Suitability of the muscle O2 resaturation parameters most used for assessing reactive hyperemia: a near-infrared spectroscopy study

Abstract Background There is a spectrum of possibilities for analyzing muscle O2 resaturation parameters for measurement of reactive hyperemia in microvasculature. However, there is no consensus with respect to the responsiveness of these O2 resaturation parameters for assessing reactive hyperemia. Objectives This study investigates the responsiveness of the most utilized muscle O2 resaturation parameters to assess reactive hyperemia in the microvasculature of a clinical group known to exhibit impairments of tissue O2 saturation (StO2). Methods Twenty-three healthy young adults, twenty-nine healthy older adults, and thirty-five older adults at risk of cardiovascular disease (CVD) were recruited. Near-infrared spectroscopy (NIRS) was used to assess StO2 after a 5-min arterial occlusion challenge and the following parameters were analyzed: StO2slope_10s, StO2slope_30s, and StO2slope_until_baseline (upslope of StO2 over 10s and 30s and until StO2 reaches the baseline value); time to StO2baseline and time to StO2max (time taken for StO2 to reach baseline and peak values, respectively); ∆StO2reperfusion (the difference between minimum and maximum StO2 values); total area under the curve (StO2AUCt); and AUC above the baseline value (StO2AUC_above_base). Results Only StO2slope_10s was significantly slower in older adults at risk for CVD compared to healthy young individuals (p < 0.001) and to healthy older adults (p < 0.001). Conversely, time to StO2max was significantly longer in healthy young individuals than in older adult at CVD risk. Conclusions Our findings suggest that StO2slope_10s may be a measure of reactive hyperemia, which provides clinical insight into microvascular function assessment.


INTRODUCTION
Endothelial dysfunction is an abnormality of the vascular system that predicts a cardiovascular event. 1,2 Assessment of reactive hyperemia in microcirculation has been used as a measure of microvascular function. 3 Microvascular dysfunction may lead to rarefaction of downstream vessels and consequent reduction in the number of capillaries, contributing to hypertension and cardiovascular events. 4 Moreover, downstream arterial disease may reduce muscle blood supply, leading to fatigue, cramp, discomfort, or pain in limbs during daily activities. 5 Therefore, measuring microvascular function may be important for identification of pathophysiologic mechanisms that impair adequate tissue blood perfusion.
Over the past decades, near-infrared spectroscopy (NIRS) combined with a vascular occlusion test (VOT) has been used for assessment of tissue O 2 saturation (StO 2 ). Since NIRS detects changes in oxygenated and deoxygenated hemoglobin in the tissues using absorption of near-infrared light at a specific wavelength, 6,7 oxygenation kinetics in the microcirculation (i.e., arterioles, venules, and capillaries) can be noninvasively assessed and used to measure reactive hyperemia in microcirculation. 6,8 Performing a VOT with an NIRS device fitted to the limb of interest, the StO 2 signal follows a downward path during the cuff inflation phase and then StO 2 goes rapidly upward immediately after cuff deflation (reperfusion phase). [8][9][10] The sudden increase in the StO 2 signal during reperfusion enables various StO 2 parameters to be calculated, which, in general, are interpreted as measures of microvascular function. 3 The majority of StO 2 parameters adopted in previous studies include reperfusion rate (i.e., upslope of the StO 2 signal during the initial 10 s and 30 s, and until the StO 2 signal reaches baseline values), 11-13 magnitude of reperfusion (i.e., total area under the curve [AUC] of StO 2 and the difference between the lowest and highest StO 2 value), 4,13-17 and others (e.g., time to StO 2 baseline, time to StO 2 maximum, and AUC above the baseline). 13,15,18,19 Despite a range of possibilities for interpretation of muscle O 2 resaturation parameters, the lack of standardization of which NIRS parameters are used during the reperfusion phase makes comparisons between studies difficult and/or may even lead to misinterpretation of results. For example, since the time to maximum StO 2 parameter is calculated as the time elapsed from release of the cuff to the maximum StO 2 value during reperfusion, individuals with higher O 2 extraction capacity (i.e., healthy young individuals) might exhibit a longer time to maximum StO 2 . In other words, in individuals with higher O 2 extraction, the distance between the lowest StO 2 value reached during cuff inflation and the highest StO 2 reached during the cuff deflation is longer, thereby influencing calculation of some NIRS parameters, such as time to StO 2 maximum and even area under the reperfusion curve. Therefore, the present study investigates the usefulness of the muscle O 2 resaturation parameters most utilized to assess reactive hyperemia in a clinical group known to exhibit impairments in muscle StO 2 . It was hypothesized that some NIRS parameters might be more sensitive than others (mainly those influenced by muscle O 2 extraction) for detecting abnormal reactive hyperemia in a clinical sample.

