Near-Infrared Spectroscopy During Reactive Hyperemia for the Assessment of Lower Limb Vascular Function

Abstract

Vascular diseases of the lower limb contribute substantially to the global burden of cardiovascular disease and comorbidities such as diabetes. Importantly, microvascular dysfunction can occur prior to, or alongside, macrovascular pathology, and both potentially contribute to patient symptoms and disease burden. Here, we describe a non-invasive approach using near-infrared spectroscopy (NIRS) during reactive hyperemia, which provides a standardized assessment of lower limb vascular (dys)function and a potential method to evaluate the efficacy of therapeutic interventions. Unlike alternative methods, such as contrast-enhanced ultrasound, this approach does not require venous access or sophisticated image analysis, and it is inexpensive and less operator-dependent. This description of the NIRS method includes representative results and standard terminology alongside the discussion of measurement considerations, limitations, and alternative methods. Future application of this work will improve standardization of vascular research design, data collection procedures, and harmonized reporting, thereby enhancing translational research outcomes in the areas of lower limb vascular (dys)function, disease, and treatment.

Introduction

Cardiovascular disease (CVD) is the leading contributor to global mortality¹. While myocardial infarction and stroke are the most common manifestations of CVD, vascular diseases of the lower limbs, such as peripheral arterial disease (PAD) and diabetic foot disease, contribute substantially to the personal, social, and healthcare burden of CVD^2,3,4. Importantly, these disease states are characterized by microvascular and macrovascular dysfunction⁵ that contribute to symptoms (e.g., intermittent claudication), functional impairment, poor mobility as well as social isolation and reduced quality of life⁶. Historically, upper-limb vascular assessment techniques have been used as a measure of systemic vascular function and associated cardiovascular risk; however, these methods are potentially not sensitive to local impairments in lower limb vascular function^7,8. While there is currently a range of techniques used to assess vascular function in the lower limb, such as flow-mediated dilatation (FMD) and contrast-enhanced ultrasound, each method has disadvantages and limitations, such as equipment cost, operator skill, or the need for invasive venous access. For these reasons, there is a need for standardized and effective techniques to evaluate lower limb vascular (dys)function that can be more readily implemented in research and clinical settings.

Continuous wave near-infrared spectroscopy (CW-NIRS) is a non-invasive, low-cost, and portable method that quantifies the relative changes in hemoglobin oxygenation in vivo. As the NIRS oxygenated and deoxygenated hemoglobin signals are derived from the small (<1 mm in diameter) vessels, local skeletal muscle metabolism and microvascular function are able to be evaluated⁹. Specifically, the tissue saturation index (TSI) [TSI = oxygenated hemoglobin/ (oxygenated hemoglobin + deoxygenated hemoglobin) x 100], provides a quantitative measure of tissue oxygenation⁹. When measured before, during, and after occlusion and reactive hyperemia, the changes in TSI indicate 'end-organ' vascular responsiveness, relative to the pre-occlusion baseline. Importantly, this method is sensitive to alterations in muscle microvascular responsiveness and perfusion associated with ageing¹⁰, disease progression¹¹, and clinical interventions (e.g., revascularization surgery^12,13 or exercise rehabilitation^14,15,16,17) in individuals with, or at risk of microvascular dysfunction.

The availability of NIRS systems has led to a rapid rise in the number of research studies reporting microvascular function¹⁸. However, differences in reactive hyperemia testing protocols, omission of detailed, repeatable NIRS methods, as well as a lack of uniformity in the description, presentation, and analysis of NIRS response parameters make comparisons across individual trials challenging. This limits the collation of data for meta-analysis and the formulation of clinical assessment recommendations^9,15.

Therefore, in this article, we describe our laboratory's standardized NIRS and vascular occlusion testing protocols for the assessment of lower limb reactive hyperemia. By disseminating these methods, we aim to contribute to the improved standardization and repeatability of data collection procedures and harmonized reporting.

