The Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) study demonstrated that intensive glucose control reduced the risk of developing diabetic peripheral neuropathy (DPN) and cardiovascular autonomic neuropathy (CAN). We evaluated multiple risk factors and phenotypes associated with DPN and CAN in this large, well-characterized cohort of participants with type 1 diabetes, followed for >23 years. DPN was defined by symptoms, signs, and nerve conduction study abnormalities in ≥2 nerves; CAN was assessed using standardized cardiovascular reflex tests. Generalized estimating equation models assessed the association of DPN and CAN with individual risk factors measured repeatedly. During DCCT/EDIC, 33% of participants developed DPN and 44% CAN. Higher mean HbA1c was the most significant risk factor for DPN, followed by older age, longer duration, greater height, macroalbuminuria, higher mean pulse rate, β-blocker use, and sustained albuminuria. The most significant risk factor for CAN was older age, followed by higher mean HbA1c, sustained albuminuria, longer duration of type 1 diabetes, higher mean pulse rate, higher mean systolic blood pressure, β-blocker use, estimated glomerular filtration rate <60 mL/min/1.73 m2, higher most recent pulse rate, and cigarette smoking. These findings identify risk factors and phenotypes of participants with diabetic neuropathy that can be used in the design of new interventional trials and for personalized approaches to neuropathy prevention.

Diabetic peripheral neuropathy (DPN) and cardiovascular autonomic neuropathy (CAN) are prevalent and costly complications of diabetes affecting up to 50% of patients with diabetes worldwide (1,2). DPN and CAN are also most challenging chronic complications given the independent risk for mortality, foot complications including lower limb amputations, severe pain, falls and fractures, arrhythmia, heart failure, and direct impact on daily function and quality of life (1,36). Although the increase in diabetes prevalence is largely driven by the increase in type 2 diabetes, type 1 diabetes prevalence and incidence has also increased worldwide, as evidenced in recent U.S. reports from the National Health and Nutrition Examination Survey (NHANES) and the SEARCH for Diabetes in Youth study (79). With the persistent increase in diabetes prevalence and incidence (10), neuropathy impacts medical care across a spectrum of providers, yet, to date, disease-modifying therapies are not available. Furthermore, the field has witnessed a decline in the interest of pharmaceutical companies to develop novel and effective therapies for neuropathy.

The Diabetes Control and Complications Trial (DCCT) study demonstrated in participants with type 1 diabetes that intensive glucose control, with lower hemoglobin A1c (HbA1c) levels, reduced the risk of DPN and CAN by 64% and 45%, respectively, over an average follow-up of 6.5 years (11). Moreover, the prevalence and incidence of DPN and CAN increased in both treatment groups through 14–17 years of observational follow-up in the Epidemiology of Diabetes Interventions and Complications (EDIC) study; however, these remained significantly lower in the intensive group compared with the conventional group (11,12).

The objective of the current study was to identify clinical and biochemical risk factors for DPN and CAN, independent of glycemia, in the large, well-characterized DCCT/EDIC cohort of participants with type 1 diabetes. Through its unparalleled wealth of clinical and biochemical evaluations obtained continuously for over 30 years, including sensitive and standardized DPN and CAN assessments, the DCCT/EDIC study provides a unique opportunity to comprehensively identify risk factors and phenotypes, associated both independently and jointly, with DPN and CAN.

Participants

The DCCT enrolled 1,441 participants with type 1 diabetes who were randomly assigned to either intensive or conventional diabetes therapy and followed for a mean of 6.5 years (13). Intensive therapy (n = 711) aimed to achieve near-normal glucose and HbA1c levels while conventional therapy (n = 730) was designed to prevent symptoms of hypo- or hyperglycemia with no glucose targets but an HbA1c target level <13.5% (13). Two parallel cohorts were recruited: the primary prevention cohort (n = 726, no retinopathy or kidney disease at baseline) and the secondary intervention cohort (n = 715, with mild-to-moderate nonproliferative retinopathy and urinary albumin excretion <200 mg/24 h). Baseline exclusion criteria included hypertension, hyperlipidemia, cardiovascular disease, neuropathy requiring medical intervention, and recurrent severe hypoglycemia.

Participants who were originally assigned to receive conventional therapy were taught and encouraged to adopt intensive therapy, and both original treatment groups returned to their own health care providers for ongoing diabetes care. In 1994, 96% of the surviving DCCT cohort enrolled in the EDIC observational study, an ongoing, long-term follow-up (14) of the DCCT cohort. The present analyses focus on the data obtained during the combined DCCT and EDIC study, up to the last complete DPN and CAN evaluations conducted in EDIC year 16/17 (2009/2010), for a mean total DCCT/EDIC study period of 23 years.

Risk Factors

Risk factors were assessed by standardized methods during DCCT and EDIC (14,15). Annual visits included a detailed medical history assessing demographic and behavioral risk factors and medical outcomes, and a physical examination with measurements of height, weight, waist circumference (only during EDIC), sitting blood pressure, and pulse rate (14,15). Insulin doses were self-reported and expressed as the average total daily dose in units per kilogram of body weight. Hypertension was defined as systolic blood pressure (SBP) ≥140 mmHg, diastolic blood pressure (DBP) ≥90 mmHg, documented hypertension, or the use of antihypertensive medications, and hyperlipidemia as LDL cholesterol ≥130 mg/dL or the use of lipid-lowering medications. Concurrent medication usage was collected during EDIC but not during the DCCT. Therefore, during the DCCT, hypertension and hyperlipidemia were defined using only elevated blood pressures and lipids.

HbA1c was measured by high-performance liquid chromatography quarterly during DCCT and annually during EDIC. Fasting lipids and albumin excretion rates (AER) were measured annually during DCCT and on alternate years during EDIC. AER was measured from 4-h urine samples using fluoroimmunoassay from DCCT baseline to EDIC year 18 and subsequently from spot urine samples, with AER estimated using the ratio of urine albumin and creatinine concentrations (16). The DCCT/EDIC central biochemistry laboratory performed all laboratory measurements with standardized methods and long-term controls to guard against assay drift. LDL cholesterol was calculated using the Friedewald equation (17).

Estimated glomerular filtration rates (eGFR) were calculated from serum creatinine measured annually using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation. Kidney disease was defined as an eGFR <60 mL/min/1.73 m2, microalbuminuria (AER ≥30 mg/24 on ≥2 consecutive visits), or macroalbuminuria (AER ≥300 mg/24). Standardized stereoscopic seven-field fundus photographs were obtained every 6 months during DCCT, and in one-fourth of the cohort annually during EDIC and graded centrally using the final Early Treatment of Diabetic Retinopathy Study (ETDRS) severity grading scale (18). Proliferative diabetic retinopathy was defined as 61/<61 or greater on the ETDRS scale or pan-retinal photocoagulation and clinically significant macular edema as focal photocoagulation or treatment with anti–vascular endothelial growth factor at any point during the DCCT/EDIC study.

Neuropathy Outcomes

DPN was formally assessed twice during the DCCT (baseline, year 5) and once during EDIC (year 13/14). The primary outcome for DPN was confirmed clinical neuropathy, requiring at least two abnormal findings among symptoms, sensory signs, or reflex changes consistent with a distal symmetrical polyneuropathy as assessed by a qualified neurologist, plus abnormal nerve conduction studies involving two or more nerves among the median (motor or sensory), peroneal, and sural nerves (11). The DPN outcome was sensitive to detection of primarily large- or mixed small- and large-fiber neuropathies. CAN was assessed up to five times during DCCT (baseline, years 2, 4, 6, and 8) and twice during EDIC (years 13/14 and 16/17) with R-R response to paced breathing, Valsalva maneuver, and postural changes in blood pressure under standardized conditions, as previously described (3,12). All subjects were required to fast and avoid caffeine and tobacco products, as well as prescription and over-the-counter medications (except for their usual insulin regimen), for at least 8 h prior to CAN testing. Presence of CAN was defined as either an R-R variation <15 or an R-R variation 15–19.9 in combination with a Valsalva ratio ≤1.5 or a decrease of >10 mmHg in DBP during 10 min of standing (3). Participants with at least two DPN (n = 1,386) or two CAN (n = 1,434) assessments during DCCT/EDIC were included in the analyses. DPN and CAN measurements overlapped twice during the study: DCCT baseline and EDIC year 13/14.

Statistical Considerations

Candidate risk factors were grouped into 11 blocks (Supplementary Table 1) as performed in other DCCT/EDIC studies of risk factors (1921). All covariates were measured either concurrently with the neurology assessment or within the closest visit window preceding the respective DPN or CAN assessment. For these analyses, given the known association with DPN, height was added to the second block of candidate risk factors (Demographic Physical). Generalized estimating equation (GEE) models were used to evaluate the association of neuropathy (DPN or CAN) with individual risk factors over repeated time points.

A comprehensive multivariable GEE regression model was evaluated for both DPN and CAN using similar model-building techniques previously described by the DCCT/EDIC study (19). Given the large number of risk factors, variables were entered into the GEE model one block at a time in the order displayed in Supplementary Table 1. After each block was added, a variable was deleted if it was not nominally significant (P < 0.10) (19). After the last block was entered, the final multivariable model was fit using the selected covariates and variables significant at P < 0.05 were retained. Height and weight were more strongly associated with DPN than BMI in both unadjusted and minimally adjusted models. Given its strong dependence on height and weight, BMI was not entered into the model with the second block of candidate risk factors. Similarly, pulse pressure was not entered into the model with the fifth block of variables since it is a function of both SBP and DBP.

A sensitivity analysis was conducted, using a backward elimination modeling technique, where variables significant at P < 0.10 were retained at each step (see Supplementary Data). The odds ratios (ORs) and unsigned covariate Z-test value are presented, the latter to differentiate covariate effects with P < 0.0001 (two-sided), equivalent to a |Z| ≥ 3.89.