Participants
Twenty-three healthy young individuals, 29 healthy older adults, and 35 older adults at risk of CVD were recruited to participate in the study. For the younger group, inclusion criteria were being healthy (not presenting CVD risk factors) and age < 35 years. For the healthy older adult group, individuals had to be healthy and ≥ 60 years of age. The inclusion criteria for older adults at high risk for CVD were age ≥ 60 years and possessing at least four CVD risk factors, according to a previous study demonstrating that presence of three or more CVD risk factors is related to microvascular dysfunction. 10,20 The following CVD risk factors were used as inclusion criteria: triglycerides ≥ 150 mg/dL; total-cholesterol ≥ 200 mg/dL; LDL-cholesterol ≥ 150 mg/dL, HDLcholesterol ≤ 40 mg/dL for men and ≤ 50 mg/dL for women; high-blood glucose ≥ 120 mg/dL and/ or taking oral hypoglycemic medications; and high de doença cardiovascular. Conclusões: Nossos achados sugerem que StO2slope_10s pode ser uma medida de hiperemia reativa, que fornece informações clínicas sobre a avaliação da função microvascular. systolic blood pressure (SBP) ≥ 135 mm Hg, and/or high diastolic blood pressure (DBP) ≥ 90 mm Hg, and/or taking anti-hypertensive medications. The exclusion criteria included human immunodeficiency virus (HIV); cancer; rheumatoid arthritis; smoking; and chronic obstructive pulmonary disease; as well as participants not being engaged in any exercise program. All experimental procedures were performed after explaining the nature of the study and obtaining written consent from participants. The study was conducted in accordance with declaration of Helsinki

Experimental protocol
After 12 h fasting, the participants arrived at the laboratory, where they underwent anthropometric measurement (weight, height, and forearm skin-fold) before blood samples were taken. Blood samples were collected only at baseline in order to determine the CVD risk factors (i.e., fasting blood glucose, triglycerides, cholesterol, etc.) of the participants enrolled on the study. Afterwards, participants rested for 10 min in the supine position on an examination table, followed by assessment of the NIRS parameters for level of muscle oxygen saturation. The experimental procedures were conducted between 8:00 and 10:00 a.m. and participants were recommended to avoid caffeine and foods rich in nitrates and nitrites for 24 h before the laboratory visit.

Tissue O 2 saturation measurement
Tissue O 2 saturation (%StO 2 ) was assessed using an NIRS system (PortaMon, Artinis, Medical Systems) connected to a personal computer via Bluetooth for data acquisition (10 Hz). Analogue-to-digital conversion and subsequent analysis of the raw data (i.e., no filter was used) was conducted using native software (Oxysoft version 2.1.6; Artinis Medical Systems), as previously described by Oliveira et al. 7,10 In brief, the NIRS device was placed on the forearm flexor muscle (flexor carpi radialis), 2 cm below the medial epicondyle of the humerus, and muscle %StO 2 was recorded continuously throughout the VOT (30 s at baseline, 5 min of occlusion, and 2 min of reperfusion). The cuff was placed over the arm, 2 cm above the medial epicondyle of the humerus and a cuff pressure of 250 mm Hg was used during the occlusion period ( Figure 1). To evaluate muscle StO 2 during ischemia and reperfusion, the following NIRS parameters were used in the statistical analysis: (i) StO 2 during baseline (StO 2base , %) was calculated as the average over the 30 s before cuff occlusion; (ii) the lowest StO 2 value reached during cuff inflation (StO 2min ) and the highest StO 2 value reached following cuff deflation (StO 2max ); (iii) time for the StO 2 signal to reach its peak after cuff release (time to StO 2max ); (iv) time for the StO 2 signal to reach the pre-occlusion baseline value after cuff release (time to StO 2base ); (v) difference between StO 2min and the StO 2max after cuff deflation (ΔStO 2reperfusion ); (vi) the area under the reperfusion curve (AUC) above the baseline (StO 2AUCabove_base ); (vii) the total AUC over 2 minutes of reperfusion (StO 2AUCt ); (viii) upslope of StO 2 signal over a 10 s reperfusion window immediately following cuff deflation (StO 2slope_10s ); (ix) upslope of StO 2 signal over a 30 s reperfusion window immediately following cuff deflation (StO 2slope_30s ); and (x) upslope of StO 2 signal over the time elapsed between StO 2min and reaching pre-occlusion baseline StO 2 values ( Figure 2).

