OBJECTIVE

Intensive glycemic control reduces the risk of kidney, retinal, and neurologic complications in type 1 diabetes (T1D), but whether it reduces the risk of lower-extremity complications is unknown. We examined whether former intensive versus conventional glycemic control among Diabetes Control and Complications Trial (DCCT) participants with T1D reduced the long-term risk of diabetic foot ulcers (DFUs) and lower-extremity amputations (LEAs) in the subsequent Epidemiology of Diabetes Interventions and Complications (EDIC) study.

RESEARCH DESIGN AND METHODS

DCCT participants (n = 1,441) completed 6.5 years on average of intensive versus conventional diabetes treatment, after which 1,408 were enrolled in EDIC and followed annually over 23 years for DFU and LEA occurrences by physical examination. Multivariable Cox proportional hazard regression models estimated associations of DCCT treatment assignment and time-updated exposures with DFU or LEA.

RESULTS

Intensive versus conventional glycemic control was associated with a significant risk reduction for all DFUs (hazard ratio 0.77 [95% CI 0.60, 0.97]) and a similar magnitude but nonsignificant risk reduction for first-recorded DFUs (0.78 [0.59, 1.03]) and first LEAs (0.70 [0.36, 1.36]). In adjusted Cox models, clinical neuropathy, lower sural nerve conduction velocity, and cardiovascular autonomic neuropathy were associated with higher DFU risk; estimated glomerular filtration rate <60 mL/min/1.73 m2, albuminuria, and macular edema with higher LEA risk; and any retinopathy and greater time-weighted mean DCCT/EDIC HbA1c with higher risk of both outcomes (P < 0.05).

CONCLUSIONS

Early intensive glycemic control decreases long-term DFU risk, the most important antecedent in the causal pathway to LEA.

Diabetes greatly increases the risk of lower-extremity amputations (LEAs), with relative risk estimates ranging from 7.4 to 41.3 in people with and without diabetes (1). Diabetic foot ulcers (DFUs) remain the principal antecedent to LEA (2), preceding ∼80% of LEAs and leading to amputation in 15% of occurrences (3). Although some complications have declined in frequency in people with type 1 diabetes (T1D) with more recent disease onset, the frequency of LEA has remained constant (4,5).

Hyperglycemia increases the risk of multiple micro- and macrovascular complications believed to lead to DFU and LEA (6). Several prospective studies have assessed associations between diabetic complications and risk of lower-extremity outcomes (79). However, none of these studies were performed prospectively in people with T1D enrolled at an early stage of illness with few or no complications, and none included an annual physical examination, including targeted foot examination, with extensive neurovascular testing that uses a common protocol over a prolonged follow-up period. To our knowledge, no randomized controlled trial has been conducted with adequate statistical power to demonstrate whether intensive glycemic control alters the risk of DFU or LEA in T1D.

The Diabetes Control and Complications Trial (DCCT) (1983–1993) was the first controlled clinical trial to demonstrate definitively the effectiveness of intensive glycemic control in reducing the occurrence and progression of several diabetic microvascular complications, including retinopathy, kidney disease, and neuropathy (10). In the current analyses, we examine the effects of intensive versus conventional glycemic control during the DCCT on the subsequent risk of DFU and LEA in the Epidemiology of Diabetes Interventions and Complications (EDIC) study, the observational follow-up of DCCT. EDIC participants were followed for >23 years, and the contributions of other diabetes characteristics and complications to risk for DFU and LEA were also studied (11).

Between 1983 and 1989, 1,441 participants age 13–39 years with a T1D duration of 1–15 years enrolled in DCCT, a multicenter controlled clinical trial designed to compare the effects of diabetes intensive treatment (INT) and conventional treatment (CON) on diabetes complications (10). Approximately one-half of the cohort (n = 711) was randomly assigned to receive intensive therapy, consisting of three or more daily insulin injections or insulin infusion pump use, with the goal of maintaining glycemic levels as close to the nondiabetic range (HbA1c <6.05% [42.6 mmol/mol]) as safely possible. The remainder (n = 730) were assigned to conventional therapy, consisting of one to two daily injections of insulin without specific glycemic targets. Subjects with a history of cardiovascular disease, hypertension, dyslipidemia, or neuropathy requiring treatment were not eligible to participate. The research protocol was approved by the institutional review boards at all sites, and all participants provided written informed consent.

In 1993, after an average 6.5 years of follow-up (range 3–9 years), 1,422 subjects (99% of the original cohort) completed a closeout visit. Participants in the CON group were encouraged to adopt intensive therapy, and all participants were referred back to their health care providers for ongoing diabetes care. In 1994, 96% of the surviving cohort agreed to participate in the EDIC observational study (1994 to present) designed to evaluate the longer-term effects of glycemic control on the risk of both micro- and macrovascular complications (11). After 23 years of follow-up (up to 2017), 1,190 participants (93% of the surviving and 83% of the original DCCT cohort) continue to be followed in this ongoing study. This analysis includes data on 1,408 participants with follow-up between EDIC years 1 and 23.

Evaluations

Annual EDIC visits included a detailed medical history and physical examination with measurements of height, weight, sitting blood pressure, and Doppler arm/leg systolic blood pressure to calculate the ankle-brachial pressure index (ABI) (11). Blood samples were collected annually and assayed centrally for HbA1c by using high-performance ion-exchange liquid chromatography (12). DCCT cumulative glycemic exposure was obtained from the updated mean HbA1c during the DCCT, while the DCCT/EDIC time-weighted mean HbA1c was used to reflect the cumulative glycemic exposure during both DCCT and EDIC. Fasting lipids (triglycerides, total and HDL cholesterol) and albumin excretion rate (AER) were measured in alternate years during EDIC and evaluated centrally. LDL cholesterol was calculated using the Friedewald equation (13). AER was measured from 4-h urine samples from DCCT baseline to EDIC year 18 and subsequently from spot urine samples, with AER estimated using the ratio of urine albumin-to-creatinine concentrations (14). Estimated glomerular filtration rate (eGFR) was calculated from serum creatinine measured annually in combination with age, sex, and race using the Chronic Kidney Disease Epidemiology Collaboration equation. Kidney disease was defined as a sustained reduction in eGFR (<60 mL/min/1.73 m2) or sustained microalbuminuria (AER ≥30 mg/24 h) on two or more consecutive visits.