Each risk factor could be included in the model as a fixed baseline covariate (e.g., baseline HbA1c), time-dependent covariate using the most recent measurement (e.g., current HbA1c), or time-dependent covariate using the updated mean of all follow-up values since randomization (e.g., updated mean HbA1c representing the cumulative mean of prior HbA1c values up to each visit). To account for different measurement frequencies of HbA1c during DCCT (every 3 months) and EDIC (every 12 months), the updated mean was computed by weighting each value by the time interval since the last measurement. All analyses were performed using SAS software (version 9.4; SAS Institute, Cary, NC).

Data and Resource Availability

Data collected for the study are available to the public through the National Institute of Diabetes and Digestive and Kidney Disease repository (https://repository.niddk.nih.gov/home/).

Over 23 years of mean follow-up, 33% of participants developed DPN and 44% CAN. At baseline, 90% of the cohort had neither DPN nor CAN; 5% had DPN only, 4% had CAN only, and 1% had both. By EDIC year 13/14, 54% of the cohort had no neuropathy; 13% had DPN only, 16% had CAN only, and 16% had both.

At DCCT baseline, 53% of the participants were men, median age was 27 years, duration of diabetes was 4 years, HbA1c was 8.80% (73 mmol/mol), and 18% were smokers. The baseline characteristics of the DCCT/EDIC cohort among those who did and did not develop DPN at any time during the study are shown in Table 1. The baseline characteristics for participants who did and did not develop CAN are presented in Table 2. Nominally significant baseline factors that were associated with higher odds of DPN at any time during DCCT/EDIC included male sex, older age, greater height, weight, and BMI in both men and women, moderate or strenuous physical activity, family history of type 2 diabetes, higher SBP and DBP, higher pulse rate, higher total cholesterol, triglycerides, and LDL cholesterol levels, longer duration of type 1 diabetes, higher HbA1c levels independent of treatment assignment, as well as randomization to conventional versus intensive treatment and inclusion in the secondary prevention versus primary intervention cohort. Most of these risk factors were also associated with higher odds of CAN at any time during DCCT/EDIC, with the exception of sex, height, physical activity, and HDL cholesterol. In addition, cigarette smoking and occasional or regular alcohol use (Table 2) at the onset of the DCCT were associated with CAN.

Table 1

DCCT baseline characteristics of participants according to the presence or absence of any DPN over the duration of the DCCT/EDIC study, through EDIC year 13/14

OverallNoYesPer unit changeOR (95% CI)P value
N 1,386 931 455    
Design       
 Treatment group (% conventional) 51 48 57 Conventional vs. intensive 1.56 (1.26, 1.93) 0.0001 
 Cohort (% secondary) 51 45 61 Secondary vs. primary 1.90 (1.53, 2.36) <0.0001 
Physical       
 Sex (% men) 52 49 60 Men vs. women 1.51 (1.21, 1.88) 0.0002 
 Age (years) 27 (22, 32) 26 (21, 32) 29 (24, 34) 5 years 1.30 (1.21, 1.41) <0.0001 
 Adult (% ≥18 years) 87 84 91 ≥18 vs. <18 years 1.83 (1.28, 2.63) 0.0009 
 Height (cm) 171 (164, 179) 170 (163, 177) 174 (167, 182) 5 cm 1.20 (1.13, 1.27) <0.0001 
 Weight men (kg) 74 (67, 82) 73 (66, 80) 78 (70, 85) 5 kg 1.19 (1.11, 1.27) <0.0001 
 Weight women (kg) 62 (56, 69) 61 (55, 68) 64 (58, 70) 5 kg 1.20 (1.10, 1.32) <0.0001 
 BMI men (kg/m224 (22, 25) 23 (21, 25) 24 (22, 26) 1 kg/m2 1.08 (1.02, 1.14) 0.0052 
 BMI women (kg/m223 (21, 25) 23 (21, 25) 23 (22, 25) 1 kg/m2 1.09 (1.03, 1.15) 0.0021 
Behavioral       
 Cigarette smoker (%) 18 18 20 Yes vs. no 1.23 (0.94, 1.60) 0.1358 
 Occasional or regular alcohol use (%) 22 20 24 Yes vs. none 1.23 (0.95, 1.59) 0.1107 
 Moderate or strenuous activity (%) 70 68 74 Yes vs. sedentary 1.31 (1.03, 1.66) 0.0264 
Family history       
 Hypertension (%) 56 54 58 Yes vs. no 1.16 (0.93, 1.44) 0.1872 
 Myocardial infarction (%) 49 47 52 Yes vs. no 1.18 (0.95, 1.46) 0.1366 
 Type 1 diabetes (%) 14 14 15 Yes vs. no 1.22 (0.90, 1.66) 0.2041 
 Type 2 diabetes (%) 12 Yes vs. no 1.44 (1.04, 2.00) 0.0292 
Blood pressure       
 SBP (mmHg) 114 (106, 122) 112 (106, 120) 116 (110,124) 5 mmHg 1.11 (1.06, 1.17) <0.0001 
 DBP (mmHg) 72 (68, 80) 70 (66, 78) 76 (68, 80) 5 mmHg 1.16 (1.09, 1.24) <0.0001 
 Pulse pressure (mmHg) 40 (34, 48) 40 (34, 48) 40 (36, 48) 5 mmHg 1.03 (0.98, 1.09) 0.2658 
 Pulse rate (bpm) 76 (68, 84) 76 (68, 82) 76 (70, 84) 1 bpm 1.02 (1.01, 1.03) <0.0001 
Lipids       
 Total cholesterol (mg/dL) 174 (152, 196) 172 (152, 194) 176 (155, 202) 10 mg/dL 1.05 (1.02, 1.08) 0.0038 
 Triglycerides (mg/dL) 70 (55, 93) 69 (54, 92) 73 (57, 98) 10 mg/dL 1.03 (1.01, 1.05) 0.0032 
 HDL cholesterol (mg/dL) 49 (42, 57) 50 (42, 58) 47 (41, 56) 10 mg/dL 0.92 (0.84, 1.00) 0.0614 
 LDL cholesterol (mg/dL) 106 (90, 127) 104 (89, 125) 110 (92, 131) 10 mg/dL 1.06 (1.02, 1.10) 0.0011 
History of diabetes       
 Duration of type 1 diabetes (years) 4 (2, 9) 4 (2, 8) 6 (3, 10) 1 year 1.08 (1.06, 1.11) <0.0001 
 Stimulated C-peptide (nmol/L⋅100)       
  Duration <5 years 12 (4, 25) 12 (4, 26) 13 (4, 24) 1 year 0.99 (0.98, 1.01) 0.3172 
  Duration ≥5 years 3 (3, 4) 3 (3, 4) 3 (3, 3) 1 year 1.01 (0.97, 1.05) 0.6864 
 Insulin dose (units/kg/day) 0.64 (0.50, 0.80) 0.64 (0.50, 0.81) 0.63 (0.49, 0.77) 1 unit/kg/day 0.88 (0.57, 1.35) 0.5552 
 HbA1c (%) 8.80 (7.84, 10.08) 8.63 (7.70, 9.74) 9.20 (8.23, 10.60) 1% 1.31 (1.22, 1.40) <0.0001 
 HbA1c (mmol/mol) 72.7 (62.2, 86.7) 70.8 (60.7, 83.0) 77.1 (66.5, 92.4) 1 mmol/mol 1.31 (1.22, 1.40) <0.0001 
OverallNoYesPer unit changeOR (95% CI)P value
N 1,386 931 455    
Design       
 Treatment group (% conventional) 51 48 57 Conventional vs. intensive 1.56 (1.26, 1.93) 0.0001 
 Cohort (% secondary) 51 45 61 Secondary vs. primary 1.90 (1.53, 2.36) <0.0001 
Physical       
 Sex (% men) 52 49 60 Men vs. women 1.51 (1.21, 1.88) 0.0002 
 Age (years) 27 (22, 32) 26 (21, 32) 29 (24, 34) 5 years 1.30 (1.21, 1.41) <0.0001 
 Adult (% ≥18 years) 87 84 91 ≥18 vs. <18 years 1.83 (1.28, 2.63) 0.0009 
 Height (cm) 171 (164, 179) 170 (163, 177) 174 (167, 182) 5 cm 1.20 (1.13, 1.27) <0.0001 
 Weight men (kg) 74 (67, 82) 73 (66, 80) 78 (70, 85) 5 kg 1.19 (1.11, 1.27) <0.0001 
 Weight women (kg) 62 (56, 69) 61 (55, 68) 64 (58, 70) 5 kg 1.20 (1.10, 1.32) <0.0001 
 BMI men (kg/m224 (22, 25) 23 (21, 25) 24 (22, 26) 1 kg/m2 1.08 (1.02, 1.14) 0.0052 
 BMI women (kg/m223 (21, 25) 23 (21, 25) 23 (22, 25) 1 kg/m2 1.09 (1.03, 1.15) 0.0021 
Behavioral       
 Cigarette smoker (%) 18 18 20 Yes vs. no 1.23 (0.94, 1.60) 0.1358 
 Occasional or regular alcohol use (%) 22 20 24 Yes vs. none 1.23 (0.95, 1.59) 0.1107 
 Moderate or strenuous activity (%) 70 68 74 Yes vs. sedentary 1.31 (1.03, 1.66) 0.0264 
Family history       
 Hypertension (%) 56 54 58 Yes vs. no 1.16 (0.93, 1.44) 0.1872 
 Myocardial infarction (%) 49 47 52 Yes vs. no 1.18 (0.95, 1.46) 0.1366 
 Type 1 diabetes (%) 14 14 15 Yes vs. no 1.22 (0.90, 1.66) 0.2041 
 Type 2 diabetes (%) 12 Yes vs. no 1.44 (1.04, 2.00) 0.0292 
Blood pressure       
 SBP (mmHg) 114 (106, 122) 112 (106, 120) 116 (110,124) 5 mmHg 1.11 (1.06, 1.17) <0.0001 
 DBP (mmHg) 72 (68, 80) 70 (66, 78) 76 (68, 80) 5 mmHg 1.16 (1.09, 1.24) <0.0001 
 Pulse pressure (mmHg) 40 (34, 48) 40 (34, 48) 40 (36, 48) 5 mmHg 1.03 (0.98, 1.09) 0.2658 
 Pulse rate (bpm) 76 (68, 84) 76 (68, 82) 76 (70, 84) 1 bpm 1.02 (1.01, 1.03) <0.0001 
Lipids       
 Total cholesterol (mg/dL) 174 (152, 196) 172 (152, 194) 176 (155, 202) 10 mg/dL 1.05 (1.02, 1.08) 0.0038 
 Triglycerides (mg/dL) 70 (55, 93) 69 (54, 92) 73 (57, 98) 10 mg/dL 1.03 (1.01, 1.05) 0.0032 
 HDL cholesterol (mg/dL) 49 (42, 57) 50 (42, 58) 47 (41, 56) 10 mg/dL 0.92 (0.84, 1.00) 0.0614 
 LDL cholesterol (mg/dL) 106 (90, 127) 104 (89, 125) 110 (92, 131) 10 mg/dL 1.06 (1.02, 1.10) 0.0011 
History of diabetes       
 Duration of type 1 diabetes (years) 4 (2, 9) 4 (2, 8) 6 (3, 10) 1 year 1.08 (1.06, 1.11) <0.0001 
 Stimulated C-peptide (nmol/L⋅100)       
  Duration <5 years 12 (4, 25) 12 (4, 26) 13 (4, 24) 1 year 0.99 (0.98, 1.01) 0.3172 
  Duration ≥5 years 3 (3, 4) 3 (3, 4) 3 (3, 3) 1 year 1.01 (0.97, 1.05) 0.6864 
 Insulin dose (units/kg/day) 0.64 (0.50, 0.80) 0.64 (0.50, 0.81) 0.63 (0.49, 0.77) 1 unit/kg/day 0.88 (0.57, 1.35) 0.5552 
 HbA1c (%) 8.80 (7.84, 10.08) 8.63 (7.70, 9.74) 9.20 (8.23, 10.60) 1% 1.31 (1.22, 1.40) <0.0001 
 HbA1c (mmol/mol) 72.7 (62.2, 86.7) 70.8 (60.7, 83.0) 77.1 (66.5, 92.4) 1 mmol/mol 1.31 (1.22, 1.40) <0.0001 