Protocol

All methods described here have been approved by the human research ethics committee of the University of the Sunshine Coast. Furthermore, all participants gave their written informed consent to participate in the measurements outlined in this protocol. Please note, vascular occlusion testing in the lower limb is contra-indicated in individuals who have previously had a revascularization procedure involving a vascular graft or stenting of the femoral or popliteal arteries. After preparing the equipment, the participant is instructed to rest in a supine position for 10 min. At this point, NIRS data collection commences, with an initial 2 min period, allowing for stability of NIRS signals to be achieved. Baseline data are then collected for 1 min, at which point a cuff located at the thigh is promptly inflated to achieve arterial occlusion. Occlusion is maintained for 5 min before the cuff is rapidly deflated. Data collection continues throughout the reactive hyperemia period until signals have recovered to baseline. Figure 1 depicts an overview of the reactive hyperemia protocol, and the detailed steps are provided below. The equipment used for the study are listed in the Table of Materials.

Figure 1: Schematic outlining NIRS reactive hyperemia measurement protocol and timings. NIRS: near-infrared spectroscopy. Please click here to view a larger version of this figure.

1. Equipment preparation

NOTE: Various NIRS, cuff inflation/occlusion, and data collection systems can be used to obtain the representative results outlined below. It is important that investigators consult their own specific user manuals and are aware of unique software, calibration, ambient light, and participant/cohort-specific considerations.

1. Ensure that all measurements are performed in a quiet, temperature-controlled room (21-23 °C).
2. Confirm that all equipment and materials (see Table of Materials) are available.
3. Ensure that the CW-NIRS device, and the computer that the device is transmitting to, are fully charged and powered on.
4. Check that the NIRS device is paired with the computer displaying the NIRS data via Bluetooth and that measurement parameters of the NIRS unit are defined according to the experimental site and study design (an example: define the differential pathlength factor (DPF) of the near-infrared light).
   NOTE: Most of the NIRS protocol steps and system settings are based on a combination of manufacturer's recommendations, researchers' experience, and expert/consensus opinion. Be aware that the values generated during measurement depend on the particular device and probes used as well as the settings chosen in the device software. There is also potential for a high degree of inter-individual variability in the signals collected. Furthermore, to allow for improved comparison between studies, it is important to report NIRS instrumentation particulars (e.g., include probe design, source-detector separation distances and wavelengths used), system settings, probe position/orientation in relation to muscle geometry and the analysis parameters/data treatment when publishing results from these measurements.
5. Enter the participant's data according to NIRS device software particulars.
6. Connect the rapid cuff inflator to the air and power sources.

2. Participant preparation

1. Ensure that the participant has read the participant information statement and consent form and has given their consent to participate in the measurement prior to commencement.
2. Explain to the participant what to expect during the measurement while they remove their shoes and socks and stand stationary for relevant anthropometric measurements.
3. Take a skinfold measurement (in triplicate) over the planned NIRS measurement site (e.g., on the medial aspect at the point of maximal calf circumference).
   NOTE: This allows for the depth of skin and adipose tissue (routinely referred to as adipose tissue thickness - ATT) to be confirmed in relation to NIRS signal penetration depth. Alternatively, ultrasound can be used to determine adipose tissue thickness. Always keep in mind that the depth of the NIRS signal/measurement is approximately half of the receiver-transmitter distance⁹.
4. To maximize signal quality, check the planned site of the NIRS transmitter/receiver probe for any hair that may absorb light and remove hair by shaving if necessary.
   NOTE: If the skin is compromised, for example, from cellulitis, or if edema is present, consider the appropriateness of measurement, as these issues relate to potentially reduced signal quality, signal penetration depth/tissue from which measurement is taken, as well as participant wound sterility.
5. Take measurements of and/or mark the position of planned NIRS probe placement in relation to relevant anatomical landmarks (e.g., superior surface of the medial condyle) to allow for accurate placement of the NIRS probe between and within participants (dependent on study design and probe design/shape) as oxygenation responses can display large heterogeneity between different muscles or even within regions of the same muscle.
6. Ask the participant to lie in a supine position on an examination plinth or bed. The participant then rests for 10 min.
7. Place a cuff around the thigh, proximal to the knee, ensuring that the tubes do not come into contact with the calf or the NIRS device (Figure 2).
8. Elevate the leg (~10 cm) with the foot and ankle on a foam support, leaving the lower leg stable and accessible for measurements (Figure 3).
   NOTE: If the test is required to be conducted in both legs over multiple occasions, the test order is randomized (before initial testing), and the order of testing is maintained for subsequent assessments for each participant.
9. Turn on the rapid cuff inflator module.
   NOTE: Ensure that the mode and pressure settings of the rapid cuff inflator are consistent with the manufacturer's instructions and vascular occlusion protocol specifications.
10. Turn on the cuff inflator air source, check that air can travel through the hose, and connect the hose to the thigh cuff.
11. Ensure that the hose of the cuff inflator is still not in contact with the calf.
12. Fix the NIRS transmitter/receiver probe securely on the skin overlying the measurement site(s) (routinely the medial aspect of the gastrocnemius; however, other sites such as the dorsum of the foot and tibialis anterior are also utilized depending on the study and participant specifics).
    NOTE: Some NIRS systems allow for the possibility of multiple site measurements simultaneously.
13. Cover the probe with black kinesiology tape or something similar, being careful to seal the edges to prevent ambient light from affecting the NIRS signal quality/values (Figure 3).
    NOTE: When fixing the probe and the adhesive covering, be sure to eliminate probe movement, but avoid compressing skin/adipose tissue/muscle.

