Assessing the Impact of BMI and Glycaemic Control on Cardiac Autonomic Neuropathy in Patients with Early and Long-standing Diabetes Mellitus: A Window of Opportunity

Abstract

Background

Cardiac autonomic neuropathy (CAN) is a well-known, yet underreported, complication of Diabetes mellitus (DM). CAN assessment can be performed using heart rate variability (HRV) and Ewing’s battery of tests. Severity of CAN may be affected by multiple factors like glycaemic control, obesity and co-existing distal neuropathy. However, the relationship between CAN and these factors remains underexplored.

Purpose

We conducted the present study to assess the relationship between the prevalence of CAN in patients with DM and associated factors like obesity and glycaemic control.

Methods

Clinically diagnosed patients of DM were recruited. They were divided into duration <5 years (early ‘Dur < 5’) and ≥5 years (long-standing, ‘Dur ≥ 5’) and evaluated using HRV and Ewing’s battery of tests. Apparently, healthy young adult controls were recruited as controls. Cardiac autonomic tone and reactivity parameters were compared between the groups.

Results

We recruited 50 patients with DM (25 ‘Dur < 5’ and 25 ‘Dur ≥ 5’) and 34 healthy controls. Controls were significantly younger than the patients. Height, weight and BMI were comparable between the groups. Patients had impaired cardiac autonomic tone but preserved response to reactivity tests when compared with controls. Normal BMI and good glycaemic control, indicated by lower values of HbA1C, may be responsible for these results.

Conclusions

Patients with DM demonstrated reduced tone but preserved autonomic reactivity, even when compared with younger, apparently healthy controls. Normal BMI and relatively lower HbA1C values may contribute to this phenomenon. Such patients may be identified and targeted for early clinical intervention for ANS salvage.

Keywords

Autonomic neuropathy autonomic function body mass index Diabetes mellitus heart rate variability

Introduction

Cardiac autonomic neuropathy (CAN) is one of the hallmarks of Diabetes mellitus (DM). The initial symptoms are innocuous, such as resting tachycardia and exercise intolerance and therefore often go undiagnosed.^{1, 2} While CAN does not manifest as clinically overt symptoms in the initial stages of DM, underlying damage can pose serious consequences in the long run. CAN predisposes to the risk of arrhythmias, hemodynamic perturbations in the operating room, silent myocardial ischaemia and sudden death.³ Therefore, CAN identification bears an important role in the management of patients with DM.

Cardiac autonomic reflex tests (CARTs) provide a cost-effective, reliable and patient-friendly means of diagnosing CAN. These consist of five simple tests—orthostasis, deep breathing, Valsalva manoeuvre, cold pressor test and hand grip test (HGT). Blood pressure and/or heart rate responses to these tests are compared with standardised cut-offs proposed by Clarke and Ewing.⁴ Cumulative scores can be computed to classify CAN as Early, Borderline or Definite.⁵ Combined with heart rate variability (HRV)—a measure of autonomic tone—CARTs can provide a comprehensive picture of the state of underlying CAN in patients with DM. These tests have been found to be repeatable across multiple studies^6–8 and therefore find wide applicability for the diagnosis of CAN in DM.

Previous literature has reported the prevalence of CAN ranging from 20% to 65%¹ in patients with DM. Literature from our country reports this prevalence figure ranging from 40% to 80% in patients with DM.^9–12 The majority of literature reports, wherein such high prevalence metrics have been reported, consist of patients with long-standing DM. However, this figure is variable in patients who are newly diagnosed or in the early phase of the disease. Prevalence in this subgroup has been reported to be from as low as 2.1%–7.3%¹³ to as high as 30%.¹⁴ In this subgroup of patients, prevalence can be affected by a multitude of factors—ranging from duration of disease, higher age, nature of glycaemic control and co-existing peripheral neuropathy. There is a scarcity of literature comparing the extent of CAN in patients with DM of recent onset versus those with long-standing disease. Therefore, we attempted to explore the relationship between autonomic indices in patients with DM with different durations of the disease and associated factors like glycaemic control and obesity.

Method

The study was approved by the Institute Ethics Committee (IEC), All India Institute of Medical Sciences, Jodhpur, India vide letter number AIIMS/IEC/3386 dated 12 March 2021. Our study was cross-sectional and observational in nature. STROBE® guidelines were followed for the study. Clinically diagnosed patients with DM were recruited from the outpatient department of Endocrinology and Metabolism of our hospital. The patients were diagnosed using the American Diabetes Association (ADA) criteria.¹⁵ Patients of either sex, between 18 and 50 years, were included. We excluded patients with comorbidities or taking medications that were likely to affect the autonomic nervous system, any pre-existing cardiovascular or neurological disease and patients with a history of substance abuse. Apparently healthy young adults of either sex, aged 18–50 years, were chosen as controls.

After obtaining written informed consent, patients were requested to report to the autonomic function laboratory of our department, in the forenoon between 9 am and 1 pm, 2 hours after a light meal. They were requested to abstain from tea/coffee/any form of tobacco on the morning of the test. Also, they were requested to avoid heavy exercise or any form of strenuous physical activity on the day before the test. Upon arrival at the lab, they were requested to empty their bladder and the study protocol was explained to them in detail. A brief history was taken to assess the presence of complaints suggestive of involvement of the ANS.