Data are based on 1,386 participants with at least two DPN assessments during DCCT/EDIC. Medians (first quartile, third quartile) or % are presented. ORs and P values were generated using GEE models with no adjustment for other factors.

Table 2

DCCT baseline characteristics of participants according to the presence or absence of any CAN over the course of the DCCT/EDIC study, through EDIC year 16/17

OverallNoYesPer unit changeOR (95% CI)P value
N 1,434 803 631    
Design       
 Treatment group (% conventional) 51 49 53 Conventional vs. intensive 1.36 (1.11, 1.66) 0.0025 
 Cohort (% secondary) 50 44 57 Secondary vs. primary 1.72 (1.41, 2.10) <0.0001 
Physical       
 Sex (% men) 53 54 52 Men vs. women 0.92 (0.75, 1.13) 0.4286 
 Age (years) 27 (22, 32) 26 (21, 31) 29 (23, 34) 5 years 1.39 (1.29, 1.50) <0.0001 
 Adult (% ≥18 years) 86 83 91 ≥18 vs. <18 years 2.03 (1.45, 2.85) <0.0001 
 Height (cm) 171 (164, 179) 171 (164, 178) 171 (164, 180) 5 cm 1.02 (0.97, 1.07) 0.4951 
 Weight men (kg) 74 (67, 82) 73 (66, 81) 76 (69, 84) 5 kg 1.14 (1.07, 1.21) <0.0001 
 Weight women (kg) 62 (56, 69) 61 (55, 68) 63 (57, 69) 5 kg 1.09 (1.01, 1.18) 0.0251 
 BMI men (kg/m224 (22, 25) 23 (21, 25) 24 (22, 26) 1 kg/m2 1.10 (1.04, 1.15) 0.0004 
 BMI women (kg/m223 (21, 25) 23 (21, 25) 23 (21, 25) 1 kg/m2 1.07 (1.02, 1.12) 0.0107 
Behavioral       
 Cigarette smoker (%) 18 16 21 Yes vs. no 1.77 (1.38, 2.28) <0.0001 
 Occasional or regular alcohol use (%) 22 19 24 Yes vs. none 1.31 (1.03, 1.65) 0.0261 
 Moderate or strenuous activity (%) 70 67 73 Yes vs. sedentary 1.23 (0.98, 1.53) 0.0698 
Family history       
 Hypertension (%) 56 55 58 Yes vs. no 1.22 (1.00, 1.50) 0.0528 
 Myocardial infarction (%) 49 49 49 Yes vs. no 1.00 (0.81, 1.22) 0.9666 
 Type 1 diabetes (%) 14 14 15 Yes vs. no 1.07 (0.80, 1.43) 0.6297 
 Type 2 diabetes (%) 11 Yes vs. no 1.54 (1.12, 2.11) 0.0073 
Blood pressure       
 SBP (mmHg) 114 (106, 122) 112 (106, 120) 116 (108, 122) 5 mmHg 1.07 (1.03, 1.12) 0.0021 
 DBP (mmHg) 72 (68, 80) 72 (66, 80) 74 (68, 80) 5 mmHg 1.13 (1.07, 1.20) <0.0001 
 Pulse pressure (mmHg) 40 (34, 48) 40 (35, 48) 40 (34, 48) 5 mmHg 1.00 (0.95, 1.05) 0.9673 
 Pulse rate (bpm) 76 (68, 84) 74 (68, 80) 78 (72, 84) 1 bpm 1.03 (1.02, 1.04) <0.0001 
Lipids       
 Total cholesterol (mg/dL) 174 (152, 196) 171 (150, 195) 177 (156, 201) 10 mg/dL 1.09 (1.06, 1.12) <0.0001 
 Triglycerides (mg/dL) 70 (55, 94) 68 (54, 90) 72 (56, 98) 10 mg/dL 1.04 (1.02, 1.06) 0.0002 
 HDL cholesterol (mg/dL) 49 (42, 58) 50 (42, 58) 48 (42, 57) 10 mg/dL 0.99 (0.91, 1.07) 0.7660 
 LDL cholesterol (mg/dL) 107 (91, 127) 104 (88, 125) 110 (92, 130) 10 mg/dL 1.10 (1.06, 1.13) <0.0001 
History of diabetes       
 Duration of type 1 diabetes (years) 4 (2, 9) 4 (2, 8) 5 (2, 10) 1 year 1.07 (1.05, 1.10) <0.0001 
 Stimulated C-peptide (nmol/L⋅100)       
  Duration <5 years 12 (4, 25) 12 (4, 24) 13 (5, 27) 1 year 1.00 (0.99, 1.01) 0.4487 
  Duration ≥5 years 3 (3, 4) 3 (3, 4) 3 (3,3) 1 year 0.99 (0.95, 1.03) 0.6122 
 Insulin dose (units/kg/day) 0.64 (0.50, 0.80) 0.65 (0.50, 0.82) 0.63 (0.50, 0.79) 1 unit/kg/day 0.79 (0.52, 1.20) 0.2679 
 HbA1c (%) 8.80 (7.82, 10.10) 8.64 (7.70, 9.81) 9.00 (8.01, 10.30) 1% 1.25 (1.18, 1.33) <0.0001 
 HbA1c (mmol/mol) 72.7 (62.0, 86.9) 70.9 (60.7, 83.7) 74.9 (64.7, 89.1) 1 mmol/mol 1.25 (1.18, 1.33) <0.0001 
OverallNoYesPer unit changeOR (95% CI)P value
N 1,434 803 631    
Design       
 Treatment group (% conventional) 51 49 53 Conventional vs. intensive 1.36 (1.11, 1.66) 0.0025 
 Cohort (% secondary) 50 44 57 Secondary vs. primary 1.72 (1.41, 2.10) <0.0001 
Physical       
 Sex (% men) 53 54 52 Men vs. women 0.92 (0.75, 1.13) 0.4286 
 Age (years) 27 (22, 32) 26 (21, 31) 29 (23, 34) 5 years 1.39 (1.29, 1.50) <0.0001 
 Adult (% ≥18 years) 86 83 91 ≥18 vs. <18 years 2.03 (1.45, 2.85) <0.0001 
 Height (cm) 171 (164, 179) 171 (164, 178) 171 (164, 180) 5 cm 1.02 (0.97, 1.07) 0.4951 
 Weight men (kg) 74 (67, 82) 73 (66, 81) 76 (69, 84) 5 kg 1.14 (1.07, 1.21) <0.0001 
 Weight women (kg) 62 (56, 69) 61 (55, 68) 63 (57, 69) 5 kg 1.09 (1.01, 1.18) 0.0251 
 BMI men (kg/m224 (22, 25) 23 (21, 25) 24 (22, 26) 1 kg/m2 1.10 (1.04, 1.15) 0.0004 
 BMI women (kg/m223 (21, 25) 23 (21, 25) 23 (21, 25) 1 kg/m2 1.07 (1.02, 1.12) 0.0107 
Behavioral       
 Cigarette smoker (%) 18 16 21 Yes vs. no 1.77 (1.38, 2.28) <0.0001 
 Occasional or regular alcohol use (%) 22 19 24 Yes vs. none 1.31 (1.03, 1.65) 0.0261 
 Moderate or strenuous activity (%) 70 67 73 Yes vs. sedentary 1.23 (0.98, 1.53) 0.0698 
Family history       
 Hypertension (%) 56 55 58 Yes vs. no 1.22 (1.00, 1.50) 0.0528 
 Myocardial infarction (%) 49 49 49 Yes vs. no 1.00 (0.81, 1.22) 0.9666 
 Type 1 diabetes (%) 14 14 15 Yes vs. no 1.07 (0.80, 1.43) 0.6297 
 Type 2 diabetes (%) 11 Yes vs. no 1.54 (1.12, 2.11) 0.0073 
Blood pressure       
 SBP (mmHg) 114 (106, 122) 112 (106, 120) 116 (108, 122) 5 mmHg 1.07 (1.03, 1.12) 0.0021 
 DBP (mmHg) 72 (68, 80) 72 (66, 80) 74 (68, 80) 5 mmHg 1.13 (1.07, 1.20) <0.0001 
 Pulse pressure (mmHg) 40 (34, 48) 40 (35, 48) 40 (34, 48) 5 mmHg 1.00 (0.95, 1.05) 0.9673 
 Pulse rate (bpm) 76 (68, 84) 74 (68, 80) 78 (72, 84) 1 bpm 1.03 (1.02, 1.04) <0.0001 
Lipids       
 Total cholesterol (mg/dL) 174 (152, 196) 171 (150, 195) 177 (156, 201) 10 mg/dL 1.09 (1.06, 1.12) <0.0001 
 Triglycerides (mg/dL) 70 (55, 94) 68 (54, 90) 72 (56, 98) 10 mg/dL 1.04 (1.02, 1.06) 0.0002 
 HDL cholesterol (mg/dL) 49 (42, 58) 50 (42, 58) 48 (42, 57) 10 mg/dL 0.99 (0.91, 1.07) 0.7660 
 LDL cholesterol (mg/dL) 107 (91, 127) 104 (88, 125) 110 (92, 130) 10 mg/dL 1.10 (1.06, 1.13) <0.0001 
History of diabetes       
 Duration of type 1 diabetes (years) 4 (2, 9) 4 (2, 8) 5 (2, 10) 1 year 1.07 (1.05, 1.10) <0.0001 
 Stimulated C-peptide (nmol/L⋅100)       
  Duration <5 years 12 (4, 25) 12 (4, 24) 13 (5, 27) 1 year 1.00 (0.99, 1.01) 0.4487 
  Duration ≥5 years 3 (3, 4) 3 (3, 4) 3 (3,3) 1 year 0.99 (0.95, 1.03) 0.6122 
 Insulin dose (units/kg/day) 0.64 (0.50, 0.80) 0.65 (0.50, 0.82) 0.63 (0.50, 0.79) 1 unit/kg/day 0.79 (0.52, 1.20) 0.2679 
 HbA1c (%) 8.80 (7.82, 10.10) 8.64 (7.70, 9.81) 9.00 (8.01, 10.30) 1% 1.25 (1.18, 1.33) <0.0001 
 HbA1c (mmol/mol) 72.7 (62.0, 86.9) 70.9 (60.7, 83.7) 74.9 (64.7, 89.1) 1 mmol/mol 1.25 (1.18, 1.33) <0.0001 

