Improved extremity tissue oxygenation with short-term exposure to textiles embedded with far infrared light emitting thermoactive particles in patients with diabetes mellitus

Abstract

Use of far infrared (FIR) energy may improve peripheral circulation. This study aimed to determine whether FIR delivered through textiles improves peripheral microcirculation as measured by transcutaneous oximetry (TcPO₂) in individuals with diabetes mellitus (DM).

Methods

A single-center, prospective, randomized, placebo-controlled, double-blinded, crossover study of 32 subjects with either type 1 or type 2 DM. An active FIR wrap followed by a placebo wrap (or vice versa) was applied to the arm, calf, ankle, and forefoot for 60 min each with continuous TcPO₂ measurements. The treatment effect of the active versus placebo wrap was estimated using a linear mixed effect model adjusted for period, sequence, baseline value, and anatomic site.

Results

The active FIR wrap increased mean TcPO₂ at the arm (2.6 ± 0.8 mmHg, p = .002), calf (1.5 ± 0.7 mmHg, p = .03), and ankle (1.7 ± 0.8 mmHg, p = .04) and composite of all sites (1.4 ± 0.5 mmHg, p = .002) after 60 min. The estimated treatment effect was significant for the active FIR wrap at the calf (1.5 ± 0.7 mmHg, p = .045) and in composite of all sites (1.2 ± 0.5 mmHg, p = .013).

Conclusion

Short-term exposure to FIR textiles improves peripheral tissue oxygenation in patients with diabetes.

Key message

• • FIR radiation delivered through textiles rapidly increases transcutaneous oxygen of distal extremities in individuals with diabetes.

Introduction

Diabetes mellitus (DM) and its associated peripheral arterial disease can lead to foot ulcerations, infections, the need for revascularization procedures, and amputations. The microvascular disease associated with diabetes is characterized by dysregulated blood flow to the skin and peripheral sympathetic nerves, which, in turn, deranges vasodilation and capillary permeability. This, in combination with atherosclerotic macrovascular disease, can cause functional ischemia in the lower extremities of patients with diabetes. ¹ Glycemic control and optimized cardiovascular risk factors are the mainstays of therapy to prevent diabetic microvascular disease. Despite advances in diabetes care over the past three decades, diabetic foot disease remains a common complication. ² Up to one quarter of individuals with diabetes will develop a foot ulcer during their lifetime,^2,3 a risk that is further compounded by sensory loss associated with peripheral neuropathy, which afflicts more than half of the population living with diabetes. In the United States, foot ulcers remain one of the leading causes of hospitalization for those with diabetes ⁴ and is estimated to cost up to 3 billion dollars in excess of diabetes-related cost. ⁵ There is therefore an urgent need for novel strategies to help prevent this morbid and costly complication.

Far infrared (FIR) light emits energy that can be harnessed as heat and has been demonstrated to improve microcirculatory function in a variety of conditions.^6–8 Recently, garments with FIR emitting fibers have been developed, with several studies in healthy volunteers demonstrating local increases in circulation after their wear. FIR emitting textiles can be used continuously for longer periods with low-risk ⁹ and are a promising non-invasive, adjunctive therapy for patients with circulatory disease.^10–14 An unpublished pilot study in patients with diabetes showed that wearing stockings made with thermoactive particles has a promising effect on pedal circulation. ¹⁵ In people with diabetic neuropathy, FIR therapy has been shown to improve peripheral protective sensation, restore temperature discrimination, and reduce pain in the foot.^12,16,17

In this pilot study, we hypothesized that exposure to FIR, using textile wraps containing woven thermoactive particles including silicon dioxide and aluminum oxide as part of a proprietary blend (Circufiber®), augments oxygen tension and therefore microcirculation in the arm, lower legs, and feet, of subjects with diabetes, as measured by transcutaneous oximetry (TcPO₂).^18–20 Our study was a prospective, randomized, placebo-controlled, double-blinded, crossover study in individuals with type 1 or 2 DM but without overt peripheral arterial disease. In our placebo-controlled crossover design, each participant wore the active FIR wrap or placebo wrap with serial measurements of TcPO₂ taken over a period of 60 min and, following a brief washout period and 15-min break, then switched to the opposite wrap for a second round of serial measurements of TcPO₂ for another 60 min. The sequence of wrap placement was randomly assigned.