Figure 2: Example of the occlusive cuff placement at the thigh. (A) From above. (B) From the side. Please click here to view a larger version of this figure.

Figure 3: Example of near-infrared spectroscopy probe position. (A) Probe attached to shaved skin at medial gastrocnemius. (B) Probe placement while ankle in foam support to allow access and ensure stability. (C) Ambient light shielding in place. Please click here to view a larger version of this figure.

3. Baseline data collection

1. Ask the participant to remain relaxed, refrain from talking, and keep their leg(s) still for the duration of the data collection.
2. Start the NIRS device recording via the computer interface.
   NOTE: If offline data collection is enabled and preferred, start the NIRS device directly.
3. Allow for at least 2 min of data to be acquired prior to initiating the start of the measurement. This ensures steady-state signals are achieved prior to data acquisition.
4. Monitor the computer screen to ensure data signal integrity and physiologically plausible values. For example, Data Acquisition (DAQ) values provide information regarding signal quality and the amount of ambient light detected by the NIRS device. They should be kept at an acceptable range throughout the measurement.
5. If, after 2 min, no fluctuations in the NIRS signal are noted (e.g., due to movement artifact), set the NIRS data baseline by pushing the corresponding button on the computer/device/syncing unit. This baseline reflects the comparative starting point of the measurement. The changes in oxygenated and deoxygenated hemoglobin will be interpreted relative to this baseline.
   NOTE: During data collection, insert software-based event markers of specific milestones, namely start and end of baseline, start and end of cuff occlusion etc., to assist in data analysis.
6. Collect at least 1 min of baseline data, again ensuring that no fluctuations or movement artifacts occur during this time.
   NOTE: Fluctuations in baseline data will affect the interpretation of some potential variables, such as hyperemic reserve (see representative results section).

4. Vascular occlusion

1. Set cuff pressure to 200 mmHg on the rapid cuff inflator before switching to cuff mode (or relevant setting on alternate systems).
   NOTE: While some authors recommend inflation to 250 mmHg at the thigh^9,19, in our experience, this is not well tolerated by some participants, leading to discontinuation of the measurement. Inflation to 200 mmHg is sufficient, in most participant groups, to occlude arterial inflow at rest while being tolerable and not resulting in bruising of the skin. To ensure effective occlusion is attained at 200 mmHg, our group routinely uses strain gauge plethysmography simultaneously with NIRS to confirm the absence of downstream blood flow during the period of cuff inflation.
2. Advise the participant on what to expect (discomfort, tingling sensations, etc.) during the process of inflating the cuff to 200 mmHg for a period of 5 min, followed by a total cuff deflation. Remind the participant again to remain relaxed, refrain from talking, and keep their leg(s) immobile for the duration of the data collection.
   NOTE: It is our experience that the first 30-60 s of the occlusion period is the least comfortable for participants. When participants let their legs relax, they tend to find the occlusion more tolerable.
3. When ready to initiate cuff inflation, mark the end of the baseline period in NIRS software.
4. Inflate the thigh cuff to a suprasystolic pressure of 200 mmHg, or alternatively to 220 mmHg in rare cases where 200 mmHg has proven ineffective.
5. Monitor the computer or screen to ensure data integrity during the occlusion period.
6. As the end of the 5 min of cuff occlusion approaches, prepare the participant by reminding them to keep their leg as still as possible and refrain from talking for approximately 3 min after cuff deflation (whilst the data from the vascular responses to reactive hyperemia are being collected).
   NOTE: This reminder to the participant to remain still is important, as participants may be tempted to move the limb to relieve/ease discomfort as circulation is restored.

