Secondary Logo

Journal Logo

ONLINE EXCLUSIVES

Skin Autofluorescence as a Novel and Noninvasive Technology for Advanced Glycation End Products in Diabetic Foot Ulcers: A Systematic Review and Meta-analysis

Varikasuvu, Seshadri Reddy PhD; Varshney, Saurabh MS; Sulekar, Harish MS, MCh

Author Information
Advances in Skin & Wound Care: November 2021 - Volume 34 - Issue 11 - p 1-8
doi: 10.1097/01.ASW.0000792932.01773.d5
  • Free

Abstract

INTRODUCTION

With a projected 592 million diagnoses by the year 2035, diabetes has become a major health concern globally. Approximately 15% of patients with diabetes develop foot wounds, and the reported lifetime incidence of diabetic foot ulcers (DFUs) ranges from 10% to 25%. Long-standing diabetes, chronic hyperglycemia, increased glycation, increased advanced glycation end products (AGEs), and diabetes-related complications are risk factors for developing DFUs.1–4 Recent studies have proposed measuring skin autofluorescence (SAF) as a novel technique for assessing tissue accumulation of AGEs.5 The level of SAF may be associated with the risk of vascular complications and mortality in diabetes.6,7 However, studies to date are limited and inconsistent about SAF levels and its use in patients with DFUs.8–15 Therefore, in this meta-analysis, one of the objectives was to compare SAF levels between patients with and without DFUs.

Despite standard treatment and management procedures, a substantial number of patients with DFUs undergo amputation, affecting their mobility; risking high morbidity and mortality; and causing profound burden on their financial, family, personal, and social domains.16,17 Accordingly, this investigation was also conducted to determine any significant association between SAF and DFU risk.

Of note, measuring circulatory AGEs is less reproducible and of limited clinical use because plasma AGEs do not accurately reflect tissue AGE accumulation.18 In contrast, SAF measurements using AGE-reader technology provide a novel and noninvasive method for assessing tissue AGEs by exploiting their fluorescent properties.5 Further, SAF levels exhibit strong associations with collagen-linked fluorescence and accumulated tissue AGE content in skin biopsies,5,18 glycemic indices such as glycated hemoglobin (HbA1c), and several diabetic complications.6,7 Therefore, the authors hypothesized that SAF measurements reflecting in vivo AGE accumulation, metabolic memory, and glycemic control over longer time periods could be useful as a novel technique in patients with DFU.

The current health costs associated with DFUs are substantial,16,17,19 but the SAF technology provides an opportunity to reduce the healthcare burden. In addition to its reproducibility over traditional invasive techniques, SAF measurements are simple to obtain.10,20 Accordingly, the use of SAF as a novel marker of AGEs for early identification of patients at risk of DFU might reduce clinical burdens and social and economic stress.

METHODS

The PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analysis) criteria were followed in the conduct and reporting of results for this review. This study is part of a protocol registered with the International Prospective Register of Systematic Reviews (PROSPERO no. CRD42020161661).

The literature search was primarily conducted in the PubMed/MEDLINE database and other electronic search engines including Google Scholar, The Cochrane Library, Science Direct, Author Mapper, and China National Knowledge Infrastructure by two researchers independently. In addition, a manual search for relevant studies was performed in the bibliographies of the retrieved citations. No retrieval filters such as publication types, time, and language were applied.

The main search strategy used the keywords “skin autofluorescence”[All Fields] AND (“diabetic foot”[MeSH Terms] OR (“diabetic”[All Fields] AND “foot”[All Fields]) OR “diabetic foot”[All Fields] OR (“diabetic”[All Fields] AND “foot”[All Fields] AND “ulcer”[All Fields]) OR “diabetic foot ulcer”[All Fields]). The search was conducted using both the MESH words and general text words. The keywords used for literature search included advanced glycation end products, autofluorescence, skin autofluorescence, diabetes, diabetic foot, diabetic wound, diabetic foot ulcer, diabetic foot syndrome. Authors identified and removed duplicates based on identical authors, affiliated institutions, participants, and reported data.