Data are based on 1,434 participants with at least two CAN assessments during DCCT/EDIC. Medians (first quartile, third quartile) or % are presented. ORs and P values were generated using GEE models with no adjustment for other factors.

Time-Dependent Characteristics

The longitudinal ORs for DPN and CAN per unit change in the current or the updated mean value of each time-dependent covariate, without adjustments, are shown in Supplementary Table 2. During follow-up, greater weight and BMI, higher SBP and DBP and pulse pressure, higher pulse rate, higher total cholesterol, triglycerides, and LDL cholesterol levels, higher HbA1c, and a higher prevalence of hypertension and hyperlipidemia, medication use, and kidney disease were all significantly associated with higher odds of DPN and CAN. DPN was also significantly associated with greater height and lower HDL cholesterol while CAN was associated with cigarette smoking, sedentary activity, and higher daily insulin dose.

Minimally Adjusted Baseline and Time-Dependent Covariate Effects

After adjustment for age and updated mean HbA1c, the following risk factors were significantly associated with higher odds of DPN: secondary prevention versus primary intervention cohort, male sex, greater height, weight, and BMI, higher SBP and DBP and pulse pressure, higher pulse rate, higher triglycerides, lower HDL cholesterol, longer duration of type 1 diabetes, higher daily insulin dose, a higher prevalence of hypertension, and any kidney disease (Table 3). Total cholesterol, LDL cholesterol, and hyperlipidemia were no longer significantly associated with odds of DPN after adjustment for age and updated mean HbA1c.

Table 3

ORs for DPN and CAN per unit change in each baseline or time-dependent covariate in separate longitudinal GEE models, minimally adjusted for age and updated mean HbA1c

