Skip to main content
Download PDF
Download ePub

The association between diabetes and dermal microvascular dysfunction non-invasively assessed by laser Doppler with local thermal hyperemia: a systematic review with meta-analysis

Cardiovascular Diabetology volume 16, Article number: 11 (2017) Cite this article

Abstract

Background/Introduction

Diabetes and cardiovascular disease develop in concert with metabolic abnormalities mirroring and causing changes in the vasculature, particularly the microcirculation. The microcirculation can be affected in different parts of the body of which the skin is the most easily accessible tissue.

Purpose

The association between diabetes and dermal microvascular dysfunction has been investigated in observational studies. However, the strength of the association is unknown. Therefore we conducted a systematic review with meta-analysis on the association between diabetes and dermal microvascular dysfunction as assessed by laser Doppler/laser speckle contrast imaging with local thermal hyperaemia as non-invasive indicator of microvascular functionality.

Methods

PubMed and Ovid were  systematically searched for eligible studies through March 2015. During the first selection, studies were included if they were performed in humans and were related to diabetes or glucose metabolism disorders and to dermal microcirculation. During the second step we selected studies based on the measurement technique, measurement location (arm or leg) and the inclusion of a healthy control group. A random effects model was used with the standardised mean difference as outcome measure. Calculations and imputation of data were done according to the Cochrane Handbook.

Results

Of the 1445 studies found in the first search, thirteen cross-sectional studies were included in the meta-analysis, comprising a total of 857 subjects. Resting blood flow was similar between healthy control subjects and diabetes patients. In contrast, the microvascular response to local skin heating was reduced in diabetic patients compared to healthy control subjects [pooled effect of −0.78 standardised mean difference (95% CI −1.06, −0.51)]. This effect is considered large according to Cohen’s effect size definition. The variability in effect size was high (heterogeneity 69%, p < 0.0001). However, subgroup analysis revealed no difference between the type and duration of diabetes and other health related factors, indicating that diabetes per se causes the microvascular dysfunction.

Conclusion

Our meta-analysis shows that diabetes is associated with a large reduction of dermal microvascular function in diabetic patients. The local thermal hyperaemia methodology may become a valuable non-invasive tool for diagnosis and assessing progress of diabetes-related microvascular complications, but standardisation of the technique and quality of study conduct is urgently required.

Background

Diabetes contributes to and accelerates cardiovascular disease [1, 2] and macrovascular complications, such as atherosclerosis [3], coronary artery disease [4, 5] and peripheral artery disease [6]. However, the most frequently diagnosed complications in diabetic patients are related to the microcirculation, including foot ulcers [7], retinopathy [8], neuropathy [9], diabetic dermopathy [10] and diminished wound healing [11]. These diabetes-related microvascular dysfunctions can eventually lead to more severe complications, illustrating the necessity to detect microvascular dysfunction in an early stage and identify diabetic patients at risk.

Diabetes may affect the microcirculation in different parts of the body from kidney to eyes and skin, but its function is most easily accessible in the latter. Moreover, skin microvascular function may already be affected in an early stage of the disease as has been assessed by invasive methodology [12].

Microvascular dysfunction used to be determined in feet and toes, as complications are first seen in these body parts. The microcirculation in feet and toes (similar to hands and fingers) can, however, fluctuate substantially due to the presence of arteriovenous anastomoses [13, 14]. Tissue perfusion at these skin sites may therefore not be the most sensitive indicators of the severity and progress of the disease. The forearm and lower leg show a more stable microcirculation and measurements are less invasive. Therefore, the functionality of dermal microcirculation is nowadays more often measured in the forearm and lower leg.

The most commonly used methods to measure microcirculation of the skin are laser Doppler flowmetry (LDF) [15] and laser speckle contrast imaging (LSCI) [16, 17]. These techniques measure the microvascular perfusion, are non-invasive, and provide a continuous measurement [17–20]. Surprisingly, resting dermal microvascular perfusion is apparently not affected in Diabetes, irrespective of the progress of the disease [21, 22]. The functionality of the microvessels may be a more sensitive indicator of complications, and different stimuli are used to determine microvascular response/reactivity [23]. Frequently iontophoresis is used in conjunction with LDF or LSCI. Although iontophoresis is generally considered as being safe, mild local allergic reactions and skin irritations have been observed in some subjects [24]. Another limitation of iontophoresis is that the drug delivery is influenced by skin resistance and this varies considerably between subjects and across different skin areas due to low-resistance pathways in the skin such as sweat ducts or hair follicles [24]. In addition, pH, ion competition in the buffer solution and biological factors such as age, gender, skin hydration and temperature can affect the drug delivery by iontophoresis and thereby contribute to the variation of the measured response. Moreover the iontophoresis procedure is rather time consuming and needs to be repeated at regular intervals [25]. An alternative is local skin heating, which is a stimulus that can be applied to induce local thermal hyperaemia (LTH) due to vasodilation. As LTH is non-invasive, has a good reproducibility and allows to asses different mechanisms causing vasodilation, LTH is one of the most commonly used patient-friendly reactivity tests [20, 23, 26]. Already in the early 90s, changes in dermal blood flow caused by diabetes have been measured with LDF in response to local heating [27]. In the “Methods” section both techniques are described in more details.

The association between diabetes mellitus and microvascular dysfunction as measured by LDF/LSCI with LTH has been investigated in observational studies. However, the strength of the association is unknown. Therefore the aim of this study was to review the strength of the association between diabetes and dermal microvascular dysfunction assessed by LDF/LSCI with LTH. Furthermore the difference in dermal microvascular function between T1DM patients and T2DM patients was assessed.