Data acquisition was performed in a noise-free, humidity and temperature-controlled environment. Skin was thoroughly cleaned with alcohol-based swabs. Adhesive ECG electrodes were applied in Lead II configuration for the purpose of ECG acquisition, and a digital stethograph was placed across the 4th intercostal space to record respiratory movements. The signals were acquired using a digital physiograph (Biopac MP 150™ system, Biopac Inc., USA). A sampling rate of 1 kHz was used for signal acquisition. A band-pass filter was used for the ECG signal to ensure optimum signal quality and remove noise. Supine rest of 5 minutes was provided, followed by acquisition of data for another 5 minutes for the purpose of HRV estimation. During HRV data acquisition, patients were requested to lie still and breathe normally. Data was visualised in real time to avoid any artefacts/ectopics in the 5-minute period. We computed standard time domain (SDNN, RMSSD, SDSD and pNN50%) and frequency domain indices (Total power, LF-power, HF-power and LF/HF ratio) as per guidelines proposed by the Task Force.¹⁶

Since ectopics are excluded by default for the purpose of HRV analysis, the RR intervals chosen are referred to as NN intervals, or normal-to-normal intervals. SDNN represents the standard deviation of normal-to-normal (NN) intervals in the data segment. RMSSD is the root mean square of the successive differences between RR intervals. SDSD is the standard deviation of the differences between RR intervals. pNN50% is the percentage of RR intervals varying by more than 50 ms in the 5-minute window. For the purpose of frequency domain parameters, spectral analysis was performed using the Fast Fourier transform (FFT) of the interbeat interval data. Total power represents the overall absolute power in the frequency band 0–0.4 Hz. Low frequency (LF-power) and high frequency (HF-power) represent the powers in the frequency bands 0.04–0.15 Hz and 0.15–0.4 Hz, respectively. LF/HF ratio is an index of sympatho-vagal balance.¹⁷ HRV analysis was performed using LabChart Pro™ software version 8 (AD Instruments, Australia) and the aforementioned time and frequency domain indices were computed.

This was followed by CARTs as per a battery of tests suggested by Ewing and colleagues.⁵ Lying to standing test (LST), Deep breathing test (DBT), HGT and Cold pressor test (CPT) were performed as per the standard protocol described elsewhere.^{18, 19} Change in systolic blood pressure (∇SBP-LST) and 30:15 ratio were computed for LST. The averaged difference of maximum and minimum heart rate during deep breathing (∇HR-DBT) and ratio of averaged maximum and minimum RR intervals (E:I ratio) were computed for DBT. Change in diastolic blood pressure was computed for HGT (∇DBP-HGT) and CPT (∇DBP-CPT), respectively. Previously published blood pressure and heart rate cut-offs were used for comparison.²⁰ ∇SBP-LST > 10 mm Hg and 30:15 ratio <1.04, ∇HR-DBT < 15 beats/min and E:I ratio < 1.21 and ∇DBP-HGT < 10 mm Hg and ∇DBP-CPT < 10 mm Hg were considered as abnormal. The Valsalva manoeuvre was excluded from the battery since it is known to increase the risk of retinopathy in patients with pre-existing DM.^21–23 All tests were performed and analysed by a single observer to prevent inter-observer bias.

We divided patients with DM into two groups—disease duration <5 years and ≥5 years and labelled them as patients with ‘early’ DM and ‘long-standing’ DM, respectively. This classification was taken based on cut-offs proposed in earlier studies.^{23, 24} Values were tabulated in a spreadsheet programme. MedCalc™ software version 23.2.6 (MedCalc Software Ltd, Ostend, Belgium; https://www.medcalc.org; 2025) was used for performing statistical analysis. The Shapiro-Wilk test was used to assess the normality of the distribution of parameters. Values were expressed as mean ± SD or Median (Interquartile range) based on the distribution. Based upon the Gaussian fit of parameters, one-way ANOVA or Kruskal-Wallis test was used to compare autonomic indices between the groups—healthy, DM < 5 years and DM ≥ 5 years. A p value of <.05 was taken as statistically significant.

Results

We recruited 50 patients with DM and 34 healthy young adults as controls. Out of 50, 25 subjects had DM duration <5 years, classified as ‘early DM’, and 25 patients had duration ≥5 years, classified as ‘long-standing DM’. The classification of patients into ‘early’ and ‘long-standing’ DM was based on 5-year disease duration; this cut-off was chosen in accordance with a similar classification used in previous studies.^{25, 26} These two groups are referred to as ‘Dur < 5’ and ‘Dur ≥ 5’, henceforth. The patients with DM (<5 years and ≥5 years) were older than the control group (mean age = 36.36 ± 7.77 and 35.96 ± 9.51 years vs. 29.65 ± 5.65 years, p = .001). Both the patient groups and healthy controls had comparable height, weight and Body Mass Index (BMI) (p = .367, .789 and .232, respectively). Patients in ‘Dur < 5’ had relatively lower HbA1C as compared to patients in ‘Dur ≥ 5’ group, but the values were not statistically significant. The values are summarised in Table 1.

