Sudoscan^® reclassifies cardiovascular risk in patients with type 2 diabetes mellitus according to the ESC 2023

Abstract

Objectives

The objective of this study is to determine if the detection of DAN through Sudoscan^® can reclassify cardiovascular risk in patients with T2DM according to the European Society of Cardiology guidelines 2023 (ESC 2023) compared to other risk scales.

Methods

A retrospective cross-sectional analytical study was conducted on 161 patients with T2DM who had undergone Sudoscan^® in a Northern Mexico Hospital between December 2022 and July 2023. We compared the cardiovascular risk with SCORE-2, SMART, ADVANCE, ASCVD plus, and Globo Risk scales.

Results

Estimated cardiovascular risk according to the ESC 2023 without consideration of DAN was low for 91 (57%), moderate for 53 (33%), high for 11 (7%) and very high for 22 (3%) of patients. While inclusion of DAN resulted in low risk in 81 (51%), moderate in 46 (28%9, high in 9(5%) and very high risk for 25 ((16%), p = 0.004). The majority of patients were classified as low or moderate risk using other scales such as ASCVD plus, SCORE-2, Globo Risk, ADVANCE, and SMART.

Conclusions

Sudoscan^® enhances cardiovascular risk assessment in T2DM by accurately diagnosing microvascular complications, ensuring comprehensive patient evaluation.

Introduction

The global prevalence of type 2 diabetes mellitus (T2DM) is increasing rapidly, affecting 537 million individuals worldwide in 2021, which makes it an important public health challenge [1–3]. In Mexico, the prevalence of T2DM is 18.3%, approximately 14.3 million people, according to the ENSANUT survey in 2022 [4]. This condition is related to complex pathophysiological bases caused by chronic hyperglycemia and insulin resistance, which is associated atherogenic dyslipidemia, hypertension, and a proinflamatory state. These are determinants for the development of cardiovascular disease and premature death [5, 6].

Distal sensory polyneuropathy (DSN) is one of the most common complications in T2DM, associated with an increased risk of cardiovascular disease and a variety of symptoms [7, 8]. The involvement of multiple systems typically becomes clinically evident in the advanced stages of DSN, which is when it is detected with tuning fork and monofilament test [5–7, 9, 10]. Furthermore, the diabetic autonomic neuropathy (DAN), which can manifest in several ways, such as pupillary, gastroparesis, cardiac, genitourinary, and sudomotor dysfunction [10, 11]. The recognition of sudomotor dysfunction, specifically hypohidrosis or anhidrosis and especially in hands and feet, could represent an early sign of DAN [6, 7].

Sudoscan ^® is an innovative device, developed in 2005 by the company Impeto Medical (France), which evaluates the function of the sudomotor system by detecting dysfunction of the sweat glands innervated by unmyelinated C fibers of the autonomic nervous system, through skin electric conductance based on glandular stimulation [6, 7]. Being a non-invasive, practical, and rapid validated method it has the advantage of identifying neuropathic involvement in early stages [2, 11–17]. This device has been previously used for screening cardiac autonomic neuropathy (CAN) but its direct association with cardiovascular risk calculation has not been evaluated [18]. CAN originates from damage to sympathetic and parasympathetic nerve fibers that supply the heart and blood vessels, resulting to disruptions in cardiovascular autonomic regulation. The advancement of CAN may have a greater impact on the development of cardiovascular disease compared to traditional risk factors [19, 20].

Individuals with T2DM face a two to four times higher risk of developing cardiovascular disease (CVD) throughout their lifetime. Among the manifestations are coronary artery disease (CAD), stroke, heart failure (HF), atrial fibrillation (AF), and peripheral artery disease (PAD) [1, 21]. Multiple scales have been developed to assess cardiovascular risk according to different conditions, backgrounds, and patient-specific factors. For the European population, there’s the SCORE-2 scale designed for individuals without a history of cardiovascular disease but with T2DM the SMART scale is for those with a history of cardiovascular disease and the ADVANCE scale is for individuals with T2DM [1, 22, 23]. On the other hand, for the North American population, there’s the ASCVD Plus scale and for the general population, the Globo Risk scale [24–26]. However, the currently proposed and validated cardiovascular risk estimation calculators do not consider important variables related to T2DM that contribute to a higher risk for heart disease such neuropathy [1].

The aim of this study was to determine if the detection of DAN through use of Sudoscan can reclassify cardiovascular risk in patients with T2DM according to the ESC 2023 and other risk scales. Additionally, we wanted to identify the prevalence of DAN affection in the study population and study its difference between genders. We hypothesize that incorporating the DAN in cardiovascular risk assessment would lead to a greater number of patients being categorized as high or very high risk compared to those assessed without considering autonomic neuropathy.