Methods

Selected measurement technique for dermal blood flow

LDF and LSCI are based on the same principle: laser light is directed towards target tissue, usually the skin. Light is scattered back on red blood cells and is then collected and analysed by optical probes [18]. The outcome is presented as blood flow or flux in arbitrary perfusion units, which are proportional to the microvascular perfusion but do not represent actual perfusion values. Nevertheless, these methods have a high sensitivity for the detection of relative changes in blood flow, are well validated, and are particularly used to determine change in perfusion induced by a stimulus at different sites of the skin [16–18, 20, 28, 29]. As shortly described in the introduction, LTH is caused by vasodilation due to local skin heating. The rise in local skin perfusion due to the vasodilation is directly proportional to the skin temperature and reaches its maximum when a local skin temperature of 44 °C is kept for at least 20 min and up to 50 min [23, 30]. The LTH response shows two perfusion peaks which are mediated by independent mechanisms: (1) the initial peak during the first 10 min depends predominantly on local sensory nerves and is mediated by an axon reflex relying on calcitonin-gene-related peptide and substance P, (2) the plateau reached after 20–30 min of heating is primarily mediated by nitric oxide [20, 30–32].

Iontophoresis is a technique in which vasoactive substances are transdermally applied. The underlying principle is the transfer of charged vasoactive drugs using a low-intensity electric current. Several drugs can be used for iontophoresis such as bradykinin, methacholine, and substance P. However, most frequently used drugs are acetylcholine (ACh) and sodium nitroprusside (SNP), which generate an endothelium-dependent and -independent vasodilatation, respectively [23].

In this review, we have focussed on the LTH-skin response in diabetic patients, representing a relatively popular technique, and offering a patient-friendly alternative to most other procedures to assess microvascular dysfunction. Iontophoresis will be discussed to put the result of the meta-analysis into perspective.

Search strategy

The search strategy was developed and designed by PPD, DF and RD. A systematic search of the databases PubMed and Ovid was conducted including studies up to the 12th March 2015. All possible terms for skin, microcirculation, and glucose metabolism disorders were used as search terms in order to identify all possibly relevant studies. Search terms that were related to the exposure were: skin, dermal, dermis, cutaneous, nailfold, microcirc*, endothel*, microvasc*, microvascular function, iontophoresis, acetylcholine, Ach, sodium nitroprusside, SNP, L-NMMA, local thermal hyper*, heat*, blood flow, perfusion, capillary, vasodilation, laser Doppler, videocapillaroscopy, laser speckle, and Doppler. Search terms in relation to diabetes mellitus were: diabetes, diabetic, DM, insulin and glucose, insulin resistan*, pre-diabet*, HOMA, HbA1c, insulin sensitiv*, hyperglyc*, OGTT, oral glucose tolerance test, glucose challenge, and glucose load. Additionally, the search string contained the terms: human, adults, adolescents, subjects, participants, and volunteers. The search was filtered on human studies, articles written in English, and search terms mentioned in the title or abstract.

Study selection

A two-step approach was followed to identify eligible studies. Both observational- and experimental-studies were deemed eligible, whereas only baseline results were used from experimental studies if included. The flowchart of the selection procedure is depicted in Fig. 1. First, two reviewers (PPD and RD) independently screened title and abstract of all retrieved studies (n = 1445) to identify potentially eligible studies. In a second step, full-texts of the studies were scrutinized by PPD and DF independently to judge eligibility based on the following selection criteria: (1) Laser Doppler flowmeter or LSCI were used to measure microcirculation of the skin, (2) LTH was used as stimulus, (3) measurements were done on arm or leg (not hands, feet, fingers, or toes), (4) studies had both a diabetes group and a (healthy) matched control group, (5) studies were not conducted in patients with any relevant concomitant disease (e.g. no heart failure, dialysis patients). When inconclusive, eligibility was discussed (PPD, RD and DF) until consensus was reached. When relevant results (e.g. mean values, duration of heating) were missing or incomplete, the authors of these studies were contacted to obtain the missing information. If authors did not respond, the study was excluded from the meta-analysis.

Fig. 1
figure 1

Flowchart of study selection

Full size image

Data extraction and meta-analysis

Baseline characteristics and population details were extracted from all included studies: type of diabetes, methodology to determine diabetes, population size, gender distribution in both diabetes—and control group, age distribution, and BMI. Secondly, information regarding methodology was extracted to identify differences in studies: type of study, specific type of laser Doppler used, duration and temperature of heat stimulus, and location(s) of measurement.

Results were presented in different outcome units by the included studies. Blood flow was presented in arbitrary perfusion units (PU), flux, milliliter × minute−1 × 100 g of tissue−1, and Volts (1PU is a pre-defined electrical signal in mV). As the blood flow was presented in different units, the results were calculated into standardised mean differences [33]:

$$Standardised \,mean \,difference = \frac{{{\text{Increase in diabetes group}} - {\text{Increase in control group}}}}{Pooled \,standard \,deviation}$$

Mean values and standard deviations of basal flow and peak flow were used to calculate the standardised mean difference for the response to LTH as measured with laser Doppler/LSCI between diabetic patients and healthy control subjects [34]. The SMD is an outcome measure that is difficult to interpret as it does not give an absolute difference or ratio. Cohen designed an effect size index to interpret the importance of the found effect [35]. According to Cohen, an effect size of 0.2 SMD is considered small, 0.5 SMD medium, and 0.8 SMD large.

The meta-analysis was performed in Review Manager 5.3. The I 2 was determined to check for heterogeneity between studies [36]. If the I 2 was 50% or higher, it was assumed there was substantial heterogeneity [37]. If heterogeneity was assumed, subgroup analysis was performed. A random effects model was used in case heterogeneity was not explained by the subgroup analysis.

When the standard deviations of the absolute changes from baseline were not available from individual studies, the missing standard deviations were imputed as described in detail in the Cochrane Handbook [38–40]. The correlation coefficients were calculated from an unpublished validation study performed by Unilever R&D Vlaardingen presenting standard deviations for basal flow (77.6 AU), peak flow (peak 1: 210.1 AU; plateau: 238.4 AU), and change (peak 1 vs baseline: 148.2 AU; plateau vs baseline: 201.0 AU). By using these imputed correlation coefficients, the standard deviations of the change from basal flow to peak flow were calculated for the studies missing the standard deviations.