Table 1.

Anthropometric and Clinical Characteristics of Study Subjects.

S. No	Parameter	Healthy	DM < 5 Years	DM ≥ 5 Years	p Value
1	Number	34	25	25	–
2	Males, females	21,13	13,12	19,6	–
3	Age (yrs)	29.65 ± 5.65	36.36 ± 7.77	35.96 ± 9.51	.001
4	Height (cm)	164.85 ± 10.85	162.32 ± 9.22	166.40 ± 10.92	.367
5	Weight (kg)	67.18 ± 16.45	66.32 ± 11.80	64.51 ± 14.96	.789
6	BMI (kg/m²)	24.45 ± 3.9	23.19 ± 4.66	25.27 ± 4.43	.232
7	HbA1C	–	7.80 (7.05–8.50)	8.30 (7.38–9.20)	.19
8	Duration of DM (yrs)	–	2 (1.00–2.63)	6.00 (5.00–10.00)	<.0001

Note: Values expressed as Mean ± SD and Median (Interquartile range). One-Way ANOVA was used to compare all parameters, except HbA1C and duration of DM, which was compared using Mann-Whitney U test. DM – Diabetes mellitus. Values in bold and italic represent statistically significant parameters.

CAN Assessment

We observed that HRV indices—time domain as well as frequency domain—were significantly different in patients with DM as compared to healthy controls. SDNN, RMSSD, SDSD and pNN50% were significantly lower in patients with DM (p = .000002, .000002, .000004 and .000006, respectively, Figure 1) when compared with healthy controls (Figure 1). The same pattern was observed in frequency domain indices as well, with Total power, LF-power and HF-power being significantly lower in patients with DM (p = .000006, .000024 and .000003, respectively). For all the aforementioned parameters, HRV indices were lowest in patients with long-standing DM, ‘Dur ≥ 5’, followed by patients with ‘early DM’, ‘Dur < 5’ and highest in healthy controls. The only exception was the LF/HF ratio, which was highest in patients with long-standing DM, ‘Dur ≥ 5’, followed by ‘early DM’, ‘Dur < 5’ and lowest in healthy controls (p = .023). The values are summarised in Figure 1.

Figure 1.

Notes: Values represented as box and whisker plots. Individual data points are represented as hollow circles. Hollow squares represent outliers. All time domain (SDNN, RMSSD, SDSD and pNN50%) and frequency domain indices (Total power, LF-power and HF-power), with the exception of LF/HF ratio, were significantly lower in patients with diabetes mellitus duration ≥5 years, when compared with diabetes mellitus duration <5 years and healthy controls (p = .000002, .000002, .000004 and .000006 for time domain indices and p = .000006, .000024, .000003 for frequency domain indices, respectively, Kruskal-Wallis test with Dunn’s post hoc test). SDNN, standard deviation of NN intervals; RMSSD, root mean square of the differences between NN intervals; SDSD, standard deviation of the differences between NN intervals; and pNN50%, percentage of NN intervals varying by more than 50 ms with respect to the previous interval. Total power, the overall variance of all NN intervals in the 5-minute segment; LF, Power in the low frequency (0.04–0.15 Hz) band; HF, power in the high frequency (0.15–0.4 Hz) band; LF/HF, ratio of LF-power and HF-power. Dur < 5, duration of diabetes mellitus less than 5 years; Dur ≥ 5, duration of diabetes mellitus ≥5 years.

Autonomic reactivity indices were derived from LST, DBT, HGT and CPT. ∇SBP-LST and 30:15 ratios were comparable between the groups (p = .724 and .525, respectively). While there was a statistically significant difference in ∇HR-DBT and E:I ratio (p = .0127 and .0033, respectively), the median values were within normal limits, as per previously published criteria.²⁰ A similar trend was observed for the rise in diastolic blood pressure on the cold pressor test (∇DBP-CPT). Median value of DBP rise after CPT was higher in healthy controls, as compared to ‘early’ and ‘long-standing’ DM (p = .0024). However, median values were within acceptable cut-offs. Rise in diastolic pressure on HGT (∇DBP-HGT) was comparable between the groups—healthy controls, ‘early’ DM and ‘long-standing’ DM (p = .058). Interestingly, the rise in DBP after hand grip exercise was higher in ‘long-standing’ DM than ‘early’ DM group and healthy controls. The values are summarised in Figure 2.

Figure 2.

Notes: Values represented as box and whisker plots. Individual data points are represented as hollow circles. Hollow squares represent outliers. There was no significant difference between ∇SBP-LST and 30:15 ratio between the groups (p = .724 and .545, respectively). ∇HR-DBT and E:I were significantly different between the groups (p = .0127 and .0033, respectively), but median values were within normal limits. There was no significant difference between ∇DBP-CPT between the groups (p = .058), whereas a significant difference was observed in ∇DBP-CPT values between the groups (p = .0024). Kruskal-Wallis test with post hoc Dunn’s test was used for comparison. Dur < 5 – duration of diabetes mellitus less than 5 years, Dur ≥ 5 – duration of diabetes mellitus ≥5 years.