Neurologic evaluations were conducted in EDIC year 13/14. Confirmed clinical neuropathy was defined as a combination of the presence of signs and symptoms consistent with distal symmetrical polyneuropathy plus nerve conduction study abnormalities involving two or more nerves among the median, peroneal, and sural nerves (15). Cardiovascular autonomic neuropathy (CAN) was evaluated with standardized cardiovascular reflex tests in EDIC years 13/14 and 16/17 (16). CAN was defined as either an R-R variation <15, an R-R variation 15–19.9 combined with a Valsalva ratio ≤1.5, or a decrease of >10 mmHg in diastolic blood pressure upon standing (16). Presence of retinopathy and macular edema were assessed by standardized seven-field stereoscopic fundus photography in 25% of participants every year during EDIC and centrally graded. Retinopathy severity was determined using the Early Treatment Diabetic Retinopathy Study (ETDRS) scale (11).

DFUs and LEAs

DFU occurrence was not recorded during the DCCT. During the annual EDIC physical examination, feet were inspected for abnormalities and ulcers. The presence of a DFU was assessed by examination and defined in the EDIC manual of operations as a traumatic or nontraumatic excavation or loss of subcutaneous tissue in the foot with evidence of inflammation and/or infection that required medical or surgical treatment by a health professional in an office or hospital setting. We identified first-recorded and subsequent DFU occurrences during follow-up. As a DFU may not heal within 1 year, a DFU noted on two consecutive annual examinations in the same limb may indicate one nonhealing ulcer or a healed first-recorded DFU followed by an ipsilateral recurrence. We defined a second or subsequent DFU in the same foot only if present in nonconsecutive years or if located on the contralateral foot in the year after the first-recorded DFU. LEA at any level, including single toe, was identified as present during the annual physical examination beginning in October 2005 (EDIC year 12). The occurrence of multiple amputations in the same participant between EDIC years 13 and 23 could not be determined from EDIC data, as amputation laterality and ipsilateral reamputation were not recorded.

Statistical Methods

Participant characteristics by treatment group at the end of DCCT were summarized. Cox proportional hazards regression models were used to estimate associations of intensive diabetes treatment during the DCCT with the risk of the first-recorded DFU and first LEA separately during EDIC and under the counting process formulation of Anderson and Gill with robust SEs to examine the association of former diabetes treatment assignment with the risk of first or all DFUs (17). In each case, we tested and confirmed that the assumption of proportional hazards was not violated. The association between treatment assignment and first-recorded DFU was visualized graphically using a Kaplan-Meier plot. In an exploratory analysis, we investigated whether the effect of the intervention varied before and after 10 years of follow-up by adding a treatment-by-time interaction to the regression model.

Cox proportional hazards regression models were also used to evaluate the associations of covariates measured at DCCT closeout and repeatedly throughout EDIC (time dependent) with the risk of first-recorded DFU and first LEA separately. In analyses involving DCCT closeout covariates, regression models were adjusted for the potential confounding effects of DCCT closeout age and sex; in analyses involving time-varying covariates, regression models were adjusted for DCCT closeout age, sex, duration of diabetes, and the DCCT/EDIC time-weighted mean HbA1c as a time-dependent covariate. All analyses were conducted using R 3.4.0 statistical software (R Foundation for Statistical Computing, Vienna, Austria). Statistical significance was defined as a two-sided P < 0.05.

Of the 711 participants randomly assigned to the INT and the 730 participants assigned to the CON group, 699 and 709, respectively, were subsequently followed in EDIC. Participants at DCCT closeout were on average ∼33 years old with an average diabetes duration of ∼12 years. Participants in both treatment groups were similar at DCCT closeout for multiple characteristics, including smoking, blood pressure, serum creatinine, and plasma lipids. At the end of DCCT, several differences emerged associated with INT, including lower mean HbA1c and AER; lower frequency of moderate nonproliferative and proliferative retinopathy, macular edema, confirmed clinical neuropathy, and CAN; and greater mean BMI (Table 1). ABI measurements began in EDIC year 1. The frequency of ABI <0.9 or >1.3 was similar between treatment groups.

Table 1

Demographic characteristics of DCCT participants who participated in EDIC follow-up at DCCT closeout in 1993 by diabetes treatment assignment (N = 1,408)

CharacteristicINT (n = 699)CON (n = 709)
Age (years) 33.4 (7.0) 32.8 (7.0) 
Female sex 344 (49) 329 (46) 
Medical history   
 Duration of diabetes (years) 12.2 (4.9) 11.7 (4.8) 
 Current smoking 160 (23) 159 (22) 
Physical examination findings   
 BMI (kg/m226.6 (4.3) 25.0 (3.1) 
 Blood pressure (mmHg)   
  Systolic 116.5 (11.5) 116.5 (11.9) 
  Diastolic 74.8 (8.7) 74.3 (8.9) 
Laboratory values   
 HbA1c (%) 7.4 (1.2) 9.1 (1.6) 
 Sustained eGFR <60 mL/min/1.73 m2 1 (0) 2 (0) 
 AER   
  Median (IQR), mg/24 h 8.6 (5.8–14.4) 10.1 (5.8–20.2) 
  ≥30 mg/24 h 71 (10) 121 (17) 
 Serum creatinine (mg/dL) 0.7 (0.1) 0.7 (0.2) 
 Plasma lipids (mg/dL)   
  Total cholesterol 180.3 (30.5) 184.0 (37.6) 
  HDL cholesterol 50.9 (12.9) 51.8 (13.0) 
  LDL cholesterol 112.5 (27.1) 114.6 (32.0) 
  Triglycerides 84.3 (52.7) 87.9 (50.7) 
 Retinopathy   
  None/mild nonproliferative 560 (80) 489 (69) 
  Moderate nonproliferative 110 (16) 149 (21) 
  Severe nonproliferative 3 (0) 21 (3) 
  Proliferative 26 (4) 50 (7) 
 Macular edema 24 (3) 47 (7) 
 Any confirmed clinical neuropathy 64 (9) 121 (17) 
 Sural nerve conduction velocity (m/s) 45.5 (6.2) 43.2 (6.0) 
 Presence of CAN 43 (6) 57 (8) 
 ABI*   
  <0.9 46 (7) 50 (7) 
  0.9–1.3 593 (83) 595 (84) 
  >1.3 10 (1) 7 (1) 
CharacteristicINT (n = 699)CON (n = 709)
Age (years) 33.4 (7.0) 32.8 (7.0) 
Female sex 344 (49) 329 (46) 
Medical history   
 Duration of diabetes (years) 12.2 (4.9) 11.7 (4.8) 
 Current smoking 160 (23) 159 (22) 
Physical examination findings   
 BMI (kg/m226.6 (4.3) 25.0 (3.1) 
 Blood pressure (mmHg)   
  Systolic 116.5 (11.5) 116.5 (11.9) 
  Diastolic 74.8 (8.7) 74.3 (8.9) 
Laboratory values   
 HbA1c (%) 7.4 (1.2) 9.1 (1.6) 
 Sustained eGFR <60 mL/min/1.73 m2 1 (0) 2 (0) 
 AER   
  Median (IQR), mg/24 h 8.6 (5.8–14.4) 10.1 (5.8–20.2) 
  ≥30 mg/24 h 71 (10) 121 (17) 
 Serum creatinine (mg/dL) 0.7 (0.1) 0.7 (0.2) 
 Plasma lipids (mg/dL)   
  Total cholesterol 180.3 (30.5) 184.0 (37.6) 
  HDL cholesterol 50.9 (12.9) 51.8 (13.0) 
  LDL cholesterol 112.5 (27.1) 114.6 (32.0) 
  Triglycerides 84.3 (52.7) 87.9 (50.7) 
 Retinopathy   
  None/mild nonproliferative 560 (80) 489 (69) 
  Moderate nonproliferative 110 (16) 149 (21) 
  Severe nonproliferative 3 (0) 21 (3) 
  Proliferative 26 (4) 50 (7) 
 Macular edema 24 (3) 47 (7) 
 Any confirmed clinical neuropathy 64 (9) 121 (17) 
 Sural nerve conduction velocity (m/s) 45.5 (6.2) 43.2 (6.0) 
 Presence of CAN 43 (6) 57 (8) 
 ABI*   
  <0.9 46 (7) 50 (7) 
  0.9–1.3 593 (83) 595 (84) 
  >1.3 10 (1) 7 (1) 