DPNCAN
*OR (95% CI)P valueOR (95% CI)P value
Design      
 Treatment group (conventional vs. intensive) 1.05 (0.83, 1.32) 0.7022 1.12 (0.91, 1.38) 0.2883 
 Cohort (secondary vs. primary) 2.07 (1.66, 2.57) <0.0001 1.81 (1.49, 2.21) <0.0001 
Physical      
 Sex (men vs. women) 1.51 (1.21, 1.88) 0.0003 0.85 (0.70, 1.04) 0.1164 
 Adult (≥18 vs. <18 years) 1.05 (0.66, 1.67) 0.8291 0.76 (0.50, 1.16) 0.1997 
 Height (per 5 cm) 1.21 (1.14, 1.28) <0.0001 0.96 (0.91, 1.01) 0.1134 
 Weight (per 5 kg) 1.09 (1.06, 1.13) <0.0001 1.02 (0.99, 1.05) 0.1299 
 Mean weight (per 5 kg) 1.12 (1.08, 1.17) <0.0001 1.02 (0.98, 1.06) 0.2686 
 BMI (per 1 kg/m21.02 (1.00, 1.05) 0.0612 1.03 (1.01, 1.04) 0.0079 
 Mean BMI (per 1 kg/m21.04 (1.01, 1.07) 0.0090 1.03 (1.01, 0.106) 0.0174 
Behavioral      
 Cigarette smoker (yes vs. no) 0.98 (0.75, 1.29) 0.8932 1.39 (1.10, 1.76) 0.0057 
 Occasional or regular alcohol use (yes vs. none) 1.14 (0.93, 1.41) 0.2158 0.85 (0.72, 1.00) 0.0462 
 Moderate or strenuous activity (yes vs. sedentary) 1.08 (0.89, 1.31) 0.4471 0.86 (0.74, 1.00) 0.0543 
Family history      
 Hypertension (yes vs. no) 1.17 (0.94, 1.46) 0.1514 1.25 (1.03, 1.53) 0.0276 
 Myocardial infarction (yes vs. no) 1.17 (0.94, 1.45) 0.1503 0.97 (0.80, 1.19) 0.7897 
 Type 1 diabetes (yes vs. no) 1.15 (0.83, 1.57) 0.4048 0.95 (0.71, 1.28) 0.7468 
 Type 2 diabetes (yes vs. no) 1.22 (0.87, 1.73) 0.2544 1.22 (0.89, 1.66) 0.2164 
Blood pressure      
 SBP (per 5 mmHg) 1.08 (1.04, 1.12) 0.0003 1.09 (1.06, 1.13) <0.0001 
 Mean SBP (per 5 mmHg) 1.17 (1.10, 1.24) <0.0001 1.24 (1.17, 1.30) <0.0001 
 DBP (per 5 mmHg) 1.07 (1.01, 1.13) 0.0184 1.05 (1.00, 1.09) 0.0462 
 Mean DBP (per 5 mmHg) 1.21 (1.11, 1.32) <0.0001 1.21 (1.11, 1.31) <0.0001 
 Pulse pressure (per 5 mmHg) 1.07 (1.02, 1.12) 0.0035 1.11 (1.07, 1.14) <0.0001 
 Pulse rate (per 1 bpm) 1.01 (1.00, 1.02) 0.0048 1.03 (1.03, 1.04) <0.0001 
 Mean pulse rate (per 1 bpm) 1.03 (1.02, 1.04) <0.0001 1.07 (1.06, 1.09) <0.0001 
 Hypertension (yes vs. no) 2.03 (1.56, 2.64) <0.0001 1.94 (1.61, 2.34) <0.0001 
Medications      
 ACE inhibitor use (yes vs. no) 1.54 (1.20, 1.97) 0.0008 1.51 (1.25, 1.82) <0.0001 
 ARB use (yes vs. no) 1.24 (0.87, 1.77) 0.2438 1.45 (1.13, 1.87) 0.0038 
 β-Blocker use (yes vs. no) 2.88 (1.79, 4.64) <0.0001 2.77 (1.95, 3.92) <0.0001 
 Calcium channel blocker use (yes vs. no) 1.41 (0.88, 2.26) 0.1524 1.70 (1.21, 2.40) 0.0022 
 Lipid-lowering agent use (yes vs. no) 1.15 (0.89, 1.48) 0.2740 1.36 (1.12, 1.65) 0.0020 
Lipids      
 Total cholesterol (per 10 mg/dL) 0.98 (0.95, 1.01) 0.1607 1.00 (0.98, 1.02) 0.9612 
 Mean total cholesterol (per 10 mg/dL) 1.02 (0.98, 1.05) 0.4334 1.06 (1.02, 1.10) 0.0027 
 Triglycerides (per 20% increase mg/dL) 1.06 (1.02, 1.10) 0.0054 1.07 (104, 1.10) <0.0001 
 Mean triglycerides (per 20% increase mg/dL) 1.09 (1.04, 1.15) <0.0001 1.14 (1.09, 1.18) <0.0001 
 HDL cholesterol (per 10 mg/dL) 0.89 (0.83, 0.93) 0.0034 0.96 (0.91, 1.01) 0.1113 
 Mean HDL cholesterol (per 10 mg/dL) 0.88 (0.81, 0.96) 0.0052 0.95 (0.88, 1.03) 0.2238 
 LDL cholesterol (per 10 mg/dL) 0.98 (0.95, 1.01) 0.1804 0.99 (0.96, 1.02) 0.4320 
 Mean LDL cholesterol (per 10 mg/dL) 1.02 (0.98, 1.06) 0.4137 1.04 (1.00, 1.08) 0.0719 
 Hyperlipidemia (yes vs. no) 1.02 (0.83, 1.26) 0.8568 1.27 (1.08, 1.51) 0.0052 
History of diabetes      
 Duration of type 1 diabetes (per 1 year) 1.11 (1.08, 1.13) <0.0001 1.09 (1.06, 1.12) <0.0001 
 Stimulated C-peptide (per 1 nmol/L*100)      
  Duration <5 years 0.99 (0.98, 1.00) 0.1345 1.00 (0.98, 1.01) 0.7650 
  Duration ≥5 years 1.00 (0.96, 1.05) 0.8726 0.98 (0.94, 1.02) 0.3685 
 Insulin dose (per 1 unit/kg/day) 1.19 (0.82, 1.73) 0.3718 1.26 (0.99, 1.62) 0.0648 
 Mean insulin dose (per 1 unit/kg/day) 2.07 (1.25, 3.43) 0.0046 1.88 (1.18, 3.00) 0.0083 
Kidney disease      
 eGFR (per 5 mL/min/1.73 m20.93 (0.90, 0.97) <0.0001 0.92 (0.90, 0.95) <0.0001 
  eGFR <60 mL/min/1.73 m2 (yes vs. no) 2.56 (1.36, 4.80) 0.0034 5.95 (3.36, 10.52) <0.0001 
  Any eGFR <60 mL/min/1.73 m2 (yes vs. no) 2.44 (1.45, 4.10) 0.0007 3.24 (2.12, 4.99) <0.0001 
 AER (per 20% increase mg/24 h) 1.06 (1.04, 1.07) <0.0001 1.06 (1.04, 1.07) <0.0001 
  Sustained AER ≥30 mg/24 h (yes vs. no) 2.45 (1.89, 3.17) <0.0001 2.36 (1.86, 3.00) <0.0001 
  Any sustained AER ≥30 mg/24 h (yes vs. no) 2.50 (1.98, 3.15) <0.0001 2.87 (2.32, 3.53) <0.0001 
  AER ≥300 mg/24 h (yes vs. no) 3.81 (2.42, 6.01) <0.0001 3.25 (2.17, 4.87) <0.0001 
  Any AER ≥300 mg/24 h (yes vs. no) 4.02 (2.81, 5.75) <0.0001 3.30 (2.36, 4.59) <0.0001 
Hypoglycemia      
 Coma and/or seizure (yes vs. no) 1.09 (0.61, 1.97) 0.7636 1.03 (0.73, 1.45) 0.8697 
 Requiring assistance (yes vs. no) 1.36 (0.96, 1.92) 0.0837 1.13 (0.88, 1.45) 0.3264 
DPNCAN
*OR (95% CI)P valueOR (95% CI)P value
Design      
 Treatment group (conventional vs. intensive) 1.05 (0.83, 1.32) 0.7022 1.12 (0.91, 1.38) 0.2883 
 Cohort (secondary vs. primary) 2.07 (1.66, 2.57) <0.0001 1.81 (1.49, 2.21) <0.0001 
Physical      
 Sex (men vs. women) 1.51 (1.21, 1.88) 0.0003 0.85 (0.70, 1.04) 0.1164 
 Adult (≥18 vs. <18 years) 1.05 (0.66, 1.67) 0.8291 0.76 (0.50, 1.16) 0.1997 
 Height (per 5 cm) 1.21 (1.14, 1.28) <0.0001 0.96 (0.91, 1.01) 0.1134 
 Weight (per 5 kg) 1.09 (1.06, 1.13) <0.0001 1.02 (0.99, 1.05) 0.1299 
 Mean weight (per 5 kg) 1.12 (1.08, 1.17) <0.0001 1.02 (0.98, 1.06) 0.2686 
 BMI (per 1 kg/m21.02 (1.00, 1.05) 0.0612 1.03 (1.01, 1.04) 0.0079 
 Mean BMI (per 1 kg/m21.04 (1.01, 1.07) 0.0090 1.03 (1.01, 0.106) 0.0174 
Behavioral      
 Cigarette smoker (yes vs. no) 0.98 (0.75, 1.29) 0.8932 1.39 (1.10, 1.76) 0.0057 
 Occasional or regular alcohol use (yes vs. none) 1.14 (0.93, 1.41) 0.2158 0.85 (0.72, 1.00) 0.0462 
 Moderate or strenuous activity (yes vs. sedentary) 1.08 (0.89, 1.31) 0.4471 0.86 (0.74, 1.00) 0.0543 
Family history      
 Hypertension (yes vs. no) 1.17 (0.94, 1.46) 0.1514 1.25 (1.03, 1.53) 0.0276 
 Myocardial infarction (yes vs. no) 1.17 (0.94, 1.45) 0.1503 0.97 (0.80, 1.19) 0.7897 
 Type 1 diabetes (yes vs. no) 1.15 (0.83, 1.57) 0.4048 0.95 (0.71, 1.28) 0.7468 
 Type 2 diabetes (yes vs. no) 1.22 (0.87, 1.73) 0.2544 1.22 (0.89, 1.66) 0.2164 
Blood pressure      
 SBP (per 5 mmHg) 1.08 (1.04, 1.12) 0.0003 1.09 (1.06, 1.13) <0.0001 
 Mean SBP (per 5 mmHg) 1.17 (1.10, 1.24) <0.0001 1.24 (1.17, 1.30) <0.0001 
 DBP (per 5 mmHg) 1.07 (1.01, 1.13) 0.0184 1.05 (1.00, 1.09) 0.0462 
 Mean DBP (per 5 mmHg) 1.21 (1.11, 1.32) <0.0001 1.21 (1.11, 1.31) <0.0001 
 Pulse pressure (per 5 mmHg) 1.07 (1.02, 1.12) 0.0035 1.11 (1.07, 1.14) <0.0001 
 Pulse rate (per 1 bpm) 1.01 (1.00, 1.02) 0.0048 1.03 (1.03, 1.04) <0.0001 
 Mean pulse rate (per 1 bpm) 1.03 (1.02, 1.04) <0.0001 1.07 (1.06, 1.09) <0.0001 
 Hypertension (yes vs. no) 2.03 (1.56, 2.64) <0.0001 1.94 (1.61, 2.34) <0.0001 
Medications      
 ACE inhibitor use (yes vs. no) 1.54 (1.20, 1.97) 0.0008 1.51 (1.25, 1.82) <0.0001 
 ARB use (yes vs. no) 1.24 (0.87, 1.77) 0.2438 1.45 (1.13, 1.87) 0.0038 
 β-Blocker use (yes vs. no) 2.88 (1.79, 4.64) <0.0001 2.77 (1.95, 3.92) <0.0001 
 Calcium channel blocker use (yes vs. no) 1.41 (0.88, 2.26) 0.1524 1.70 (1.21, 2.40) 0.0022 
 Lipid-lowering agent use (yes vs. no) 1.15 (0.89, 1.48) 0.2740 1.36 (1.12, 1.65) 0.0020 
Lipids      
 Total cholesterol (per 10 mg/dL) 0.98 (0.95, 1.01) 0.1607 1.00 (0.98, 1.02) 0.9612 
 Mean total cholesterol (per 10 mg/dL) 1.02 (0.98, 1.05) 0.4334 1.06 (1.02, 1.10) 0.0027 
 Triglycerides (per 20% increase mg/dL) 1.06 (1.02, 1.10) 0.0054 1.07 (104, 1.10) <0.0001 
 Mean triglycerides (per 20% increase mg/dL) 1.09 (1.04, 1.15) <0.0001 1.14 (1.09, 1.18) <0.0001 
 HDL cholesterol (per 10 mg/dL) 0.89 (0.83, 0.93) 0.0034 0.96 (0.91, 1.01) 0.1113 
 Mean HDL cholesterol (per 10 mg/dL) 0.88 (0.81, 0.96) 0.0052 0.95 (0.88, 1.03) 0.2238 
 LDL cholesterol (per 10 mg/dL) 0.98 (0.95, 1.01) 0.1804 0.99 (0.96, 1.02) 0.4320 
 Mean LDL cholesterol (per 10 mg/dL) 1.02 (0.98, 1.06) 0.4137 1.04 (1.00, 1.08) 0.0719 
 Hyperlipidemia (yes vs. no) 1.02 (0.83, 1.26) 0.8568 1.27 (1.08, 1.51) 0.0052 
History of diabetes      
 Duration of type 1 diabetes (per 1 year) 1.11 (1.08, 1.13) <0.0001 1.09 (1.06, 1.12) <0.0001 
 Stimulated C-peptide (per 1 nmol/L*100)      
  Duration <5 years 0.99 (0.98, 1.00) 0.1345 1.00 (0.98, 1.01) 0.7650 
  Duration ≥5 years 1.00 (0.96, 1.05) 0.8726 0.98 (0.94, 1.02) 0.3685 
 Insulin dose (per 1 unit/kg/day) 1.19 (0.82, 1.73) 0.3718 1.26 (0.99, 1.62) 0.0648 
 Mean insulin dose (per 1 unit/kg/day) 2.07 (1.25, 3.43) 0.0046 1.88 (1.18, 3.00) 0.0083 
Kidney disease      
 eGFR (per 5 mL/min/1.73 m20.93 (0.90, 0.97) <0.0001 0.92 (0.90, 0.95) <0.0001 
  eGFR <60 mL/min/1.73 m2 (yes vs. no) 2.56 (1.36, 4.80) 0.0034 5.95 (3.36, 10.52) <0.0001 
  Any eGFR <60 mL/min/1.73 m2 (yes vs. no) 2.44 (1.45, 4.10) 0.0007 3.24 (2.12, 4.99) <0.0001 
 AER (per 20% increase mg/24 h) 1.06 (1.04, 1.07) <0.0001 1.06 (1.04, 1.07) <0.0001 
  Sustained AER ≥30 mg/24 h (yes vs. no) 2.45 (1.89, 3.17) <0.0001 2.36 (1.86, 3.00) <0.0001 
  Any sustained AER ≥30 mg/24 h (yes vs. no) 2.50 (1.98, 3.15) <0.0001 2.87 (2.32, 3.53) <0.0001 
  AER ≥300 mg/24 h (yes vs. no) 3.81 (2.42, 6.01) <0.0001 3.25 (2.17, 4.87) <0.0001 
  Any AER ≥300 mg/24 h (yes vs. no) 4.02 (2.81, 5.75) <0.0001 3.30 (2.36, 4.59) <0.0001 
Hypoglycemia      
 Coma and/or seizure (yes vs. no) 1.09 (0.61, 1.97) 0.7636 1.03 (0.73, 1.45) 0.8697 
 Requiring assistance (yes vs. no) 1.36 (0.96, 1.92) 0.0837 1.13 (0.88, 1.45) 0.3264 