5. Reactive hyperemia

1. After 5 min of occlusion has passed, rapidly deflate the thigh cuff completely (to 0 mmHg). Simultaneously, mark the end of the occlusive period in the NIRS software. The reactive hyperemia response, due to the resumption of blood flow and related factors, will be visible in the NIRS software display (Figure 4).
2. After a minimum of 3 min post-occlusion, or alternatively, after the NIRS data has returned to baseline, mark the end of the recovery period in NIRS software and stop the measurement.
   NOTE: The duration of recovery depends, in part, on the research or clinical question being investigated and, therefore, the specific NIRS parameters chosen for analysis, as well as the potential for concomitant measures (such as FMD) that may require a longer recovery duration.
3. Initiate data saving and exportation of NIRS results for data treatment and analysis.

6. Follow up procedures

1. Remove the NIRS device(s) and cuff from the participant.
2. If required, clean the NIRS device (and occlusion cuff) following manufacturer's instructions and relevant hygiene standards.
3. Inspect the NIRS device to ensure transmitter/receiver integrity and battery performance for future measurements.

Representative Results

Near-infrared spectroscopy
Continuous wave near-infrared spectroscopy devices measure relative changes in oxygenated (O₂Hb) and deoxygenated (HHb) hemoglobin, which reflect local O₂ delivery and utilization via light-emitting sources and photodetectors, set specific distances apart. Wavelengths of light between ~700 nm and 850 nm are emitted, corresponding with the peak absorbency of O₂Hb and HHb. Once near-infrared light has penetrated skeletal muscle, the scatter and absorption of light are dependent on the wavelength. Skeletal muscle is a heterogeneous tissue, and so the scattering and absorption coefficients, as well as the given pathlength through which light travels, cannot be quantified. Therefore, CW-NIRS technology uses the modified Lambert-Beer Law, which includes a differential pathlength factor (DPF), and enables the relative concentrations of O₂Hb and HHb to be calculated. In addition to O₂Hb and HHb, CW-NIRS derived measures include total hemoglobin (THb = O₂Hb + HHb), a marker of blood volume / total signal strength and tissue saturation index (TSI = O₂Hb/THb*100). The TSI is generated by spatially resolved spectroscopy (SRS) technology, where photons are measured at multiple spacings from the source, improving precision⁹ and providing a quantitative measure of tissue oxygenation. In addition, due to the multiple source-detector distances, SRS enhances the contribution of deeper tissues, while reducing the contribution of more superficial tissue, like skin/adipose tissue, to the NIRS signals. It is also worth noting that TSI is alternately referred to as regional oxygen saturation (rSO₂), tissue oxygenation index (TOI), or muscle tissue oxygen saturation (StO₂) in the literature. It is therefore suggested that going forward, a single term be adopted to standardize nomenclature and reduce confusion.

It is further recommended that all NIRS traces collected during research be analyzed and reported⁹. It is also important to appreciate that CW spectrometers do not provide absolute concentrations of chromophores in the interrogated tissues because the actual path length of light is unknown. These devices instead provide changes relative to a predetermined baseline. Therefore, as noted by Cornelis et al.¹⁵, it is recommended that NIRS signals are analyzed as changes from a time point of interest. It is additionally recommended that the amplitudes and slopes of responses are the focus of reporting and interpretation, as these variables are less sensitive to confounding issues such as adipose tissue thickness or signal-to-noise ratios.

NIRS signals during occlusion and reactive hyperemia
This article will focus predominantly on describing TSI signal responses during occlusion and reactive hyperemia, as this SRS measure is less prone to changes in cutaneous signal contributions and measurement error when compared to the raw O₂Hb and HHb signals.