Materials and methods

Study subjects

Adults with either type 1 or type 2 DM, followed in an outpatient endocrinology clinic were recruited from April 2021 through July 2022. The inclusion criteria were adults age between 18 and 80 years old and diagnosis of DM (either Type 1 or Type 2). The exclusion criteria were: chronic kidney disease with eGFR <30 mlmin/1.73 m², use of systemic corticosteroids within the previous 3 months, alcohol or any illicit substance abuse, has ever received immunosuppressive agents, has ever undergone radiation therapy, has ever used cytotoxic agents, pregnant, breastfeeding or attempting to become pregnant, history of known peripheral arterial disease or lower extremity amputation, any neurologic disease or other condition preventing normal ambulation, history of significant trauma to the lower extremities, major orthopedic or neurological damage to the lower extremities, history of saphenous venous graft harvesting, any ulcer or open wound on the lower extremities, inability to lie flat for up to 90 min at a time, non-English speaker, any COVID19-like symptoms within the past 21 days, and COVID19 positivity within the past 21 days. Participants underwent ankle brachial index (ABI) measurements and those with severe peripheral arterial disease (ABI <0.4) were excluded. In total, we recruited 39 subjects; 1 was excluded for ABI <0.4, 6 withdrew prior to randomization, 1 withdrew after the first half of the crossover trial. An intention-to-treat analysis was performed on the 32 participants who consented and were randomized for the study (Figure 1). The research protocol was approved by the Yale University Human Investigations Committee, our institutional review board.

Figure 1.

CONSORT diagram of enrollment, participation, and analysis.

Intervention and outcome

In our pilot set-up, we observed that wearing an actual sock containing woven thermoactive particles tended to compress the oximetry sensors, leading to an immediate and false decrease in TcPO₂. Therefore, in this study, we used wraps made of the exact same textile used in the socks, allowing us to leave the sensors in place and minimize compressive interference. The active FIR and placebo wraps were identical with the exception that the placebo wrap did not contain the proprietary blend of thermoactive particles including a mixture of 9 minerals with a predominance of silicon dioxide and aluminum oxide woven into the fabric. We further allowed for re-equilibration of the sensors after placing the wraps at the start of the active phase. Of note, previous reports using FIR socks did not comment on how this technical challenge was addressed and its impact on the magnitude of change in TcPO₂ measurements.

Strict COVID-19 protocols were followed throughout the trial including masking of all subjects and all study personnel. Subjects were seated on a reclined examination table with legs extended and supported by leg rest. Following skin cleaning and shaving, a transcutaneous oximetry sensor lead was individually placed on the arm, lower leg, ankle, and foot (Figure 1). With the subjects in a relaxed and comfortable position, 30 min of baseline data were recorded. Following the baseline period, and while keeping the sensors in place, the subject’s arm, lower leg, ankle, and foot were wrapped in either an active FIR wrap or a placebo wrap for the initial phase of the crossover trial (Figure 2). Randomization of the wrap sequence (active-placebo or placebo-active) was chosen by a computerized random number generator. There was a five-minute period of re-equilibration after wrap application during the active phase. TcPO₂ measurements were taken continuously and recorded at least every 5 min for 60 min during the active phase using the PeriFlux 5000 radiometer (Radiometer America, Inc). Subjects were given a 15-min break wherein the wrap and sensors were removed. In the second phase of the crossover trial, the sensors were replaced as prior with 30 min of stabilization followed by placement of the second wrap for another active phase lasting 60 min with TcPO₂ measured as prior. Subjects were left in a supine position and encouraged to minimize movement to minimize sensor interference during the active and stabilization phases of the protocol.

Figure 2.

Placement of transcutaneous oximetry leads on anatomic sites. (a) Placement of sensor on arm without wrap. (b) Arm in textile wrap over sensor. (c) Placement of sensor lead on partially wrapped calf, ankle and forefoot. (d) Leg in textile wrap over sensors.

