Lower extremity arterial disease (LEAD) is one of the most common complications of diabetes and harms the peripheral arteries via multiple pathways [
Considering the serious health hazards of LEAD, a comprehensive identification of preventable risk factors of LEAD is important to improving patients’ quality of life and reducing the associated medical costs. Smoking, ageing, race/ethnicity, increased levels of inflammatory markers, homocysteinaemia, and abdominal obesity are currently identified as risk factors for LEAD [
We used clinical data from the Department of Nephrology and Endocrinology, PLA 148th Hospital. Of the 1025 inpatients (from January 2010 to December 2012), we excluded 25 inpatients with type 1 DM and 11 adult inpatients with latent autoimmune diabetes, and we recruited 989 (507 men and 482 women) as our participants.
We collected data regarding each participant’s gender, age, occupation, region of residence, alcohol and smoking habits, T2DM duration, and LEAD status.
T2DM was defined according to the American Diabetes Association criteria [
An alcohol user was defined as a regular drinker who consumed alcohol almost every day and had regularly consumed alcohol for more than half a year. This definition was used because the average alcohol consumption of occasional drinkers is difficult to determine [
This information was collected by a primary nurse. The patients’ answers to the questions on alcohol use were confirmed by the patients and their relatives to ensure the accuracy of the information. Central obesity was defined as a waist circumference (WC) > 90 cm in men and >80 cm in women [
SPSS version 19.0 was used to analyse the data. The significance level for all tests was set at a two-tailed
PSM [
The Committee for Medical Ethics of the Chinese PLA General Hospital examined and approved our study. Before completing the questionnaire, each involved participant signed an informed consent form.
Nine hundred and eighty-nine (507 men and 482 women) inpatients were involved in our study before PSM. The average age was 56.8 ± 11.6 years (range: 14–93 years). The average ages of patients who did and did not consume alcohol were 52.3 ± 11.3 years (range: 28–85 years) and 57.5 ± 11.3 years (range: 14–93 years), respectively. The general characteristics (age, gender, origin, occupation, smoking status, BMI, and central obesity) of the participants are shown in Table
Demographic characteristics of the participants according to alcohol use before and after PSM.
Group | Number (%) | Alcohol use (before PSM) | Alcohol use (after PSM) | ||||
---|---|---|---|---|---|---|---|
Total |
Yes ( |
No ( |
|
Yes ( |
No ( |
|
|
Age (years) | <0.001 | 0.534 | |||||
≤40 | 81 (8.3) | 22 (15.9) | 59 (7.1) | 16 (13.4) | 15 (12.6) | ||
60–69 | 529 (54.5) | 79 (57.2) | 450 (54.0) | 68 (57.1) | 76 (63.9) | ||
≥70 | 361 (37.2) | 37 (26.8) | 324 (38.9) | 35 (29.4) | 28 (23.5) | ||
Gender | <0.001 | 1.000 | |||||
Male | 498 (51.3) | 136 (98.6) | 362 (43.5) | 117 (98.3) | 117 (98.3) | ||
Female | 473 (48.7) | 2 (1.4) | 471 (56.5) | 2 (1.7) | 2 (1.7) | ||
Occupation | 0.01 | 0.562 | |||||
White collar | 103 (10.6) | 22 (15.9) | 81 (9.7) | 84 (70.6) | 77 (64.7) | ||
Light physical labourer | 117 (12.0) | 23 (16.7) | 94 (11.3) | 19 (16.0) | 25 (21.0) | ||
Hard physical labourer | 751 (77.3) | 93 (67.4) | 658 (79.0) | 16 (13.4) | 17 (14.3) | ||
Region | 0.756 | 0.408 | |||||
Shandong province | 940 (96.8) | 133 (96.4) | 807 (96.9) | 117 (98.3) | 115 (96.6) | ||
Other province | 31 (3.2) | 5 (3.6) | 26 (3.1) | 2 (1.7) | 4 (3.4) | ||
Smoker | <0.001 | 0.512 | |||||
Yes | 182 (18.7) | 90 (65.2) | 92 (11.0) | 71 (59.7) | 66 (55.5) | ||
No | 789 (81.3) | 48 (34.8) | 741 (89.0) | 48 (40.3) | 53 (44.5) | ||
BMI | 0.038 | 0.783 | |||||
<24.00 | 368 (37.9) | 39 (28.3) | 329 (39.5) | 36 (30.3) | 32 (26.9) | ||
24.00–27.99 | 388 (40.0) | 62 (44.9) | 326 (39.1) | 50 (42.0) | 55 (46.2) | ||
≥28.00 | 215 (22.1) | 37 (26.8) | 178 (21.4) | 33 (27.7) | 32 (26.9) | ||
Central obesity | <0.001 | 0.697 | |||||
Yes | 625 (64.4) | 66 (47.8) | 559 (67.1) | 60 (50.4) | 57 (47.9) | ||
No | 346 (35.6) | 72 (52.2) | 274 (32.9) | 59 (49.6) | 62 (52.1) | ||
Mean ± SD | |||||||
Age | 52.3 ± 11.3 | 57.5 ± 11.3 | <0.001 | 53.2 ± 11.9 | 51.9 ± 11.9 | 0.401 | |
Duration of T2DM | 5.4 ± 6.0 | 7.5 ± 6.6 | 0.001 | 5.8 ± 6.3 | 5.9 ± 6.1 | 0.937 | |
BMI | 25.8 ± 3.7 | 25.3 ± 4.1 | 0.143 | 25.8 ± 3.9 | 25.9 ± 3.9 | 0.861 | |
WC | 90.9 ± 8.4 | 88.3 ± 8.8 | 0.001 | 90.9 ± 8.7 | 90.7 ± 8.4 | 0.825 |
After PSM, 238 participant pairs were matched, and the two groups were balanced for age, gender, occupation, smoking status, BMI, central obesity, and T2DM duration (with and without alcohol consumption: 5.8 ± 6.3 years versus 5.9 ± 6.1 years, resp.;
According to the logistic regression analysis, patients who consumed alcohol had a higher risk of LEAD (OR: 2.75, 95% CI: 1.11–6.80) than patients who did not after adjusting for age, gender, region, occupation, smoking status, BMI, WC, and T2DM duration. Regarding the risk of LEAD after adjusting for alcohol use ≤8 U/day and >8 U/day, the odds ratios (ORs) were 2.07 (95% confidence interval (CI): 0.78–5.54,
OR (95% CI) of LEAD in participants according to alcohol use.