A selection of the NIRS TSI variables evaluated during occlusion and reactive hyperemia is depicted in Figure 4. Baseline TSI represents the average TSI before the initiation of the vascular occlusion (cuff inflation), routinely for 1 min. The desaturation rate during occlusion is represented by Slope 1. Due to the arterial occlusion, reductions in TSI indicated by Slope 1 can be attributed to oxygen utilization/the resting muscle metabolic rate. The TSI_MIN is the lowest TSI value obtained during occlusion. The TSI_IMAG(difference between baseline TSI and TSI_MIN) represents the magnitude of ischemia induced by occlusion (and the stimulus for vasodilation and post-occlusion hyperemia). Slope 2 indicates the reperfusion rate after cuff release and represents the reactive hyperemia/microvascular reactivity response. Alternatively, the reperfusion slope can be described as a half-time. The TSI_MAX is the highest TSI value obtained after cuff release. The reperfusion magnitude is calculated as the difference between TSI_MAX and TSI_MIN, and the time to TSI_MAX is calculated as the difference in time (s) from TSI_MIN to TSI_MAX. The TSI reactive hyperemia area under the curve (AUC) is calculated from the return to baseline after cuff release for 1 min, 2 min, or 3 min. Lastly, the hyperemic reserve, representing the change in TSI above baseline, can be calculated as the difference between TSI_MAX and the baseline TSI, expressed as a percentage^9,10,12,20.

Figure 4: Tissue saturation index (TSI) signal during a NIRS vascular occlusion test with TSI variables of interest during reactive hyperemia. Baseline TSI represents the average TSI before the initiation of cuff inflation. Slope 1 represents the desaturation rate during occlusion. The TSI_MIN is the lowest TSI value obtained during occlusion. The TSI_IMAG (difference between baseline TSI and TSI_MIN) represents the magnitude of ischemia induced by occlusion. Slope 2 indicates the reperfusion rate after cuff release. The TSI_MAX is the highest TSI value obtained after cuff release. The reperfusion magnitude is calculated as the difference between TSI_MAX  and TSI_MIN, and the time to TSI_MAX is calculated as the difference in time (s) from TSI_MIN to TSI_MAX. The TSI reactive hyperemia area under the curve (AUC) is calculated from the return to baseline after cuff release for 1 min, 2 min, or 3 min. The hyperemic reserve, representing the change in TSI above baseline, can be calculated as the difference between TSI_MAX and the baseline TSI, expressed as a percentage. Please click here to view a larger version of this figure.

In addition to TSI variables, it is also possible to estimate tissue oxygen uptake (mVO₂) during arterial cuff occlusion, given the absence of changes in blood volume, by calculating the slopes of change in HHb and O₂Hb during the initial linear portion of these parameter's responses to occlusion (Slope 1), based on the assumption that the rate of disappearance of O₂Hb (and/or the rate of appearance of HHb) equates to the rate of oxygen utilization by the muscle under interrogation.

Utilizing NIRS to evaluate microvascular responsiveness in clinical populations
Importantly, NIRS measurements during reactive hyperemia testing can differentiate the microvascular responses of apparently healthy participants from those with varying degrees of dysfunction. Differences in responsiveness between an apparently healthy individual and an individual with PAD are seen in Figure 5. Specifically, slower rates of reperfusion, longer recovery times, smaller maximal hyperemic responses, and differences in variables such as time to TSI_MAX and TSI_AUC are observed in a representative participant with PAD, compared to the healthy control²⁰. Of note, NIRS variables such as those described in this article are also used during standardized clinical exercise and occlusion testing to evaluate (micro)vascular responses to a specific treatment or physical activity stimulus or to determine the result of routine exercise training in healthy¹⁷ and CVD populations²¹. For a recent summary of NIRS measurement parameters and results from studies utilizing NIRS in lower limb peripheral vascular disease during reactive hyperemia and exercise, please refer to the systematic review by Joseph et al.¹⁸.

Lastly, when designing research that incorporates NIRS measures such as those displayed in Figure 4 and Figure 5, be aware that large interindividual variability in absolute values has been shown in apparently healthy and clinical populations, making it challenging to, for example, define absolute values for PAD diagnosis based on NIRS data²².

Figure 5: Tissue saturation index (TSI) signal during reactive hyperemia testing. Tissue saturation index (TSI) signal evidencing the ability of NIRS to distinguish between the vascular responsiveness of an apparently healthy male participant and a male participant with peripheral artery disease (Ankle Brachial Index = 0.5) during reactive hyperemia testing. *Start of NIRS baseline; ^#Start of arterial occlusion; ^End of arterial occlusion. Please click here to view a larger version of this figure.

Suboptimal measurements
As described in the protocol section above, once NIRS signals are being recorded, it is important to visually inspect the data traces for movement artifacts prior to setting the NIRS baseline for all traces. Figure 6 depicts movement artifacts in the O₂Hb and THb traces.