Studies reporting the SAF values in patients with DFU compared with participants without DFU were included in the primary outcome analysis. The primary outcome was the difference in SAF between participants with or without DFUs. Studies reporting unadjusted or adjusted odds ratios (ORs) or hazard ratios (HRs) with their 95% confidence interval (CI) for DFU were included to estimate a pooled OR for the association of SAF with DFU risk. All study types reporting SAF in patients with DFU were included. Non-DFU groups consisted of patients with diabetes but without DFUs. The exclusion criteria were studies reporting SAF in diabetes and associated complications other than DFUs and studies with no SAF data such as reviews, editorials, commentaries, and letters.

Data Extraction and Quality Assessment

Two researchers independently and thoroughly screened the full texts of all relevant studies on SAF in DFUs to extract the following information: first author names, country and year of publication, study type and follow-up duration, sample sizes of men and women, age and body mass index (BMI), disease duration, glycemic indices, SAF method and its values in the DFU and non-DFU groups, several covariates used in adjustment, and the unadjusted and/or multivariate adjusted ORs for the association of SAF with DFU risk.

Study quality was assessed using the Newcastle-Ottawa Scale21 based on three domains: selection (4 points maximum), comparability (2 points maximum), and outcome (3 points maximum). Scores ranged from 0 to 9 points; studies scoring 7 or more were considered good; 4–6 points, fair; and 3 or fewer points, poor.

Two authors independently performed data extraction and quality assessment. Any disagreements were resolved in discussion with other authors. If any additional relevant information was required, the studies’ corresponding authors were contacted using the emails provided.

Data Analysis

The standardized mean difference (SMD) and its 95% CI were computed to determine the differences in SAF between DFU and non-DFU groups. In case of significant between-study heterogeneity, a random-effects model was used to compute SMD; otherwise, a fixed-effects model was used. The effect size for SMD values was presented as a Z score. Any Z score with P < .05 was considered statistically significant. Between-study heterogeneity was examined with the Cochrane Q statistic and expressed as percentages of I2. P < .10 or I2 statistic of >50% indicated a significant heterogeneity.

Metaregression analysis was performed using country, sample size, age, BMI, sex, HbA1c, and diabetes duration as covariates. A one-study leave-out sensitivity analysis was performed to test the robustness of this meta-analysis. All comparisons were two-tailed, and analyses were conducted using Open-Meta software (Tufts Medical Center, Boston, Massachusetts) and Review Manager Software version 5.3 (Cochrane Collaboration, Copenhagen, Denmark). A meta-analysis of ORs was conducted to determine the association of SAF with DFU risk. Pooled unadjusted and adjusted ORs with 95% CIs were obtained. Statistical significance was considered P < .05. This meta-analysis of ORs was performed using Comprehensive meta-analysis software version 3 (Biostat, Englewood, New Jersey).

RESULTS

Search Results

The PRISMA flow diagram for this meta-analysis is shown in Figure 1. A total of 677 records were preliminarily identified. After screening the titles and abstracts of all relevant articles and removal of duplicates, eight potential articles were considered for full-text review. Of these, two articles were excluded for not having relevant data, and six studies were included in the final analysis.8–13

Figure 1
Figure 1:
STUDY FLOW DIAGRAM

Study Characteristics

Table 1 summarizes the characteristics of the five included studies8–12 with 611 participants that compared SAF between DFU and non-DFU groups. All of these were prospective studies published from 2005 to 2019 in English. The individual study sample sizes ranged from 16 to 70 and 23 to 170 in the DFU and non-DFU groups, respectively. The study participants from three studies comprised patients with both types 1 and 2 diabetes.8,11,12 The patients with diabetes in one study8 had Charcot neuroarthropathy, a challenging form of diabetic foot syndrome, and the presence of metabolic syndrome was reported in another study.13 Men outnumbered women in all of the included studies. In the DFU group, the mean/median age of DFU patients ranged from 57 to 68 years, and the BMI ranged from 23 to 35 kg/m2. Diabetes duration ranged from 11 to 17 years, and HbA1c values ranged from 7.4 to 9.2. The non-DFU group participants were matched with their DFU counterparts in age (except one study10), sex, HbA1c, and diabetes duration. The other covariates such as smoking status, hypertension, insulin use, angiotensin-converting enzyme inhibitors/angiotensin-receptor blockers, and the amount of micro- and macroangiopathy are summarized in Table 1.