Data analysis

Demographic and clinical characteristics of subjects were presented using descriptive statistics including medians and interquartile ranges (IQR) for continuous variables, or frequencies and percentages for categorical variables. TcPO₂ measures were displayed as means and standard errors. In this crossover design, a carryover effect was possible and inherent in the sequence of the wrap. Thus, we evaluated the carryover effect and period effect by analyzing the variance in a general linear model including treatment, period and sequences nested in each subject. For the main outcomes, we performed linear mixed effect models adjusted for period, sequence, baseline TcPO₂, and anatomic site as fixed effects and participant as the random effect to assess the difference in TcPO₂ between baseline and 60 min of exposure to the active FIR wrap or placebo, and to estimate the treatment effect by comparing the mean change in TcPO₂ between the active FIR versus placebo wrap. Statistical tests were two-tailed and an alpha of 0.05 was considered as statistically significant. All analyses were preformed using SAS 9.4 (Cary, NC). Participants, investigators, data collectors, and data analysts were blinded until the completion of data analysis.

For subgroup analysis, data from certain anatomic sites were excluded if the oximetry sensor disconnected during the active period or there was otherwise evidence of sensor malfunction. In pre-trial design, we calculated a sample size of 44 participants to achieve 80% power with 0.05 two-side type 1 error for a 6 mmHg in TcPO₂ difference between treatment groups. The trial was discontinued after two years of recruitment due to lengthy delays and ongoing challenges posed by COVID-19 leading to staffing and location limitations.

Results

Baseline characteristics of subjects

Table 1 shows the baseline characteristics of subjects. The median age was 58 years (IQR: 52–68 years). There was a preponderance of female (75%) and White (84.4%) subjects. Seventy-five percent of participants had type 2 diabetes and 25% had type 1 diabetes. Their median hemoglobin A1c was 6.8%. Diabetic complications were identified in 21.9% of subjects, with a distribution of retinopathy (15.6%), nephropathy (6.3%) and cardiovascular disease (6.3%). More than half the participants were overweight or obese. All participants who completed the trial had normal range ankle brachial indices (ABI) at enrollment.

Table 1.

Baseline characteristics of subjects with diabetes mellitus. Data are represented by median and interquartile range or relative frequency. BMI: Body mass index; LDL: low density lipoprotein.

	Overall [N = 32]
Age
	58.0 (52.0–68.0)
Sex
Female	24 (75%)
Male	8 (25%)
Race
White	27 (84.4%)
Non-White	5 (15.6%)
Ethnicity
Hispanics	1 (3.1%)
non-Hispanics	31 (96.9%)
BMI
Median	29.9 (27.2–35.6)
Normal	4 (12.5%)
Overweight	12 (37.5%)
Obesity	16 (50.0%)
Diabetes mellitus
Type 1 DM	8 (25.0%)
Type 2 DM	24 (75.0%)
Complications
Yes	7 (21.9%)
Retinopathy	5 (15.6%)
Nephropathy	2 (6.3%)
CVD	2 (6.3%)
Ankle brachial index
Right, median	1.3 (1.2–1.5)
Left, median	1.2 (1.1–1.4)
Average, median	1.3 (1.2–1.4)
Hemoglobin A1c%
Median	6.8 (6.5–7.6)
LDL
Median	76.0 (65.0–88.0)
Wrap sequence
AB	16 (50%)
BA	16 (50%)

Higher unadjusted mean TcPO₂ with active FIR wrap

The unadjusted mean TcPO₂ over the 60-min exposure between active FIR wrap and placebo wrap at any site was not significantly different but trended higher for the active wrap compared to placebo at all individual sites (Figure 3). We visualized a divergence in mean TcPO₂ between the active and placebo wraps at 40 minutes at the arm, ankle, calf, and composite of all sites, with a steeper rise in TcPO₂ for the active wrap (Figure 3). To approximate the distribution of a sock, we observed the mean TcPO₂ of the calf and ankle together and observed it to be higher for the active compared to placebo wrap, also with a noticeably greater rate of increase around 40 min (Supplementary Figure 1). The forefoot was excluded from this lower limb composite analysis due to unreliable TcPO₂ readings with challenging placement of the oximetry sensor in this anatomical area due to lack of soft tissue and difficulty in consistently finding a flat surface.

Figure 3.

Mean and standard error of TcPO₂ measures over 60 min of the arm, ankle, calf, forefoot, and composite of all sites.