Data are mean (SD) for continuous variables and n (%) for categorical variables unless otherwise indicated. IQR, interquartile range.

*

ABI was initially measured at EDIC year 1.

A total of 195 participants developed at least one DFU during 23 years of EDIC follow-up, with 48 developing multiple DFUs (two ulcers n = 31, three ulcers n = 11, four or more ulcers n = 6). More than one DFU in the same limb was identified in 21 participants on the basis of our criterion of recording in nonconsecutive years, with the remaining 27 participants having more than one ulcer in different limbs.

DFU occurred in 86 participants in the INT group and 109 in the CON group, for incidences of 6.0 and 7.8 per 1,000 person-years, respectively. The risk of first-recorded DFU was nominally, but not statistically significantly lower in the INT participants (hazard ratio [HR] 0.78 [95% CI 0.59, 1.03]). All DFUs, including first and subsequent lesions, totaled 117 among INT participants and 153 among CON participants, for incidences of 7.3 and 9.6 per 1,000 person-years, respectively. The risk of all DFUs was statistically significantly lower by 23% in the INT participants (0.77 [0.60, 0.97]).

As seen in the Kaplan-Meier curve in Fig. 1, cumulative incidence of first-recorded DFU was relatively low and similar between former treatment groups until ∼10 years of EDIC follow-up, when the curves began to diverge. Given the separation of the DFU cumulative incidence curves at 10 years, we estimated period-specific HRs for ≤10 years and >10 years. A significantly lower risk of first-recorded DFU seen after 10 years was associated with INT versus CON (HR 0.63 [95% CI 0.41, 0.97]) but not before 10 years (0.92 [0.63, 1.34]). The interaction term of DCCT treatment with time during EDIC follow-up was not significant (P = 0.19).

Figure 1

Kaplan-Meier curves of the association between former INT and CON groups and cumulative incidence of first-recorded DFU over time. Before EDIC, participants were followed, on average, for 6.5 years while participating in DCCT.

Figure 1

Kaplan-Meier curves of the association between former INT and CON groups and cumulative incidence of first-recorded DFU over time. Before EDIC, participants were followed, on average, for 6.5 years while participating in DCCT.

Close modal

There were 12 prior amputations reported at EDIC year 13, the year that amputation data collection began. An additional 24 amputations occurred between EDIC years 13 and 23. Of the 36 amputations, 15 occurred in the INT group and 21 in the CON group, for incidences of 1.0 and 1.4 per 1,000 person-years, respectively. The risk of amputation among INT participants was 30% lower than in CON participants, but this difference was not statistically significant (HR 0.70 [95% CI 0.36, 1.36]).

Several characteristics measured at DCCT closeout and adjusted for age, sex, and diabetes duration were associated with a higher risk of both DFU and LEA (Table 2). Statistically significant risk factors in common for DFU and LEA were DCCT closeout HbA1c, DCCT mean HbA1c, and confirmed clinical neuropathy. Greater age, AER ≥30 mg/24 h, CAN, and any retinopathy were associated with higher DFU risk, while greater BMI, current smoking, higher serum triglyceride concentration, macular edema, and ABI <0.9 were associated with higher LEA risk. Following adjustment for DCCT mean HbA1c, age, AER ≥30 mg/24 h, CAN, and any retinopathy remained statistically significantly associated with DFU risk, while age, BMI, serum triglyceride concentration, and macular edema were statistically significantly associated with LEA risk.

Table 2

Associations of covariates at DCCT closeout with risk of first-recorded DFU or LEA during EDIC follow-up

