Abstract
Aims/Introduction
We investigated the association of polyneuropathy (PN) with all‐cause and cardiovascular (CV) mortality and with cardiovascular disease (CVD) events stratified by diabetes status.
Materials and Methods
This prospective cohort study used the UK Biobank. Polyneuropathy was defined based on nurse‐led interviews or ICD codes for polyneuropathy. Cox proportional hazards models were used to investigate the association of polyneuropathy with clinical outcomes.
Results
A total of 459,127 participants were included in the analysis. Polyneuropathy was significantly associated with all‐cause and cardiovascular mortality, and with CVD events even after adjusting for CVD risk factors across all diabetes statuses. Metabolic parameters HbA1c, waist circumference, BMI and the inflammatory parameter C‐reactive protein showed significant mediation effects for the association between polyneuropathy and CVD. Adherence to a favorable lifestyle was associated with a lower risk of all‐cause and cardiovascular mortality regardless of polyneuropathy status.
Conclusions
Polyneuropathy was associated with all‐cause and cardiovascular mortality, and with CVD events in subjects with diabetes or prediabetes, even those having normal glucose tolerance. This study suggests the importance of polyneuropathy as a risk factor for death and highlights the necessity of early diagnosis and lifestyle intervention for those with type 2 diabetes and polyneuropathy.
Keywords: Cardiovascular diseases, Diabetes mellitus, Polyneuropathies
Polyneuropathy was associated with all‐cause and cardiovascular mortality, and with CVD events in subjects with diabetes or prediabetes, even those having normal glucose tolerance.
INTRODUCTION
Polyneuropathy (PN) is a disorder of the peripheral nerves which presents with tingling, numbness, burning, and pain in distal parts of the arms and legs. While one of the most common cause of polyneuropathy is diabetes, accounting for 32–53% of cases 1 , polyneuropathy can affect subjects with normal glucose tolerance (NGT) or prediabetes as well 2 , 3 .
Most previous studies have focused on diabetic polyneuropathy, and have suggested high mortality and cardiovascular disease (CVD) incidence rates in patients with diabetic polyneuropathy 4 , 5 , 6 . There have been few studies focusing on the clinical outcomes of polyneuropathy in individuals without diabetes. One prospective cohort study including US adults showed that polyneuropathy was significantly associated with all‐cause mortality in subjects without diabetes 7 . However, whether polyneuropathy is significantly associated with cardiovascular (CV) mortality or various subtypes of CVD in those without diabetes or with prediabetes has not been investigated. In addition, no study has investigated potential mediators between polyneuropathy and CVD, nor the potential effect of a lifestyle on mortality in polyneuropathy. Considering the association of adherence to a favorable lifestyle (abstinence from smoking, ideal BMI, physical activity at goal, and healthy eating habits) for the sake of cardiovascular health with lower risk of all‐cause and cardiovascular mortality 8 , 9 , 10 , it is intriguing to consider the possibility of a similar effect of favorable lifestyle on mortality in polyneuropathy.
Therefore, we aimed to examine the association of polyneuropathy with all‐cause and cardiovascular mortality and with CVD events including both atherosclerotic CVD (ASCVD) and non‐ASCVD events, all stratified by diabetes status (normal glucose tolerance, prediabetes, and type 2 diabetes), using the UK Biobank. In addition, we analyzed the mediation effect of metabolic and inflammatory parameters in the association between polyneuropathy and CVD. Lastly, we examined the association of adherence to a favorable lifestyle with all‐cause and cardiovascular mortality stratified by polyneuropathy status.
MATERIALS AND METHODS
Study population
This prospective cohort study is the UK Biobank comprising approximately 500,000 middle‐aged UK residents aged 40–69 years from 2006 to 2010 and followed up with participants for subsequent health‐related outcomes 11 . Baseline biological samples were collected and touch screen questionnaires answered as described previously 12 . The UK Biobank received ethics approval from the Northwest Multi‐Centre Research Ethics Committee (reference number 11/NW/03820). All participants provided written informed consent before enrollment in the study, which was conducted in accordance with the principles of the Declaration of Helsinki. In our analysis, we excluded those with missing data (n = 43,276) and those with secondary polyneuropathy based on ICD codes (n = 102): G62.0 (drug‐induced polyneuropathy), G62.1 (alcoholic polyneuropathy), G62.2 (polyneuropathy due to other toxic agents), G62.8 (other specified polyneuropathies), G63.0 (polyneuropathy in infectious and parasitic diseases classified elsewhere), G63.1 (polyneuropathy in neoplastic disease), G63.3 (polyneuropathy in other endocrine and metabolic diseases), G63.4 (polyneuropathy in nutritional deficiency), G63.5 (polyneuropathy in systemic connective tissue disorders), G63.6 (polyneuropathy in other musculoskeletal disorders), and G63.8 (polyneuropathy in other diseases classified elsewhere). Ultimately, in total 459,127 participants were included in the analysis.
Ascertainment of mortality and CVD events
The main outcomes included all‐cause mortality, cardiovascular mortality, events of coronary artery disease (CAD), ischemic stroke, peripheral artery disease (PAD), heart failure (HF), and atrial fibrillation/flutter. Cause and date of death were extracted from death certificates held by the National Health Service (NHS) Information Centre (England and Wales) and by the NHS Central Register Scotland (Scotland). Date and cause of hospital admission were identified through record linkage to Health Episode Statistics and the Scottish Morbidity Records. Causes of death were classified according to the ICD‐10. Cardiovascular mortality was defined as any death related to a CVD event. CVD events were defined as a composite of coronary artery disease, ischemic stroke, peripheral artery disease, heart failure, and atrial fibrillation/flutter, and were ascertained using (1) the self‐report at baseline enrollment, (2) ICD‐10 codes in hospital admission records, and (3) the first occurrence of a CVD event in hospital in‐patient records, death records, cancer registry, and primary care records. Coronary artery disease, ischemic stroke, and peripheral artery disease were classified as ASCVD. Heart failure and atrial fibrillation/flutter were classified as non‐ASCVD. Detailed definitions of coronary artery disease, ischemic stroke, peripheral artery disease, heart failure, and atrial fibrillation/flutter are provided in Table S1. Follow‐up ended on date of death, date of end of follow‐up for the assessment center attended, or the first date of CVD‐related hospitalization.