Several studies contained multiple diabetes groups, e.g. a non-neuropathic diabetes group and a neuropathic diabetes group. These groups were analysed as if they were separate studies. The shared control group, the healthy control subjects, was evenly divided among the two diabetes groups and used as two separate control groups as described in the Cochrane Handbook [38]. The outcome is continuous, therefore the mean change was the same in both control groups.

A priori variables were determined that could have an effect on the outcome. These variables were used to define subgroup analyses. Regarding participant characteristics age (young to old), type of diabetes (type 1 vs type 2), BMI (normal weight to obesity), duration of diabetes (short duration to long duration), and glycated haemoglobin (low HbA1c value to high HbA1c value) were determined as possible influential variables. Regarding methodology the duration of the heat stimulus (<20 min vs >20 min) and location of measurement (forearm vs tibia vs quadriceps) were deemed to be of possible influence.

Eligible studies were also evaluated on quality of conduct and/or reporting. For the quality assessment we used a list of quality criteria that was developed by Downs and Black to assess the quality of observational studies [41]. The criteria were adapted for use for our analysis, involving only studies without medical intervention (Additional file 1). The highest possible quality score was 17 points. The quality scores are presented in Table 1.

Table 1 Participant characteristics of studies included in meta-analysis on LTH
Full size table

Results

Meta-analysis

Selection process

The systematic search yielded a total of 1445 studies. During the first selection, studies were excluded that did not meet the criteria (Fig. 1), resulting in the selection of 314 studies. After the second selection procedure, 22 studies were selected as eligible for inclusion in the meta-analysis. The reference lists of these 22 studies were searched for eligible studies, which yielded one additional study. Finally, a number of the eligible studies were excluded from the meta-analysis due to crucial missing results or incorrect methods. Another reason of exclusion was that LTH was not the only stimulus and that therefore the effects of LTH could not be separated from other stimuli. This resulted in the inclusion of 13 studies in the meta-analysis.

Overview of included studies

All 13 included studies were analytical cross-sectional studies, investigating the association between a risk factor and outcome at a single point in time. These studies contained a total of 857 participants with the number of participants in each study ranging from 20 to 150 (Table 1).

All studies used a laser Doppler flowmeter. Basal flow was standardised to a normal temperature. These temperatures ranged from 30 to 35 °C. The temperature to obtain maximum vasodilation was 44 °C in most studies. Only two studies used a lower temperature, i.e. 43 °C [42] and 42 °C [43]. The included studies differed on several methodological characteristics. Duration of the heating to obtain maximum vasodilation: in five studies the skin was heated for less than 20 min [42, 44–47], while in eight studies heating was applied for 20 min or more [43, 48–54]. There were three locations of measurements: eight studies measured on the forearm [42, 43, 46–48, 52–54], three on the pretibial surface of the leg [49–51], and two on the quadriceps muscle [44, 45].

Pooled overall risk estimate

Baseline dermal blood flow did not differ between control and diabetes group (p = 0.51). The pooled overall estimate showed a lower microvascular response to local thermal stimulus of −0.78 SMD (95% CI −1.06, −0.51) in diabetic patients compared to control subjects (Fig. 2). The mean estimates of all studies showed a lower SMD in the diabetic patients compared to the control group of which eleven subgroups out of ten studies reached significance [43–45, 47, 48, 50–54].

Fig. 2
figure 2

Forest plot of pooled overall effect of microvascular function in diabetic patients versus control subjects based on a random effects model. Studies containing subgroups (for instance Arora 1998a and Arora 1998b) were studies presenting two different diabetes groups. The a-subgroups contained diabetic patients with additional complications and the b-subgroups diabetic patients without complications

Full size image

Heterogeneity analysis

A heterogeneity of 69% (p < 0.0001) was found in the meta-analysis. Therefore heterogeneity was assumed and subgroup analyses were performed. However, the latter could not explain the heterogeneity and a random effects model was used.

Subgroup analyses

A priori, several variables were selected to be of possible influence on the effect. There were no differences between studies conducted on T1DM, T2DM, or both types of diabetes. The studies that did not specify the type of diabetes [44, 45] were significantly different, with an SMD of −2.52 (95% CI −3.40, −1.64, I2 0%) (Fig. 3). However, this difference may also be explained by the unique location of the measurement [44, 45]. Only in these studies blood flow was determined on the quadriceps and showed a significant different effect size compared to studies conducted on the forearm (−0.71 SMD, 95% CI −1.01, −0.41) or pretibial surface (−0.58 SMD, 95% CI −0.99, −0.16) (Fig. 4). Subgroup analyses for the variables age, BMI, duration of diabetes, HbA1c, and duration of heating showed no significant effects (data not shown).

Fig. 3
figure 3

Forest plot of subgroup analysis for microvascular function according to type of diabetes

Full size image
Fig. 4
figure 4

Forest plot of subgroup analysis for microvascular function according to location of measurement

Full size image

Quality of conduct of the eligible studies was moderate (10 studies scored 6–11 out of 17; Table 1).

Evidence from systematic literature review

Iontophoresis

The first selection of studies (n = 314) was checked for studies eligible for a systematic review of the microvascular reactivity after iontophoresis with ACh and SNP. Based on the criteria mentioned in the “Methods” section 22 studies were suitable for inclusion. 6 of the 22 studies did not make a distinction in the types of diabetes [54–59], 6 were conducted in T1DM [52, 53, 60–63] and 11 in T2DM patients [12, 43, 63–70].