DFULEA
Model 1 adjusted
HR (95% CI)
Model 2 adjusted
HR (95% CI)
Model 1 adjusted
HR (95% CI)
Model 2 adjusted
HR (95% CI)
Male sex 1.23 (0.92, 1.63) 1.25 (0.94, 1.66) 1.82 (0.91, 3.65) 1.93 (0.96, 3.90) 
Age (per 10 years) 1.34 (1.09, 1.66) 1.46 (1.19, 1.80) 1.52 (0.92, 2.51) 1.85 (1.14, 3.00) 
BMI (per 5 kg/m21.02 (0.85, 1.22) 1.07 (0.89, 1.28) 1.57 (1.15, 2.13) 1.88 (1.38, 2.54) 
Current smoking 1.13 (0.82, 1.57) 1.03 (0.74, 1.44) 2.02 (1.02, 4.00) 1.66 (0.84, 3.31) 
Duration of diabetes (per 5 years) 1.08 (0.94, 1.25) 1.13 (0.97, 1.30) 1.18 (0.85, 1.63) 1.40 (0.99, 1.98) 
Systolic blood pressure (per 10 mmHg) 1.05 (0.93, 1.19) 1.02 (0.90, 1.15) 1.24 (0.94, 1.65) 1.11 (0.85, 1.45) 
LDL cholesterol (per 10 mg/dL) 1.00 (0.95, 1.05) 0.97 (0.92, 1.02) 1.01 (0.90, 1.14) 0.96 (0.86, 1.08) 
HDL cholesterol (per 5 mg/dL) 1.01 (0.95, 1.07) 1.01 (0.95, 1.07) 0.93 (0.80, 1.08) 0.92 (0.79, 1.08) 
Triglycerides (per 20 mg/dL) 1.04 (0.99, 1.09) 1.02 (0.96, 1.07) 1.13 (1.05, 1.23) 1.11 (1.02, 1.22) 
Mean DCCT HbA1c (per 10% increment) 1.23 (1.14, 1.34) Same as model 1 1.73 (1.41, 2.12) Same as model 1 
DCCT closeout HbA1c only (per 10% increment) 1.20 (1.12, 1.29) — 1.60 (1.34, 1.91) — 
Sustained eGFR <60 mL/min/1.73 m2 * * * * 
Sustained AER ≥30 mg/24 h 2.13 (1.47, 3.10) 1.67 (1.13, 2.45) 2.11 (0.88, 5.05) 1.05 (0.42, 2.61) 
Any confirmed clinical neuropathy 1.52 (1.08, 2.14) 1.23 (0.87, 1.75) 2.86 (1.40, 5.83) 1.74 (0.85, 3.58) 
Sural nerve conduction velocity (per 5 m/s) 0.90 (0.78, 1.02) 0.95 (0.82, 1.09) 0.89 (0.62, 1.28) 1.09 (0.74, 1.61) 
Any CAN 1.60 (1.13, 2.26) 1.43 (1.01, 2.03) 1.48 (0.65, 3.37) 1.10 (0.47, 2.56) 
Any retinopathy     
 None/mild nonproliferative 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 
 Moderate nonproliferative 1.65 (1.14, 2.39) 1.33 (0.92, 1.93) 1.36 (0.55, 3.35) 0.83 (0.34, 2.01) 
 Severe nonproliferative 1.01 (0.32, 3.24) 0.71 (0.22, 2.28) 1.57 (0.20, 12.38) 0.59 (0.07, 4.84) 
 Proliferative 1.68 (0.92, 3.08) 1.15 (0.62, 2.14) 2.39 (0.72, 7.99) 0.95 (0.28, 3.27) 
Macular edema 1.40 (0.85, 2.31) 1.39 (0.84, 2.29) 4.01 (1.74, 9.23) 4.15 (1.78, 9.67) 
ABI     
 <0.9 0.84 (0.46, 1.55) 0.79 (0.43, 1.46) 2.62 (1.01, 6.80) 2.27 (0.86, 5.95) 
 0.9–1.3 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 
 >1.3 0.76 (0.19, 3.07) 0.74 (0.18, 2.98) * * 
DFULEA
Model 1 adjusted
HR (95% CI)
Model 2 adjusted
HR (95% CI)
Model 1 adjusted
HR (95% CI)
Model 2 adjusted
HR (95% CI)
Male sex 1.23 (0.92, 1.63) 1.25 (0.94, 1.66) 1.82 (0.91, 3.65) 1.93 (0.96, 3.90) 
Age (per 10 years) 1.34 (1.09, 1.66) 1.46 (1.19, 1.80) 1.52 (0.92, 2.51) 1.85 (1.14, 3.00) 
BMI (per 5 kg/m21.02 (0.85, 1.22) 1.07 (0.89, 1.28) 1.57 (1.15, 2.13) 1.88 (1.38, 2.54) 
Current smoking 1.13 (0.82, 1.57) 1.03 (0.74, 1.44) 2.02 (1.02, 4.00) 1.66 (0.84, 3.31) 
Duration of diabetes (per 5 years) 1.08 (0.94, 1.25) 1.13 (0.97, 1.30) 1.18 (0.85, 1.63) 1.40 (0.99, 1.98) 
Systolic blood pressure (per 10 mmHg) 1.05 (0.93, 1.19) 1.02 (0.90, 1.15) 1.24 (0.94, 1.65) 1.11 (0.85, 1.45) 
LDL cholesterol (per 10 mg/dL) 1.00 (0.95, 1.05) 0.97 (0.92, 1.02) 1.01 (0.90, 1.14) 0.96 (0.86, 1.08) 
HDL cholesterol (per 5 mg/dL) 1.01 (0.95, 1.07) 1.01 (0.95, 1.07) 0.93 (0.80, 1.08) 0.92 (0.79, 1.08) 
Triglycerides (per 20 mg/dL) 1.04 (0.99, 1.09) 1.02 (0.96, 1.07) 1.13 (1.05, 1.23) 1.11 (1.02, 1.22) 
Mean DCCT HbA1c (per 10% increment) 1.23 (1.14, 1.34) Same as model 1 1.73 (1.41, 2.12) Same as model 1 
DCCT closeout HbA1c only (per 10% increment) 1.20 (1.12, 1.29) — 1.60 (1.34, 1.91) — 
Sustained eGFR <60 mL/min/1.73 m2 * * * * 
Sustained AER ≥30 mg/24 h 2.13 (1.47, 3.10) 1.67 (1.13, 2.45) 2.11 (0.88, 5.05) 1.05 (0.42, 2.61) 
Any confirmed clinical neuropathy 1.52 (1.08, 2.14) 1.23 (0.87, 1.75) 2.86 (1.40, 5.83) 1.74 (0.85, 3.58) 
Sural nerve conduction velocity (per 5 m/s) 0.90 (0.78, 1.02) 0.95 (0.82, 1.09) 0.89 (0.62, 1.28) 1.09 (0.74, 1.61) 
Any CAN 1.60 (1.13, 2.26) 1.43 (1.01, 2.03) 1.48 (0.65, 3.37) 1.10 (0.47, 2.56) 
Any retinopathy     
 None/mild nonproliferative 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 
 Moderate nonproliferative 1.65 (1.14, 2.39) 1.33 (0.92, 1.93) 1.36 (0.55, 3.35) 0.83 (0.34, 2.01) 
 Severe nonproliferative 1.01 (0.32, 3.24) 0.71 (0.22, 2.28) 1.57 (0.20, 12.38) 0.59 (0.07, 4.84) 
 Proliferative 1.68 (0.92, 3.08) 1.15 (0.62, 2.14) 2.39 (0.72, 7.99) 0.95 (0.28, 3.27) 
Macular edema 1.40 (0.85, 2.31) 1.39 (0.84, 2.29) 4.01 (1.74, 9.23) 4.15 (1.78, 9.67) 
ABI     
 <0.9 0.84 (0.46, 1.55) 0.79 (0.43, 1.46) 2.62 (1.01, 6.80) 2.27 (0.86, 5.95) 
 0.9–1.3 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 
 >1.3 0.76 (0.19, 3.07) 0.74 (0.18, 2.98) * * 

Model 1 adjustment: age, sex, and diabetes duration at DCCT closeout, as appropriate. Model 2 adjustment: age, sex, diabetes duration at DCCT closeout, and mean DCCT HbA1c, as appropriate. Boldface indicates statistically significant associations. Ref, referent.