Active FIR wrap increases TcPO₂ in arm, ankle, calf, and composite in adjusted model

Given the crossover design of this study, we tested for carry-over and period effects and did not find a significant trend for either. Using a linear mixed effect model adjusting for period effect, sequence effect, and baseline value, we observed a statistically significant increase in the mean TcPO₂ from baseline to 60 min for the active FIR wrap at all sites except the forefoot: arm (2.6 ± 0.8 mmHg, p = .002), calf (1.5 ± 0.7 mmHg, p = .03), and ankle (1.7 ± 0.8 mmHg, p = .04) (Table 2). In contrast, there was no significant change over time with the placebo wrap. We performed a composite analysis of all the anatomic sites, with sites added as co-variates to the model, and found a statistically significant increase in the mean difference of TcPO₂ over the exposure period for the active wrap (1.4 ± 0.5 mmHg, p = .002) (Table 2). To evaluate the potential use of FIR textiles as a sock in patients with diabetic foot disease, we pooled data from the lower extremity sites, excluding the forefoot (due to the aforementioned unreliable TcPO₂ readings) and observed a significant mean increase in TcPO₂ over 60 min from the ankle and calf, when analyzed together (1.5 ± 0.5 mmHg, p = .005), which was also not observed for the placebo wrap (Supplementary Table 1).

Table 2.

Estimated treatment effect using linear mixed effect model.

Site	FIR active TcPO2 (60 min–5 min)		Placebo TcPO2 (60 min–5 min)		Estimated treatment effect FIR active versus placebo
	mmHg (SE)	p-value	mmHg (SE)	p-value	mmHg (SE)	p-value
Arm 1	2.6 (0.8)	0.002	1.2 (0.8)	0.14	1.4 (1.2)	0.25
Ankle 1	1.7 (0.8)	0.04	−0.2 (0.8)	0.76	2.0 (1.3)	0.14
Calf 1	1.5 (0.7)	0.03	0.04 (0.6)	0.94	1.5 (0.7)	0.045
Forefoot 1	−0.13 (0.84)	0.88	0.06 (0.80)	0.94	−0.19 (1.17)	0.87
Composite 2	1.4 (0.5)	0.002	0.2 (0.4)	0.65	1.2 (0.5)	0.013

¹Model adjusted with period effect, sequence effect, and baseline value (t0).

²Model adjusted with period effect, sequence effect, and baseline value (t0), and body part.

60-min exposure active FIR wrap has a positive treatment effect

To determine whether the observed increases in TcPO₂ were attributable to the thermoactive FIR properties of the wrap, we estimated the treatment effect versus placebo and found significantly positive effects for the active wrap at the calf (1.5 vs 0.04 mmHg, p = .045) as well as in composite analysis (1.4 vs 0.2 mmHg, p = .013) (Table 2). The estimated treatment effect was also positive for the active wrap in composite without the forefoot (1.9 vs 0.3 mmHg, p = .005), and ankle and calf (1.5 vs. −0.2 mmHg, p = .02) (Supplementary Table 1). While the changes at the arm and ankle were numerically larger with the FIR textile wrap, the differences versus placebo wrap did not achieve statistical significance.

Discussion

Infrared light is a form of electromagnetic energy with wavelengths longer than that of visible light, which can be categorized into three groups according to wavelength. Far infrared (FIR) is thought to transfer energy subcutaneously in the form of heat, perceived by thermoreceptors in the skin as radiant heat. In the context of circulatory diseases, through thermal and non-thermal effects, FIR energy has been shown to increase arterial blood flow and peripheral circulation, promote capillary dilatation and improve endothelial function. ⁶ While the exact mechanisms are unknown, studies of FIR therapy in diabetes suggest an effect through nitric oxide regulation, reduced oxidative stress and inflammation, and, over time, increased neovascularization. ²¹