Several studies not differentiating between types of diabetes showed a diminished response to ACh [54, 55, 57, 59] and SNP [54, 55, 57] iontophoresis in diabetic patients. However, the differences seem to be more marked and more significant after ACh iontophoresis than after SNP iontophoresis [55, 57]. Studies comparing T1DM patients with healthy controls show similar results. The reduction in blood flow due to diabetes seems to be clearer after iontophoresis with ACh than after iontophoresis with SNP [52, 53, 61]. Next to a reduction in peak vasodilation, Katz et al. [60] showed that time to peak vasodilation after ACh iontophoresis was doubled in diabetic subjects compared to controls, while there was no difference after SNP iontophoresis. Three studies reported a difference in response to both ACh and SNP between patients with T2DM and control subjects [12, 43, 64]. The differences were similar in all three studies and microvascular response was shown to be reduced by a third for both vasoactive substances. A study [63] comparing the microvascular reactivity of patients with T1DM, T2DM, and healthy control subjects, observed a reduced response to ACh in subjects with T2DM, but did not report a difference in response to SNP. In contrast, two studies reported a significant difference between patients with T2DM and control subjects for SNP iontophoresis but not for ACh iontophoresis [68, 69]. The response to SNP iontophoresis was reduced by a third in diabetic patients compared to control subjects. Three studies, however, observed no differences between T2DM patients and healthy control subjects [65–67]. Caballero et al. [12] described a reduced response, assessed as peak vasodilation and percentage increase over baseline, to ACh and SNP in patients with impaired glucose tolerance and relatives of diabetic patients as compared to healthy control subjects. Although not significant, the response gradually decreased from relatives of diabetic patients to subjects with impaired glucose tolerance to diabetic patients [12]. Two of three studies with diabetic patients with microvascular complications (albuminuria, neuropathy or retinopathy) observed at least a trend towards an additional impairment in microvascular function as compared to diabetic patients without complications [54, 69].

Discussion

The present meta-analysis provides a number of important observations. First, our meta-analyses shows an effect size of the association between diabetes mellitus and microvascular dysfunction as assessed by LTH response that can be considered to be nearly large according to Cohen’s effect size index [35]. Secondly, subgroup analysis showed a difference in effect size between locations of measurement. The fact that the response reported in studies conducted in the quadriceps muscle was more reduced than in the forearm or pretibial surface could suggest that assessment of dermal microvascular function at the quadriceps would be the preferred choice to discriminate healthy from diabetic subjects. However, these findings are limited by the small subgroup size for the quadriceps (n = 40) and being conducted by only one research group [44, 45]. The relatively small standard deviation reported by this group affects the SMD size effect [44, 45] and could either represent the excellence of measurement or a typical smaller variation specifically for the quadriceps as measurement site.

Thirdly, no differences in effect size were found between T1DM and T2DM or whether both types of diabetes were included, which may indicate that the microvascular function is impaired by diabetes per se and not by a factor common to either T1DM or T2DM. Patients with T1DM tend to be younger, leaner, have had diabetes for a longer period and a worse glycemic control compared to T2DM patients [71] and thus could have a poorer microvascular health. However, T2DM patients tend to be overweight and are older in age [71], characteristics that could also impair the microcirculation. Our observation is confirmed by the fact that sensitivity analyses for age, BMI, duration of diabetes, and HbA1c revealed no explanation for the heterogeneity of the data.

Likewise, no differences were found in subgroup analysis for duration of heating. However, the comparison is limited since the data was not reported optimally in a number of studies. The heating procedure should at least take 40 min to allow detection of the nerve-axon reflex-related first peak within the first 10 min, followed by a nitric oxide-dependent plateau-phase at the end of the procedure [72]. In a number of studies the plateau-phase was likely not reached yet. These issues give possible explanations for the lack in differences in subgroup-analyses.

It is interesting to note that four studies included two different diabetes groups: a group of diabetic patients with additional microvascular complications such as dermopathy and a group of matched diabetic patients without these complications [47, 49, 51, 54]. Although not considered a priori, these studies allowed us to differentiate between diabetic patients with and without additional microvascular complications. The findings of these studies indicate that the diabetes group with additional complications had an attenuated response compared to the groups without these complications. Even though the observed difference between these groups was not significant, the attenuated response in diabetic patients with additional complications may support our hypothesis that a longer duration of diabetes and additional complications in diabetic patients are associated with a larger degree of impairment of the microvascular reactivity. In line with our assumption, other groups reported an additional worsening of microvascular function in diabetic patients with microvascular complications [73–75].

As LTH causes vasodilation by different mechanisms, a nerve-axon reflex and a nitric oxide-mediated peak [72], this reactivity test was deemed to successfully assess which aspect is impaired in diabetes. However, due to the lack of guidelines on reactivity tests with a heat stimulus, studies differed in duration of heating. Therefore the difference between the two peaks could not be assessed and the underlying mechanism resulting in microvascular dysfunction in diabetes could not be reviewed. Overall, consensus on standardisation of the microvascular LTH response is required to qualify the method as a patient-friendly diagnostic tool.

The quality of study conduct was moderate. This may reflect either the quality of the study design, but may also be affected by (poor) description and reporting of the study. Therefore, the quality assessment was not used for excluding studies. The quality assessment has revealed that the conduct (or reporting thereof) of dermal microvascular function as assessed by the LTH response in diabetic patients requires improvement. This conclusion supports the recommendation to strive for standardization of the methodology.

Multiple studies reported a reduction in vasodilation in diabetic patients after iontophoresis of particularly ACh [52, 53, 55, 57, 61–63]. This implies that the microvascular dysfunction in diabetic patients is essentially endothelium-dependent. Also, the impairment appears to be more evident in type 1 diabetes than in type 2 diabetes. This may be due to the longer duration of diabetes in type 1 diabetic patients and the worse glycaemic control, which may enhance the microvascular impairment, as an increase in HbA1c values was found to be correlated with a decrease in microvascular response [43, 47, 64, 76, 77]. This correlation is in line with the observation by Caballero et al. [12] that microvascular reactivity gradually decreased from relatives of diabetic patients to subjects with impaired glucose tolerance to diabetic patients. In addition, two studies with diabetic patients with microvascular complications reported at least a trend towards an additional impairment in microvascular function as compared to diabetic patients without complications [54, 69]. However, those effects were not significant and the evidence for a stepwise impairment of microvascular function from subjects at risk for diabetes to diabetic patients with complications is rather limited.

Strengths and limitations

This review has several strengths. It is the first systematic review on dermal microvascular dysfunction in diabetes and provides a clear insight in this association. Many studies have been conducted addressing this topic, but general consensus was not created yet. Furthermore, we used the PRISMA Statement [78] and the Cochrane Handbook [37] as guidance for this systematic review. This ensures a clear structure of our review and objective methods for data-extraction and -analysis.