*

No foot ulcer or amputation occurred; HR cannot be estimated.

Model 2 comparison of moderate to severe to none/mild nonproliferative retinopathy: DFU P = 0.04, LEA P = 0.56.

Model 2 comparison of ABI <0.9 and >1.3 with ABI 0.9–1.3: DFU P = 0.80, LEA P = 0.14.

Assessing updated time-dependent exposure measurements and adjusting for age, sex, and diabetes duration at DCCT closeout, risk factors in common for both DFU and LEA now included time-weighted mean DCCT/EDIC HbA1c, serum triglyceride concentration, sustained eGFR <60 mL/min/1.73 m2, AER ≥30 mg/24 h, confirmed clinical neuropathy, lower sural nerve conduction velocity, CAN, any retinopathy, and macular edema (Table 3). In addition, current smoking was associated with LEA risk. After adjustment for time-weighted mean DCCT/EDIC HbA1c, the only statistically significant risk factor remaining in common between DFU and LEA was any retinopathy. Confirmed clinical neuropathy, lower sural nerve conduction velocity, and CAN remained associated with higher DFU risk, while sustained eGFR <60 mL/min/1.73 m2, AER ≥30 mg/24 h, and macular edema statistically significantly predicted LEA risk.

Table 3

Associations of time-dependent covariates measured beginning in EDIC year 1 with the risk of first-recorded DFU or LEA during EDIC follow-up, adjusted for DCCT/EDIC time-weighted mean HbA1c and diabetes duration at DCCT closeout

DFULEA
Model 1 adjusted
HR (95% CI)
Model 2 adjusted
HR (95% CI)
Model 1 adjusted
HR (95% CI)
Model 2 adjusted
HR (95% CI)
BMI (per 5 kg/m21.02 (0.87, 1.20) 1.00 (0.85, 1.18) 0.98 (0.74, 1.29) 0.92 (0.70, 1.22) 
Current smoking 1.23 (0.84, 1.81) 1.03 (0.70, 1.52) 2.18 (1.00, 4.76) 1.44 (0.62, 3.31) 
Systolic blood pressure (per 10 mmHg) 1.07 (0.99, 1.15) 1.02 (0.95, 1.09) 1.06 (0.85, 1.34) 0.95 (0.80, 1.13) 
LDL cholesterol (per 10 mg/dL) 1.00 (0.96, 1.04) 0.97 (0.93, 1.01) 1.00 (0.91, 1.10) 0.93 (0.86, 1.01) 
HDL cholesterol (per 5 mg/dL) 1.01 (0.97, 1.05) 1.02 (0.98, 1.07) 0.95 (0.87, 1.04) 0.99 (0.91, 1.08) 
Triglycerides (per 20 mg/dL) 1.03 (1.00, 1.06) 1.01 (0.97, 1.04) 1.05 (1.02, 1.08) 1.02 (0.97, 1.07) 
Cumulative mean HbA1c (per 10% increment) 1.33 (1.29, 1.37) Same as model 1 2.37 (2.12, 2.64) Same as model 1 
Sustained eGFR <60 mL/min/1.73 m2 1.87 (1.19, 2.92) 1.36 (0.88, 2.11) 5.46 (2.36, 12.64) 2.63 (1.30, 5.34) 
Sustained AER ≥30 mg/24 h 1.64 (1.19, 2.25) 1.23 (0.89, 1.72) 4.27 (2.06, 8.83) 2.23 (1.04, 4.78) 
Confirmed clinical neuropathy 2.71 (1.85, 3.97) 2.18 (1.50, 3.18) 5.25 (1.55, 17.85) 2.08 (0.56, 7.74) 
Sural nerve conduction velocity (per 5 m/s) 0.78 (0.69, 0.88) 0.84 (0.75, 0.95) 0.60 (0.48, 0.74) 0.81 (0.61, 1.09) 
Presence of CAN 2.07 (1.47, 2.90) 1.68 (1.22, 2.32) 4.46 (1.50, 13.23) 2.36 (0.87, 6.39) 
Any retinopathy     
 None/mild nonproliferative 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 
 Moderate nonproliferative 1.47 (1.00, 2.15) 1.16 (0.79, 1.72) 0.63 (0.16, 2.46) 0.37 (0.10, 1.42) 
 Severe nonproliferative 0.22 (0.03, 1.58) 0.18 (0.02, 1.31) * * 
 Proliferative 1.70 (1.15, 2.52) 1.19 (0.78, 1.80) 4.59 (1.94, 10.90) 1.77 (0.76, 4.10) 
Macular edema 1.66 (1.20, 2.30) 1.31 (0.96, 1.78) 5.58 (2.29, 13.58) 3.13 (1.35, 7.26) 
ABI     
 <0.9 1.25 (1.00, 1.56) 1.16 (0.93, 1.45) 1.79 (0.88, 3.64) 1.34 (0.64, 2.84) 
 0.9–1.3 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 
 >1.3 0.88 (0.56, 1.37) 0.79 (0.47, 1.32) 0.47 (0.11, 1.97) 0.26 (0.03, 1.98) 
DFULEA
Model 1 adjusted
HR (95% CI)
Model 2 adjusted
HR (95% CI)
Model 1 adjusted
HR (95% CI)
Model 2 adjusted
HR (95% CI)
BMI (per 5 kg/m21.02 (0.87, 1.20) 1.00 (0.85, 1.18) 0.98 (0.74, 1.29) 0.92 (0.70, 1.22) 
Current smoking 1.23 (0.84, 1.81) 1.03 (0.70, 1.52) 2.18 (1.00, 4.76) 1.44 (0.62, 3.31) 
Systolic blood pressure (per 10 mmHg) 1.07 (0.99, 1.15) 1.02 (0.95, 1.09) 1.06 (0.85, 1.34) 0.95 (0.80, 1.13) 
LDL cholesterol (per 10 mg/dL) 1.00 (0.96, 1.04) 0.97 (0.93, 1.01) 1.00 (0.91, 1.10) 0.93 (0.86, 1.01) 
HDL cholesterol (per 5 mg/dL) 1.01 (0.97, 1.05) 1.02 (0.98, 1.07) 0.95 (0.87, 1.04) 0.99 (0.91, 1.08) 
Triglycerides (per 20 mg/dL) 1.03 (1.00, 1.06) 1.01 (0.97, 1.04) 1.05 (1.02, 1.08) 1.02 (0.97, 1.07) 
Cumulative mean HbA1c (per 10% increment) 1.33 (1.29, 1.37) Same as model 1 2.37 (2.12, 2.64) Same as model 1 
Sustained eGFR <60 mL/min/1.73 m2 1.87 (1.19, 2.92) 1.36 (0.88, 2.11) 5.46 (2.36, 12.64) 2.63 (1.30, 5.34) 
Sustained AER ≥30 mg/24 h 1.64 (1.19, 2.25) 1.23 (0.89, 1.72) 4.27 (2.06, 8.83) 2.23 (1.04, 4.78) 
Confirmed clinical neuropathy 2.71 (1.85, 3.97) 2.18 (1.50, 3.18) 5.25 (1.55, 17.85) 2.08 (0.56, 7.74) 
Sural nerve conduction velocity (per 5 m/s) 0.78 (0.69, 0.88) 0.84 (0.75, 0.95) 0.60 (0.48, 0.74) 0.81 (0.61, 1.09) 
Presence of CAN 2.07 (1.47, 2.90) 1.68 (1.22, 2.32) 4.46 (1.50, 13.23) 2.36 (0.87, 6.39) 
Any retinopathy     
 None/mild nonproliferative 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 
 Moderate nonproliferative 1.47 (1.00, 2.15) 1.16 (0.79, 1.72) 0.63 (0.16, 2.46) 0.37 (0.10, 1.42) 
 Severe nonproliferative 0.22 (0.03, 1.58) 0.18 (0.02, 1.31) * * 
 Proliferative 1.70 (1.15, 2.52) 1.19 (0.78, 1.80) 4.59 (1.94, 10.90) 1.77 (0.76, 4.10) 
Macular edema 1.66 (1.20, 2.30) 1.31 (0.96, 1.78) 5.58 (2.29, 13.58) 3.13 (1.35, 7.26) 
ABI     
 <0.9 1.25 (1.00, 1.56) 1.16 (0.93, 1.45) 1.79 (0.88, 3.64) 1.34 (0.64, 2.84) 
 0.9–1.3 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 1.0 (Ref) 
 >1.3 0.88 (0.56, 1.37) 0.79 (0.47, 1.32) 0.47 (0.11, 1.97) 0.26 (0.03, 1.98) 