Statistical analysis
One-way ANOVA was used to identify significant differences between the characteristics of the participants at baseline and tissue StO 2 parameters during the VOT. An a priori power analysis was conducted (G*Power v. 3.0.1) for a specific test (ANOVA: Fixed effects, omnibus, one-way). Based on a statistical power (1 -β) of 0.80, an effect size of 0.46, and an overall significance level of 0.05, 51 participants would be needed to detect a statistical difference. The effect size was based on a previous study that found d=0.93 (or f=0.46; d=2f) when assessing reperfusion rates in older individuals and young participants. 17 The present study enrolled 87 participants. Additional post hoc tests with Bonferroni adjustment were performed when appropriate. Statistical significance was set at a P value ≤ 0.05 and the results were expressed as mean ± standard deviation (SD). A commercially

RESULTS
Baseline characteristics of the participants and medications used are shown in Table 1. Healthy young individuals exhibited significant lower body mass index than healthy older adults (P = 0.029) and than older adults at risk for CVD (P = 0.016). Blood glucose (P < 0.001), total cholesterol (P = 0.004), triglycerides (P = 0.049), and systolic blood pressure (P = 0.001) were significantly higher in older adults at risk for CVD than in healthy young individuals. Moreover, blood glucose (P < 0.001), triglycerides (P = 0.026), and systolic blood pressure (P = 0.012) were significantly higher than in healthy older adults. Table 2. Significantly slower StO 2slope _ 10s (P < 0.001), and StO 2slope _ 30s (P < 0.001), and blunted ΔStO 2reperfusion (P < 0.001), StO 2AUCabove_base (P < 0.001), StO 2AUCt (P < 0.001), StO 2max (P = 0.035), and StO 2 min (P < 0.001) were observed in older adults at CVD risk compared to healthy young individuals. Additionally, StO 2slope _ 10s was significantly slower (P < 0.001) in older adults at risk for CVD compared to their healthy counterparts. Time to StO 2max was significantly slower (P = 0.047) in healthy young individuals than older adults at risk of CVD.

DISCUSSION
Given the widespread utilization of NIRS devices and the spectrum of possibilities for analyzing muscle O 2 resaturation parameters as a measurement of reactive hyperemia in the microvasculature of different populations, the present study investigated the effectiveness of the O 2 resaturation parameters most frequently used for evaluating reactive hyperemia in healthy older adults or older adults with CVD risk factors. It was hypothesized that analyzing the effectiveness of the most widespread NIRS parameters for detecting aging and CVD risk factors might help  Muscle O 2 resaturation rate parameters (StO 2slope _ 10s , StO 2slope _ 30s , and StO 2slope _ until_base ) are largely adopted to assess reactive hyperemia in hypertensive individuals, 4 gestational diabetic women, 14 and older adults at risk for CVD disease, 7,10,17 among other clinical groups. 13,18 Our findings demonstrated that only StO 2slope _ 10s was significantly slower in older adults with CVD risk factors compared to healthy older adults and to healthy young individuals, suggesting that StO 2slope_10s may be a more sensitive NIRS parameter for assessing reactive hyperemia. The StO 2slope _ 10s has typically been adopted to assess O 2 resaturation rate, since previous studies have reported that this parameter is correlated with reactive hyperemia measured in the brachial artery 20 and with FMD response. 8,11 It has been postulated that the StO 2slope _ 10s may be a stimulus for FMD response given that FMD is evoked by increasing blood flow after cuff release. 8,20,21 In addition, utilizing the StO 2slope_10s may be considered an appropriate way to analyze O 2 reperfusion rate given the linearity of the StO 2 signal over the initial 10 s after cuff release. 22 Time to StO 2max may be mainly affected by the degree of muscle O 2 extraction (StO 2min ), since this recovery time parameter measures the time interval between the StO 2min (the lowest StO 2 value reached during cuff inflation) and StO 2max (the highest StO 2 value reached during cuff deflation). Therefore, individuals with higher muscle O 2 extraction capacity (mitochondrial function) will reach a lower StO 2 value during the cuff inflation phase (StO 2min ), which makes the time to StO 2max longer for those individuals with higher O 2 extraction capacity. 17,23 In support of this observation, a significantly longer time to StO 2max was found in healthy individuals compared to older adults at CVD risk, which may have been the result of a lower StO 2 value reached during cuff inflation (StO 2min ), combined with the higher StO 2 value reached during the cuff deflation phase (StO 2max ).
Previous studies have failed to show significant differences in time to StO 2max in individuals with metabolic syndrome 24 and older adults at risk for CVD. 10 However, Kragelj et al. 18 found a significantly longer time to StO 2max in individuals with peripheral vascular disease (a clinical population) compared to healthy participants. Interestingly, the groups investigated in their study 18 did not exhibit a significant difference in O 2 extraction parameters, which may explain the expected longer time to StO 2max in individuals with peripheral vascular disease. Therefore, adoption of the recovery time parameters (i.e., time to StO 2max ) should be recommended when the clinical population evaluated is not assumed to exhibit impairment in muscle O 2 extraction; otherwise, data interpretation and comparison may be rendered incorrect.
The current study also demonstrated a significant reduction in ΔStO 2reperfusion and StO 2AUCt in healthy older adults and those with CVD risk factors, compared to healthy young individuals. However, no significant difference was observed in these parameters in healthy older adults compared to older adults at risk for CVD, suggesting that ΔStO 2reperfusion and StO 2AUCt are not sensitive parameters for detecting the effect of CVD risk factors on reactive hyperemia. The ΔStO 2reperfusion and StO 2AUCt parameters can be interpreted as measures of the magnitude of O 2 resaturation, since the lowest and highest StO 2 value (amplitude) within the reperfusion phase (cuff deflation) are considered for calculating these parameters. However, it is important to note that, in common with time to StO 2max , these parameters may also be affected by parameters of muscle O 2 extraction during cuff deflation (i.e. StO 2min ).
Corroborating this idea, Rosenberry et al. 17 demonstrated that the impaired StO 2AUCt observed in older adults at CVD risk in comparison to healthy young individuals was eliminated when the level of muscle O 2 extraction (StO 2min ) was standardized by varying occlusion times (3-, 4-, and 5-min cuff occlusion) across the two age groups. It is likely that differences between ΔStO 2reperfusion and StO 2AUCt parameters do not only necessarily reflect impairment in O 2 resaturation magnitude (reactive hyperemia), but also in muscle O 2 extraction during occlusion (mitochondrial dysfunction). 7,17 In light of these findings, since NIRS continuously monitors muscle O 2 saturation throughout the VOT, assessment of ΔStO 2reperfusion and StO 2AUCt could be performed controlling the level of tissue ischemia (varying the occlusion times), 17 which would be crucial to better interpret ΔStO 2reperfusion and StO 2AUCt . Alternatively, AUC above_base is not affected by muscle StO 2min and its utilization might be recommendable.