Table 1 - STUDY CHARACTERISTICS
Study Author, Year, Country, Study Type Group n (M/F; T1/T2DM) Age, y; BMI, HbA 1c , % Clinical Characteristics, n Other Characteristics, n Matching Newcastle-Ottawa Scale
Araszkiewicz et al,8 2015, Poland, prospective case-control DFU 70 (54/16; 17/53) 59, 31.6, 7.6 DD, 16 y; HT, 47; smokers, 8; eGFR, 75.9; CRP, 3.8;VPT-L, 35.6; VPT-R, 37 Retinopathy, 56; nephropathy, 32; neuropathy, 70; macroangiopathy, 30 Age, sex, BMI, DD, HT, smoking, TC, TG, LDL-C, HDL-C, macroangiopathy 6
Non-DFU 70 (54/16;18/52) 60, 31.9, 8.4 DD, 15 y; HT, 53;
smokers, 15; eGFR, 86.3; CRP, 1.9; VPT-LF [V], 20; VPT-RF [V], 18.9
Retinopathy, 30; nephropathy, 14; neuropathy, 31; microangiopathy, 33
Fernando et al,9 2019, Australia, prospective case-control DFU 16 (10/6; 0/16) 63.4, 35.4, 7.4 DD, 15.1 y; HT, 14; smokers, 1; INS use, 8; eGFR, 66.1; CRP, 4.1 DL, 11; stroke, 1; IHD, 5; CHD, 1; CPD, 4; CLD, 2; CRI, 4 Age, sex, ethnicity, BMI, DD, HbA1c, HT, smoking, TC/HDL-C, CRP 7
Non-DFU 63 (44/19; 0/63) 63.3, 35.4, 7.2 DD, 10.8 y; HT, 41; smokers, 5; INS use, 16; eGFR, 78.5; CRP, 2.8 DL, 41; stroke, 2; IHD: 15; CHD, 8; CPD, 11; CLD, 5; CRI, 8
Hu et al,10 2012, China, prospective cross-sectional (follow-up) DFU 25 (15/10; NA) 68.2, 22.92, 9.25, DD, 11.07 y; smoke index, 200; CRP, 35.57; FBG, 9.65; VPT, 25.69; ABI-L: 0.90; ABI-R: 0.92 NA Sex, BMI, WC, smoke index, PBG, HbA1c, BUN 7
Non-DFU 170 (96/74; NA) 57.01, 23.57,9.47 DD, 6.71 y; smoke index, 274.58; CRP, 9.42; FBG, 10.43; VPT, 17.29; ABI-L, 1.08; ABI-R: 1.08
Meerwaldt et al,11 2005, the Netherlands, prospective cross-sectional DFU 24 (13/10; 5/18) 57, 30, 8.2 DD, 17 y, smokers, 10; ABI, 1.17; Wagner score (1/2/3/4), 5/8/7/4 Retinopathy, 21; MAU, 13; CHD, 4; INS use: 18; ACEi, 13; diuretics, 10; antioxidants, 6 Age, sex, BMI, DD, HbA1c, BP, Cr, TC, TG, smoking, INS, antioxidants, CHD 7
Non-DFU 23 (13/10; 8/15) 53,28.4, 7.6 DD, 13 y; smokers, 7; ABI, 1.23; Wagner score (1/2/3/4), none Retinopathy, 5; MAU, 3; CHD, 2; INS use, 19; ACEi, 7; diuretics, 3; antioxidants, 5
Vouillarmet et al,12 2013, France, prospective (follow-up) DFU 66 (45/11; 10/56) 63.3, 29.5, 7.8 DD, 17 y Retinopathy, 55%; nephropathy, 50%; macroangiopathy, 49%; INS use, 73%; statins, 67%; ACEi, 76% DD 7
Non-DFU 84, NA NA, NA, 8.4 NA NA
Abbreviations: ABI, ankle-brachial index; ACEi, angiotensin-converting enzyme inhibitors; ARBs, angiotensin II receptor blockers; BMI, body mass index (kg/m2); BUN, blood urea nitrogen; CHD, chronic heart disease; CLD, chronic liver disease; CPD, chronic pulmonary disease; Cr, creatinine; CRI, chronic renal insufficiency; CRP, C-reactive protein (mg/dL); DD, duration of diabetes (y); DFU, diabetic foot ulcer; DL, dyslipidemia; e-GFR, estimated glomerular filtration rate (mL/min per 1.73 m2); F, female; FBG, fasting blood glucose (mmol/L); HbA1c, glycated hemoglobin (%); HDL, high-density lipoproteins; HT, hypertension; IHD, ischemic heart disease; INS, insulin; LDL, low-density lipoprotein; M, male; MAU, microalbuminuria; NA, not available; SAF, skin autofluorescence; T1/T2DM, type1/type 2 diabetes mellitus; TC, total cholesterol; TG, triglycerides; VPT-L, vibration perception threshold-left foot; VPT-R, vibration perception threshold—right foot; WC, waist circumference.