Model 1 adjustment: age, sex, and diabetes duration at DCCT closeout, as appropriate. Model 2 adjustment: age, sex, diabetes duration at DCCT closeout, and time-weighted mean DCCT/EDIC HbA1c, as appropriate. Boldface indicates statistically significant associations. Ref, referent.

*

No amputations occurred; HR cannot be estimated.

Model 2 comparison of moderate to severe to none/mild nonproliferative retinopathy: DFU P = 0.01, LEA P = 0.0001.

Model 2 comparison of ABI <0.9 and >1.3 with ABI 0.9–1.3: DFU P = 0.12, LEA P = 0.12.

These results demonstrate that a period of intensive glycemic control early in the course of T1D reduced the risk of DFU during subsequent long-term follow-up in the DCCT/EDIC cohort. To our knowledge, this is the first demonstration of a reduction in DFU risk associated with a randomized intervention to improve glycemic control. By extension, intensive glycemic control would also be expected to reduce amputation risk, given the key role that DFU plays in leading to amputation, although our results did not confirm this likely because of the low power to detect associations given the very few amputations observed. The delayed effect of intensive glycemic control on risk of DFU and LEA after the DCCT intervention provides another example of the phenomenon of metabolic memory as observed with other complications in the DCCT/EDIC (18).

Higher risk of DFU in the former CON compared with INT group was observed ∼11 years into EDIC (Fig. 1), an average of 17 years from DCCT baseline. Similar mean HbA1c levels throughout EDIC were observed in both groups, reflecting similar management of glycemia after DCCT (18). A statistically significant association was seen between DCCT treatment and DFU incidence after 10 years of EDIC follow-up, although the importance of this difference is unclear because of the lack of a significant treatment-by-time interaction. Of note, the magnitude of effect of prior INT was similar for all DFU and LEA outcomes, with HRs ranging from 0.70 to 0.78, suggesting that the lack of statistically significant associations with risk of first-recorded DFU and LEA may reflect the low power due to the smaller number of events.

The observed associations of HbA1c over time with risks of DFU and LEA in this analysis support the importance of hyperglycemia in their development and are consistent with published observational research. A prospective study of U.S. veterans with diabetes demonstrated a 10% higher risk of DFU per 1% increase in HbA1c (19). A meta-analysis that included 14 prospective studies of LEA in diabetes demonstrated a 25% higher risk for each 1% increase in HbA1c and found that this association did not differ significantly in T1D versus type 2 diabetes (T2D) (20). The Pittsburgh Epidemiology of Diabetes Complications study provided further evidence in T1D of an association between higher HbA1c and higher risk of a combined definition of lower extremity complication that included ulcer and amputation (21).

In addition to DCCT mean HbA1c and HbA1c measured at DCCT closeout, multiple factors measured during EDIC significantly predicted higher DFU risk, including older age, albuminuria, confirmed clinical neuropathy, CAN, and any retinopathy. After adjustment for DCCT mean HbA1c, all these risk factors remained significantly associated with DFU, except for confirmed clinical neuropathy. The only significant risk factor in common measured at DCCT closeout for both DFU and LEA was confirmed clinical neuropathy, but this association became nonsignificant after adjustment for DCCT mean HbA1c. Although DFU is recognized as the antecedent event in the majority of LEAs, there are clear differences between the risk profiles of these outcomes.

Given the duration of EDIC follow-up available for this analysis, we conducted additional analyses that included time-updated measurements to assess temporal change in risk factors for DFU and LEA. These additional analyses revealed several factors not associated with DFU and LEA risk at DCCT closeout that emerged as statistically significant risk factors during follow-up, even after adjustment for time-weighted mean DCCT/EDIC HbA1c (Table 3). These emergent risk factors included several diabetes complications, such as any retinopathy for LEA, confirmed clinical neuropathy and lower nerve conduction velocity for DFU, and kidney disease and albuminuria for LEA.