Ascertainment of variables
Anthropometric indices such as height, body weight, BMI, waist circumference (WC), systolic blood pressure (SBP), and diastolic blood pressure (DBP) were measured according to a standard protocol. Ethnicity was categorized as White, Asian, Black, mixed, and other. Information concerning smoking status, alcohol frequency, physical activity, eating habits, medical history, and medication was collected via touch screen questionnaire or in‐person interview during the initial assessment visit. Lifestyle behavior factors included alcohol frequency, smoking status, eating habits, and physical activity. Smoking status was classified as non‐smoker or current smoker. Alcohol frequency was classified as below once or twice a week, three or four times a week, or daily. Participants were considered as undertaking regular physical activity if they reported more than 5 days/week of moderate activity or more than 3 days/week of vigorous activity. Regarding eating habits, we referred to recommendations on dietary priorities for cardiovascular health 13 . Participants were considered as having healthy eating habits if they adhered to at least half of the dietary recommendations for cardiovascular health. Lifestyle behaviors were grouped into unfavorable (0 or 1 healthy lifestyle factor), intermediate (2 healthy lifestyle factors), and favorable (≥3 healthy lifestyle factors).
The procedures for the collection, processing, and archiving of biological samples have been described and validated previously 14 . HbA1c was measured by a Variant II Turbo Analyzer (Bio‐Rad). Glucose, lipid profiles, and inflammatory markers were measured by an AU5800 clinical chemistry analyzer (Beckman Coulter). The eGFR was calculated by the Chronic Kidney Disease Epidemiology Collaboration 15 . Details about these measurements are available online (https://biobank.ndph.ox.ac.uk/showcase/showcase/docs/biomarker_issues.pdf and https://biobank.ndph.ox.ac.uk/showcase/showcase/docs/serum_hb1ac.pdf).
At baseline, polyneuropathy was defined based on the nurse‐led interview or ICD codes for polyneuropathy: G62.9 (polyneuropathy, unspecified), G63.2 (diabetic polyneuropathy), E114 (non‐insulin‐dependent diabetes mellitus with neurological complications), and E144 (unspecified diabetes mellitus with neurological complications). The UK Biobank questionnaire is administered in two sequential parts during the assessment visit: the touch screen questionnaire and the nurse‐led interview. Following the questionnaire, the nurse‐led interview was performed to address medical history. All responses from the touchscreen questionnaire were flagged to the nurse. The nurses also asked individuals directly about their medical history. Detailed information about the nurse‐led interview is available from pages 17–20 of the protocol 16 and page 12 of the interview manual 17 . Type 2 diabetes was defined based on information from self‐reports, diagnostic codes, the first occurrence before enrollment, use of anti‐hyperglycemic medication, and/or HbA1c ≥ 6.5% (48 mmol/mol). Prediabetes was defined when not meeting the criteria of type 2 diabetes as above and having a HbA1c of 5.7% (39 mmol/mol) to 6.4% (48 mmol/mol). A normal glucose tolerance was defined when not meeting criteria of type 2 diabetes as above and having a HbA1c < 5.7% (39 mmol/mol) 18 , 19 . Prevalent comorbidities at baseline were defined based on self‐report, hospitalization records, first occurrence before enrollment, or medication. Detailed definitions of diabetes and comorbidities are described in Table S1.
Statistical analysis
Data were expressed as the mean ± standard deviation or number (%). Categorical variables were compared using chi‐square test, and continuous variables were compared using anova. Associations of polyneuropathy with all‐cause and cardiovascular mortality and with CVD events (coronary artery disease, ischemic stroke, peripheral artery disease, heart failure, and atrial fibrillation/flutter) stratified by diabetes status were analyzed using Cox proportional hazards models. For all‐cause mortality and cardiovascular mortality analyses, three incremental models were used. Model 1 was adjusted for age, sex, and ethnicity. Model 2 was adjusted for model 1 plus BMI, waist circumference, smoking status, physical activity, eating habits, and alcohol intake. Model 3 was adjusted for model 2 plus SBP, DBP, triglyceride, LDL‐cholesterol, HDL‐cholesterol, and eGFR. Model 4 was adjusted for model 3 plus cancer history. To analyze the associations between polyneuropathy and CVD events (coronary artery disease, ischemic stroke, peripheral artery disease, heart failure, atrial fibrillation/flutter), covariates in model 4 were adjusted. For survival analyses, participants were censored at the date of death, the date of follow‐up end (January 31, 2018 for England and Wales; November 30, 2016 for Scotland), or the date of follow‐up loss. To assess the proportional hazard assumption, log minus log plots and Schoenfeld residuals were used. For subtypes of CVD analyses, participants with each subtype of CVD history were excluded.
For causal mediation analysis, the R package mediation was used to examine the mediation effect for contributions of metabolic and inflammatory factors such as HbA1c, BMI, waist circumference, C‐reactive protein (CRP), blood pressure, and eGFR. Each mediation analysis was performed with 1,000 simulations using a quasi‐Bayesian method to estimate the variance. All analyses were performed using R (3.7.0). A two‐sided P < 0.05 was defined statistically significant without correcting for multiple comparisons.
RESULTS
Baseline characteristics
Of the 502,505 participants recruited to the UK Biobank, 459,127 were included in this study. Baseline characteristics of the participants by diabetes and polyneuropathy status are summarized in Table 1. Overall prevalence rates of prediabetes and type 2 diabetes were 14.1% and 4.7%, respectively, while the overall prevalence of polyneuropathy was 0.9% (0.7% in those with normal glucose tolerance, 1.0% in those with prediabetes, and 3.0% in those with type 2 diabetes). Those with polyneuropathy were older, were more frequently male, had regular physical activity less frequently, had a higher level of HbA1c, and had lower levels of LDL‐cholesterol and eGFR. Those with polyneuropathy also had higher prevalences of hypertension, dyslipidemia, and cancer.
Table 1.