FIR emitting textiles are interwoven with thermoactive mineral particles. They are purported to work by absorbing natural body heat and then re-emitting energy back to the skin as FIR light. ⁶ One of the major attractions for using FIR as adjunctive therapy is its low-to-no risk safety profile. ⁹ Two major advantages of FIR textiles over previously used non-textile therapeutic modalities such as ceramic heaters and sauna therapy are: (1) the potential for longer-term and continuous application with textiles and (2) no potential to overheat the skin. In healthy individuals, FIR shirts have been shown to increase TcPO₂ by 5–8% compared to placebo. ¹⁴ Healthy men and women who wore gloves and socks made with thermoactive particles were shown to improve mean TcPO₂ levels in unpublished studies.^10,11 In another unpublished study of high risk diabetic subjects, wearing FIR emitting gloves and stockings was associated with an 8–12% increase in TcPO₂ within 40–50 min. ¹⁵ Use of a FIR warming blanket for 20 min improved microcirculation, as measured by laser Doppler flow, in subjects with type 2 diabetes. ²² Together, these studies suggest that FIR therapy delivered through garments or textiles have promising potential to improve peripheral circulation in patients with diabetes, but overall there is a paucity of peer reviewed clinical studies in individuals with diabetes. Given that diabetic foot disease due to microcirculatory dysfunction is a major and costly complication of diabetes, use of FIR textiles and garments as a complementary approach to improving microcirculation is attractive.

Our study found that Circufiber® textile with thermoactive particles (active wrap) significantly increases transcutaneous oxygen in the upper and lower extremities after 60 min of exposure compared to baseline in individuals with type 1 and type 2 diabetes while the placebo had no effect on increasing TcPO₂. The active FIR wrap versus placebo showed a significant treatment effect but not all sites, likely due to inadequate power. This proof-of-concept study showed that the FIR-emitting fibers in the active wrap can improve tissue oxygenation relatively quickly in the distal extremities of individuals with diabetes in comparison to placebo. These findings suggest that more prolonged exposure to the FIR textile, such as being worn daily as a diabetic sock, could potentially lead to sustained improvements in microcirculation.

This study had many strengths. Participants included individuals with both type 1 and type 2 diabetes. Wraps were used in this study, as opposed to socks, to avoid compressive effects that would interfere with transcutaneous oximetry measurements and allowed us to leave the sensors in place when applying the wraps. We used transcutaneous oximetry as a non-invasive tool to approximate microcirculation in the distal extremities. It is commonly used to predict wound healing potential and for qualification of hyperbaric oxygen therapy in patients with diabetic foot disease and has even been proposed as a useful tool to screen patients with diabetes at risk for foot ulceration, ²³ but its technical challenges may limit its use in longer term studies given the sensitivity of the leads to disturbance by compression or motion. In addition, prolonged skin exposure to the leads may cause erythema or skin reaction when exceeding 60–90 min of use. While there are inherent limitations to a crossover design study, we found no carry-over or sequence effects. Limitations of the study include technical challenges in using the TcPO₂ probe in certain anatomical sites, over long periods of time, and in avoiding lead removal and compression when a garment is applied. Our study was also limited by a narrow diversity of study participants and short exposure period to the intervention. Despite these challenges and a smaller sample size, we still observed a significant improvement with the active FIR wrap over placebo. Additional studies may address these limitations using alternative modalities to measure transcutaneous oxygen, particularly on the plantar surface of the forefoot or distal foot, and over extended periods of continuous application of the FIR fabric. Since many diabetic foot ulcers occur in the distal forefoot including the toes, a more accurate assessment of efficacy in this anatomic area would be important in future studies. Neuropathy was an exclusion criteria in this study and therefore the effect of FIR on diabetic neuropathy could not be assessed, but would be an area of interest for additional studies.

We have demonstrated that exposure to a textile with woven thermoactive particles that emit FIR radiation improved transcutaneous oxygenation in the arm and leg, suggesting an improvement in microcirculation. Future studies with such garments could include randomized prospective trials of patients with diabetes and known vascular disease at an elevated risk for diabetic foot complications to assess the clinical effects of FIR on TcPO₂ in that setting. Moreover, longer term trials could ultimately test the clinical efficacy of this intervention on ulcer formation, ulcer healing, need for surgical intervention, peripheral neuropathy, and quality of life measures including pain scale. In addition, mechanistic studies to better understand the cellular changes leading to improved perfusion are warranted. Circufiber® garments are a promising, non-invasive, and low risk complementary therapy to improve microcirculation in diabetes with the potential to decrease the risk of diabetic vascular complications.

Appendix.

Abbreviations

ABI

ankle brachial index

BMI

body mass index

DM

diabetes mellitus

FIR

far infrared

IQR

interquartile range

LDL:

low density lipoprotein

TcPO₂

transcutaneous oximetry

ORCID iDs

Diana Athonvarangkul https://orcid.org/0000-0001-9697-5248

Adam Mayerson https://orcid.org/0000-0001-9336-3828