedure, there were two vessels run off to the foot as confirmed by angiography (Figure 2C,2D). Despite occlusion of post tibial and plantar artery, marked improvement is visually observed (as shown by increased yellow signal) in angiosomes at the heel and plantar of the treated limb after successful revascularisation of ATA, DPA and peroneal, possibly due to reperfusion of collaterals.

Figure 2.

Case 1 is a 61-year-old Indian male who presented with a left big toe wound. Original photograph and SFDI map (A) pre-angioplasty and (B) post-angioplasty. Image of angiographic run (C) pre-angioplasty and (D) post-angioplasty. SFDI, spatial frequency domain imaging; StO₂, tissue oxygen saturation.

Prior to discharge, he underwent a left big toe amputation. A repeat duplex ultrasound scan revealed restenosis at anterior tibial and DPA 3 months post-index angioplasty. A repeated intervention offered but subject was keen on conservative management instead as his claudication symptoms were resolved post-angioplasty. At last follow-up 9 months after index angioplasty, his wound has healed and is asymptomatic without any claudication or rest pain (Rutherford 0).

StO₂ decrease in Case 2

Case 2 is an 84-year-old Chinese female presenting with Rutherford 5 disease for a left shin ulcer. She underwent PTA to treat femoropopliteal disease that extended from the superficial femoral artery to P3 of the popliteal artery. The lesion was severely calcified and required concomitant atherectomy before angioplasty with a plain and a drug-coated balloon. There was one vessel runoff to the foot in the angiogram by the end of procedure. As measured by SFDI, unlike the expected trend, StO₂ decreased post-PTA from 0.61 to 0.31 (−49.2%) (Figure 3A,3B). These contrasting results may be attributed to slow flow phenomenon, which is commonly observed after use of paclitaxel drug-eluting balloon (DEB) during the procedure, potentially arising from particulate embolization (14). There was one vessel runoff to the foot in the angiogram by the end of procedure, which shows an inadequate increase in visual blood flow (Figure 3C,3D). Despite this, her left shin ulcer has epithelialized one-month post-op and did not require any dressing (Rutherford 0).

Figure 3.

Case 2 is an 84-year-old Chinese female who presented with a left shin ulcer. Original photograph and SFDI map (A) pre-angioplasty and (B) post-angioplasty. Image of angiographic run (C) pre-angioplasty and (D) post-angioplasty. SFDI, spatial frequency domain imaging; StO₂, tissue oxygen saturation.

In the event of discordant results, discrepancies should be reconciled with other clinical findings such as toe pressures. Serial measurements are needed to determine whether the decreased StO₂ represents a temporary post-intervention effect or an early indicator of impaired microvascular recovery.

StO₂ increase in Case 3

Case 3 is a 65-year-old Indian male who presented with a right second toe wound (Rutherford 5). PTA was performed on superficial femoral artery and ATA. SFDI showed substantial improvement in StO₂ from 0.16 to 0.46 (+187.5%), indicating successful revascularisation and improved tissue oxygenation post-PTA (Figure 4A,4B). There was one vessel runoff to foot in the angiogram at the end of the procedure (Figure 4C,4D). Like in Case 1, visible improvement (marked by increased yellow signal) to plantar is observed. Unfortunately, perfusion to angiosomes at toes does not appear to have much change visually.

Figure 4.

Case 3 is a 65-year-old Indian male who presented with a right 2nd toe wound. Original photograph and SFDI map (A) pre-angioplasty and (B) post-angioplasty. Image of angiographic run (C) pre-angioplasty and (D) post-angioplasty. SFDI, spatial frequency domain imaging; StO₂, tissue oxygen saturation.

At the last follow-up 4 months post-op, the wound remains unhealed and a repeat duplex ultrasound scan revealed re-occlusion of the ATA.