Nevertheless, our meta-analysis also has several limitations. Importantly, we observed heterogeneity between studies. Meta-analyses include studies with differences in study design and additionally are prone to be affected by confounding. The number of subjects (n = 857) included was limited, whilst the variance between studies in characteristics of the participants and methodology used may have been too large to reveal the major contributors to the heterogeneity. Laser Doppler and LSCI in combination with local thermal hyperaemia are frequently used methods. Nevertheless, there are no official guidelines available for these measurements and a better standardization is required [73]. Cracowski et al. [20] developed ten suggestions which should be followed to minimize the variability of these measurements. The lack of official guidelines explains why studies differed on methodological aspects such as location of measurement, duration of heating, maximum temperature or outcome measure. The development and implementation of official guidelines for laser Doppler measurements in combination with LTH would improve the reproducibility of the measurement and the comparability of data between studies.

Another limitation was the difference in outcome measures between studies. Not only the unit was different, the absolute values of these outcome measures were also different in range and for this reason we used the SMD. SMD provides an indication of the strength of the association between diabetes and dermal microvascular dysfunction, but cannot be used to determine the absolute difference between diabetic patients and control subjects.

Furthermore, the quality of reporting of the selected studies was not optimal. For example, the gender of the subjects or use of medication is not reported in all publications. This is reflected in the scores the studies received during the quality assessment. Moreover, ten eligible studies were excluded from the analysis due to missing data and no response from the authors, or due to the use of an additional stimulus besides local heating. The excluded studies showed similar results as the included studies and therefore it is unlikely that the exclusion of these studies substantially influenced the findings.

Conclusions

Diabetic patients show an impaired dermal microvascular hyperaemic response to local heating compared to healthy subjects. An important issue for future research is the implementation of guidelines for microvascular LTH response to strengthen the validity of this tool for assessing (progress of) diabetes-related microvascular complications. Further LDF/LSCI studies including higher number of healthy subjects, prediabetic subjects and diabetic patients with and without complications would be required to demonstrate the stepwise impairment of microvascular function in diabetes and to confirm the applicability of LDF/LSCI as a diagnostic tool.

Abbreviations

Ach:

acetylcholine

BP:

blood pressure

DM:

diabetes mellitus

HbA1c :

glycosylated hemoglobin

LDF:

laser Doppler flowmetry

LSCI:

laser speckle contrast imaging

LTH:

local thermal hyperaemia

SD:

standard deviation

SEM:

standard error of the mean

SMD:

standard mean difference

SNP:

sodium nitroprusside

T1DM:

type 1 diabetes mellitus

T2DM:

type 2 diabetes mellitus

References

  1. Tryggestad JB, Willi SM. Complications and comorbidities of T2DM in adolescents: findings from the TODAY clinical trial. J Diabetes Complications. 2015;29:307–12.

    Article  PubMed  Google Scholar 

  2. Paneni F, Costantino S, Cosentino F. Insulin resistance, diabetes, and cardiovascular risk. Curr Atheroscler Rep. 2014;16:419.

    Article  PubMed  Google Scholar 

  3. Paneni F, Beckman JA, Creager MA, Cosentino F. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part I. Eur Heart J. 2013;34:2436–43.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Tousoulis D, Kampoli AM, Stefanadis C. Diabetes mellitus and vascular endothelial dysfunction: current perspectives. Curr Vasc Pharmacol. 2012;10:19–32.

    Article  CAS  PubMed  Google Scholar 

  5. Tousoulis D, Kampoli AM, Papageorgiou N, Papaoikonomou S, Antoniades C, Stefanadis C. The impact of diabetes mellitus on coronary artery disease: new therapeutic approaches. Curr Pharm Des. 2009;15:2037–48.

    Article  CAS  PubMed  Google Scholar 

  6. Beckman JA, Paneni F, Cosentino F, Creager MA. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part II. Eur Heart J. 2013;34:2444–52.

    Article  PubMed  Google Scholar 

  7. Yazdanpanah L, Nasiri M, Adarvishi S. Literature review on the management of diabetic foot ulcer. World J Diabetes. 2015;6:37–53.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Sayin N, Kara N, Pekel G. Ocular complications of diabetes mellitus. World J Diabetes. 2015;6:92–108.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Vinik AI, Nevoret ML, Casellini C, Parson H. Diabetic neuropathy. Endocrinol Metab Clin North Am. 2013;42:747–87.

    Article  PubMed  Google Scholar 

  10. Morgan AJ, Schwartz RA. Diabetic dermopathy: a subtle sign with grave implications. J Am Acad Dermatol. 2008;58:447–51.

    Article  PubMed  Google Scholar 

  11. Arya AK, Tripathi R, Kumar S, Tripathi K. Recent advances on the association of apoptosis in chronic non healing diabetic wound. World J Diabetes. 2014;5:756–62.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Caballero AE, Arora S, Saouaf R, Lim SC, Smakowski P, Park JY, et al. Microvascular and macrovascular reactivity is reduced in subjects at risk for type 2 diabetes. Diabetes. 1999;48:1856–62.

    Article  CAS  PubMed  Google Scholar 

  13. Thoresen M, Walloe L. Skin blood flow in humans as a function of environmental temperature measured by ultrasound. Acta Physiol Scand. 1980;109:333–41.

    Article  CAS  PubMed  Google Scholar 

  14. Vanggaard L, Kuklane K, Holmer I, Smolander J. Thermal responses to whole-body cooling in air with special reference to arteriovenous anastomoses in fingers. Clin Physiol Funct Imaging. 2012;32:463–9.

    Article  PubMed  Google Scholar 

  15. Berardesca E, Leveque JL, Masson P. EEMCO guidance for the measurement of skin microcirculation. Skin Pharmacol Appl Skin Physiol. 2002;15:442–56.

    Article  PubMed  Google Scholar 

  16. Briers JD. Laser Doppler, speckle and related techniques for blood perfusion mapping and imaging. Physiol Meas. 2001;22:R35–66.