Table 2 summarizes the characteristics of the three included studies10,12,13 with 437 participants that reported the relationship between SAF and DFU risk. All of these were recently published English-language studies. The percentage of DFU cases across these studies ranged from 13 to 44. All three studies followed the International Diabetes Federation, American Diabetes Association, or International Working Group on the Diabetic Foot criteria. These studies reported the association of SAF with DFU in the form of unadjusted and adjusted ORs with 95% CI.

Table 2 - CHARACTERISTICS OF STUDIES INCLUDED IN THE META-ANALYSIS OF ODDS RATIOS
Variable Hu et al,10,14 2012 Monami et al,13 2008 Vouillarmet et al,12 2013
Country China Italy France
Study type Cross-sectional (follow-up) Cross-sectional Prospective (follow-up)
NOS 7 6 7
Total n 195 92 150
DFU (%) 12.8 23.9 44
Non-DFU (%) 87.2 76.1 56
Criteria IWGDF IDF ADA and IWGDF
Outcome Unadjusted and adjusted ORs Unadjusted and adjusted ORs Unadjusted and adjusted ORs
Univariate association of SAF with DFU OR (95% CI), 3.08 (1.62–5.80) OR (95% CI), 4.4 (2.1–9.0) OR (95% CI), 2.63 (1.48–4.66)
Other univariate associations with DFU Age, DD, BUN, Cr, CRP, TG, HDL, LDL, ABI-L, ABI-R, VPT MAU, CRI, IHD, retinopathy, neuropathy, lower limb arteriopathy Sex (women), HbA1c
Independent association of SAF with DFU OR (95% CI), 2.55 (1.10–5.91) HR (95% CI), 3.4 (1.6–7.3) OR (95% CI), 3.22 (1.53–6.79)
Adjusted covariates Age, DD, BUN, Cr, CRP, TG, HDL, LDL, ABI-L, ABI-R, VPT Age, HbA1c Age, sex, BMI, DD, HbA1c, retinopathy, nephropathy, CVD, LEAD, statin, ACEi/ARBs treatment
Other independent associations with DFU BUN, TG, ABI-R, CRP CRI, neuropathy, lower limb arteriopathy HbA1c
Abbreviations: ABI, ankle-brachial index; ACEi, angiotensin-converting enzyme inhibitors; ADA, American Diabetes Association; ARBs, angiotensin II receptor blockers; BMI, body mass index; CI, confidence interval; Cr, creatinine; CRI, chronic renal insufficiency; CRP, C-reactive protein; CVD, cardiovascular disease; DD, duration of diabetes; DFU, diabetic foot ulcer; HbA1c, glycated hemoglobin; HDL, high-density lipoprotein; IDF, International Diabetes Federation; IWGDF, International Working Group on the Diabetic Foot; LDL, low-density lipoprotein; LEAD, lower-extremity arterial disease; MAU, microalbuminuria; NOS, Newcastle-Ottawa Scale; SAF, skin autofluorescence; OR, odds ratio; SAF, skin autofluorescence; TG, triglycerides; VPT, vibration perception threshold; WC, waist circumference.