Discussion

In the present work, we assessed HRV and autonomic reactivity indices in patients with DM and young healthy adults. We divided patients with DM into two groups—those with disease duration <5 years and ≥5 years. There was a significant decline in HRV indices, both time domain and frequency domain, in patients with DM as compared to controls. However, the autonomic reactivity indices, derived from CARTs, were found to be within normal limits (Figures 1 and 2). These findings are suggestive of the fact that patients with DM may have preserved responses to CARTs, suggestive of them being in early stages of CAN.

CAN is a well-known, yet underreported, complication of DM. Impairment of HRV is suggested to be the earliest sign of subclinical CAN.²⁷ Vagal involvement occurs early in the course of the disease, responsible for parasympathetic dysfunction. With subsequent involvement of the sympathetic limb, full-blown CAN manifests. As per recommendations of ADA released in 2025, screening for CAN should be performed at the time of diagnosis in patients with Type 2 DM and within 5 years of diagnosis in patients with Type 1 DM. Thereafter, annual screening for CAN is recommended for all patients with DM.²⁸ Therefore, it is recommended to institute CAN assessment as a part of the routine workup of patients with DM.

Co-existing CAN increases the risk of complications in DM, especially cardiovascular disease (CVD). Initial signs may be subtle, consisting of resting tachycardia and exercise intolerance, and therefore often go undetected and unreported. The vagus nerve is involved early in the course of disease, leading to diminished parasympathetic activity and sympathetic dominance.³ Therefore, patients present with increased resting heart rate in the early phase of CAN due to DM. In the long run, concomitant involvement of the sympathetic limb also happens. Exercise intolerance may result due to a variety of factors—poor glucose uptake by exercising muscles, subnormal augmentation of cardiac output in response to increased demand and vascular changes limiting muscle perfusion.^29–31 However, serious features such as orthostatic hypotension and QT prolongation can have sinister consequences. Patients of DM with CAN are at increased risk of coronary artery disease, including myocardial infarction, sudden cardiac death, heart failure and arrhythmias.^{1, 2} Therefore, CAN assessment adds clinical value in the diagnosis and management of patients with DM. This assessment can be easily performed in the clinical physiology laboratory.

Assessment of CAN is commonly done using Ewing’s battery of tests. The initial battery consisted of orthostatic challenge (active lying to standing or passive head up tilt test, in patients who cannot stand briskly and/or without support), metronomic deep breathing, Valsalva manoeuvre and HGT.^{4, 5} Over time, the cold pressor test^{20, 32–35} has been incorporated as an integral component of this battery. These tests are labelled collectively as CARTs. The blood pressure and heart rate responses to these tests are graded based on the magnitude. The cumulative score derived from the tests can be used for stratification as early, definite or severe, based on criteria proposed by Ewing and colleagues.⁵ As per the AIIMS New Delhi criteria, CAN is classified as early or definite. Abnormal response in one test or borderline response in two tests is suggestive of early CAN. Two abnormal tests are suggestive of definite CAN. The detailed classification mechanism is discussed elsewhere.³⁴ Other scores, such as Bellavere scores,³⁶ are also used, although sparingly.

There are multiple risk factors which have been reported to precipitate CAN in patients with DM. These include poor glycaemic control, higher age, obesity, and co-existing issues such as hypertension and peripheral distal neuropathy.² Duration of disease is another important risk factor for the development of CAN in DM. In this context, the recent onset of disease is likely to be associated with a lower incidence of CAN. This premise has been explored in multiple studies, studying autonomic function in patients with early/recent onset of DM. However, the definition of ‘early’ remains a matter of debate. In some papers, ‘newly diagnosed’ DM has been defined as disease onset within one year of presenting to the clinics.³⁷ Even in this group, authors have reported considerable prevalence of CAN, with figures as high as ~40%. However, many patients may present to the clinics after the first year of symptom onset. In such instances, a median duration of 5 years of symptoms has been defined as ‘early’ DM.^{25, 26} We adopted this cut-off in the present work and compared autonomic function indices between patients with ‘early’ and those with ‘long-standing’ disease.

Herein, we observed an interesting pattern wherein CARTs demonstrated values within normal ranges. These patients only had impairment of HRV, suggestive of subclinical involvement of the ANS. This finding is important from a clinical standpoint, since, contrary to previous reports in DM, these patients had preserved response to CARTs. We attempted to explore the reason behind this phenomenon.

As stated previously, obesity and glycaemic control are important factors in the pathogenesis of DM. It has been observed that longer duration of disease and poorer glycaemic control may increase the prevalence of CAN in patients with DM.³⁸ The relationship between BMI and DM is complex. The conventional view is that higher BMI is associated with increased risk of DM and complications due to the disease.^{39, 40} However, some studies have also challenged this relationship, demonstrating no effect of BMI or an inverse relationship between BMI and DM. A Tehran-based study demonstrated no association between baseline BMI and risk of DM. However, the authors reported that an increase in BMI with time led to an increased risk of DM.⁴¹ Another study performed in China demonstrated an inverse relationship between BMI and mortality in patients with DM.⁴² Recent literature is challenging the notion of BMI as a predictor of risk of DM, since such risk can be affected by ethnicity and geographical differences.⁴³ Therefore, the relationship between DM and BMI continues to evolve.