Discussion

These cases suggest that real-time insights into tissue oxygenation after endovascular revascularisation can be measured with SFDI. The increase in StO₂ in Cases 1 and 3 confirms the effectiveness of angioplasty in restoring perfusion. This observation is expected with the treatment of flow-limiting lesions. However, the post-procedural StO₂ decline in Case 2, who received DEB therapy, raises concerns. While DEB prevents restenosis by inhibiting neointimal hyperplasia, its immediate impact on microvascular oxygenation remains unclear (15). The decline in StO₂ may result from slow flow phenomenon, caused by transient drug particle embolization post-treatment with DEBs, which can obstruct downstream capillaries before clearing over time (14). Additionally, the eluted drug’s pharmacological effects may induce temporary vasoconstriction or endothelial dysfunction, impairing oxygen delivery (16). Another possibility is a delayed therapeutic response, where initial endothelial injury from the procedure precedes long-term vascular remodeling and improved perfusion. These findings underscore the complexity of assessing perfusion changes in patients undergoing DEB angioplasty, highlighting the need for longitudinal monitoring to distinguish transient effects from potential complications such as early restenosis or impaired tissue healing.

Using NIR imaging, Schremmer et al. [2024] reported a correlation between improved StO₂ specifically at wound area and reduction of wound area (17). However, the follow-up measurements used were not taken immediately post-op. Like Case 2, it was observed in several studies that some patients present with an initial reduction in StO₂ post-revascularisation, which later increased at follow-up one to four months after procedure, suggesting the possibility for changes in perfusion post-revascularisation not being immediately apparent (5,17,18). Future studies with longer follow-up are required to validate this observation.

While angiography remains the standard method for assessing perfusion changes, its interpretation is often subjective and dependent on the clinician’s expertise. SFDI emerges as a beneficial complementary tool by providing objective, quantitative perfusion metrics across a large imaging area in near real-time. Prior clinical studies have further demonstrated that SFDI provides rapid, reliable, and wide-field assessment of plantar foot perfusion in patients with and without diabetes, supporting its feasibility as a clinical tool for lower-limb vascular evaluation (9). Unlike conventional methods such as ABI and TcPO₂, which may not fully capture microvascular changes, SFDI generates spatially resolved StO₂ maps that can enhance post-procedural monitoring and guide treatment decisions. In comparison to NIRS, which provides real-time perfusion data through contact-based point measurements and is highly dependent on sensor placement, SFDI offers non-contact, two-dimensional mapping of tissue oxygenation, avoiding discomfort and direct contact with open wounds. Hyperspectral imaging with the TIVITA tissue camera system is also used in assessing the StO₂ in PAD subjects in a non-invasive, non-contact manner, but the StO₂ is measured only up to a depth of 0.8 mm (19). SFDI is particularly advantageous for evaluating surface perfusion and microvasculature up to a depth of 3 mm, though it does not extend to deeper vessels. In timepoint comparison studies, such as pre- and post-PTA assessments within the same subject, pigmentation does not affect the outcomes; however, it may introduce variability in interparticipant comparisons. Furthermore, optimal image acquisition requires a low-light environment to minimize background interference. In our study, occasional blackened regions appeared in intra-operative images due to ambient light exposure. These regions were excluded during region-of-interest selection prior to quantitative analysis and therefore did not affect calculated StO₂, but they did impact immediate visualization of the perfusion maps in the operating theatre, which should be considered during clinical implementation.

One limitation of this study is that measurements were only performed before and after the interventions. Immediate post-procedural microcirculatory status may be influenced by transient factors such as edema, vasospasm, or pharmacological effects of endovascular procedures and may not reflect stabilized or peak perfusion improvement (20). To better understand perfusion patterns and temporal changes in StO₂, a larger prospective study incorporating additional time points or serial measurements over a defined follow-up period is required. Doing so would also allow for an in-depth analysis of whether changes in StO₂ have predictive value for long-term outcomes such as wound healing and limb salvage.

Conclusions

In conclusion, by integrating SFDI alongside the angiography, clinicians can obtain a more comprehensive view of perfusion dynamics, ultimately improving post-angioplasty assessment and optimizing patient outcomes.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Ethics approval was obtained from SingHealth Centralised Institutional Review Board (CIRB Ref 2021/2648). All procedures performed in this study were in accordance with the ethical standards of SingHealth Centralised Institutional Review Board, and with the Helsinki Declaration and its subsequent amendments. Written informed consent was obtained from the patients for publication of this article and accompanying images. A copy of the written consent is available for review by the editorial office of this journal.