The associations of multiple risk factors reflecting dysfunction in different organ systems with lower-extremity outcomes support a multifactorial etiology for DFU and LEA (2). The loss of protective sensation as well as foot deformities because of both motor and sensory neuropathy increase the likelihood of abnormal bony prominences, pressure points, and unnoticed tissue injury (22). The association between CAN and higher DFU risk may be due to skin microvascular dysfunction, as the autonomic nervous system in part regulates skin perfusion (23). Whether associations between lower-extremity complications and vision-threatening macular edema and proliferative diabetic retinopathy are markers of overall microvascular damage or related to poor self-care because of visual impairment is unknown. Longitudinal research has demonstrated that poor vision is associated with a higher risk of DFU and LEA (8,24). In this study, kidney disease, as reflected by sustained albumin excretion or eGFR <60 mL/min/1.73 m2, was associated with these lower-extremity outcomes, but the mechanism underlying this association is unknown. Chronic kidney disease is a well-known risk factor for atherosclerosis in general and involves peripheral arteries in particular (25). However, atherosclerosis affecting peripheral arteries in this cohort was uncommon (6–7%) at the start of EDIC as measured by ABI <0.9. The association of DFU with kidney disease may be explained in part by confounding by hyperglycemia given the reduction in associations between DFU and eGFR following adjustment for time-weighted mean DCCT/EDIC HbA1c. Of note, however, the association between LEA and kidney disease persisted even after adjustment for time-weighted mean DCCT/EDIC HbA1c.

ABI <0.9 measured at the start of EDIC was significantly associated with higher amputation risk compared with a normal ABI (0.9–1.3), but ABI <0.9 measured repeatedly during EDIC follow-up was not. A possible explanation may be that the young individuals with low ABI at the start of EDIC had accelerated atherosclerosis that advanced more rapidly to occlusive disease and tissue loss than those whose ABI diminished over follow-up in association with aging. These findings contrast with T2D, where reports have shown an association between low ABI and higher risk of both ulcer and amputation (8,24). Prior research in the DCCT/EDIC did not demonstrate an association of random assignment to CON in the DCCT and the development of ABI <0.9 during EDIC follow-up (26).

Current smoking measured at DCCT closeout or during EDIC follow-up was associated with a significantly higher risk of LEA but not DFU. A higher risk of amputation among smokers is well established among people with diabetes, but few studies have demonstrated a similar association with DFUs (27), with only 2 of 17 reports finding a statistically significant association between current smoking and active or past DFU (28). Both reports of a significant association were conducted among people with T2D only.

The Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial demonstrated a reduced risk of LEA (HR 0.69 [95% CI 0.48, 0.99]) in people with T2D randomly assigned to intensive versus standard glycemic control for a mean duration of 3.7 years, a shorter treatment period compared with DCCT (29). The effect diminished after adjustment for mean postrandomization HbA1c, demonstrating that improved glycemic control mainly contributed to the preventive effect.

In the DCCT/EDIC, lower-extremity complications were not designated as major outcomes, which limits this current analysis. A duration criterion is often used in research to distinguish DFU from minor skin trauma. No such duration criterion was used in EDIC, but the stipulation that the lesion required medical or surgical treatment was designed to eliminate recording of minor skin trauma. Whether an ulcer was persistent or recurrent on the same limb was inferred on the basis of whether it was recorded in consecutive years, potentially missing the occurrence of a new ulcer on the same limb between annual examinations. Amputation level was not recorded, thereby preventing analyses of major versus minor amputations, although such analyses would likely not have been informative because of low amputation number. The ABI is a well-established clinical surrogate measurement of peripheral artery disease, but it is limited by falsely high readings because of vessel incompressibility caused by medial artery calcification seen in diabetes as well as less accuracy than established gold standard measurements (30). In addition, microvascular dysfunction not measurable by the ABI resulting in impaired skin blood flow because of neuropathy, endothelial dysfunction, or anatomic changes may contribute to the pathogenesis of DFU independent of macrovascular function (23). Direct measurements of skin blood flow were not available for this analysis but might be partially inferred from the tests of cardiovascular autonomic function that were performed, which demonstrated higher DFU risk with CAN. The observational analyses we conducted were adjusted for well-characterized covariate measurements obtained from a rigorous protocol. Nevertheless, the potential exists for bias as a result of residual confounding.

These analyses demonstrate the association between early intensive glycemic control and reduced incidence of DFU among people with T1D. In addition, multiple well-established diabetes complications are associated with risk of both DFU and LEA. DFU can be added to the list of complications potentially preventable by intensive glycemic treatment, further reinforcing the importance of optimal glycemic control implemented as early as possible for individuals with T1D to prevent this outcome.

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

This article contains supplementary material online at https://doi.org/10.2337/figshare.17049872.

*

A complete list the DCCT/EDIC Research Group investigators and members as of 1 July 2021 can be found in the supplementary material online.

Funding. The 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 grants U01-DK-094176 and U01-DK-094157) and through support by the National Eye Institute (1993–2007), National Institute of Neurological Disorders and Stroke, General Clinical Research Centers Program, and Clinical Translational Science Center Program (2006 to present). Additional support for this DCCT/EDIC collaborative study was provided by an unrestricted fund from the Northwest Kidney Centers. VA Puget Sound provided support for E.J.B. and I.H.d.B.’s participation in this research.

Duality of Interest. Industry contributors have provided free or discounted supplies or equipment to support participants’ adherence to the study: Abbott Diabetes Care, Animas, Bayer Diabetes Care, Becton Dickinson, Eli Lilly, Extend Nutrition, Insulet Corporation, Lifescan, Medtronic Diabetes, Nipro Home Diagnostics, Nova Diabetes Care, Omron, Perrigo Diabetes Care, Roche Diabetes Care, and Sanofi. No other potential conflicts of interest relevant to this article were reported.

Industry contributors have had no role in the DCCT/EDIC study.

Author Contributions. E.J.B., L.R.Z., and I.H.d.B. designed and conducted the statistical analyses, prepared the tables and figures, and wrote the manuscript. B.H.B., R.P.-B., C.C.C., G.M.L., R.G.-K., and B.Z. designed the analysis and reviewed and edited the manuscript. E.J.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.

Data Sharing. Data collected for the DCCT/EDIC study through 30 June 2017 are available to the public through the National Institute of Diabetes and Digestive and Kidney Diseases Central Repository (https://repository.niddk.nih.gov/studies/edic). Data collected in the current cycle (July 2017 to June 2022) will be available within 2 years after the end of the funding cycle.