In the present work, we assessed whether ‘normal’ BMI and ‘lower’ HbA1C may preclude/delay the onset of CAN in patients with DM. The patients of our study, those with DM < 5 years and DM ≥ 5 years, had normal BMI (mean values being 23.19 ± 4.66 and 25.27 ± 4.43 kg/m², respectively). This may be one of the contributing factors to their normal response to CARTs. Another factor that may be responsible is the relatively lower values of HbA1C observed in the patients with DM. Median HbA1C values in Dur < 5 and Dur ≥ 5 were 7.80 g% and 8.30 g%, respectively. These findings are supported by Karthikayen and colleagues,¹¹ wherein patients of DM with a mean HbA1C of 8 g% did not show any CAN. Devi and colleagues⁴⁴ were also unable to find features of CAN in patients with a median HbA1C of 6.8 g%. Similar observations have been reported by Mirg and colleagues,¹² wherein the group with a mean HbA1C of 9.38 g% did not show laboratory features of CAN. Verma and colleagues²⁴ have also reported a HbA1C of less than 8 g% to be suggestive of fair control in patients with DM. Therefore, good glycaemic control, as demonstrated by relatively lower HbA1C, as per literature-based cut-off values, may have had a protective effect on patients with DM in our study. Elevated levels of HbA1C, suggestive of poor glycaemic control present with a higher prevalence of CAN.

Duration of disease is another important factor, which may concomitantly affect the incidence of CAN in patients with DM. Patients with a short duration of the disease may not demonstrate features of CAN. Akbar and colleagues⁴⁵ have reported the average duration for development of CAN from diagnosis to be about 10–12 years in a sample of 202 patients with DM. Similar trends have been reported in the aforementioned studies from our country,^{10, 12, 44} wherein CAN was absent in patients with a mean duration of DM being 2.5, 5.5 and 3.13 years, respectively. The median durations of disease in ‘Dur < 5’ and ‘Dur ≥ 5’ were 2 and 6 years, respectively (Table 1). However, poor glycaemic control may offset the protective effect of a shorter duration of disease.

Based on our findings, we may infer that normal BMI with lower HbA1C, as compared to previous reports, may be responsible for the normal response to CARTs in both ‘early’ and ‘long standing’ DM patients. The presence of impaired HRV indices may be suggestive of subclinical CAN. These findings reinforce the notion that good glycaemic control and weight control may provide protective effects against CAN in patients with DM. The findings are also useful from a clinical standpoint. We may recommend that routine screening of patients with DM, using biochemical and CARTs with HRV, may help identify the subgroups with ‘subclinical’ CAN. Since these patients are in early stages of neuropathy, clinicians may institute pharmacotherapy or other therapeutic measures, so as to delay/reverse CAN and prevent its potential complications, some of which may be life-threatening. Such interventions have been documented in the literature^46–48 and may be beneficial from a therapeutic perspective.

One interesting observation that merits discussion is the higher rise in DBP after hand grip exercise in patients with ‘long-standing’ DM, when compared with ‘early’ DM and healthy controls. There may be two possible reasons behind this phenomenon. DM has been shown to manifest with parasympathetic withdrawal and sympathetic overactivity with the progression of disease.^{49, 50} We observed a higher LF/HF ratio, which may be an indicator of such an imbalance. However, the validity of the LF/HF ratio as a standalone marker of sympatho-vagal balance has been challenged recently.⁵¹ Another probable reason behind this exaggerated response may be increased vascular stiffness in patients with DM. DM has been known to cause vascular damage, leading to increased arterial stiffness in the long run.^{52, 53} We hypothesise that this may be another factor probably responsible for the exaggerated diastolic blood pressure response to the HGT. However, we did not have the necessary equipment to measure vascular stiffness in the present work. Therefore, in the absence of an objective marker of arterial stiffness measurement in the present work, we are unable to verify our proposed hypothesis.

There are some limitations to the present work. First, we did not have a sufficient number of patients with Type 1 and Type 2 DM to explore the differences in CAN between the groups. Also, we did not have the tools and resources to quantify the co-existence of peripheral neuropathy, nephropathy or other biochemical parameters. Further work is needed to establish the role of these factors in the pathogenesis of CAN. The third potential limitation was the age difference between the patients and apparently healthy controls. There was a significant difference between the ages of the groups—apparently, health controls were younger than patients. But it was an insightful finding, wherein patients, despite being older than the control group, had preserved autonomic reactivity parameters with only impairment in cardiac autonomic tone. We could not compare the effect of different types of drugs and adherence to pharmacotherapy in the present work. We suggest that this avenue can serve as a direction for future work.

Conclusion

Despite the present work having an exploratory focus, we believe that it provides valuable insights into the factors governing CAN in patients with DM. Our findings reinforce the potential protective effects of BMI and good glycaemic control, reflected by lower HbA1C values, on autonomic reactivity parameters. The study extends our knowledge on the relationship between good glycaemic control and weight control, and prevention of CAN in patients with DM. In addition, identification of such patients with DM, with early/subclinical CAN, may provide opportunities for lifestyle modification and therapeutic interventions for salvage of the ANS. This will reduce the risk of comorbidities and complications and may potentially be very useful from a clinical perspective.