Figure 6: Example of movement artifact in NIRS signal during the pre-baseline data collection period. ^#Start of NIRS pre-baseline period. THb: Total hemoglobin. O₂Hb: Oxygenated hemoglobin. Please click here to view a larger version of this figure.

The protocol described above utilizes a pressure of 200 mmHg to occlude arterial flow in the lower limb. Very occasionally, this pressure is not sufficient to occlude blood flow, which compromises the hyperemic response. This may occur, for example, in a participant with grade 3 hypertension. In addition to using other technology, such as strain gauge plethysmography, to confirm the cessation of blood flow, it is important to be observant of any variance from the predicted NIRS responses after cuff occlusion (e.g., Figure 7, Panel B: O₂Hb signal rising (failing to decrease due to hypoxia and tissue oxygen utilization as expected) and THb also rising (not remaining relatively linear due to stable blood volume). Noting these deviations from expected responses allows early intervention to cease the current measurement/ permits the occlusion pressure to be increased until it is evident that the artery has been occluded in subsequent measurements.

Figure 7: Example of total hemoglobin (THb) and oxygenated hemoglobin (O₂Hb) traces. (A) Successful arterial occlusion and (B) unsuccessful arterial occlusion. ^#Start of arterial occlusion. Please click here to view a larger version of this figure.

Discussion

This article outlines standardized procedures for the assessment of lower limb reactive hyperemia using CW-NIRS TSI to evaluate microvascular function. This protocol has been refined by examination of cuff occlusion duration on response magnitude, NIRS test-retest reliability during reactive hyperemia, as well as the level of agreement between NIRS and other methods of microvascular evaluation such as contrast-enhanced ultrasound^23,24. A longer cuff-occlusion duration corresponds with a lower TSI_MIN²³, as seen in Figure 8 and Figure 9, reflecting a greater ischemic stimulus. Compared with 1 min and 3 min cuff-occlusion durations, a greater magnitude of response was also demonstrated following 5 min cuff-occlusion, indicating an increased vasodilatory response to a larger hypoxic stimulus. These are important considerations when targeting a specific degree of ischemic stimulus, for example, as differences in TSI_MIN have been found to be a function of age when using a fixed occlusion timeframe²⁵. Furthermore, test-retest reliability of NIRS-derived post-occlusive reactive hyperemia measures has been established using a 5 min cuff occlusion duration^24,26.

Figure 8: Effect of occlusion duration on tissue saturation index traces in the medial gastrocnemius of older adults during cuff occlusion and reactive hyperemia (n = 13). Please click here to view a larger version of this figure.

Figure 9: Effect of occlusion duration on minimum tissue saturation index (TSI_MIN; means ±SD) in older adults (n = 13). TSI_MIN in the medial gastrocnemius muscle following 1 min, 3 min, and 5 min of thigh cuff occlusion. *Significant difference from 1 min; ^#significant difference from 3 min. Please click here to view a larger version of this figure.

Considerations/troubleshooting
Even though CW-NIRS systems are relatively easy to use, certain practical issues need to be considered, including the fact that CW-NIRS probes are vulnerable to ambient light contamination. It is therefore imperative that data indicators reflecting the integrity of the NIRS signal are monitored during measurements and that appropriate skin contact and shielding are maintained to enable physiologically accurate values to be obtained. Additionally, if data collection depends on Bluetooth connectivity, it is important to ensure that the manufacturer's instructions regarding Bluetooth are adhered to in order to avoid connection failures.

Concerning the reactive hyperemia method, although rare, a participant may report that the rapid inflation of the ischemia-inducing cuff causes intolerable localized discomfort. In these individuals, it is our group's experience that inflating the cuff to 100 mmHg before gradually inflating to 200 mmHg over a period of 5 s is better tolerated. However, if discomfort persists despite this alteration, the measure is discontinued.

Methodologically, as highlighted by Rosenberry and Nelson²⁵, when interpreting reactive hyperemia via CW-NIRS responses, it may be important to distinguish between 'reactive' and 'responsive' hyperemia. Measurements taken ≤30 s following cuff occlusion provide insight into the reactive phase of reperfusion (blood flow response arising from the ischemic stimulus and resultant microvascular dilation). Beyond this initial phase, the hyperemia and resultant shear stress contribute to vessel dilation and enhanced limb blood flow (responsive hyperemia). From a mechanistic point of view, because NIRS provides a measurement of skeletal muscle metabolism, it could be necessary to consider that impaired skeletal muscle oxidative capacity, for example, mitochondrial dysfunction, may impact the time course of measurement parameters. Lastly, as remarked previously, researchers may also need to take into account the degree of ischemic insult achieved in a set period of cuff occlusion and the resultant length/magnitude of physiologically mediated hyperemia that follows when deciding on when to end data collection, as well as when interpreting and comparing data between participants²⁵.