All three studies estimated SAF using AGE-Reader technology (DiagnOptics Technologies BV, Groningen, the Netherlands) to estimate the accumulation of AGEs with fluorescence properties in skin. In this process, at room temperature, the emission light illuminates the normal skin of the forearm without scars or pigmentation in the range of 300 to 420 nm. Then, excitation light is reflected from the skin with a wavelength of 300 to 600 nm. Both the lights are assessed with a spectrometer, and the emission light divided by the excitation light gives SAF values in arbitrary units.15 Averages of triplicate SAF measurements taken at the forearm were reported in all the studies.

The quality of the included studies was good with scores greater than 5.

Primary Outcome

Figure 2 shows the cumulative results of this meta-analysis combining the evidence from five included studies that included the primary outcome.8–12 Despite significant between-study heterogeneity (I2 = 68%, P = .01), the random-effects model showed significantly higher levels of SAF in DFU groups when compared with the non-DFU groups. The pooled SMD was 0.67 (95% CI, 0.32–1.01). The overall effect size for SMD was Z = 3.81 (P = .0001).

Figure 2
Figure 2:
FOREST PLOT COMPARING SAF BETWEEN GROUPS

The metaregression analysis performed with several covariates showed that variables such as year of publication, HbA1c (Figure 3), and combining HbA1c with age or BMI or diabetes duration were a significant (P < .05) source of between-study heterogeneity (data not shown). The sensitivity analysis showed that no single study significantly influenced the combined effect size, which remained stable after leaving out each study (Figure 4). Because there were fewer than eight studies, a funnel plot analysis for publication bias was not conducted.

Figure 3
Figure 3:
METAREGRESSION PLOT WITH GLYCATED HEMOGLOBIN AS COVARIATE
Figure 4
Figure 4:
SENSITIVITY ANALYSIS FOREST PLOT

Secondary Outcome

With no significant between-study heterogeneity detected for the meta-analysis of unadjusted (P = .43) and adjusted ORs (P = .94), the fixed-effects meta-analysis showed SAF levels to be significantly and independently associated with DFU in both unadjusted (OR, 3.16; 95% CI, 2.18–4.57; Z = 6.13; P < .001) and adjusted models (OR, 3.07; 95% CI, 1.95-4.81; Z = 4.89; P < .001; Figures 5 and 6, respectively).

Figure 5
Figure 5:
META-ANALYSIS OF UNADJUSTED ODDS RATIOS
Figure 6
Figure 6:
META-ANALYSIS OF ADJUSTED ODDS RATIOS

DISCUSSION

This systematic review and meta-analysis on SAF in patients with DFU is the first of its kind. The results of this meta-analysis indicated a significantly higher level of SAF in patients with DFU when compared with non-DFU cases. The higher SAF showed an independent association with DFU risk.