Data are based on 1,386 participants with at least two DPN assessments or 1,434 participants with at least two CAN assessments during DCCT/EDIC. ORs and P values were generated using GEE models, adjusting for age and updated mean HbA1c. Triglyceride and AER values were log transformed, and the ORs are presented per 20% increase in the covariate (1.2β). ARB, angiotensin II receptor blocker.

*

B, baseline value; C, current value; M, updated mean value; A, cumulative incidence (e.g., any use). Covariates classified as C, M, and A enter into the analyses as time-dependent covariates.

For CAN, secondary prevention versus primary intervention cohort, greater BMI, behavioral risk factors (cigarette smoking, alcohol use, sedentary activity), higher SBP and DBP and pulse pressure, higher pulse rate, higher total cholesterol and triglycerides levels, longer duration of type 1 diabetes, higher daily insulin dose, a higher prevalence of hypertension and hyperlipidemia, and any kidney disease were associated with higher odds of CAN after adjustment for age and updated mean HbA1c (Table 3). LDL cholesterol was no longer significant after adjustment for age and updated mean HbA1c.

The use of ACE inhibitors and β-adrenoceptor blocking medications (β-blockers) during EDIC was significantly associated with higher odds of DPN and CAN.

Final Multivariate Models

The final multivariable GEE models for DPN and CAN, with covariates listed in order of the unsigned covariate Z-test values (or P values) to designate relative importance in the models, are shown in Table 4. For DPN, higher mean HbA1c was the most significant risk factor (OR = 1.63 per 1 HbA1c percentage point, 95% CI 1.51, 1.77, Z = 11.81, P < 0.0001), followed closely by older age (OR = 1.42 per 5 years, 95% CI 1.30, 1.55, Z = 8.05, P < 0.0001). Other nominally significant factors, in order of significance, were longer duration of type 1 diabetes (Z = 7.66, P < 0.0001), greater height (Z = 7.31, P < 0.0001), presence of any AER ≥300 mg/24 hour (Z = 4.12, P < 0.0001), higher mean pulse rate (Z = 3.92, P < 0.0001), β-blocker use (Z = 3.38, P = 0.0007), and presence of any sustained AER ≥30 mg/24 hour (Z = 2.76, P = 0.0058).

Table 4

Final multivariable GEE models for DPN and CAN as a function of baseline and time-dependent covariates

DPN OR (95% CI) Z-test value P value 
Mean HbA1c (per 1%) 1.63 (1.51, 1.77) 11.81 <0.0001 
Age (per 5 years) 1.42 (1.30, 1.55) 8.05 <0.0001 
Duration of type 1 diabetes (per 1 year) 1.10 (1.08, 1.13) 7.66 <0.0001 
Height (per 5 cm) 1.27 (1.19, 1.35) 7.31 <0.0001 
Any AER ≥300 mg/24 h (yes vs. no) 2.44 (1.60, 3.73) 4.12 <0.0001 
Mean pulse rate (per 1 bpm) 1.03 (1.01, 1.04) 3.92 <0.0001 
β-Blocker use (yes vs. no) 2.50 (1.47, 4.24) 3.38 0.0007 
Any sustained AER ≥30 mg/24 h (yes vs. no) 1.48 (1.12, 1.96) 2.76 0.0058 
CAN OR (95% CI) Z-test value P value 
Age (per 5 years) 1.50 (1.39, 1.62) 10.43 <0.0001 
Mean HbA1c (per 1%) 1.27 (1.17, 1.37) 6.04 <0.0001 
Any sustained AER ≥30 mg/24 h (yes vs. no) 1.95 (1.57, 2.43) 5.96 <0.0001 
Duration of type 1 diabetes (per 1 year) 1.07 (1.05, 1.10) 5.73 <0.0001 
Mean pulse rate (per 1 bpm) 1.04 (1.02, 1.06) 4.62 <0.0001 
Mean SBP (per 5 mmHg) 1.11 (1.05, 1.17) 3.75 0.0002 
β-Blocker use (yes vs. no) 2.01 (1.37, 2.93) 3.60 0.0003 
eGFR <60 mL/min/1.73 m2 (yes vs. no) 2.88 (1.59, 5.19) 3.51 0.0005 
Pulse rate (per 1 bpm) 1.02 (1.00, 1.03) 2.84 0.0045 
Cigarette smoker (yes vs. no) 1.31 (1.04, 1.66) 2.28 0.0226 
DPN OR (95% CI) Z-test value P value 
Mean HbA1c (per 1%) 1.63 (1.51, 1.77) 11.81 <0.0001 
Age (per 5 years) 1.42 (1.30, 1.55) 8.05 <0.0001 
Duration of type 1 diabetes (per 1 year) 1.10 (1.08, 1.13) 7.66 <0.0001 
Height (per 5 cm) 1.27 (1.19, 1.35) 7.31 <0.0001 
Any AER ≥300 mg/24 h (yes vs. no) 2.44 (1.60, 3.73) 4.12 <0.0001 
Mean pulse rate (per 1 bpm) 1.03 (1.01, 1.04) 3.92 <0.0001 
β-Blocker use (yes vs. no) 2.50 (1.47, 4.24) 3.38 0.0007 
Any sustained AER ≥30 mg/24 h (yes vs. no) 1.48 (1.12, 1.96) 2.76 0.0058 
CAN OR (95% CI) Z-test value P value 
Age (per 5 years) 1.50 (1.39, 1.62) 10.43 <0.0001 
Mean HbA1c (per 1%) 1.27 (1.17, 1.37) 6.04 <0.0001 
Any sustained AER ≥30 mg/24 h (yes vs. no) 1.95 (1.57, 2.43) 5.96 <0.0001 
Duration of type 1 diabetes (per 1 year) 1.07 (1.05, 1.10) 5.73 <0.0001 
Mean pulse rate (per 1 bpm) 1.04 (1.02, 1.06) 4.62 <0.0001 
Mean SBP (per 5 mmHg) 1.11 (1.05, 1.17) 3.75 0.0002 
β-Blocker use (yes vs. no) 2.01 (1.37, 2.93) 3.60 0.0003 
eGFR <60 mL/min/1.73 m2 (yes vs. no) 2.88 (1.59, 5.19) 3.51 0.0005 
Pulse rate (per 1 bpm) 1.02 (1.00, 1.03) 2.84 0.0045 
Cigarette smoker (yes vs. no) 1.31 (1.04, 1.66) 2.28 0.0226 

Data are based on 1,386 participants with at least two DPN assessments or 1,434 participants with at least two CAN assessments during DCCT/EDIC. Odds ratios and P values were generated using GEE models. Covariates are listed in the order of significance as indicated by the Z-test value.

In the multivariable GEE model for CAN (Table 4), older age (OR 1.50 per 5 years, 95% CI 1.39, 1.62, Z = 10.43, P < 0.0001) was the most significant risk factor, followed by higher mean HbA1c (OR 1.27 per 1 HbA1c percentage point, 95% CI 1.17, 1.37, Z = 6.04, P < 0.0001). Other nominally significant factors, in order of significance, were presence of any sustained AER ≥30 mg/24 h (Z = 5.96, P < 0.0001), longer duration of type 1 diabetes (Z = 5.73, P < 0.0001), higher mean pulse rate (Z = 4.62, P < 0.0001), higher mean SBP (Z = 3.75, P = 0.0002), β-blocker use (Z = 3.60, P = 0.0003), presence of eGFR <60 mL/min/1.73 m2 (Z = 3.51, P = 0.0005), higher pulse rate (Z = 2.84, P = 0.0045), and cigarette smoking (Z = 2.28, P = 0.0226).

A sensitivity analysis was conducted starting with the complete set of variables in the model and using a backward elimination approach to select covariates (Supplementary Table 3). This technique selected a similar set of covariates for the DPN with the exception that mean pulse rate and any sustained AER ≥30 mg/24 h were not included in the backward elimination model while mean daily insulin dose and sex were marginally significant. For CAN, mean SBP was not included in the backward elimination model whereas the most recent AER was as a continuous measurement.

Supplementary Table 4 presents the final model build for DPN and CAN including variables for diabetic retinopathy. In both models, a similar set of risk factors was retained in the final selection in addition to any proliferative diabetic retinopathy for both DPN and CAN and any clinically significant macular edema for CAN.