Experimental considerations
Of note, the NIRS system is a relevant tool that can be easily used during vascular surgery, as it continuously assesses muscle O 2 saturation. For example, a previous study conducted at our laboratory showed that prolonged occlusion of limb blood flow (i.e., tourniquet application during limb surgery) causes transient microvascular dysfunction that can be assessed by NIRS-derived reperfusion upslope. 25 Additionally, this study also demonstrated that using a local nitric oxide donor (transdermal nitroglycerin) administered during prolonged forearm cuff occlusion protected against this microvascular dysfunction, 25 which is relevant for the field of vascular surgery. Moreover, NIRS systems have been widely utilized to detect impaired microvascular reactivity in many clinical populations, such as hypertension, older adults at risk for CVD, and patients with metabolic syndrome. 4,10,24 Although NIRS systems are largely utilized to assess oxygenation of the forearm muscles, one limitation of this technique is the discrete volume of a specific muscle (2 to 6 cm) and the superficial area of skeletal muscle that NIRS examines. 26 Furthermore, another important factor that should be considered is that adipose tissue thickness can affect NIR light penetration into the muscle of interest and, consequently, the muscle oxygenation levels measured. However, it is important to point out that no significant differences were observed in skin fold thickness between groups, ensuring that the depth of light penetration (~18 mm) sufficiently interrogates the forearm flexor muscles. 26,27 Additionally, it is noteworthy that all participants in this study did not undergo a clinical examination by a clinician prior to muscle oxygenation measurements. Although we are aware that it would be important for all participants to undergo clinical examination, the main purpose of this study was to investigate the effectiveness of tissue oxygenation parameters in detecting impaired reactive hyperemia in the microvasculature of a population previously known to have microvascular damage. 3,4,7,9 Therefore, the absence of clinical examination does not compromise the results of this study.
In conclusion, the findings of this study suggest that StO 2slope_10s was the most sensitive NIRS parameter for assessing reactive hyperemia. Additionally, the magnitude of reperfusion (i.e., StO 2AUCt ) and recovery time (i.e., time to StO 2max ) parameters should be interpreted with caution when assessing clinical populations, since muscle O 2 extraction may affect those parameters. The results of the present study will be helpful for future studies that investigate the effect of an external factor (disease, means of treatment, etc.) on muscle O 2 resaturation parameters.