Now widely accepted, SAF levels are a novel and noninvasive measure reflective of tissue AGE accumulation in diabetes and related complications.5–7 As a function of hyperglycemia and nonenzymic glycation, the formation and accumulation of AGEs have been well established in the pathogenesis of diabetes and related complications.22,23 Higher levels of AGEs and their binding to AGE receptors lead to the activation of the nuclear factor κβ signaling cascade, which in turn causes protein functional loss, cellular dysfunction, and matrix degeneration.22–25 In tissue, increased glycation has been reported to influence collagen cross-linking, alter mechanical response, cause mechanical stress, increase injury risk, and lead to DFUs.24,26

Although the incidence of vascular complications may vary across included studies, one study reported a similar rise in SAF levels in patients with DFUs and microvascular and macrovascular complications with no significant difference between them.15 However, SAF levels may increase with the number of vascular complications and their severity, as reported previously.27 Liu et al15 found an independent association between SAF with micro- and macrovascular complications in patients with DFU. However, because of the lack of a control group and DFU risk assessment data, this was not included in the meta-analysis.

After adjusting for other diabetic complications, Vouillarmet et al12 reported a significantly higher level of SAF among patients with DFU compared with the non-DFU group. In addition, the higher SAF was shown to correlate significantly with vibration perception threshold, an indicator of protective sensation loss. Hu et al10 reported that SAF significantly correlates with age, waist circumference, diabetes duration, C-reactive protein, and vibration perception threshold. Even after adjusting for these factors, the SAF level remained significantly high in the DFU group.

Monami et al13 reported that, in addition to age and HbA1c, higher SAF levels are associated with adiposity and metabolic syndrome. Meerwaldt et al11 reported that age, HbA1c, and microalbuminuria independently affect 60% of the variance in SAF levels in patients with diabetes, whereas the severity of foot ulcers (assessed by Wagner score) independently explained 50% of SAF variance. The higher SAF levels in patients with DFU were negatively correlated with sensory nerve conduction velocity and amplitudes (median and sural nerves) and motor conduction velocity and amplitudes (median and peroneal nerves).11 Fernando et al9 reported a perfect within-site agreement, poor between-site agreement, and moderate to substantial agreement between left- and right-side measurements of SAF at six sites (left and right arms/legs and feet) in patients with DFUs. The SAF measures (median [95% CI]) taken at the plantar surface approximately 1 cm distal to the third metatarsal head were similar at the right foot (5.2 [4–5.5] vs 5.2 [5–5.4]) and slightly higher at the left foot (5.1 [3.6–5.7] vs 4.8 [4.5–5.4]) in their DFU group versus non-DFU groups. This study9 was the first to report SAF measurements of the foot in patients with DFU by placing the foot over the AGE reader, a technique that requires further investigation.

These results are further supported by the metaregression analysis showing a significant association between HbA1c levels and SAF; the higher the HbA1c, the more the SAF (P = .006, Figure 3). In line with this, three of the included studies found positive correlations between SAF and HbA1c.10,11,13 There are studies showing significant positive associations between SAF and age, BMI, waist circumference, and duration of diabetes in patients with DFU.10–15 Further, the metaregression models using HbA1c in combination with age (P = .004), BMI (P = .02), and diabetes duration (P = .007) demonstrated a significant influence.

The secondary outcome results of this meta-analysis indicated an independent association of SAF levels with DFU risk and support existing evidence in the relevant literature. Two of the included studies followed up study participants.12,14 Hu et al14 reported that, by the end of 1 year, more primary events (death, a second hospitalization, history of surgery, and history of dialysis) were recorded in the DFU group than the non-DFU group (35% vs 27.9%), whereas stable blood glucose control was reported in 41.2% of the DFU versus 46.8% of non-DFU follow-ups. Vouillarmet et al12 followed 58 patients with DFU for 2 months, providing standard wound care (debridement, daily dressing, and offloading), and reported a significant prognostic value of SAF on DFU healing (OR, 3.26; 95% CI, 1.26–8.46; P = .015). Further, SAF was controlled in the 74% of follow-up patients with DFU after complete healing.

The association of HbA1c with DFU risk is controversial and unclear. Although no significant association was reported by Hu et al,10 an inverse association was found by Vouillarmet et al12 between HbA1c and DFU risk. In contrast, Monami et al13 did not report any OR of HbA1c for DFU risk. However, in view of several studies suggesting SAF as a measure of tissue AGE accumulation, oxidative stress, and long-term glycemic/metabolic memory associated with HbA1c,6,12,23 this meta-analysis suggests and supports the use of SAF as a novel and noninvasive technology to assess AGE accumulation in relation to DFU risk.