Materials and methods

This was a retrospective observational cross-sectional analytical study in patients with T2DM over 40 years old treated at the Metabolic Clinic of Hospital Clinica Nova located in Northern Mexico and who had undergone the Sudoscan skin conductance study using the Sudoscan device between December 2022 and July 2023. The research adhered to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines [27]. Approval for the study was obtained from the local institutional review board (IRB number: 02102023-CN-MI-CI), and it was conducted in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki) regarding human experimentation. Due to the retrospective nature of the study, consent form was waived.

With a non-probabilistic convenience sampling, we included patients of both genders, aged 40 to 76 years, who had access to medical care from Hospital Clinica Nova, diagnosis of T2DM, and previously evaluated with Sudoscan reported in the study period as well as complete and up-to-date medical records. The sole exclusion criterion considered was pregnancy.

From the medical records of patients with T2DM who attended follow-up and control at the Metabolic Clinic, we collected demographic data such as gender, race and age (years). We collected vital signs such as, the last systolic and diastolic blood pressure (mmHg) measured by sphygmomanometer and reported in medical visits, as well as somatometric data such as weight (kg), height (cm), and Body Mass Index (BMI).

Also, related to medical history, we obtained smoking habits (positive/negative) and presence of obesity or overweight in the study period; personal history of T2DM and years since diabetes diagnosis, as well as hypertension, dyslipidemia, coronary artery disease, cerebrovascular disease, and/or peripheral arterial disease, aortic aneurysm, atrial fibrillation, and retinopathy. These variables were reported in a dichotomous manner. From the treatment section, we included the history of treatment for systemic arterial hypertension, dyslipidemia and current prescription of antithrombotics, all documented in a dichotomic manner.

From the laboratory studies section, we gathered the following variables, all measured within a six-month timeframe of the Sudoscan measurement: HDL-cholesterol (HDL-c, mg/dl), LDL-cholesterol (LDL-c, mg/dl), total cholesterol (mg/dl), triglycerides (mg/dl), C-reactive protein (CRP, mg/L), glycated hemoglobin type A (HbA1c, %), and serum creatinine (mg/dl). Measurements were conducted using the chemiluminescence method with the Cobas C model 501 (Roche, Hitachi, Japan). Additionally, the creatinine albumin ratio was calculated by dividing albumin concentration (mg) by creatinine concentration (g).

We also included the results of the Sudoscan^® (Impeto Medical, EZS 01750010193, Paris, France) study conducted between December 2022 and July 2023. Sudomotor dysfunction was determined when the result was less than 60 microSiemens (µS) in feet and/or less than 70 microSiemens (µS) in hands. With this, individual cardiovascular risk was estimated using the SCORE-2, SMART, ADVANCE, ASCVD plus, Globo Risk scales and ESC score system from 2023 guidelines, and the results were added to the aforementioned database (1,15–18).

Statistical analysis

All variables were explored, and the normality of the data was evaluated through histograms and the Kolmogorov-Smirnov test. Descriptive statistics were performed using frequencies and percentages for categorical variables, whereas median and interquartile range were used for the description of quantitative variables because data had a non-parametric distribution. Chi-square tests were used for the comparison of categorical variables. The comparison of quantitative variables was conducted using the U Mann-Whitney test due to non-normal distribution. Missing completely at random values were computed through complete case analysis. A p-value < 0.05 was considered significant, and the analysis was performed using R v. 4.0.3 software.

Results

This study initially involved a total of 198 patients. After applying the selection criteria, the analysis was conducted on 161 patients, of which 86 (53%) of the patients were male. In the comparison between genders, there were no differences in age (Table 1). Although, there were difference in systolic and diastolic blood pressures, serum creatinine, total cholesterol and HDL-c (p-<0.05). History of aortic aneurysm, atrial fibrillation, or current prescription of antithrombotic treatment were not found among the participants.

Table 1.