Abbreviations

CAN: Cardiac autonomic neuropathy

DM: Diabetes mellitus

CART: Cardiac autonomic reflex tests

ECG: Electrocardiogram

HRV: Heart rate variability

IEC: Institute Ethics Committee

SDNN: Standard deviation of normal-to-normal intervals

RMSSD: Root mean square of successive differences

SDSD: Standard deviation of differences between adjacent NN intervals

pRR50: Percentage of RR intervals varying by more than 50 milliseconds

LF: Low frequency

HF: High frequency

LST: Lying to standing test

DBT: Deep breathing test

HGT: Hand grip test

CPT: Cold pressor test

∇SBP-LST: Change in systolic blood pressure on lying to standing test

∇HR-DBT: Averaged difference of maximum and minimum heart rate during deep breathing

E:I ratio: Ratio of averaged maximum and minimum RR intervals

∇DBP-HGT: Change in diastolic blood pressure after hand grip test

∇DBP-CPT: Change in diastolic blood pressure after cold pressor test

ADA: American Diabetes Association

CVD: Cardiovascular diseases

Footnotes

Acknowledgements

The authors acknowledge the support provided by Mr Shiv Kumar, Laboratory staff, for his help in data acquisition.

Authors’ Contribution

Conceptualisation: SS and MS conceptualised the idea. MS, SS were involved in data curation, and MS performed formal analysis. Methodology and Validation were done by SS, OLB and RGS. The original draft was written by SS and MS. Reviewing and editing were done by SS, OLB and RS. All authors have critically reviewed and approved the final draft and are responsible for the content and similarity index of the manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Data Availability

Data will be made available upon reasonable request.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

ICMJE Statement

This article complies with the International Committee of Medical Journal Editors (ICMJE) uniform requirements for the manuscript.

Statement of Ethics

The study was approved by the Institute Ethics Committee, All India Institute of Medical Sciences, Jodhpur, India vide letter number AIIMS/IEC/3386 dated 12 March 2021.

ORCID iDs

Mamta Shobhawat

Shival Srivastav

References

Spallone

, Ziegler

, Freeman

, . Cardiovascular autonomic neuropathy in diabetes: Clinical impact, assessment, diagnosis, and management. Diabetes Metab Res Rev 2011; 27(7): 639–653. DOI: 10.1002/dmrr.1239.

Bissinger

Cardiac autonomic neuropathy: Why should cardiologists care about that?

J Diabetes Res 2017; 2017: 5374176. DOI: 10.1155/2017/5374176.

Balcıoğlu

and Müderrisoğlu

Diabetes and cardiac autonomic neuropathy: Clinical manifestations, cardiovascular consequences, diagnosis and treatment. World J Diabetes 2015; 6(1): 80–91. DOI: 10.4239/wjd.v6.i1.80.

Clarke

and Ewing

DJ.

Cardiovascular reflex tests in the natural history of diabetic autonomic neuropathy. N Y State J Med 1982; 82(6): 903–908.

Ewing

, Martyn

, Young

, . The value of cardiovascular autonomic function tests: Ten years’ experience in diabetes. Diabetes Care 1985; 8(5): 491–498.

Schumann

, Lukas

, Rieger

, . One-week test-retest recordings of resting cardiorespiratory data for reliability analysis. Sci Data 2025; 12(1): 12. DOI: 10.1038/s41597-024-04303-y.

Schumann

, Lukas

, Rieger

, . One-week test-retest stability of heart rate variability during rest and deep breathing. Physiol Meas 2025; 13(2). DOI: 10.1088/1361-6579/adae51.

Jose

, Srivastav

and Mehta

Investigating the visit-to-visit repeatability of autonomic neuropathy assessment measures in healthy subjects. Ann Neurosci 2025; 42(1): e09727531241310034. DOI: 10.1177/09727531241310034.

Pappachan

, Sebastian

, Bino

, . Cardiac autonomic neuropathy in diabetes mellitus: Prevalence, risk factors and utility of corrected QT interval in the ECG for its diagnosis. Postgrad Med J 2008; 84(990): 205–210. DOI: 10.1136/pgmj.2007.064048.

10.

Bhuyan

, Baro

, Sarma

, . A study of cardiac autonomic neuropathy in patients with type 2 diabetes mellitus: A Northeast India experience. Indian J Endocrinol Metab 2019; 23(2): 246–250. DOI: 10.4103/ijem.IJEM_336_18.

11.

Karthikeyan

, Manjunath

, Latha

, . Prevalence of cardiac autonomic neuropathy in patients with type 2 diabetes mellitus. J Fam Med Prim Care 2023; 12(9): 2070–2074. DOI: 10.4103/jfmpc.jfmpc_609_23.

12.

Mirg

, Pednekar

, Kungar

, . A study of cardiac autonomic neuropathy in patients with diabetes mellitus. J Assoc Physicians India 2024; 72(9): 32–36. DOI: 10.59556/japi.72.0667.

13.