Strengths and Limitations

The included studies were of considerable heterogeneity, and the adjusted covariates differ between studies. Although heterogeneity cannot be eliminated in a meta-analysis, it was addressed by applying a random-effects model. Further, although this meta-analysis of ORs produced no significant between-study heterogeneity, the presence of clinical heterogeneity is unavoidable. Other limitations pertain to the small number of studies included and the resultant lack of funnel plot analysis for publication bias. However, the one-study leave-out sensitivity analysis strengthens the robustness of these results.

Of note, US federal law restricts the AGE reader to investigational use only. Currently, there are no FDA-approved noninvasive skin AGE reader devices available for commercial distribution, and their use is limited to clinical trials.

Recommendations for Future Study

It is known that susceptibility to infections is increased in diabetic wounds,28 and CRP levels vary between infected and uninfected DFUs.29 These findings, along with evidence of significant associations between SAF and CRP,8–10 provide promising evidence for future studies using AGE reader technology to assess tissue accumulation of AGEs and the utility of SAF measurements in DFU risk assessment and management. However, because no studies report threshold values of SAF for DFUs, further studies on the diagnostic utilities and prognostic uses of SAF in DFUs in relation to early detection and wound healing are needed.

CONCLUSIONS

This meta-analysis suggests that the SAF value is significantly higher in DFU cases compared with non-DFU cases, indicating that the presence of foot ulcers in patients with diabetes is related to higher SAF values. Further, the higher SAF measurements were significantly and independently associated with DFU risk. Accordingly, these SAF measurements could be useful as a novel and noninvasive indicator of DFU risk. Of course, DFU healing depends on treating increased plantar pressure and infection as well as correcting vascular compromise and patient adherence to treatment.