Personal history and laboratory test of the study population

Variables	Total, n = 161	Women, n = 76	Men, n = 85	p-value
Age (years) 1	54 (14)	56 (14)	53 (15)	0.3
Smoking habit1	20 (12)	7 (9.2)	13 (15)	0.2
Overweight1	45 (28)	18 (24)	27 (32)	0.293
Obesity1	97 (60)	50 (66)	47 (55)	0.199
T2DM1	161 (100)	76 (100)	85 (100)	-
Evolution of T2DM (years)2	10 (12)	11 (12)	10 (10)	0.055
Dyslipidemia2	134 (84)	60 (80)	74 (88)	0.193
Hypertension 1	81 (50)	36 (47)	45 (53)	0.529
Coronary artery disease1	8 (5.0)	4 (5.3)	4 (4.7)	> 0.9
Cerebrovascular disease1	2 (1.2)	1 (1.3)	1 (1.2)	> 0.9
Peripheral arterial disease1	2 (1.2)	1 (1.3)	1 (1.2)	> 0.9
Retinopathy1	27 (17)	13 (17)	14 (16)	> 0.9
Body Mass index (kg/m2) 2	31.0 (7.0)	32.0 (7.3)	30.0 (6.0)	0.2
Systolic Blood Pressure (mmHg) 2	125 (16)	120 (13)	130 (20)	0.015
Diastolic Blood Pressure (mmHg) 2	80 (10)	71 (10)	80 (14)	0.001
HbA1c (%)2	7.60 (1.80)	7.55 (1.60)	7.70 (2.10)	0.3
Creatinine (mg/dl) 2	0.79 (0.29)	0.64 (0.21)	0.88 (0.20)	< 0.001
Albumin creatinine ratio (mg/g) 2	9 (22)	8 (13)	10 (46)	0.2
Triglycerides (mg/dl) 2	151.31 (112.37)	147 (110.86)	155.4 (120.19	0.477
Total cholesterol (mg/dl) 2	161 (68)	167 (69)	154 (55)	0.045
HDL-c (mg/dl) 2	44 (13)	47 (15)	42 (13)	0.007
LDL-c (mg/dl) 2	99 (55)	101 (67)	95 (45)	0.2
CRP (mg/L) 2	0.65 (0.90)	0.63 (0.73)	0.65 (1.00)	0.8
Treatment for hypertension 1	81 (50)	36 (47)	45 (53)	0.529
Treatment for dyslipidemia1	102 (64)	48 (63)	54 (64)	1.00

Abbreviations: T2DM – Type 2 Diabetes Mellitus; HbA1c - Hemoglobin type A being separated on cation exchange chromatography; HDL-c – HDL cholesterol; LDL-c – LDL cholesterol; CRP – C Reactive Protein

¹ Frequency and percentage (%) p-value obtained through Pearson’s Chi-squared test

² Median and Interquartile Range (IQR) p-value obtained through U Mann-Whitney test

Men showed significantly higher levels of systolic blood pressure (median (IQR): 125(16), men 130(20) vs. women 120(13), p = 0.015) and diastolic blood pressure (median (IQR) 80(10), men 80(14) vs. women 71(10), p = 0.001).

Regarding laboratory tests, women showed significantly higher levels of median (IQR) total cholesterol (women 167 (69) vs. men 154 (55), p = 0.045], and high-density lipoprotein cholesterol (HDL-c) [women 47 (15) vs. men 42 (13), p = 0.007). Men showed significant higher levels of median (IQR) blood creatinine (men 0.88 (0.20) vs. women 0.64 (0.21), p = < 0.001). Other variables such as HbA1c, albumin creatinine ratio, C reactive protein, triglycerides and LDL-c had a similar distribution without significant difference between genders.

Eighty-one (50%) patients were on treatment for hypertension, and 102 (64%) patients were on treatment for dyslipidemia, showing no significant difference between genders. Non patients were reported to be on antithrombotic treatment. Medical history and laboratory tests are shown in Table 2.

Table 2.

Sudoscan^® test results on hands and feet

Variables		Total, n = 161	Women, n = 76	Men, n = 85	p-value
Hands	Measurement (µS) 1	69 (20)	71 (21)	68 (17)	0.4
	Patients with neuropathic involvement2	85 (53)	36 (47)	49 (58)	0.2
Feet	Measurement (µS) 1	71 (19)	71 (19)	70 (18)	0.8
	Patients with neuropathic involvement2	77 (48)	34 (45)	43 (51)	0.5

¹ Median and Interquartile Range (IQR) p-value obtained through U Mann-Whitney test

² Frequency and percentage (%) p-value obtained through Pearson’s Chi-squared test

The median (IQR) hands conductance measured in microSiemens was 69 (20), while the feet conductance was 71 (19). Nonetheless, nerve conduction studies in hands were abnormal in 85 (53%) patients, while in feet they were abnormal in 77 (48%). The total number of patients found with a neuropathic involvement was 109 (67%). None of these variables showed significant differences between genders. Table 2 presents the assessment of skin conductance by Sudoscan test applied to hands and feet.