Baseline characteristics according to the status of diabetes and polyneuropathy
| Total | NGT | P value | Prediabetes | P value | Type 2 diabetes | P value | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| PN (−) | PN (+) | PN (−) | PN (+) | PN (−) | PN (+) | |||||
| (n = 459,127) | (n = 370,034) | (n = 2,638) | (n = 64,316) | (n = 618) | (n = 20,886) | (n = 635) | ||||
| Age (years) | 57.0 ± 8.1 | 56.3 ± 8.2 | 59.6 ± 7.3 | <0.001 | 60.1 ± 7.0 | 61.6 ± 6.3 | <0.001 | 60.6 ± 6.8 | 61.2 ± 6.7 | 0.045 |
| Sex, n (%) | ||||||||||
| Women | 249,264 (54.3) | 204,842 (55.4) | 1,164 (44.1) | <0.001 | 35,213 (54.7) | 258 (41.7) | <0.001 | 7,612 (36.4) | 175 (27.6) | <0.001 |
| Men | 209,863 (45.7) | 165,192 (44.6) | 1,474 (55.9) | 29,103 (45.3) | 360 (58.3) | 13,274 (63.6) | 460 (72.4) | |||
| Ethnicity, n (%) | ||||||||||
| White | 434,618 (95.1) | 354,734 (96.3) | 2,548 (97.1) | 0.184 | 58,030 (90.8) | 571 (93.3) | 0.058 | 18,165 (87.4) | 570 (90.3) | 0.007 |
| Asian | 9,907 (2.2) | 5,785 (1.6) | 30 (1.1) | 2,625 (4.1) | 11 (1.8) | 1,420 (6.8) | 36 (5.7) | |||
| Black | 6,118 (1.3) | 3,433 (0.9) | 19 (0.7) | 1,924 (3.0) | 20 (3.3) | 712 (3.4) | 9 (1.4) | |||
| Mixed | 2,591 (0.6) | 1,974 (0.5) | 13 (0.5) | 466 (0.7) | 4 (0.7) | 126 (0.6) | 8 (1.3) | |||
| Others | 3,869 (0.8) | 2,619 (0.7) | 13 (0.5) | 862 (1.3) | 6 (1.0) | 361 (1.7) | 8 (1.3) | |||
| BMI (kg/m2) | 27.4 ± 4.8 | 26.9 ± 4.4 | 27.1 ± 4.5 | 0.003 | 28.9 ± 5.2 | 29.3 ± 5.6 | 0.074 | 31.6 ± 5.8 | 33.4 ± 6.1 | <0.001 |
| Waist circumference (cm) | 90.3 ± 13.5 | 88.6 ± 12.7 | 91.5 ± 13.0 | <0.001 | 94.3 ± 13.7 | 98.2 ± 14.9 | <0.001 | 103.5 ± 13.8 | 110.0 ± 15.0 | <0.001 |
| Current smoker, n (%) | 48,091 (10.5) | 35,436 (9.6) | 271 (10.3) | 0.234 | 9,938 (15.5) | 98 (15.9) | 0.838 | 2,273 (10.9) | 75 (11.8) | 0.500 |
| Alcohol intake, n (%) | ||||||||||
| Below once or twice a week | 257,750 (56.2) | 200,252 (54.2) | 1,394 (53.0) | <0.001 | 40,686 (63.5) | 387 (62.8) | 0.471 | 14,565 (69.8) | 468 (73.4) | 0.009 |
| Three or four times a week | 106,622 (23.3) | 90,346 (24.5) | 587 (22.3) | 12,254 (19.1) | 111 (18.0) | 3,253 (15.6) | 71 (11.2) | |||
| Daily | 93,855 (20.5) | 78,796 (21.3) | 651 (24.7) | 11,152 (17.4) | 118 (19.2) | 3,040 (14.6) | 98 (15.4) | |||
| Regular physical activity, n (%) | 316,828 (69.0) | 260,521 (70.4) | 1,745 (66.1) | <0.001 | 41,730 (64.9) | 383 (62.0) | 0.143 | 12,144 (58.1) | 305 (48.0) | <0.001 |
| Lifestyle behavior, n (%) | ||||||||||
| Favorable | 237,381 (54.1) | 203,006 (57.1) | 1,351 (53.9) | <0.001 | 25,842 (42.9) | 240 (41.3) | 0.158 | 6,784 (35.1) | 158 (27.0) | <0.001 |
| Intermediate | 148,209 (33.8) | 116,296 (32.7) | 844 (33.7) | 22,973 (38.1) | 212 (36.5) | 7,667 (39.7) | 217 (37.1) | |||
| Unfavorable | 53,000 (12.1) | 35,990 (10.1) | 311 (12.4) | 11,492 (19.1) | 129 (22.2) | 4,868 (25.2) | 210 (35.9) | |||
| SBP (mmHg) | 139.8 ± 19.7 | 138.8 ± 19.6 | 140.7 ± 19.3 | <0.001 | 143.4 ± 19.8 | 143.1 ± 19.0 | 0.679 | 143.8 ± 18.4 | 143.3 ± 19.6 | 0.544 |
| DBP (mmHg) | 82.2 ± 10.7 | 82.1 ± 10.7 | 82.6 ± 10.9 | 0.019 | 83.3 ± 10.8 | 83.2 ± 10.8 | 0.811 | 81.9 ± 10.2 | 80.2 ± 10.4 | <0.001 |
| Total cholesterol (mmol/L) | 5.7 ± 1.1 | 5.8 ± 1.1 | 5.6 ± 1.1 | <0.001 | 5.8 ± 1.2 | 5.6 ± 1.2 | <0.001 | 4.5 ± 1.0 | 4.3 ± 1.0 | <0.001 |
| Triglyceride (mmol/L) | 1.7 ± 1.0 | 1.7 ± 1.0 | 1.8 ± 1.1 | <0.001 | 2.0 ± 1.1 | 2.0 ± 1.2 | 0.326 | 2.2 ± 1.2 | 2.4 ± 1.4 | <0.001 |
| LDL cholesterol (mmol/L) | 3.6 ± 0.9 | 3.6 ± 0.8 | 3.5 ± 0.8 | <0.001 | 3.6 ± 0.9 | 3.5 ± 0.9 | <0.001 | 2.7 ± 0.7 | 2.6 ± 0.7 | <0.001 |
| eGFR (mL/min per 1.73 m2) | 79.0 ± 14.5 | 79.5 ± 14.1 | 77.1 ± 15.5 | <0.001 | 76.3 ± 14.7 | 74.6 ± 16.0 | 0.012 | 78.4 ± 17.1 | 73.5 ± 22.2 | <0.001 |
| HbA1c (mmol/mol) | 36.13 ± 6.78 | 33.98 ± 2.96 | 34.14 ± 3.05 | 0.006 | 40.91 ± 1.90 | 41.16 ± 2.13 | 0.004 | 51.98 ± 13.17 | 57.78 ± 16.96 | <0.001 |
| HbA1c (%) | 5.46 ± 0.62 | 5.26 ± 0.27 | 5.27 ± 0.28 | 0.006 | 5.89 ± 0.17 | 5.92 ± 0.19 | 0.004 | 6.91 ± 1.21 | 7.44 ± 1.55 | <0.001 |
| CRP (nmol/L) | 2.6 ± 4.4 | 2.3 ± 4.0 | 2.8 ± 4.7 | <0.001 | 3.6 ± 5.3 | 3.9 ± 5.2 | 0.121 | 3.4 ± 5.0 | 4.4 ± 5.8 | <0.001 |
| Hypertension, n (%) | 134,018 (29.2) | 91,749 (24.8) | 906 (34.3) | <0.001 | 25,470 (39.6) | 299 (48.4) | <0.001 | 15,061 (72.1) | 535 (83.9) | <0.001 |
| Dyslipidemia, n (%) | 83,091 (18.1) | 49,574 (13.4) | 545 (20.7) | <0.001 | 18,049 (28.1) | 207 (33.5) | 0.003 | 14,223 (68.1) | 493 (77.6) | <0.001 |
| Cancer, n (%) | 52,938 (11.5) | 41,230 (11.2) | 440 (16.7) | <0.001 | 8,451 (13.2) | 120 (19.4) | <0.001 | 2,614 (12.5) | 84 (13.1) | 0.722 |
Data are shown as mean ± standard deviation, or percentages. CRP, C‐reactive protein; DBP, diastolic blood pressure; NGT, normal glucose tolerance; PN, polyneuropathy; SBP, systolic blood pressure.