    Article  CAS  PubMed  Google Scholar 

  17. Mahe G, Humeau-Heurtier A, Durand S, Leftheriotis G, Abraham P. Assessment of skin microvascular function and dysfunction with laser speckle contrast imaging. Circ Cardiovasc Imaging. 2012;5:155–63.

    Article  PubMed  Google Scholar 

  18. Oberg PA. Laser-Doppler flowmetry. Crit Rev Biomed Eng. 1990;18:125–63.

    CAS  PubMed  Google Scholar 

  19. Fagrell B. Dynamics of skin microcirculation in humans. J Cardiovasc Pharmacol. 1985;7(Suppl 3):S53–8.

    Article  PubMed  Google Scholar 

  20. Cracowski JL, Minson CT, Salvat-Melis M, Halliwill JR. Methodological issues in the assessment of skin microvascular endothelial function in humans. Trends Pharmacol Sci. 2006;27:503–8.

    Article  CAS  PubMed  Google Scholar 

  21. Walther G, Obert P, Fdr Dutheil, Chapier R, Lesourd B, Naughton G, et al. Metabolic syndrome individuals with and without type 2 diabetes mellitus present generalized vascular dysfunction: cross-sectional study. Arterioscler Thromb Vasc Biol. 2015;35:1022–9.

    Article  CAS  PubMed  Google Scholar 

  22. Beer S, Fo Feihl, Ruiz J, In Juhan-Vague, Aillaud MF, Wetzel SG, et al. Comparison of skin microvascular reactivity with hemostatic markers of endothelial dysfunction and damage in type 2 diabetes. Vasc Health Risk Manag. 2008;4:1449–58.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Roustit M, Cracowski JL. Non-invasive assessment of skin microvascular function in humans: an insight into methods. Microcirculation. 2012;19:47–64.

    Article  PubMed  Google Scholar 

  24. Turner J, Belch JJ, Khan F. Current concepts in assessment of microvascular endothelial function using laser Doppler imaging and iontophoresis. Trends Cardiovasc Med. 2008;18:109–16.

    Article  CAS  PubMed  Google Scholar 

  25. Wright CI, Kroner CI, Draijer R. Non-invasive methods and stimuli for evaluating the skin’s microcirculation. J Pharmacol Toxicol Methods. 2006;54:1–25.

    Article  CAS  PubMed  Google Scholar 

  26. Roustit M, Millet C, Blaise S, Dufournet B, Cracowski JL. Excellent reproducibility of laser speckle contrast imaging to assess skin microvascular reactivity. Microvasc Res. 2010;80:505–11.

    Article  CAS  PubMed  Google Scholar 

  27. Stevens MJ, Edmonds ME, Douglas SL, Watkins PJ. Influence of neuropathy on the microvascular response to local heating in the human diabetic foot. Clin Sci (Lond). 1991;80:249–56.

    Article  CAS  PubMed  Google Scholar 

  28. Eun HC. Evaluation of skin blood flow by laser Doppler flowmetry. Clin Dermatol. 1995;13:337–47.

    Article  CAS  PubMed  Google Scholar 

  29. Vongsavan N, Matthews B. Some aspects of the use of laser Doppler flow meters for recording tissue blood flow. Exp Physiol. 1993;78:1–14.

    Article  CAS  PubMed  Google Scholar 

  30. Kellogg DL Jr, Zhao JL, Wu Y, Johnson JM. Antagonism of soluble guanylyl cyclase attenuates cutaneous vasodilation during whole body heat stress and local warming in humans. J Appl Physiol. 1985;2011(110):1406–13.

    Google Scholar 

  31. Minson CT. Thermal provocation to evaluate microvascular reactivity in human skin. J Appl Physiol. 2010;109:1239–46.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Minson CT, Berry LT, Joyner MJ. Nitric oxide and neurally mediated regulation of skin blood flow during local heating. J Appl Physiol. 2001;91:1619–26.

    CAS  PubMed  Google Scholar 

  33. Takeshima N, Sozu T, Tajika A, Ogawa Y, Hayasaka Y, Furukawa TA. Which is more generalizable, powerful and interpretable in meta-analyses, mean difference or standardized mean difference? BMC Med Res Methodol. 2014;14:30.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Borenstein M, Hedges L, Higgins J, Rothstein H. Introduction to meta-analysis. New York: Wiley; 2009.

    Book  Google Scholar 

  35. Cohen J. A power primer. Psychol Bull. 1992;112:155–9.

    Article  CAS  PubMed  Google Scholar 

  36. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002;21:1539–58.

    Article  PubMed  Google Scholar 

  37. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327:557–60.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Higgins J, Green S. Cochrane handbook for systematic reviews of interventions, Version 5.1.0: The Cochrane Collaboration; 2011.

  39. Follmann D, Elliott P, Suh I, Cutler J. Variance imputation for overviews of clinical trials with continuous response. J Clin Epidemiol. 1992;45:769–73.

    Article  CAS  PubMed  Google Scholar 

  40. Abrams KR, Gillies CL, Lambert PC. Meta-analysis of heterogeneously reported trials assessing change from baseline. Stat Med. 2005;24:3823–44.

    Article  PubMed  Google Scholar 

  41. Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health. 1998;52:377–84.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Aellen J, Dabiri A, Heim A, Liaudet L, Burnier M, Ruiz J, et al. Preserved capillary density of dorsal finger skin in treated hypertensive patients with or without type 2 diabetes. Microcirculation. 2012;19:554–62.

    Article  PubMed  Google Scholar 

  43. Morris SJ, Shore AC, Tooke JE. Responses of the skin microcirculation to acetylcholine and sodium nitroprusside in patients with NIDDM. Diabetologia. 1995;38:1337–44.

    Article  CAS  PubMed  Google Scholar 

  44. Petrofsky J, Paluso D, Anderson D, Swan K, Alshammari F, Katrak V, et al. The ability of different areas of the skin to absorb heat from a locally applied heat source: the impact of diabetes. Diabetes Technol Ther. 2011;13:365–72.