This study showed that low and moderate cardiovascular risk was present in a large part of the sample, on the ASCVD plus score 92 (58%) and 51 (32%) patients, respectively; on the SCORE-2 scale 86 (53%) and 69 (43%) patients; on Globo Risk score 108 (67%) and 52 (32%) patients; on ADVANCE score 78 (48%) and 45 (28%) patients; and on the SMART score, although no patient was classified in low risk, 6 (55%) were moderate risk. The mean percentage risk varied according to the scale evaluated. The degrees of cardiovascular risk estimated on ASCVD Plus, SMART, SCORE-2, Globo Risk, and ADVANCE risk scales are shown in Table 3.

Table 3.

Cardiovascular risk assessment on ASCVD Plus, SMART, SCORE-2, Globo Risk, and ADVANCE risk scales

Scale	ASCVD plus	SCORE-2	Globo Risk	ADVANCE	SMART
Average percentage risk 1	8 (13)	4.3 (4.4)	7.0 (7.0)	5.6 (6.6)	29 (14)
Risk level
Low 2	92 (58%)	86 (53%)	108 (67%)	78 (48%)	0 (0%)
Moderate 2	51 (32%)	69 (43%)	52 (32%)	45 (28%)	6 (55%)
High 2	12 (7.5%)	6 (3.7%)	1 (0.6%)	34 (21%)	3 (27%)
Very high 2	5 (3.1%)	-	-	4 (2.5%)	2 (18%)

n = 161.¹ Median and Interquartile Range (IQR). ² Frequency and percentage (%)

The evaluation of cardiovascular risk using the ESC 2023 score system had differences when considering DAN involvement. Without considering Sudoscan results, many patients were classified as low risk (57%), moderate risk (33%), followed by high (7%) and very high risk (3%). However, upon incorporating Sudoscan data, a significant shift in risk categorization was observed. The proportion of patients classified as low risk decreased to 51%, moderate risk to 28% and high risk to 5%, while the percentage classified as very high risk increased to 16% (p = 0.004). Figure 1 summarizes the distribution of risk levels among the study sample.

Fig. 1.

Cardiovascular risk assessment on ESC score system 2023 comparing consideration of Sudoscan^® test. ¹ Frequency and percentage (%). n = 161. p-value obtained through Pearson’s Chi-squared test

Discussion

In this study, the comparison between cardiovascular risk classification according to the European Society of Cardiology and the 2023 guidelines revealed a significant difference when considering the presence of DAN. Without taking DAN into account, 57% of the patients were classified with low risk, 33% with moderate risk, 7% with high risk, and 3% with very high risk. However, when DAN was taken into consideration, the distribution changed significantly, with 16% of patients reclassified as very high risk, 5% as high risk, 28% as moderate risk and 51% as low risk. Other risk scales classified many patients in low and moderate risk categories without a similar distribution, due to the different criteria for their calculation.

Risk calculators like SCORE-2, SMART, ADVANCE, ASCVD Plus, and Globo Risk do not include diabetic neuropathy in risk assessment. Additionally, they do not consider other important factors for classification, such as diabetic retinopathy, kidney damage, or other micro and macrovascular complications that are common in long-standing T2DM [22–26, 28]. In 2023, the new PREVENT scale was published for determining cardiovascular risk. This scale incorporates a larger number of laboratory and clinical criteria, including factors related to renal and metabolic impairment. However, it does not include retinopathy or neuropathy as contributing factors in the risk estimation. This omission may highlight similar limitations observed in the other scales analyzed in this study [29, 30].

In our study, we found that a significant number of patients with T2DM had abnormal nerve conduction in both their hands and feet, as detected by Sudoscan. Specifically, 53% of patients had abnormal nerve conduction in their hands, and 48% in their feet, resulting in an overall neuropathic involvement rate of 67%. In comparison, other studies conducted in larger populations report sudomotor dysfunction rates detected by Sudoscan ranging from 27.6 to 44.7%, without distinguishing between hands and feet [2, 7]. Similar prevalence rates have been reported globally, with autonomic neuropathy documented in 21–44% of patients with T2DM across various regions [6, 9, 13, 30–32]. The difference between our study and the others mentioned likely stems from our smaller sample size, which may affect the prevalence of DAN.