Associations of PN with risk of all‐cause and CV mortality stratified by diabetes status
Over a median follow‐up of 8.9 years, 18,106 (3.9%) participants died, including 3,679 (0.8%) from cardiovascular causes. The respective all‐cause and cardiovascular mortality rates were 3.2% and 0.6% in those with normal glucose tolerance but no polyneuropathy, 10.6% and 1.6% in those with normal glucose tolerance and polyneuropathy, 5.8 and 1.3% in those with prediabetes but no polyneuropathy, 14.2% and 2.4% in those with prediabetes and polyneuropathy, 9.3% and 2.9% in those with type 2 diabetes but no polyneuropathy, and 20.8% and 6.8% in those with type 2 diabetes and polyneuropathy (Table 2).
Table 2.
Event numbers, incidence rates, and hazard ratios (95% CI) of all‐cause and cardiovascular mortality according to the status of diabetes and polyneuropathy
| Events, n/N | Incidence rate per 1,000 person‐years (95% CI) | Absolute risk (%) | Crude | Model 1 | Model 2 | Model 3 | Model 4 | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | ||||
| All‐cause mortality | |||||||||||||
| NGT/PN (−) | 11,905/37,0034 | 3.65 (3.59–3.72) | 3.2 | 1 (Reference) | 1 (Reference) | 1 (Reference) | 1 (Reference) | 1 (Reference) | |||||
| NGT/PN (+) | 280/2,638 | 12.37 (10.96–13.90) | 10.6 | 3.45 (3.07–3.87) | <0.001 | 2.43 (2.11–2.81) | <0.001 | 2.52 (2.23–2.85) | <0.001 | 2.48 (2.19–2.82) | <0.001 | 2.36 (2.08–2.69) | <0.001 |
| Prediabetes/PN (−) | 3,762/64,316 | 6.74 (6.52–6.96) | 5.8 | 1.85 (1.79–1.92) | <0.001 | 1.47 (1.41–1.53) | <0.001 | 1.30 (1.25–1.35) | <0.001 | 1.28 (1.23–1.33) | <0.001 | 1.28 (1.22–1.33) | <0.001 |
| Prediabetes/PN (+) | 88/618 | 16.74 (13.42–20.62) | 14.2 | 4.67 (3.80–5.74) | <0.001 | 2.84 (2.19–3.69) | <0.001 | 2.71 (2.17–3.39) | <0.001 | 2.71 (2.15–3.41) | <0.001 | 2.62 (2.08–3.30) | <0.001 |
| Type 2 diabetes/PN (−) | 1,939/20,886 | 10.87 (10.39–11.36) | 9.3 | 3.00 (2.86–3.15) | <0.001 | 2.21 (2.09–2.34) | <0.001 | 1.84 (1.74–1.94) | <0.001 | 1.62 (1.53–1.72) | <0.001 | 1.62 (1.53–1.72) | <0.001 |
| Type 2 diabetes/PN (+) | 132/635 | 24.92 (20.85–29.55) | 20.8 | 6.90 (5.82–8.19) | <0.001 | 4.92 (4.08–5.94) | <0.001 | 3.76 (3.11–4.54) | <0.001 | 3.39 (2.79–4.13) | <0.001 | 3.42 (2.81–4.16) | <0.001 |
| Cardiovascular mortality | |||||||||||||
| NGT/PN (−) | 2,107/370,034 | 0.64 (0.61–0.67) | 0.6 | 1 (Reference) | 1 (Reference) | 1 (Reference) | 1 (Reference) | 1 (Reference) | |||||
| NGT/PN (+) | 41/2,638 | 1.75 (1.26–2.38) | 1.6 | 2.95 (2.20–3.96) | <0.001 | 2.09 (1.56–2.82) | <0.001 | 1.88 (1.35–2.60) | <0.001 | 1.93 (1.38–2.68) | <0.001 | 1.93 (1.38–2.68) | <0.001 |
| Prediabetes/PN (−) | 862/64,316 | 1.51 (1.41–1.62) | 1.3 | 2.38 (2.19–2.57) | <0.001 | 1.86 (1.71–2.01) | <0.001 | 1.50 (1.37–1.64) | <0.001 | 1.46 (1.34–1.61) | <0.001 | 1.46 (1.34–1.61) | <0.001 |
| Prediabetes/PN (+) | 15/618 | 2.74 (1.53–4.52) | 2.4 | 4.17 (2.51–6.92) | <0.001 | 2.59 (1.56–4.30) | <0.001 | 1.96 (1.11–3.47) | <0.001 | 2.09 (1.18–3.68) | 0.011 | 2.08 (1.18–3.68) | 0.011 |
| Type 2 diabetes/PN (−) | 611/20,886 | 3.33 (3.07–3.61) | 2.9 | 5.23 (4.78–5.72) | <0.001 | 3.27 (2.98–3.59) | <0.001 | 2.51 (2.26–2.79) | <0.001 | 2.23 (1.99–2.51) | <0.001 | 2.23 (1.99–2.51) | <0.001 |
| Type 2 diabetes/PN (+) | 43/635 | 7.76 (5.62–10.46) | 6.8 | 12.13 (8.97–16.41) | <0.001 | 6.70 (4.93–9.10) | <0.001 | 4.29 (3.02–6.08) | <0.001 | 3.77 (2.62–5.43) | <0.001 | 3.77 (2.62–5.43) | <0.001 |
Model 1: adjusted for age, sex, and ethnicity. Model 2: model 1 plus adjustment for BMI, smoking status, physical activity, eating habit, and alcohol intake. Model 3: model 2 plus adjustment for systolic blood pressure, diastolic blood pressure, LDL cholesterol, and eGFR. Model 4: model 3 plus adjustment for cancer history. CI, confidence interval; HR, hazard ratio; NGT, normal glucose tolerance; PN, polyneuropathy.