Ratzmann

, Raschke

, Gander

, . Prevalence of peripheral and autonomic neuropathy in newly diagnosed type II (noninsulin-dependent) diabetes. J Diabet Complications 1991; 5(1): 1–5. DOI: 10.1016/0891-6632(91)90002-7.

14.

Gateva

and Kamenov

Cardiac autonomic neuropathy in patients with newly diagnosed carbohydrate disturbances. Horm Metab Res 2022; 54(5): 308–315. DOI: 10.1055/a-1775-8251.

15.

American Diabetes Association. 2. Classification and diagnosis of diabetes: Standards of medical care in diabetes—2018. Diabetes Care 2018; 41(Suppl 1): S13–S27. DOI: 10.2337/dc18-S002.

16.

Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability: Standards of measurement, physiological interpretation, and clinical use. Eur Heart J 1996; 17(3): 354–381.

17.

Santos-de-Araújo

, Shida-Marinho

and Pontes-Silva

Heart rate variability (HRV): Checklist for observational and experimental studies. Autoimmun Rev 2022; 21(11): 103190. DOI: 10.1016/j.autrev.2022.103190.

18.

Ravichandran

, Srivastav

, Kamble

, . VEGF-A and cardiac autonomic function in newly diagnosed type 2 diabetes mellitus: A cross-sectional study at a tertiary care center. J Fam Med Prim Care 2019; 8(10): 3185–3190. DOI: 10.4103/jfmpc.jfmpc_537_19.

19.

Sampath

, Bade

, Goyal

, . Respiratory dysfunction in Parkinson’s disease: Relation with dysautonomia. Ann Indian Acad Neurol 2022; 25(4): 683–687. DOI: 10.4103/aian.aian_940_21.

20.

Tewari

, Gadia

, Kumar

, . Sympathetic-parasympathetic activity and reactivity in central serous chorioretinopathy: A case-control study. Invest Ophthalmol Vis Sci 2006; 47(8): 3474–3478. DOI: 10.1167/iovs.05-1246.

21.

Nagar

and Dhingra

Valsalva retinopathy. Postgrad Med J 2017; 93(1097): 174. DOI: 10.1136/postgradmedj-2016-134176.

22.

Rohowetz

, Patel

, Sridhar

, . Valsalva retinopathy: Clinical features and treatment outcomes. Retina (Phila Pa) 2023; 43(8): 1317–1320. DOI: 10.1097/IAE.0000000000003772.

23.

Srivastav

, Jamil

and Zeltser

Valsalva maneuver. In: StatPearls . Treasure Island, FL: StatPearls Publishing, 2025. http://www.ncbi.nlm.nih.gov/books/NBK537248/

24.

Verma

, Alam

, Ahmad

, . Effect of glycemic control and disease duration on cardiac autonomic function and oxidative stress in type 2 diabetes mellitus. J Diabetes Metab Disord 2018; 17(2): 149–158. DOI: 10.1007/s40200-018-0354-6.

25.

John

APP

, Udupa

, Avangapur

, . Cardiac autonomic dysfunctions in type 2 diabetes mellitus: An investigative study with heart rate variability measures. Am J Cardiovasc Dis 2022; 12(4): 224–232.

26.

Yan

, Tsai

, Ko

, . Quantitative examination of early diabetes by light-emitting diodes light-induced pupillary light reflex. Rev Sci Instrum 2021; 92(1): 014101. DOI: 10.1063/5.0030042.

27.

Pfeifer

, Weinberg

, Cook

, . Autonomic neural dysfunction in recently diagnosed diabetic subjects. Diabetes Care 1984; 7(5): 447–453. DOI: 10.2337/diacare.7.5.447.

28.

American Diabetes Association Professional Practice Committee. 12. Retinopathy, neuropathy, and foot care: Standards of care in diabetes—2025. Diabetes Care 2025; 48(1 Suppl 1): S252–S265. DOI: 10.2337/dc25-S012.

29.

Agashe

and Petak

Cardiac autonomic neuropathy in diabetes mellitus. Methodist DeBakey Cardiovasc J 2018; 14(4): 251–256. DOI: 10.14797/mdcj-14-4-251.

30.

Pop-Busui

Cardiac autonomic neuropathy in diabetes. Diabetes Care 2010; 33(2): 434–441. DOI: 10.2337/dc09-1294.

31.

Anderson

, Weeldreyer

, Leicht

, . Exercise intolerance in type 2 diabetes: A systematic review and meta-analysis. J Am Heart Assoc 2025; 14(5): e035721. DOI: 10.1161/JAHA.124.035721.

32.

Jyotsna

, Sahoo

, Sreenivas

, . Prevalence and pattern of cardiac autonomic dysfunction in newly detected type 2 diabetes mellitus. Diabetes Res Clin Pract 2009; 83(1): 83–88. DOI: 10.1016/j.diabres.2008.09.054.

33.

Jyotsna

, Singh

, Deepak

, . Progression of cardiac autonomic dysfunction in newly detected type 2 diabetes. Diabetes Res Clin Pract 2010; 90(1): e5–6. DOI: 10.1016/j.diabres.2010.04.009.

34.

Khandelwal

, Jaryal

and Deepak

KK.