1
Narres
M
,
Kvitkina
T
,
Claessen
H
, et al
.
Incidence of lower extremity amputations in the diabetic compared with the non-diabetic population: A systematic review
.
PLoS One
2017
;
12
:
e0182081
2
Pecoraro
RE
,
Reiber
GE
,
Burgess
EM
.
Pathways to diabetic limb amputation. Basis for prevention
.
Diabetes Care
1990
;
13
:
513
521
3
Leese
GP
,
Feng
Z
,
Leese
RM
,
Dibben
C
,
Emslie-Smith
A
.
Impact of health-care accessibility and social deprivation on diabetes related foot disease
.
Diabet Med
2013
;
30
:
484
490
4
Costacou
T
,
Orchard
TJ
.
Recent trends over time in vascular disease in type 1 diabetes: insights from the Pittsburgh Epidemiology of Diabetes Complications study
.
Cardiovasc Endocrinol Metab
2019
;
8
:
3
13
5
Miller
RG
,
Secrest
AM
,
Ellis
D
,
Becker
DJ
,
Orchard
TJ
.
Changing impact of modifiable risk factors on the incidence of major outcomes of type 1 diabetes: the Pittsburgh Epidemiology of Diabetes Complications Study
.
Diabetes Care
2013
;
36
:
3999
4006
6
Armstrong
DG
,
Boulton
AJM
,
Bus
SA
.
Diabetic foot ulcers and their recurrence
.
N Engl J Med
2017
;
376
:
2367
2375
7
Abbott
CA
,
Carrington
AL
,
Ashe
H
, et al.;
North-West Diabetes Foot Care Study
.
The North-West Diabetes Foot Care Study: incidence of, and risk factors for, new diabetic foot ulceration in a community-based patient cohort
.
Diabet Med
2002
;
19
:
377
384
8
Boyko
EJ
,
Ahroni
JH
,
Stensel
V
,
Forsberg
RC
,
Davignon
DR
,
Smith
DG
.
A prospective study of risk factors for diabetic foot ulcer
.
The Seattle Diabetic Foot Study Diabetes Care
1999
;
22
:
1036
1042
.
9
Hurley
L
,
Kelly
L
,
Garrow
AP
, et al
.
A prospective study of risk factors for foot ulceration: the West of Ireland Diabetes Foot Study
.
QJM
2013
;
106
:
1103
1110
10
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
11
Epidemiology of Diabetes Interventions and Complications (EDIC)
.
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
12
Steffes
M
,
Cleary
P
,
Goldstein
D
, et al
.
Hemoglobin A1c measurements over nearly two decades: sustaining comparable values throughout the Diabetes Control and Complications Trial and the Epidemiology of Diabetes Interventions and Complications study
.
Clin Chem
2005
;
51
:
753
758
13
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
14
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
15
Albers
JW
,
Herman
WH
,
Pop-Busui
R
, et al.;
Diabetes Control and Complications Trial /Epidemiology of Diabetes Interventions and Complications Research Group
.
Effect of prior intensive insulin treatment during the Diabetes Control and Complications Trial (DCCT) on peripheral neuropathy in type 1 diabetes during the Epidemiology of Diabetes Interventions and Complications (EDIC) Study
.
Diabetes Care
2010
;
33
:
1090
1096
16
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
17
Andersen
P
,
Gill
R
.
Cox’s regression model for counting processes, a large sample study
.
Ann Stat
1982
;
10
:
1100
1120
18
Nathan
DM
;
DCCT/EDIC Research Group
.
The diabetes control and complications trial/epidemiology of diabetes interventions and complications study at 30 years: overview
.
Diabetes Care
2014
;
37
:
9
16
19
Boyko
EJ
,
Ahroni
JH
,
Cohen
V
,
Nelson
KM
,
Heagerty
PJ
.
Prediction of diabetic foot ulcer occurrence using commonly available clinical information: the Seattle Diabetic Foot Study
.
Diabetes Care
2006
;
29
:
1202
1207
20
Adler
AI
,
Erqou
S
,
Lima
TA
,
Robinson
AH
.
Association between glycated haemoglobin and the risk of lower extremity amputation in patients with diabetes mellitus-review and meta-analysis
.
Diabetologia
2010
;
53
:
840
849
21
Olson
JC
,
Erbey
JR
,
Forrest
KY
,
Williams
K
,
Becker
DJ
,
Orchard
TJ
.
Glycemia (or, in women, estimated glucose disposal rate) predict lower extremity arterial disease events in type 1 diabetes
.
Metabolism
2002
;
51
:
248
254
22
van Schie
CH
,
Vermigli
C
,
Carrington
AL
,
Boulton
A
.
Muscle weakness and foot deformities in diabetes: relationship to neuropathy and foot ulceration in Caucasian diabetic men
.
Diabetes Care
2004
;
27
:
1668
1673
23
Chao
CY
,
Cheing
GL
.
Microvascular dysfunction in diabetic foot disease and ulceration
.
Diabetes Metab Res Rev
2009
;
25
:
604
614
24
Boyko
EJ
,
Seelig
AD
,
Ahroni
JH
.
Limb- and person-level risk factors for lower-limb amputation in the prospective Seattle Diabetic Foot Study
.
Diabetes Care
2018
;
41
:
891
898
25
Matsushita
K
,
Ballew
SH
,
Coresh
J
, et al.;
Chronic Kidney Disease Prognosis Consortium
.
Measures of chronic kidney disease and risk of incident peripheral artery disease: a collaborative meta-analysis of individual participant data
.
Lancet Diabetes Endocrinol
2017
;
5
:
718
728
26
Carter
RE
,
Lackland
DT
,
Cleary
PA
, et al.;
Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study Research Group
.
Intensive treatment of diabetes is associated with a reduced rate of peripheral arterial calcification in the diabetes control and complications trial
.
Diabetes Care
2007
;
30
:
2646
2648
27
Liu
M
,
Zhang
W
,
Yan
Z
,
Yuan
X
.
Smoking increases the risk of diabetic foot amputation: A meta-analysis
.
Exp Ther Med
2018
;
15
:
1680
1685
28
Monteiro-Soares
M
,
Boyko
EJ
,
Ribeiro
J
,
Ribeiro
I
,
Dinis-Ribeiro
M
.
Predictive factors for diabetic foot ulceration: a systematic review
.
Diabetes Metab Res Rev
2012
;
28
:
574
600
29
Goldman
MP
,
Clark
CJ
,
Craven
TE
, et al
.
Effect of intensive glycemic control on risk of lower extremity amputation
.
J Am Coll Surg
2018
;
227
:
596
604
30
Young
MJ
,
Adams
JE
,
Anderson
GF
,
Boulton
AJ
,
Cavanagh
PR
.
Medial arterial calcification in the feet of diabetic patients and matched non-diabetic control subjects
.
Diabetologia
1993
;
36
:
615
621
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/journals/pages/license.