The NIRS variables of interest are commonly accepted to reflect microvascular (dys)function; however, it is important to acknowledge that when interpreting these results, microvascular responses are dependent to some degree on macrovascular/conduit artery blood flow and function. Also, microvascular dysfunction may occur prior to, or alongside, pathological changes to the macrovasculature⁵, and microvascular dysfunction is associated with a range of cardiometabolic diseases²⁰. This micro and macrovascular pathology may therefore both contribute to why transient ischemia and the resultant hyperemic response can indicate increased cardiovascular disease risk^27,28,29. To better understand the relationship between macro and microvascular function, it can be useful to combine NIRS with measures of macrovascular function, such as FMD^10,30, a measure of endothelium-dependent shear-stress mediated dilation of conduit arteries. Additionally, it is important to acknowledge the relative recency of NIRS being used for microvascular assessment and consider the underlying physiological mechanisms that contribute to NIRS-derived variables when interpreting the clinical relevance of such measurements in relation to conduit artery function. For instance, NIRS with vascular occlusion testing is a recently validated method^31,32, compared to the well-established FMD technique. Both measures involve a comparison of baseline to post 5 min cuff occlusion; however, there are some key differences. For FMD, measurement is taken proximal to the cuff; thus, the measurement site does not become ischemic, whereas NIRS is measured distal to the cuff at a site that undergoes ischemic stress. Because the dynamic responses within both the ischemic phase and the hyperemic phase provide information on metabolic activity as well as perfusion capacity (microvascular function), it is essential to standardize the protocol to enable comparison within and between participants. With this point in mind, researchers should be aware that the analysis strategy adopted for NIRS data (for example, the time window used to determine Slope 2) can impact the relationship between NIRS and FMD³⁰. Lastly, when interpreting NIRS variables in conjunction with measures of conduit artery (macrovascular) function, alongside improvements in functional variables such as walking distance, consideration should be given to the fact that there is the distinct possibility of 'built in' redundancy of exercise responses. Specifically, both underlying micro and macrovascular mechanisms could potentially contribute to improvements in functional parameters, but the relative contributions of each may be difficult to gauge¹⁵.

Limitations
A limitation of this protocol is that vascular occlusion testing in the lower limb is contra-indicated in individuals who have previously had a revascularization procedure involving a vascular graft or stenting of the femoral or popliteal arteries. This highlights the importance of having access to relevant vascular imaging and a detailed medical history when deciding whether to perform vascular occlusion testing in certain participant populations. As mentioned previously, it is also important to realize participants may be unable to tolerate the discomfort associated with cuff inflation and/or arterial occlusion of 5 min duration and to make it clear to all participants that the measurement can be discontinued if this is the case. In our experience, while approximately 10% of participants report noticeable discomfort during the measurement, the need to terminate the measure early as a result of this discomfort has occurred in only a small number of participants (1%-2%).

A limitation of the NIRS technology is that the signals are influenced by the degree of skin pigmentation (melanin). Currently, there is no generally available technique to correct the reduction in NIRS signals due to the melanin content in the skin, limiting the application of this technique in individuals with dark skin tones^9,33. Additionally, the depth of the NIRS signal/measurement is approximately half of the receiver-transmitter (source-detector) distance. Therefore, excess adiposity over the site of interrogation reduces the relative contribution of underlying muscle tissue to the NIRS signal. It is therefore important to ensure that the depth of light penetration sufficiently interrogates the underlying muscle(s) to provide information specific to muscle blood flow/metabolism interactions⁹. This is one reason why the gastrocnemius muscle is a standard site of measurement, as subcutaneous adipose tissue rarely exceeds 1cm at this location²⁰. Alternatively, the tibialis anterior is another potential site that can be utilized³⁴.

There are multiple types of NIRS devices (frequency domain, time domain, and continuous wave) that function differently. Differences in parameters such as reperfusion slopes have been reported when comparing devices; therefore, caution is advised when comparing or combining results from different types of NIRS devices²⁵.