Pattern and prevalence of cardiovascular autonomic neuropathy in diabetics visiting a tertiary care referral center in India. Indian J Physiol Pharmacol 2011; 55(2): 119–127.

35.

Panigrahi

, Srivastava

, Garg

, . Examination of the autonomic nervous system at the bedside. Acta Neurol Belg 2025; 125(3): 625–634. DOI: 10.1007/s13760-024-02654-2.

36.

Bellavere

, Bosello

, Fedele

, . Diagnosis and management of diabetic autonomic neuropathy. Br Med J (Clin Res Ed) 1983; 287(6384): 61. DOI: 10.1136/bmj.287.6384.61-a.

37.

Siddiqui

, Dev

and Khandelwal

Cardiac autonomic neuropathy in newly diagnosed patients with type 2 diabetes mellitus. Cureus 2023; 15(10): e47366. DOI: 10.7759/cureus.47366.

38.

Dhumad

, Hamdan

, Khudhair

, . Correlation of staging and risk factors with cardiovascular autonomic neuropathy in patients with type II diabetes mellitus. Sci Rep 2021; 11(1): 3576. DOI: 10.1038/s41598-021-80962-w.

39.

Gray

, Picone

, Sloan

, . Relation between BMI and diabetes mellitus and its complications among US older adults. South Med J 2015; 108(1): 29–36. DOI: 10.14423/SMJ.0000000000000214.

40.

Huang

, Wu

, Ni

, . Global, regional, and national burden of type 2 diabetes mellitus caused by high BMI from 1990 to 2021, and forecasts to 2045: Analysis from the global burden of disease study 2021. Front Public Health 2025; 13: 1515797. DOI: 10.3389/fpubh.2025.1515797.

41.

Mehran

, Mousapour

, Khalili

, . BMI variability and incident diabetes mellitus: Tehran Lipid and Glucose Study (TLGS). Sci Rep 2022; 12(1): 18370. DOI: 10.1038/s41598-022-22817-6.

42.

Song

, Chen

, He

, . Associations of BMI with all-cause mortality in normoglycemia, impaired fasting glucose and type 2 diabetes mellitus among an elderly Chinese population: A cohort study. BMC Geriatr 2022; 22(1): 690. DOI: 10.1186/s12877-022-03382-z.

43.

Strings

, Wells

, Bell

, . The association of body mass index and odds of type 2 diabetes mellitus varies by race/ethnicity. Public Health 2023; 215: 27–30. DOI: 10.1016/j.puhe.2022.11.017.

44.

Devi

, Subramaniam

, Adole

, . Cardiovascular autonomic neuropathy in patients with type 2 diabetes mellitus and its association with serum omentin and leptin. J Fam Med Prim Care 2020; 9(6): 2926–2930. DOI: 10.4103/jfmpc.jfmpc_461_20.

45.

Akbar

, Bhandari

, Habib

, . Potential association of triglyceride glucose index with cardiac autonomic neuropathy in type 2 diabetes mellitus patients. J Korean Med Sci 2017; 32(7): 1131–1138. DOI: 10.3346/jkms.2017.32.7.1131.

46.

Syngle

, Verma

, Krishan

, . Minocycline improves peripheral and autonomic neuropathy in type 2 diabetes: MIND study. Neurol Sci 2014; 35(7): 1067–1073. DOI: 10.1007/s10072-014-1647-2.

47.

Didangelos

, Karlafti

, Kotzakioulafi

, . Vitamin B12 supplementation in diabetic neuropathy: A 1-year, randomized, double-blind, placebo-controlled trial. Nutrients 2021; 13(2): 395. DOI: 10.3390/nu13020395.

48.

Putz

, Tordai

, Hajdú

, . Vitamin D in the prevention and treatment of diabetic neuropathy. Clin Ther 2022; 44(5): 813–823. DOI: 10.1016/j.clinthera.2022.03.012.

49.

Dimitropoulos

, Tahrani

and Stevens

MJ.

Cardiac autonomic neuropathy in patients with diabetes mellitus. World J Diabetes 2014; 5(1): 17–39. DOI: 10.4239/wjd.v5.i1.17.

50.

Kanagala

and Harsoda

JM.

Sympathetic overactivity and parasympathetic impairment in type 2 diabetes: An analysis of cardiovascular autonomic functions. Cureus 2024; 16(5): e59561. DOI: 10.7759/cureus.59561.

51.

Billman

GE.

The LF/HF ratio does not accurately measure cardiac sympatho-vagal balance. Front Physiol 2013; 4: 26. DOI: 10.3389/fphys.2013.00026.

52.

Jia

and Sowers

JR.

Hypertension in diabetes: An update of basic mechanisms and clinical disease. Hypertens (Dallas Tex 1979) 2021; 78(5): 1197–1205. DOI: 10.1161/HYPERTENSIONAHA.121.17981.

53.

Lamacchia

and Sorrentino

MR.

Diabetes mellitus, arterial stiffness and cardiovascular disease: Clinical implications and the influence of SGLT2i. Curr Vasc Pharmacol 2021; 19(2): 233–240. DOI: 10.2174/1570161118666200317150359.