In crude and adjustment models, the HRs for all‐cause and cardiovascular mortality increased incrementally in those with polyneuropathy based on diabetes status relative to those without polyneuropathy. Compared with those having normal glucose tolerance but no polyneuropathy, the HRs (95% CI) for all‐cause mortality were 2.36 (2.08–2.69) in those with normal glucose tolerance and polyneuropathy, 2.62 (2.08–3.30) in those with prediabetes and polyneuropathy, and 3.42 (2.81–4.16) in those with type 2 diabetes and polyneuropathy. In addition, compared with those having normal glucose tolerance but no polyneuropathy, the HRs (95% CI) for cardiovascular mortality were 1.93 (1.38–2.68) in those with normal glucose tolerance and polyneuropathy, 2.08 (1.18–3.68) in those with prediabetes and polyneuropathy, and 3.77 (2.62–5.43) in those with type 2 diabetes and polyneuropathy (Table 2).
Among the Asian subgroup, the HR for all‐cause mortality increased incrementally with more deleterious diabetes status in those with polyneuropathy relative to those without polyneuropathy. Additionally, the HR for cardiovascular mortality was higher in those with type 2 diabetes and polyneuropathy compared with those with type 2 diabetes but no polyneuropathy (Tables S2 and S3).
Associations of PN with risk of CVD events stratified by diabetes status
Regarding CVD events, in those with polyneuropathy, the HRs for events of coronary artery disease, peripheral artery disease, heart failure, and atrial fibrillation/flutter increased incrementally with more severe diabetes status relative to those without polyneuropathy. In particular, those with type 2 diabetes and polyneuropathy had a higher risk of coronary artery disease (HR 3.64 [95% CI 2.92–4.53]), ischemic stroke (HR 3.99 [2.60–6.11]), peripheral artery disease (HR 10.55 [8.10–13.72]), heart failure (HR 4.85 [3.80–6.19]), and atrial fibrillation/flutter (HR 3.71 [2.94–4.68]), among which the HRs for peripheral artery disease were highest (Figure 1). Among the Asian subgroup, those with type 2 diabetes and polyneuropathy had higher risk of CVD events compared with those with type 2 diabetes but no polyneuropathy (Table S4).
Figure 1.
Hazard ratios (95% CI) of cardiovascular disease events according to diabetes and polyneuropathy status. Adjusted for age, sex, ethnicity, BMI, smoking status, physical activity, eating habit, alcohol intake, systolic blood pressure, diastolic blood pressure, LDL cholesterol, estimated glomerular filtration rate, and cancer history. CAD, coronary artery disease; CI, confidence interval; CVD, cardiovascular disease; HF, heart failure; HR, hazard ratio; NGT, normal glucose tolerance; PAD, peripheral artery disease; PN, polyneuropathy.
Mediation effects for overall CVD according to diabetes status
Mediation analysis revealed HbA1c, WC, BMI, triglyceride, CRP, and eGFR to feature statistically significant mediation effects for the association between polyneuropathy and CVD across all diabetes statuses. In particular, HbA1c showed principal mediation effects for polyneuropathy in type 2 diabetes, and even in prediabetes and normal glucose tolerance. Meanwhile, WC, BMI, and triglyceride showed higher mediation effects for polyneuropathy in prediabetes and normal glucose tolerance than in type 2 diabetes (Table 3).
Table 3.
Mediation effects for overall cardiovascular diseases according to diabetes status
| Mediated proportion (%) | P value | |
|---|---|---|
| NGT | ||
| HbA1c | 35.4 | <0.001 |
| Waist circumference | 20.4 | <0.001 |
| BMI | 16.9 | <0.001 |
| Triglyceride | 5.5 | <0.001 |
| CRP | 5.6 | <0.001 |
| SBP | 1.0 | <0.001 |
| eGFR | 0.1 | <0.001 |
| Prediabetes | ||
| HbA1c | 37.0 | <0.001 |
| Waist circumference | 22.1 | <0.001 |
| BMI | 19.9 | <0.001 |
| Triglyceride | 6.8 | <0.001 |
| CRP | 6.4 | <0.001 |
| SBP | 1.1 | <0.001 |
| eGFR | 0.1 | <0.001 |
| Type 2 diabetes | ||
| HbA1c | 46.2 | <0.001 |
| Waist circumference | 15.4 | <0.001 |
| BMI | 13.2 | <0.001 |
| CRP | 6.9 | <0.001 |
| Triglyceride | 3.5 | <0.001 |
| SBP | 0.6 | <0.001 |
| eGFR | 0.3 | <0.001 |
CRP, C‐reactive protein; NGT, normal glucose tolerance; SBP, systolic blood pressure.
Associations of lifestyle habits with all‐cause and CV mortality in subjects with type 2 diabetes according to PN status
In subjects with type 2 diabetes and polyneuropathy, adherence to favorable lifestyle habits was associated with a 54% reduced risk of all‐cause mortality (unfavorable vs favorable, HR 0.46 [95% CI 0.37–0.57]) and 60% reduced risk of cardiovascular mortality (unfavorable vs favorable, HR 0.40 [0.25–0.63]). In subjects with type 2 diabetes but no polyneuropathy, adherence to favorable lifestyle habits was associated with 56% reduced risk of all‐cause mortality (unfavorable vs favorable, HR 0.44 [95% CI 0.42–0.46]) and 67% reduced risk of cardiovascular mortality (unfavorable vs favorable, HR 0.33 [95% CI 0.30–0.36]) (Table 4).
Table 4.