REFERENCES

1. Zhang P, Lu J, Jing Y, Tang S, Zhu D, Bi Y. Global epidemiology of diabetic foot ulceration: a systematic review and meta-analysis. Ann Med 2017;49:106–16.
2. Guariguata L, Whiting DR, Hambleton I, Beagley J, Linnenkamp U, Shaw JE. Global estimates of diabetes prevalence for 2013 and projections for 2035. Diabetes Res Clin Pract 2014;103:137–49.
3. Guest JF, Fuller GW, Vowden P. Diabetic foot ulcer management in clinical practice in the UK: costs and outcomes. Int Wound J 2018;15:43–52.
4. The ADVANCE Collaborative Group. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med 2008;358:2560–72.
5. Fokkens BT, Smit AJ. Skin fluorescence as a clinical tool for non-invasive assessment of advanced glycation and long-term complications of diabetes. Glycoconj J 2016;33:527–535.
6. Gerrits EG, Lutgers HL, Kleefstra N, et al. Skin autofluorescence: a tool to identify type 2 diabetic patients at risk for developing microvascular complications. Diabetes Care 2008;31:517–21.
7. Meerwaldt R, Lutgers HL, Links TP, et al. Skin autofluorescence is a strong predictor of cardiac mortality in diabetes. Diabetes Care 2007;30:107–12.
8. Araszkiewicz A, Soska J, Borucka K, et al. In diabetic Charcot neuroarthropathy impaired microvascular function is related to long lasting metabolic control and low grade inflammatory process. Microvasc Res 2015;101:143–7.
9. Fernando ME, Crowther RG, Lazzarini PA, et al. Within- and between-body-site agreement of skin autofluorescence measurements in people with and without diabetes-related foot disease. J Diabetes Sci Technol 2019;13:836–46.
10. Hu H, Han CM, Hu XL, Ye WL, Huang WJ, Smit AJ. Elevated skin autofluorescence is strongly associated with foot ulcers in patients with diabetes: a cross-sectional, observational study of Chinese subjects. J Zhejiang Univ Sci B 2012;13:372–7.
11. Meerwaldt R, Links TP, Graaff R, et al. Increased accumulation of skin advanced glycation end-products precedes and correlates with clinical manifestation of diabetic neuropathy. Diabetologia 2005;48:1637–44.
12. Vouillarmet J, Maucort-Boulch D, Michon P, Thivolet C. Advanced glycation end products assessed by skin autofluorescence: a new marker of diabetic foot ulceration. Diabetes Technol Ther 2013;15:601–15.
13. Monami M, Lamanna C, Gori F, Bartalucci F, Marchionni N, Mannucci E. Skin autofluorescence in type 2 diabetes: beyond blood glucose. Diabetes Res Clin Pract 2008;79:56–60.
14. Hu H, Jin X, Jiang H, Hu X, Han C. An observational study of foot ulceration risk in diabetic patients evaluated with autofluorescence: 195 cases cross-sectional and one-year outcome. Int J Diabetes Dev Ctries 2012;330:18–22.
15. Liu C, Xu L, Gao H, et al. The association between skin autofluorescence and vascular complications in Chinese patients with diabetic foot ulcer: an observational study done in Shanghai. Int J Low Extrem Wounds 2015;14:28–36.
16. Morbach S, Furchert H, Gröblinghoff U, et al. Long-term prognosis of diabetic foot patients and their limbs: amputation and death over the course of a decade. Diabetes Care 2012;35:2021–7.
17. Rice J, Desai U, Cummings A, Birnbaum H, Skornicki M, Parsons N. Burden of diabetic foot ulcers for Medicare and private insurers. Diabetes Care 2014;37:651–8.
18. Dorrian CA, Cathcart S, Clausen J, Shapiro D, Dominiczak MH. Factors in human serum interfere with the measurement of advanced glycation end products. Cell Mol Biol (Noisy-le-grand) 1998;44:1069–79.
19. Lavery LA, Davis KE, Berriman SJ, et al. WHS guidelines update: diabetic foot ulcer treatment guidelines. Wound Repair Regen 2016;24:112–26.
20. Stirban A, Heinemann L. Skin autofluorescence—a non-invasive measurement for assessing cardiovascular risk and risk of diabetes. Eur Endocrinol 2014;10:106–10.
21. Wells GA, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Ottawa Hospital Research Institute. www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Last accessed August 7, 2021.
22. Ahmed N. Advanced glycation end products—role in pathology of diabetic complications. Diabetes Res Clin Pract 2005;67:3–21.
23. Rhee SY, Kim YS. The role of advanced glycation end products in diabetic vascular complications. Diabetes Metab J 2018;42:188–95.
24. Huijberts MS, Schaper NC, Schalkwijk CG. Advanced glycation end products and diabetic foot disease. Diabetes Metab Res Rev 2008;24:S19–24.
25. Peppa M, Stavroulakis P, Raptis SA. Advanced glycoxidation products and impaired diabetic wound healing. Wound Repair Regen 2009;17:461–72.
26. Dinh TL, Veves A. A review of the mechanisms implicated in the pathogenesis of the diabetic foot. Int J Low Extrem Wounds 2005;4:154–9.
27. Tanaka K, Tani Y, Asai J, et al. Skin autofluorescence is associated with severity of vascular complications in Japanese patients with type 2 diabetes. Diabet Med 2012;29:492–500.
28. Hirsch T, Spielmann M, Zuhaili B, et al. Enhanced susceptibility to infections in a diabetic wound healing model. BMC Surg 2008;8:5.
29. Jeandrot A, Richard JL, Combescure C, et al. Serum procalcitonin and C-reactive protein concentrations to distinguish mildly infected from non-infected diabetic foot ulcers: a pilot study. Diabetologia 2008;51:347–52.
Keywords:

autofluorescence; diabetic foot ulcer; diabetic wound; glycation; risk; skin; wound healing

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.