Different mechanisms for diabetic neuropathy have been proposed. One key mechanism involves the activation of the polyol pathway due to persistent hyperglycemia [33]. This pathway leads to increased production of sorbitol and fructose within nerve cells, resulting in osmotic stress and subsequent nerve damage. Additionally, the accumulation of sorbitol can lead to reduced levels of myoinositol, an important nerve cell membrane phospholipid, further compromising nerve function. The polyol pathway also contributes to oxidative stress, which damages nerves through protein glycosylation and activation of AGE (advanced glycation end-product) receptors [30, 31, 33, 34]. These processes ultimately result in reduced levels of neurotrophins essential for nerve survival and regeneration, as well as mitochondrial dysfunction within axons, highlighting the disease’s multifaceted pathophysiology. Persistent hyperglycemia not only triggers the polyol pathway but also induces oxidative stress, inflammation, and endothelial dysfunction, all of which contribute to neural dysfunction and autonomic disturbances. The autonomic nervous system regulates essential involuntary bodily functions, including cardiovascular activity. Therefore, dysfunction in autonomic nerves due to diabetes significantly increases cardiovascular risk. Understanding and integrating these complex mechanisms is crucial for effectively managing cardiovascular risk in patients with diabetic autonomic neuropathy [1, 33, 34]. The early detection of neuropathy is crucial due to its association with cardiovascular risk, which is a key factor in estimating the 10-year risk of developing cardiac disease. This highlights its importance in initial treatment and prognosis [1, 28]. However, the accuracy of cardiovascular risk classification is compromised by the omission of neuropathy in many existing risk calculators. Despite this, the 2023 European guidelines for risk classification include neuropathy as a consideration. The European Society of Cardiology’s SCORE-2 calculator incorporates microvascular complications, including neuropathy, retinopathy, and renal impairment, to accurately classify patients with T2DM.

The study found that a significant number of individuals had notable comorbidities. Specifically, 28% were overweight and 67% were obese. Additionally, 84% had dyslipidemia, and 50% had hypertension. Previous studies, such as the Framingham Heart Study, the INTERHEART, and INTERSTROKE studies, have shown strong connections between smoking, dyslipidemia, hypertension, diabetes, obesity, unhealthy diet, physical inactivity, alcohol consumption, psychosocial factors, and the development of cardiovascular disease [1, 35–39]. It’s worth noting that factors such as smoking, retinopathy, cardiovascular disease history, and certain kidney-metabolic health indicators were not prevalent or uncontrolled in this study. Furthermore, there were no significant differences between genders, unlike in other studies where these factors were more prevalent in women [37].

In this study, we compared the criteria for cardiovascular risk assessment across different scales and examined the impact of including DAN (diabetic autonomic neuropathy) as an assessment criterion. Additionally, we described the prevalence of sudomotor dysfunction in patients with T2DM detected by Sudoscan in the northeastern region of the country.

One limitation of this study is its retrospective cross-sectional design with non-probabilistic convenience sampling, which may introduce information bias. However, this bias was minimized through thorough examination of medical records. Another limitation is the small sample size, which can affect the outcomes; despite this, the results still showed significance. It’s important to note that cardiovascular risk calculators have an intrinsic bias based on the population they were developed in, which can affect the accuracy of risk predictions for individual patients. Future research should focus on longitudinal studies to validate these findings and investigate the effectiveness of specific interventions in reducing cardiovascular risk among patients with T2DM and neuropathy.

Conclusion

The integration of an advanced diagnostic tool, such as Sudoscan^®, enhances the reclassification of cardiovascular risk in patients with T2DM. By considering Sudoscan^®’s diagnosis as a criterion for microvascular complications, healthcare providers can more accurately assess and classify patients’ risks. This approach helps ensure that existing diabetic complications are not overlooked. Also, this helps to improve the medical care of the patients.

Abbreviations

T2DM

Type 2 diabetes mellitus

DSN

Distal sensory polyneuropathy

DAN

Diabetic autonomic neuropathy

ESC 2023

European Society of Cardiology guidelines 2023

ENSANUT

Encuesta Nacional de Salud y Nutrición

CAN

Cardiac autonomic neuropathy

CVD

Developing cardiovascular disease

CAD

Coronary artery disease

HF

Heart failure

AF

Atrial fibrillation

PAD

Peripheral artery disease

SCORE-2

Systematic Coronary Risk Evaluation 2

SMART

Second manifestations of arterial disease

ASCVD

Atherosclerotic cardiovascular disease

STROBE

Strengthening the Reporting of Observational Studies in Epidemiology

AGE

Advanced glycation end-product

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

Data will be made available on reasonable request.

Declarations

Ethical approval

The research adhered to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines. Approval for the study was obtained from the local institutional review board (CEI-EM UDEM, IRB number: 02102023-CN-MI-CI), and it was conducted in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki) regarding human experimentation. Due to the retrospective nature of the study, consent form was waived.

Disclosure

All authors read and approved the final manuscript.

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.