Hazard ratios (95% CI) of lifestyle habits with all‐cause and cardiovascular mortality in subjects with type 2 diabetes according to polyneuropathy status
| Model 1 | ||
|---|---|---|
| HR (95% CI) | P value | |
| All‐cause mortality | ||
| PN (−) | ||
| Unfavorable LH | 1 (Reference) | |
| Intermediate LH | 0.65 (0.62–0.67) | <0.001 |
| Favorable LH | 0.44 (0.42–0.46) | <0.001 |
| PN (+) | ||
| Unfavorable LH | 1 (Reference) | |
| Intermediate LH | 0.61 (0.49–0.75) | <0.001 |
| Favorable LH | 0.46 (0.37–0.57) | <0.001 |
| Cardiovascular mortality | ||
| PN (−) | ||
| Unfavorable LH | 1 (Reference) | |
| Intermediate LH | 0.57 (0.52–0.62) | <0.001 |
| Favorable LH | 0.33 (0.30–0.36) | <0.001 |
| PN (+) | ||
| Unfavorable LH | 1 (Reference) | |
| Intermediate LH | 0.65 (0.42–1.01) | <0.001 |
| Favorable LH | 0.40 (0.25–0.63) | <0.001 |
Model 1: adjusted for age, sex, and ethnicity. CI, confidence interval; HR, hazard ratio; LH, lifestyle habits; PN, polyneuropathy.
DISCUSSION
This study demonstrated that polyneuropathy is associated with an increased risk of all‐cause and cardiovascular mortality and with CVD events across all diabetes statuses, even after adjustment for CVD risk factors and cancer history. Moreover, polyneuropathy is associated with increased risk of both ASCVD and non‐ASCVD, among which the risk for peripheral artery disease was the highest. Furthermore, we found that metabolic parameters HbA1c, WC, BMI, and triglyceride and the inflammatory parameter CRP mediate the association between polyneuropathy and CVD. We also found that adherence to a favorable lifestyle is associated with a reduced risk of all‐cause and cardiovascular mortality even in subjects with type 2 diabetes and polyneuropathy.
In this study, the risks for all‐cause and cardiovascular mortality were significantly higher for those with polyneuropathy than for those without polyneuropathy regardless of diabetes status, which is in line with a previous study 7 . In type 2 diabetes, polyneuropathy is associated with an increased risk of CVD events 6 , cardiac autonomic neuropathy (CAN) 20 , foot ulcers, and amputation 21 , all of which are associated with increased mortality 22 , 23 , 24 . These conditions may contribute to a higher risk of all‐cause and cardiovascular mortality in those with type 2 diabetes and polyneuropathy compared with those having type 2 diabetes but no polyneuropathy. Meanwhile, in those with normal glucose tolerance or prediabetes and also polyneuropathy, PN‐associated comorbidities and multiple functional impairments such as difficulty in walking and climbing stairs, and also fall tendency 25 , may contribute to a higher all‐cause mortality relative to those without polyneuropathy. The presence of polyneuropathy may reflect systemic subclinical microvascular disease 26 , which is associated with an increased risk of CVD events 27 . Subclinical microvascular disease may contribute to the higher cardiovascular mortality in those with normal glucose tolerance or prediabetes and also polyneuropathy.
Among patients with normal glucose tolerance or type 2 diabetes, polyneuropathy was associated with an increased risk of ASCVD, most of all for peripheral artery disease. Considering that the risk of peripheral artery disease is increased by the presence of polyneuropathy28, and also the role of ischemia in the pathophysiology of polyneuropathy 29 , polyneuropathy may reflect early ischemia which eventually leads to peripheral artery disease. Interestingly, we found polyneuropathy to be associated with an increased risk of non‐ASCVD events such as heart failure and atrial fibrillation/flutter. Polyneuropathy is also reportedly associated with cardiac autonomic neuropathy 30 , and cardiac autonomic neuropathy in turn with increased risk of incident heart failure 31 and atrial fibrillation 32 . Dysregulation of neurotransmitters such as neuropeptide Y may mediate these associations 33 , 34 . In addition, systemic inflammation has been recognized as a common risk factor of polyneuropathy, heart failure, and arrhythmia 35 , 36 , 37 . Taken together, in the presence of polyneuropathy, cardiac autonomic neuropathy and inflammation may contribute to an increased risk of non‐ASCVD across all diabetes statuses.
Importantly, polyneuropathy could be present in subjects without diabetes. Previous studies suggest that general and abdominal obesity are significantly associated with the risk of polyneuropathy 38 , 39 , and that inflammatory cytokines or fat inflammation may play roles in the pathogenesis of polyneuropathy 35 , 40 . In this study, we found that metabolic parameters HbA1c (hyperglycemia), WC (abdominal obesity), BMI (obesity), triglyceride, and the inflammatory parameter CRP mediate the association between polyneuropathy and CVD. In addition, we found metabolic parameters such as WC, BMI, and triglyceride to exhibit higher mediation effects for polyneuropathy in normal glucose tolerance and prediabetes than in type 2 diabetes. These results suggest the necessity of early diagnosis of polyneuropathy prior to the development of diabetes, and of managing metabolic derangements or cardiovascular risk factors in those with polyneuropathy.
We found adherence to a favorable lifestyle to be associated with a reduced risk of all‐cause and cardiovascular mortality in those with type 2 diabetes regardless of polyneuropathy status. While the benefits of healthy lifestyle habits in reducing cardiovascular events have been well established 41 , it is not evident whether adherence to a favorable lifestyle contributes to reduced mortality by altering the progression of polyneuropathy. This seems a plausible hypothesis considering a previous report that intensive lifestyle intervention such as increased physical exercise and weight loss significantly decreases questionnaire‐based diabetic polyneuropathy in overweight patients with type 2 diabetes 42 .
This study has several strengths. First, this is a prospective cohort study with a large number of participants, meaning we had sufficient power to undertake analyses stratified by diabetes status. Second, this study differentiated diabetes status into normal glucose tolerance, prediabetes, and type 2 diabetes, and CVD into coronary artery disease, ischemic stroke, peripheral artery disease, heart failure, and atrial fibrillation/flutter. Additionally, the strong association between polyneuropathy and peripheral artery disease in this study supported the consensus statement 28 . Third, this study demonstrated a mediation effect of metabolic parameters in the relationship of polyneuropathy and CVD. Finally, this study suggested a beneficial effect of adherence to a favorable lifestyle on all‐cause and cardiovascular mortality. All of these give novelty to this study in comparison with prior works 7 .