    Article  PubMed  Google Scholar 

  45. Petrofsky JS, Alshammari F, Bains GS, Khowailed IA, Lee H, Kuderu YN, et al. What is more damaging to vascular endothelial function: diabetes, age, high BMI, or all of the above? Med Sci Monit. 2013;19:257–63.

    Article  PubMed  PubMed Central  Google Scholar 

  46. Forst T, Caduff A, Talary M, Weder M, Brandle M, Kann P, et al. Impact of environmental temperature on skin thickness and microvascular blood flow in subjects with and without diabetes. Diabetes Technol Ther. 2006;8:94–101.

    Article  CAS  PubMed  Google Scholar 

  47. Skrha J, Prazny M, Haas T, Kvasnicka J, Kalvodova B. Comparison of laser-Doppler flowmetry with biochemical indicators of endothelial dysfunction related to early microangiopathy in type 1 diabetic patients. J Diabetes Complicat. 2001;15:234–40.

    Article  CAS  PubMed  Google Scholar 

  48. Rendell M, Bergman T, O’Donnell G, Drobny E, Borgos J, Bonner RF. Microvascular blood flow, volume, and velocity measured by laser Doppler techniques in IDDM. Diabetes. 1989;38:819–24.

    Article  CAS  PubMed  Google Scholar 

  49. Brugler A, Thompson S, Turner S, Ngo B, Rendell M. Skin blood flow abnormalities in diabetic dermopathy. J Am Acad Dermatol. 2011;65:559–63.

    Article  PubMed  Google Scholar 

  50. Wigington G, Ngo B, Rendell M. Skin blood flow in diabetic dermopathy. Arch Dermatol. 2004;140:1248–50.

    Article  PubMed  Google Scholar 

  51. Ngo B, Wigington G, Hayes K, Huerter C, Hillman B, Adler M, et al. Skin blood flow in necrobiosis lipoidica diabeticorum. Int J Dermatol. 2008;47:354–8.

    Article  PubMed  Google Scholar 

  52. Gomes MB, Matheus AS, Tibirica E. Evaluation of microvascular endothelial function in patients with type 1 diabetes using laser-Doppler perfusion monitoring: which method to choose? Microvasc Res. 2008;76:132–3.

    Article  CAS  PubMed  Google Scholar 

  53. Matheus AS, Tibirica E, da Silva PB, Matta dFBd, Gomes MB. Uric acid levels are associated with microvascular endothelial dysfunction in patients with type 1 diabetes. Diabet Med. 2011;28:1188–93.

    Article  CAS  PubMed  Google Scholar 

  54. Arora S, Smakowski P, Frykberg RG, Simeone LR, Freeman R, LoGerfo FW, et al. Differences in foot and forearm skin microcirculation in diabetic patients with and without neuropathy. Diabetes Care. 1998;21:1339–44.

    Article  CAS  PubMed  Google Scholar 

  55. Foster W, Carruthers D, Lip GY, Blann AD. Inflammation and microvascular and macrovascular endothelial dysfunction in rheumatoid arthritis: effect of treatment. J Rheumatol. 2010;37:711–6.

    Article  CAS  PubMed  Google Scholar 

  56. Doupis J, Rahangdale S, Gnardellis C, Pena SE, Malhotra A, Veves A. Effects of diabetes and obesity on vascular reactivity, inflammatory cytokines, and growth factors. Obesity (Silver Spring). 2011;19:729–35.

    Article  CAS  PubMed  Google Scholar 

  57. Zomer E, de Ridder I, Kompa A, Komesaroff P, Gilbert R, Krum H. Effect of urotensin II on skin microvessel tone in diabetic patients without heart failure or essential hypertension. Clin Exp Pharmacol Physiol. 2008;35:1147–50.

    Article  CAS  PubMed  Google Scholar 

  58. Economides PA, Caselli A, Tiani E, Khaodhiar L, Horton ES, Veves A. The effects of atorvastatin on endothelial function in diabetic patients and subjects at risk for type 2 diabetes. J Clin Endocrinol Metab. 2004;89:740–7.

    Article  CAS  PubMed  Google Scholar 

  59. Caselli A, Hanane T, Jane B, Carter S, Khaodhiar L, Veves A. Topical methyl nicotinate-induced skin vasodilation in diabetic neuropathy. J Diabetes Complications. 2003;17:205–10.

    Article  PubMed  Google Scholar 

  60. Katz A, Ekberg K, Johansson BL, Wahren J. Diminished skin blood flow in type I diabetes: evidence for non-endothelium-dependent dysfunction. Clin Sci (Lond). 2001;101:59–64.

    Article  CAS  Google Scholar 

  61. DiMeglio LA, Tosh A, Saha C, Estes M, Mund J, Mead LE, et al. Endothelial abnormalities in adolescents with type 1 diabetes: a biomarker for vascular sequelae? J Pediatr. 2010;157:540–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  62. Koitka A, Abraham P, Bouhanick B, Sigaudo-Roussel D, Demiot C, Saumet JL. Impaired pressure-induced vasodilation at the foot in young adults with type 1 diabetes. Diabetes. 2004;53:721–5.

    Article  CAS  PubMed  Google Scholar 

  63. Kilo S, Berghoff M, Hilz M, Freeman R. Neural and endothelial control of the microcirculation in diabetic peripheral neuropathy. Neurology. 2000;54:1246–52.

    Article  CAS  PubMed  Google Scholar 

  64. Lim SC, Caballero AE, Smakowski P, LoGerfo FW, Horton ES, Veves A. Soluble intercellular adhesion molecule, vascular cell adhesion molecule, and impaired microvascular reactivity are early markers of vasculopathy in type 2 diabetic individuals without microalbuminuria. Diabetes Care. 1999;22:1865–70.

    Article  CAS  PubMed  Google Scholar 

  65. Hamdy O, Abou-Elenin K, LoGerfo FW, Horton ES, Veves A. Contribution of nerve-axon reflex-related vasodilation to the total skin vasodilation in diabetic patients with and without neuropathy. Diabetes Care. 2001;24:344–9.