There is potential for CW-NIRS to be used during surgical interventions to confirm successful revascularization; however, fluctuations in real-time NIRS signals, as well as a large degree of inter-individual variability during endovascular interventions, impede real-time confirmation of successful microvascular reperfusion, based on the NIRS curves alone. Therefore, the current associated need to rely on post-processing of data should be a consideration for investigators planning future experimental protocols³⁵.

Alternative methods
Other methods commonly used to measure lower limb vascular function, perfusion and/or reactive hyperemia include venous occlusion plethysmography, doppler ultrasound, and contrast-enhanced ultrasound. However, NIRS has distinct advantages over each of these methods. Whilst venous occlusion plethysmography is a simple and reproducible method for measuring limb blood flow via volume changes, it is an indirect measurement of blood flow, and the inflation-deflation technique leads to very low temporal resolution. In contrast, NIRS has high temporal resolution and evaluates changes in the microvasculature/tissue oxygen metabolism directly. Ultrasound measurement of blood flow via vessel diameter and Doppler velocity signal provides excellent temporal and spatial resolution; however, there are multiple analytical approaches for quantifying results, and this method is highly operator-dependent, relying on extensive skill and training. Ultrasound is also sensitive to even minor motion artifacts throughout the measurement²⁵. By comparison, NIRS is relatively easy to use and is not as sensitive to motion artifacts. Real-time contrast-enhanced ultrasound quantifies skeletal muscle microvascular responsiveness during reactive hyperemia; however, this technique is limited by the need for intravenous access, which requires medical oversight. Contrast-enhanced ultrasound also requires sophisticated image analysis, extensive skill, and training, as well as expensive ultrasound equipment and contrast agents. Initial evidence indicates that contrast-enhanced ultrasound-derived measures of post-occlusive reactive hyperemia, including skeletal muscle microvascular blood flow responsiveness, correlate with NIRS-derived measures following 5 min cuff occlusion²⁴. Lastly, due to the relative ease and affordability of NIRS, multiple users could be trained and deployed rapidly, allowing for NIRS to be used in large multi-center investigations whilst minimizing interoperator variability.

Future directions and applications
Promising applications of NIRS technology include the assessment of regional oxygenation dynamics in PAD for improved limb risk stratification and evaluation of treatment effectiveness or clinical deterioration, particularly in those individuals with symptoms masked by peripheral neuropathy and diabetic complications^15,18,22. Newer systems include multiple, relatively small, and lightweight CW-NIRS probes that collect data simultaneously, enabling evaluation of the heterogeneity of microvascular responses between sites⁹ during reactive hyperemia or exercise testing. Near-infrared spectroscopy measures have also shown promise for the prediction of wound healing in lower limb conditions such as diabetic foot disease/ulceration, when compared to standard healing criteria, with implications for improved patient care as well as reduced cost via discontinuing ineffective medical therapies earlier^36,37. Furthermore, evidence of NIRS techniques assisting in the confirmation of microvascular perfusion success in patients with critical limb ischemia after conduit artery reperfusion interventions is growing^13,35. Lastly, NIRS allows microvascular desaturation/ischemic stimuli to be targeted during clinical exercise therapy, and post-intervention NIRS also enables evaluation of changes in lower limb vascular function, providing mechanistic insights into changes in functional ability and walking capacity^15,38.

In conclusion, it is hoped that disseminating this protocol will allow improved uniform analysis and pooling of data between individual trials and disparate research groups. This more consistent approach will, in turn, allow for an enhanced understanding of lower limb vascular dysfunction and improve translational research outcomes aimed at improving the health and well-being of individuals with peripheral vascular disease.

Materials

Name	Company	Catalog Number	Comments
Cuff Inflator Air Source	Hokanson	AG101 AIR SOURCE
Elastic Cohesive Bandage	MaxoWrap	18228-BL	For blocking out ambient light
OxySoft	Artinis	3.3.341 x64
PortaLite (NIRS)	Artinis	0302-00019-00
PortaSync MKII (Remote)	Artinis	0702-00860-00	For Marking milestones during measurement
Rapid Cuff Inflator	Hokanson	E20 RAPID CUFF INFLATOR
Thigh Cuff	Hokanson	CC17
Transpore Surgical Tape	3M	1527-1	For fixing probe to skin