Nevertheless, this study also has some limitations. First, this study may underestimate the true prevalence of polyneuropathy because the UK Biobank cohort population is healthier than the average UK population due to selection bias. The lower prevalence of polyneuropathy in this study compared with previous studies 2 , 43 could also be a factor of differences in diagnosis methods; here, polyneuropathy was identified by the nurse‐led interview or ICD codes. Indeed, another study that used databases from Germany and the UK and identified polyneuropathy based on ICD codes or the original diagnosis showed a prevalence comparable to our results 44 . Second, information about clinical presentation and subtypes of polyneuropathy, along with the means of diagnosing polyneuropathy, was not available. In addition, polyneuropathy was not evaluated with the Michigan Neuropathy Screening Instrument or a neurophysiological study. These limitations make it difficult for us to elucidate the etiologies of polyneuropathy; for the purpose of this study, we considered diabetes mellitus, prediabetes, and obesity as possible etiologies 45 . Third, the duration of prediabetes and type 2 diabetes was not assessed. Lastly, because most of the cohort population were of White ethnicity, the results cannot be generalized to other ethnicities.
In conclusion, polyneuropathy is associated with all‐cause and cardiovascular mortality and with CVD events across all diabetes statuses. Metabolic parameters such as hyperglycemia, abdominal obesity, obesity, and triglyceride along with the inflammatory parameter CRP mediate the association between polyneuropathy and CVD. Adherence to a favorable lifestyle is associated with a reduced risk of all‐cause and cardiovascular mortality regardless of polyneuropathy status. Taken together, this study highlights the importance of polyneuropathy as a risk factor for death even in those without diabetes, and suggests a necessity for early diagnosis of polyneuropathy and lifestyle intervention in those with type 2 diabetes and polyneuropathy. Future studies are needed to verify the effects of lifestyle intervention on mortality or CVD events.
DISCLOSURES
The authors declare no conflict of interest.
Approval of the research protocol: This study using the UK Biobank resource was approved under application number 67855, and was conducted in accordance with the principles of the Declaration of Helsinki.
Informed consent: Informed consent was obtained from all participants.
Registry and the registration no. of the study/trial: N/A.
Animal studies: N/A.
Supporting information
Table S1 | Definitions of baseline comorbidities and cardiovascular diseases
Table S2 | Baseline characteristics of the Asian subgroup
Table S3 | Event numbers, incidence rates, and hazard ratios (95% CI) of all‐cause and cardiovascular mortality among the Asian subgroup
Table S4 | Event numbers, incidence rates, and hazard ratios (95% CI) of cardiovascular disease events among the Asian subgroup
ACKNOWLEDGMENTS
This study was supported by grants from the National Research Foundation of Korea (NRF) (Grant No. NRF‐2022R1F1A1072279). We appreciate the UK Biobank participants.
REFERENCES
- 1. Callaghan BC, Price RS, Feldman EL. Distal symmetric polyneuropathy: a review. JAMA 2015; 314: 2172–2181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Lee CC, Perkins BA, Kayaniyil S, et al. Peripheral neuropathy and nerve dysfunction in individuals at high risk for type 2 diabetes: the PROMISE cohort. Diabetes Care 2015; 38: 793–800. [DOI] [PubMed] [Google Scholar]
- 3. Ziegler D, Rathmann W, Dickhaus T, et al. Neuropathic pain in diabetes, prediabetes and normal glucose tolerance: the MONICA/KORA Augsburg surveys S2 and S3. Pain Med 2009; 10: 393–400. [DOI] [PubMed] [Google Scholar]
- 4. Coppini DV, Bowtell PA, Weng C, et al. Showing neuropathy is related to increased mortality in diabetic patients: a survival analysis using an accelerated failure time model. J Clin Epidemiol 2000; 53: 519–523. [DOI] [PubMed] [Google Scholar]
- 5. Forsblom CM, Sane T, Groop PH, et al. Risk factors for mortality in type II (non‐insulin‐dependent) diabetes: evidence of a role for neuropathy and a protective effect of HLA‐DR4. Diabetologia 1998; 41: 1253–1262. [DOI] [PubMed] [Google Scholar]
- 6. Brownrigg JR, de Lusignan S, McGovern A, et al. Peripheral neuropathy and the risk of cardiovascular events in type 2 diabetes mellitus. Heart 2014; 100: 1837–1843. [DOI] [PubMed] [Google Scholar]
- 7. Hicks CW, Wang D, Matsushita K, et al. Peripheral neuropathy and all‐cause and cardiovascular mortality in U.S. adults: a prospective cohort study. Ann Intern Med 2021; 174: 167–174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Gaye B, Canonico M, Perier MC, et al. Ideal cardiovascular health, mortality, and vascular events in elderly subjects: the Three‐City study. J Am Coll Cardiol 2017; 69: 3015–3026. [DOI] [PubMed] [Google Scholar]
- 9. Guo L, Zhang S. Association between ideal cardiovascular health metrics and risk of cardiovascular events or mortality: a meta‐analysis of prospective studies. Clin Cardiol 2017; 40: 1339–1346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Ford ES, Greenlund KJ, Hong Y. Ideal cardiovascular health and mortality from all causes and diseases of the circulatory system among adults in the United States. Circulation 2012; 125: 987–995. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Collins R. What makes UK Biobank special? Lancet 2012; 379: 1173–1174. [DOI] [PubMed] [Google Scholar]
- 12. Sudlow C, Gallacher J, Allen N, et al. UK biobank: An open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med 2015; 12: e1001779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Mozaffarian D. Dietary and policy priorities for cardiovascular disease, diabetes, and obesity: a comprehensive review. Circulation 2016; 133: 187–225. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Elliott P, Peakman TC, Biobank UK. The UK Biobank sample handling and storage protocol for the collection, processing and archiving of human blood and urine. Int J Epidemiol 2008; 37: 234–244. [DOI] [PubMed] [Google Scholar]
- 15. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009; 150: 604–612. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. UK Biobank: Protocol for a large-scale prospective epidemiological resource. Available from: https://www.ukbiobank.ac.uk/media/gnkeyh2q/study‐rationale.pdf Accessed June 20, 2023.
- 17. UK Biobank Verbal Interview stage. Available from: https://biobank.ndph.ox.ac.uk/ukb/ukb/docs/Interview.pdf Accessed June 20, 2023.