These analyses provide a thorough evaluation of numerous clinical features and risk factors, considered alone and jointly, associated with DPN and CAN in a large cohort of participants with type 1 diabetes who have been phenotyped with detailed and standardized neuropathy evaluations and risk factors obtained longitudinally over more than 20 years. Retention has been extremely high, and we have demonstrated high prevalence rates for both DPN (33%) and CAN (44%), which continued to rise over the duration of the study. Given the persistent increase in diabetes prevalence and incidence, these findings impact medical care across a spectrum of providers including general practitioners, internal medicine physicians, endocrinologists, neurologists, cardiologists, gastroenterologists, and pain specialists.

We have previously reported that intensive diabetes therapy achieving lower glucose control markedly reduced the risk of DPN and CAN during the DCCT compared with conventional therapy (13,22), an effect that persisted through much of EDIC follow-up (3,11). In concert with these prior reports, the data reported here underscore the important role of ongoing glycemic management. In multivariable models, higher mean HbA1c was the most significant risk factor for DPN and strongly associated with CAN.

Beyond glycemia and diabetes duration, our results confirm strong associations with kidney disease and elevated blood pressure for both DPN and CAN, in age and glucose adjusted models. Thirty years ago, the Pittsburgh Epidemiology of Diabetes Complications study, involving 400 participants with type 1 diabetes, reported a cross-sectional association of hypertension with DPN (23) and CAN (24). This was later confirmed in prospective analyses (25,26). Other studies of individuals with type 1 diabetes including the European Diabetes Prospective Complications Study (EURODIAB) (27) and the SEARCH for Diabetes in Youth study (28) have reported similar findings for DPN and/or CAN. In addition, hypertension is closely linked with kidney disease in type 1 diabetes (26). The current analyses suggest that hypertension and kidney disease (higher AER and lower eGFR) are independently associated with DPN and CAN. The mechanisms linking hypertension to DPN are unclear but may involve a complex set of pathways interlinking nerve function and energy production with a dysfunctional neural vascular supply (29).

In our multivariable GEE regression analyses, height was retained in the final model. The unadjusted and minimally adjusted associations between height and weight, separately, with DPN (P < 0.0001 for both) were much stronger than the association between BMI and DPN (P = 0.0497) (Table 3). We therefore decided to keep both height and weight in the second block during the multivariable modeling process, and we removed BMI. The adverse influence of increasing height on peripheral nerve function is well known and attributed in part to the tapering of axon diameter at a given distance from the cell body (30,31).

After adjusting for age and glucose control, elevated triglycerides were a risk factor for DPN and CAN, while lower HDL cholesterol levels increased DPN risk and higher total cholesterol levels increased CAN risk. Our findings are in agreement with other studies identifying dyslipidemia as a risk factor in the development and progression of CAN (24,27,28) and DPN (23,27). Traditionally, elevated triglycerides, in concert with low HDL cholesterol, are highly prevalent in type 2 diabetes. However, with the contemporary changes in clinical phenotypes in individuals with type 1 diabetes, particularly with the increase in overweight and obese individuals, metabolic syndrome and associated dyslipidemia are now more prevalent in type 1 diabetes as well (28). Observational evidence also suggests that lowering triglyceride levels with fibrates may delay development of DPN (32). The role of dyslipidemia is further supported by recent work in experimental diabetes where tissue-specific changes in fatty acid metabolism are associated with DPN, independent of glycemic control (33).

Occasional or regular alcohol use did not increase the risk of DPN over the course of the DCCT/EDIC study in any of the models despite previous research demonstrating that alcohol is a peripheral neurotoxicant (34). One study of the relationship between peripheral neuropathy and nutritional status in individuals with chronic alcohol abuse demonstrated that the total lifetime dose of alcohol is an essential independent factor for acquiring an alcohol-related peripheral neuropathy, usually after a cumulative exposure exceeding about 15 kg alcohol/kg body weight (34). In the DCCT/EDIC, the measure of alcohol use was likely imprecise to determine an effect on DPN.

Cigarette smoking has generally been considered a risk factor for DPN (35). However, we did not establish smoking as a major risk factor for DPN over the course of DCCT/EDIC study. This observation differs from the results of several other investigations (23,25,27,36) but is consistent with our earlier report that smoking was not associated with confirmed clinical neuropathy during DCCT (11). It is unclear why our results differ from these previous studies; however, our analyses may be limited since the cigarette smoking measurement available in DCCT/EDIC did not quantify pack-years of exposure. However, smoking has been consistently associated with CAN (23,25,27), in line with our current findings.

Other interesting findings of these analyses include the strong associations between both DPN and CAN and the use of β-blocking medications. Among the various classes of medications used to treat individuals with diabetes, β-blockers may directly affect measures of CAN, given their reported blunting effect on heart rate. As such, improvement in heart rate variability would be expected, particularly with β1-selective drugs that may augment vagally mediated heart rate variability (3,37). Although evidence for the benefit of β-blockers in modulating the impaired cardiac autonomic regulation during high sympathetic stress advocates for their use in patients with heart failure and/or post myocardial infarction (38), these agents are not currently recommended for the treatment of CAN (1). The strong association between use of β-blockers and DPN, demonstrated in this study, was not expected. β-Blockers are not first-line drugs in the treatment of hypertension in individuals with type 1 diabetes. Therefore, it is reasonable to assume that these associations do not actually reflect a direct adverse effect on autonomic or peripheral nerve function but are likely driven by participants with more advanced disease states and/or overall worse risk factor profiles (e.g., cardiovascular disease, heart failure, coronary artery disease, more severe hypertension) who are prescribed β-blocking agents as second- or third-line therapy. In our multivariable model, many of the same factors were also shown to contribute to DPN and CAN risk, supporting a potential confounding effect of use of β-blocker agents.

One limitation of this study is that during EDIC we only collected information on the use of major classes of medication agents (i.e., β-blockers, ACE inhibitors, angiotensin receptor blockers, statins, etc.). Unfortunately, information on the specific type of agents from each class is not available, preventing more in-depth analyses that could take into account the differential effects of β-blocker groupings. Another study limitation is that in contrast with CAN, DPN was only assessed twice during the DCCT and once again in EDIC. In addition, the concept of small-fiber neuropathy was not well established at the start of the DCCT, and thus currently accepted methods to assess primarily small-fiber dysfunction were not available.

Strengths of the current study include the detailed phenotypic characteristics, including demographics and multiple traditional and diabetes-related risk factors, in addition to the extensive, sensitive, and standardized evaluations of DPN and CAN available longitudinally in this large cohort of participants with type 1 diabetes spanning more than 20 years of follow-up. Different than many other studies of shorter duration and with smaller sample sizes, the DCCT/EDIC study has allowed for a comprehensive multifactorial evaluation of risk factors simultaneously, which is unique.

In summary, in these comprehensive analyses, we found that higher mean HbA1c and older age were the strongest risk factors for both DPN and CAN. Although hyperglycemia is strongly associated with the development of DPN and CAN, we found that other risk factors may be associated as well. Given the oberservational nature of our analyses, the risk factors we have identified are not necessarily causal in their relationship with DPN or CAN. However, we have demonstrated that individuals with more favorable risk factor profiles had lower prevalence of DPN and CAN in this type 1 diabetes population. Beyond the insights into underlying biological mechanisms described above, these findings provide critical information on the spectrum of risk factors and the phenotypes of patients with neuropathy, one of the most challenging diabetes complications. These data can be used in the design of new interventional trials, to stimulate actions to overcome a paradoxical lag in neuropathy-related drug discovery, as well as for personalized approaches to neuropathy prevention and treatment among individuals with type 1 diabetes.

Clinical trial reg. nos. NCT00360815 and NCT00360893, clinicaltrials.gov

Funding. DCCT/EDIC has been supported by cooperative agreement grants (1982–1993, 2012–2017, 2017–2022) and contracts (1982–2012) with the Division of Diabetes Endocrinology and Metabolic Diseases of the National Institute of Diabetes and Digestive and Kidney Diseases (current grant numbers U01 DK094176 and U01 DK094157) and by the National Eye Institute, the National Institute of Neurological Disorders and Stroke, the General Clinical Research Centers Program (1993–2007), and the Clinical Translational Science Center Program (2006 to present), Bethesda, MD. Industry contributors have had no role in the DCCT/EDIC study but have provided free or discounted supplies or equipment to support participants’ adherence to the study: Abbott Diabetes Care (Alameda, CA), Animas (West Chester, PA), Bayer Diabetes Care (North America Headquarters, Tarrytown, NY), Becton Dickinson (Franklin Lakes, NJ), Eli Lilly (Indianapolis, IN), Extend Nutrition (St. Louis, MO), Insulet Corporation (Bedford, MA), Lifescan (Milpitas, CA), Medtronic Diabetes (Minneapolis, MN), Nipro Home Diagnostics (Fort Lauderdale, FL), Nova Diabetes Care (Billerica, MA), Omron (Shelton, CT), Perrigo Diabetes Care (Allegan, MI), Roche Diabetes Care (Indianapolis, IN), Sanofi (Bridgewater, NJ).

Duality of Interest. No other potential conflicts of interest relevant to this article were reported.