    Article  CAS  PubMed  Google Scholar 

  66. Shrikhande G, Khaodhiar L, Scali S, Lima C, Hubbard M, Dudley K, et al. Valsartan improves resting skin blood flow in type 2 diabetic patients and reduces poly(adenosine diphosphate-ribose) polymerase activation. J Vasc Surg. 2006;43:760–70.

    Article  PubMed  Google Scholar 

  67. Rousseau P, Mahe G, Fromy B, Ducluzeau PH, Saumet JL, Abraham P. Axon-reflex cutaneous vasodilatation is impaired in type 2 diabetic patients receiving chronic low-dose aspirin. Microvasc Res. 2009;78:218–23.

    Article  CAS  PubMed  Google Scholar 

  68. Brooks BA, Franjic B, Ban CR, Swaraj K, Yue DK, Celermajer DS, et al. Diastolic dysfunction and abnormalities of the microcirculation in type 2 diabetes. Diabetes Obes Metab. 2008;10:739–46.

    Article  CAS  PubMed  Google Scholar 

  69. Brooks BA, McLennan SV, Twigg SM, Yue DK. Detection and characterisation of microcirculatory abnormalities in the skin of diabetic patients with microvascular complications. Diab Vasc Dis Res. 2008;5:30–5.

    Article  PubMed  Google Scholar 

  70. Economides PA, Caselli A, Zuo CS, Sparks C, Khaodhiar L, Katsilambros N, et al. Kidney oxygenation during water diuresis and endothelial function in patients with type 2 diabetes and subjects at risk to develop diabetes. Metabolism. 2004;53:222–7.

    Article  CAS  PubMed  Google Scholar 

  71. Ryden L, Grant PJ, Anker SD, Berne C, Cosentino F, Danchin N, et al. ESC Guidelines on diabetes, pre-diabetes, and cardiovascular diseases developed in collaboration with the EASD: the Task Force on diabetes, pre-diabetes, and cardiovascular diseases of the European Society of Cardiology (ESC) and developed in collaboration with the European Association for the Study of Diabetes (EASD). Eur Heart J. 2013;34:3035–87.

    Article  PubMed  Google Scholar 

  72. Charkoudian N. Skin blood flow in adult human thermoregulation: how it works, when it does not, and why. Mayo Clin Proc. 2003;78:603–12.

    Article  PubMed  Google Scholar 

  73. Stirban A. Microvascular dysfunction in the context of diabetic neuropathy. Curr DiabRep. 2014;14:541.

    Article  Google Scholar 

  74. Östlund Papadogeorgos N, Jörneskog G, Bengtsson M, Kahan T, Kalani M. Severely impaired microvascular reactivity in diabetic patients with an acute coronary syndrome. Cardiovasc Diabetol. 2016;15:66.

    Article  PubMed  PubMed Central  Google Scholar 

  75. Tomešová J, Gruberova J, Lacigova S, Cechurova D, Jankovec Z, Rusavy Z. Differences in skin microcirculation on the upper and lower extremities in patients with diabetes mellitus: relationship of diabetic neuropathy and skin microcirculation. Diabetes Technol Ther. 2013;15:968–75.

    Article  PubMed  Google Scholar 

  76. Taub PR, Higginbotham E, Henry RR. Beneficial and detrimental effects of glycemic control on cardiovascular disease in type 2 diabetes. Curr Cardiol Rep. 2013;15:332.

    Article  PubMed  Google Scholar 

  77. Blonde L. Benefits and risks for intensive glycemic control in patients with diabetes mellitus. Am J Med Sci. 2012;343:17–20.

    Article  PubMed  Google Scholar 

  78. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol. 2009;62:1006–12.

    Article  PubMed  Google Scholar 

Download references

Authors’ contributions

DF, PPD and RD designed the study; DF, PPD and RD collected the data; PPD analysed the data; DF, PPD and RD interpreted the data; DF and PPD drafted the manuscript; DF, PPD, LAS and RD critically revised the manuscript. All authors read and approved the final manuscript.

Acknowledgements

The authors greatly acknowledge Rouyanne T. Ras (Unilever Research and Development, Vlaardingen, The Netherlands) and Michiel de Boer (Faculty of Earth and Life Sciences, VU Amsterdam, Amsterdam, The Netherlands) for advice on the statistical analyses.

Competing interests

Research Scientists DF and RD are employed by Unilever Research and Development. Unilever markets food products some of which address cardiovascular risk factors. All other authors have no competing interests.

Availability of data and material

The datasets analysed during the current meta-analysis are available from the corresponding author on reasonable request.

Consent for publication

Not applicable; the current manuscript is a literature review.

Ethics approval and consent to participate

Not applicable; the current manuscript is a literature review.

Funding

Unilever R&D Vlaardingen is the funder of this review with meta-analysis.

Author information

Authors and Affiliations

  1. Unilever Research and Development, Vlaardingen, Olivier van Noortlaan 120, PO Box 114, 3130 AC, Vlaardingen, The Netherlands

    Dagmar Fuchs, Pepijn P. Dupon & Richard Draijer

  2. Faculty of Earth and Life Sciences, Free University Amsterdam, De Boelelaan 1085, 1081 HV, Amsterdam, The Netherlands

    Pepijn P. Dupon & Laura A. Schaap

Authors
  1. Dagmar Fuchs
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Pepijn P. Dupon
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Laura A. Schaap
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Richard Draijer
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Dagmar Fuchs.

Additional file

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Fuchs, D., Dupon, P.P., Schaap, L.A. et al. The association between diabetes and dermal microvascular dysfunction non-invasively assessed by laser Doppler with local thermal hyperemia: a systematic review with meta-analysis. Cardiovasc Diabetol 16, 11 (2017). https://doi.org/10.1186/s12933-016-0487-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12933-016-0487-1

Keywords

  • Skin perfusion
  • Microcirculation
  • Vasodilation
  • Morphology
  • Laser Doppler flowmetry
  • Laser speckle contrast imaging

Cardiovascular Diabetology

ISSN: 1475-2840