- 18. ElSayed NA, Aleppo G, Aroda VR, et al. 2. Classification and diagnosis of diabetes: Standards of Care in Diabetes‐2023. Diabetes Care 2023; 46: S19–S40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. American Diabetes Association . Diagnosis and classification of diabetes mellitus. Diabetes Care 2014; 37(Suppl 1): S81–S90. [DOI] [PubMed] [Google Scholar]
- 20. Motataianu A, Maier S, Bajko Z, et al. Cardiac autonomic neuropathy in type 1 and type 2 diabetes patients. BMC Neurol 2018; 18: 126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Boulton AJ. The diabetic foot: from art to science. The 18th Camillo Golgi lecture. Diabetologia 2004; 47: 1343–1353. [DOI] [PubMed] [Google Scholar]
- 22. Maser RE, Mitchell BD, Vinik AI, et al. The association between cardiovascular autonomic neuropathy and mortality in individuals with diabetes: A meta‐analysis. Diabetes Care 2003; 26: 1895–1901. [DOI] [PubMed] [Google Scholar]
- 23. Brownrigg JR, Davey J, Holt PJ, et al. The association of ulceration of the foot with cardiovascular and all‐cause mortality in patients with diabetes: a meta‐analysis. Diabetologia 2012; 55: 2906–2912. [DOI] [PubMed] [Google Scholar]
- 24. Tseng CH, Chong CK, Tseng CP, et al. Mortality, causes of death and associated risk factors in a cohort of diabetic patients after lower‐extremity amputation: a 6.5‐year follow‐up study in Taiwan. Atherosclerosis 2008; 197: 111–117. [DOI] [PubMed] [Google Scholar]
- 25. Hoffman EM, Staff NP, Robb JM, et al. Impairments and comorbidities of polyneuropathy revealed by population‐based analyses. Neurology 2015; 84: 1644–1651. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Hicks CW, Wang D, Daya NR, et al. Associations of cardiac, kidney, and diabetes biomarkers with peripheral neuropathy among older adults in the atherosclerosis risk in communities (ARIC) study. Clin Chem 2020; 66: 686–696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Kuller LH, Shemanski L, Psaty BM, et al. Subclinical disease as an independent risk factor for cardiovascular disease. Circulation 1995; 92: 720–726. [DOI] [PubMed] [Google Scholar]
- 28. American Diabetes Association . Peripheral arterial disease in people with diabetes. Diabetes Care 2003; 26: 3333–3341. [DOI] [PubMed] [Google Scholar]
- 29. Teunissen LL, Notermans NC, Wokke JH. Relationship between ischemia and neuropathy. Eur Neurol 2000; 44: 1–7. [DOI] [PubMed] [Google Scholar]
- 30. Voulgari C, Psallas M, Kokkinos A, et al. The association between cardiac autonomic neuropathy with metabolic and other factors in subjects with type 1 and type 2 diabetes. J Diabetes Complications 2011; 25: 159–167. [DOI] [PubMed] [Google Scholar]
- 31. Kaze AD, Yuyun MF, Erqou S, et al. Cardiac autonomic neuropathy and risk of incident heart failure among adults with type 2 diabetes. Eur J Heart Fail 2022; 24: 634–641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Agarwal SK, Norby FL, Whitsel EA, et al. Cardiac autonomic dysfunction and incidence of atrial fibrillation: results from 20 years follow‐up. J Am Coll Cardiol 2017; 69: 291–299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Ajijola OA, Chatterjee NA, Gonzales MJ, et al. Coronary sinus neuropeptide Y levels and adverse outcomes in patients with stable chronic heart failure. JAMA Cardiol 2020; 5: 318–325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Stavrakis S, Morris L, Takashima AD, et al. Circulating neuropeptide Y as a biomarker for neuromodulation in atrial fibrillation. JACC Clin Electrophysiol 2020; 6: 1575–1576. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Schlesinger S, Herder C, Kannenberg JM, et al. General and abdominal obesity and incident distal sensorimotor polyneuropathy: Insights into inflammatory biomarkers as potential mediators in the KORA F4/FF4 cohort. Diabetes Care 2019; 42: 240–247. [DOI] [PubMed] [Google Scholar]
- 36. Murphy SP, Kakkar R, McCarthy CP, et al. Inflammation in heart failure: JACC state‐of‐the‐art review. J Am Coll Cardiol 2020; 75: 1324–1340. [DOI] [PubMed] [Google Scholar]
- 37. Aviles RJ, Martin DO, Apperson‐Hansen C, et al. Inflammation as a risk factor for atrial fibrillation. Circulation 2003; 108: 3006–3010. [DOI] [PubMed] [Google Scholar]
- 38. Callaghan BC, Xia R, Reynolds E, et al. Association between metabolic syndrome components and polyneuropathy in an obese population. JAMA Neurol 2016; 73: 1468–1476. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Andersen ST, Witte DR, Dalsgaard EM, et al. Risk factors for incident diabetic polyneuropathy in a cohort with screen‐detected type 2 diabetes followed for 13 years: ADDITION‐Denmark. Diabetes Care 2018; 41: 1068–1075. [DOI] [PubMed] [Google Scholar]
- 40. Kim K, Oh TJ, Park YS, et al. Association between fat mass or fat fibrotic gene expression and polyneuropathy in subjects with obesity: a Korean metabolic bariatric surgery cohort. Front Endocrinol 2022; 13: 881093. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Barbaresko J, Rienks J, Nothlings U. Lifestyle indices and cardiovascular disease risk: A meta‐analysis. Am J Prev Med 2018; 55: 555–564. [DOI] [PubMed] [Google Scholar]
- 42. Look ARG. Effects of a long‐term lifestyle modification programme on peripheral neuropathy in overweight or obese adults with type 2 diabetes: the Look AHEAD study. Diabetologia 2017; 60: 980–988. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Katon JG, Reiber GE, Nelson KM. Peripheral neuropathy defined by monofilament insensitivity and diabetes status: NHANES 1999‐2004. Diabetes Care 2013; 36: 1604–1606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Kostev K, Jockwig A, Hallwachs A, et al. Prevalence and risk factors of neuropathy in newly diagnosed type 2 diabetes in primary care practices: a retrospective database analysis in Germany and U.K. Prim Care Diabetes 2014; 8: 250–255. [DOI] [PubMed] [Google Scholar]
- 45. Ziegler D, Papanas N, Vinik AI, et al. Epidemiology of polyneuropathy in diabetes and prediabetes. Handb Clin Neurol 2014; 126: 3–22. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1 | Definitions of baseline comorbidities and cardiovascular diseases
Table S2 | Baseline characteristics of the Asian subgroup
Table S3 | Event numbers, incidence rates, and hazard ratios (95% CI) of all‐cause and cardiovascular mortality among the Asian subgroup
Table S4 | Event numbers, incidence rates, and hazard ratios (95% CI) of cardiovascular disease events among the Asian subgroup