Author Contributions. B.H.B., R.A.G.-K., and R.P.-B. designed the analyses and wrote the manuscript. J.W.A., E.L.F., C.L.M., N.H.W., T.J.O., M.L.-V., and J.M.L. wrote portions of the manuscript and reviewed and edited the manuscript. B.H.B. conducted the statistical analyses and prepared the tables. B.H.B. had final responsibility for the decision to submit for publication. B.H.B. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

1.
Pop-Busui
R
,
Boulton
AJ
,
Feldman
EL
, et al
.
Diabetic neuropathy: a position statement by the American Diabetes Association
.
Diabetes Care
2017
;
40
:
136
154
2.
Sadosky
A
,
Mardekian
J
,
Parsons
B
,
Hopps
M
,
Bienen
EJ
,
Markman
J
.
Healthcare utilization and costs in diabetes relative to the clinical spectrum of painful diabetic peripheral neuropathy
.
J Diabetes Complications
2015
;
29
:
212
217
3.
Pop-Busui
R
,
Low
PA
,
Waberski
BH
, et al.;
DCCT/EDIC Research Group
.
Effects of prior intensive insulin therapy on cardiac autonomic nervous system function in type 1 diabetes mellitus: the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications study (DCCT/EDIC)
.
Circulation
2009
;
119
:
2886
2893
4.
Orchard
TJ
,
Dorman
JS
,
Maser
RE
, et al
.
Factors associated with avoidance of severe complications after 25 yr of IDDM. Pittsburgh Epidemiology of Diabetes Complications Study I
.
Diabetes Care
1990
;
13
:
741
747
5.
Ziegler
D
,
Zentai
CP
,
Perz
S
, et al.;
KORA Study Group
.
Prediction of mortality using measures of cardiac autonomic dysfunction in the diabetic and nondiabetic population: the MONICA/KORA Augsburg Cohort Study
.
Diabetes Care
2008
;
31
:
556
561
6.
Sacre
JW
,
Franjic
B
,
Jellis
CL
,
Jenkins
C
,
Coombes
JS
,
Marwick
TH
.
Association of cardiac autonomic neuropathy with subclinical myocardial dysfunction in type 2 diabetes
.
JACC Cardiovasc Imaging
2010
;
3
:
1207
1215
7.
Pettitt
DJ
,
Talton
J
,
Dabelea
D
, et al.;
SEARCH for Diabetes in Youth Study Group
.
Prevalence of diabetes in U.S. youth in 2009: the SEARCH for Diabetes in Youth study
.
Diabetes Care
2014
;
37
:
402
408
8.
Dabelea
D
,
Mayer-Davis
EJ
,
Saydah
S
, et al.;
SEARCH for Diabetes in Youth Study
.
Prevalence of type 1 and type 2 diabetes among children and adolescents from 2001 to 2009
.
JAMA
2014
;
311
:
1778
1786
9.
Bullard
KM
,
Cowie
CC
,
Lessem
SE
, et al
.
Prevalence of diagnosed diabetes in adults by diabetes type - United States, 2016
.
MMWR Morb Mortal Wkly Rep
2018
;
67
:
359
361
10.
International Diabetes Federation
.
IDF Diabetes Atlas, 9th edition [Internet], 2019. Available from https://diabetesatlas.org
11.
Martin
CL
,
Albers
JW
,
Pop-Busui
R
;
DCCT/EDIC Research Group
.
Neuropathy and related findings in the Diabetes Control and Complications trial/Epidemiology of Diabetes Interventions and Complications study
.
Diabetes Care
2014
;
37
:
31
38
12.
Pop-Busui
R
,
Braffett
BH
,
Zinman
B
, et al.;
DCCT/EDIC Research Group
.
Cardiovascular autonomic neuropathy and cardiovascular outcomes in the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) study
.
Diabetes Care
2017
;
40
:
94
100
13.
Nathan
DM
,
Genuth
S
,
Lachin
J
, et al.;
Diabetes Control and Complications Trial Research Group
.
The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus
.
N Engl J Med
1993
;
329
:
977
986
14.
The DCCT/EDIC Research Group
.
Epidemiology of Diabetes Interventions and Complications (EDIC). Design, implementation, and preliminary results of a long-term follow-up of the Diabetes Control and Complications Trial cohort
.
Diabetes Care
1999
;
22
:
99
111
15.
The DCCT Research Group
.
Implementation of treatment protocols in the Diabetes Control and Complications Trial
.
Diabetes Care
1995
;
18
:
361
376
16.
Younes
N
,
Cleary
PA
,
Steffes
MW
, et al.;
DCCT/EDIC Research Group
.
Comparison of urinary albumin-creatinine ratio and albumin excretion rate in the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications study
.
Clin J Am Soc Nephrol
2010
;
5
:
1235
1242
17.
Friedewald
WT
,
Levy
RI
,
Fredrickson
DS
.
Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge
.
Clin Chem
1972
;
18
:
499
502
18.
Diabetes Control and Complications Trial (DCCT)/Epidemiology of Diabetes Interventions and Complications (EDIC) Research Group
.
Effect of intensive diabetes therapy on the progression of diabetic retinopathy in patients with type 1 diabetes: 18 years of follow-up in the DCCT/EDIC. Diabetes 2015;64:631–642
19.
Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Research Group
.
Risk factors for cardiovascular disease in type 1 diabetes
.
Diabetes
2016
;
65
:
1370
1379
20.
Hainsworth
DP
,
Bebu
I
,
Aiello
LP
, et al.;
Diabetes Control and Complications Trial (DCCT)/Epidemiology of Diabetes Interventions and Complications (EDIC) Research Group
.
Risk factors for retinopathy in type 1 diabetes: the DCCT/EDIC study
.
Diabetes Care
2019
;
42
:
875
882
21.
Perkins
BA
,
Bebu
I
,
de Boer
IH
, et al.;
Diabetes Control and Complications Trial (DCCT)/Epidemiology of Diabetes Interventions and Complications (EDIC) Research Group
.
Risk factors for kidney disease in type 1 diabetes
.
Diabetes Care
2019
;
42
:
883
890
22.
The DCCT Research Group
.
The effect of intensive diabetes therapy on measures of autonomic nervous system function in the Diabetes Control and Complications Trial (DCCT)
.
Diabetologia
1998
;
41
:
416
423
23.
Maser
RE
,
Steenkiste
AR
,
Dorman
JS
, et al
.
Epidemiological correlates of diabetic neuropathy. Report from Pittsburgh Epidemiology of Diabetes Complications Study
.
Diabetes
1989
;
38
:
1456
1461
24.
Maser
RE
,
Pfeifer
MA
,
Dorman
JS
,
Kuller
LH
,
Becker
DJ
,
Orchard
TJ
.
Diabetic autonomic neuropathy and cardiovascular risk. Pittsburgh Epidemiology of Diabetes Complications Study III
.
Arch Intern Med
1990
;
150
:
1218
1222
25.
Forrest
KY
,
Maser
RE
,
Pambianco
G
,
Becker
DJ
,
Orchard
TJ
.
Hypertension as a risk factor for diabetic neuropathy: a prospective study
.
Diabetes
1997
;
46
:
665
670
26.
Stella
P
,
Ellis
D
,
Maser
RE
,
Orchard
TJ
.
Cardiovascular autonomic neuropathy (expiration and inspiration ratio) in type 1 diabetes. Incidence and predictors
.
J Diabetes Complications
2000
;
14
:
1
6
27.
Tesfaye
S
,
Chaturvedi
N
,
Eaton
SE
, et al.;
EURODIAB Prospective Complications Study Group
.
Vascular risk factors and diabetic neuropathy
.
N Engl J Med
2005
;
352
:
341
350
28.
Dabelea
D
,
Stafford
JM
,
Mayer-Davis
EJ
, et al.;
SEARCH for Diabetes in Youth Research Group
.
Association of type 1 diabetes vs type 2 diabetes diagnosed during childhood and adolescence with complications during teenage years and young adulthood
.
JAMA
2017
;
317
:
825
835
29.
Feldman
EL
,
Nave
KA
,
Jensen
TS
,
Bennett
DLH
.
New horizons in diabetic neuropathy: mechanisms, bioenergetics, and pain
.
Neuron
2017
;
93
:
1296
1313
30.
Rivner
MH
,
Swift
TR
,
Malik
K
.
Influence of age and height on nerve conduction
.
Muscle Nerve
2001
;
24
:
1134
1141
31.
Stetson
DS
,
Albers
JW
,
Silverstein
BA
,
Wolfe
RA
.
Effects of age, sex, and anthropometric factors on nerve conduction measures
.
Muscle Nerve
1992
;
15
:
1095
1104
32.
Davis
TM
,
Yeap
BB
,
Davis
WA
,
Bruce
DG
.
Lipid-lowering therapy and peripheral sensory neuropathy in type 2 diabetes: the Fremantle Diabetes Study
.
Diabetologia
2008
;
51
:
562
566
33.
Sas
KM
,
Kayampilly
P
,
Byun
J
, et al
.
Tissue-specific metabolic reprogramming drives nutrient flux in diabetic complications
.
JCI Insight
2016
;
1
:
e86976
34.
Estruch
R
,
Nicolás
JM
,
Villegas
E
,
Junqué
A
,
Urbano-Márquez
A
.
Relationship between ethanol-related diseases and nutritional status in chronically alcoholic men
.
Alcohol Alcohol
1993
;
28
:
543
550
35.
Śliwińska-Mossoń
M
,
Milnerowicz
H
.
The impact of smoking on the development of diabetes and its complications
.
Diab Vasc Dis Res
2017
;
14
:
265
276
36.
Clair
C
,
Cohen
MJ
,
Eichler
F
,
Selby
KJ
,
Rigotti
NA
.
The effect of cigarette smoking on diabetic peripheral neuropathy: a systematic review and meta-analysis
.
J Gen Intern Med
2015
;
30
:
1193
1203
37.
Aronson
D
,
Burger
AJ
.
Effect of beta-blockade on heart rate variability in decompensated heart failure
.
Int J Cardiol
2001
;
79
:
31
39
38.
Yancy
CW
,
Jessup
M
,
Bozkurt
B
, et al
.
2017 ACC/AHA/HFSA focused update of the 2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America
.
J Card Fail
2017
;
23
:
628
651
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at https://www.diabetesjournals.org/content/license.

Supplementary data