|
Model A | Model B | Model C | |
---|---|---|---|---|
OR (95% CI) | OR (95% CI) | OR (95% CI) | ||
Alcohol use | ||||
No (reference) | 9 (7.6) | 1 | 1 | 1 |
Yes | 21 (17.6) | 2.62 (1.15–5.99) | 2.61 (1.09–6.23) | 2.75 (1.11–6.80) |
|
0.022 | 0.031 | 0.028 | |
Alcohol consumption | ||||
No (reference) | 9 (7.6) | 1 | 1 | 1 |
≤8 U/day | 14 (15.7) | 2.28 (0.94–5.54) | 1.95 (0.76–5.02) | 2.07 (0.78–5.54) |
>8 U/day | 7 (23.3) | 3.72 (1.26–11.01) | 6.33 (1.89–21.15) | 6.35 (1.78–22.65) |
|
0.012 | 0.004 | 0.005 | |
Alcohol use duration | ||||
No (reference) | 9 (7.6) | 1 | 1 | 1 |
≤20 years | 8 (15.1) | 2.17 (0.79–5.99) | 2.25 (0.86–5.90) | 2.41 (0.88–6.60) |
>20 years | 13 (19.7) | 3.00 (1.21–7.45) | 3.40 (1.13–10.23) | 3.48 (1.09–11.15) |
|
0.017 | 0.015 | 0.019 | |
Continuous | ||||
No (reference) | 1 | 1 | 1 | |
Alcohol consumption (U) | 1.06 (1.00–1.12) | 1.10 (1.03–1.18) | 1.11 (1.04–1.19) | |
|
0.048 | 0.004 | 0.003 | |
No (reference) | 1 | 1 | 1 | |
Alcohol use duration (years) | 1.03 (1.01–1.06) | 1.02 (0.99–1.04) | 1.02 (0.99–1.05) | |
|
0.013 | 0.054 | 0.055 |
Model A: crude model; model B: adjusted for age, gender, region, and occupation; model C: adjusted for age, gender, region, occupation, smoking status, BMI, WC, and T2DM duration.
When alcohol consumption was analysed as a continuous outcome, models A to C showed that increased alcohol consumption was associated with an increased risk of LEAD (all
The gender imbalance between alcohol consumers and nonconsumers is striking. We performed an analysis on male patients only (the number of alcohol-consuming women was too low for a separate analysis of female patients). Compared with male patients who did not consume alcohol, male patients who consumed alcohol had a higher risk of LEAD (OR: 3.17, 95% CI: 1.25–8.09) after adjusting for age, region, occupation, smoking status, BMI, WC, and T2DM duration. Regarding the risk of LEAD after adjusting for alcohol use ≤8 U/day and >8 U/day, the ORs were 2.43 (95% CI: 0.89–6.68,
OR (95% CI) of LEAD in male participants according to alcohol use.
|
Model A | Model B | Model C | |
---|---|---|---|---|
OR (95% CI) | OR (95% CI) | OR (95% CI) | ||
Alcohol use | ||||
None (reference) | 8 (6.8) | 1 | 1 | 1 |
Yes | 21 (17.9) | 2.98 (1.26–7.04) | 3.10 (1.25–7.67) | 3.17 (1.25–8.09) |
|
0.013 | 0.015 | 0.016 | |
Alcohol consumption | ||||
None (reference) | 8 (6.8) | 1 | 1 | 1 |
≤8 U/day | 14 (16.1) | 2.61 (1.04–6.54) | 2.36 (0.88–6.26) | 2.43 (0.89–6.68) |
>8 U/day | 7 (23.3) | 4.15 (1.37–12.58) | 7.32 (2.10–25.42) | 7.03 (1.91–25.84) |
|
0.007 | 0.002 | 0.003 | |
Alcohol use duration | ||||
None (reference) | 8 (6.8) | 1 | 1 | 1 |
≤20 years | 8 (15.1) | 2.42 (0.86–6.85) | 3.88 (1.25–12.07) | 3.93 (1.19–12.96) |
>20 years | 13 (20.3) | 3.47 (1.36–8.90) | 2.74 (1.01–7.39) | 2.82 (1.01–7.91) |
|
0.009 | 0.042 | 0.039 | |
Continuous | ||||
None (reference) | 1 | 1 | 1 | |
Alcohol consumption (U) | 1.06 (1.00–1.13) | 1.11 (1.04–1.18) | 1.11 (1.04–1.19) | |
|
0.036 | 0.003 | 0.003 | |
None (reference) | 1 | 1 | 1 | |
Alcohol use duration (years) | 1.04 (1.01–1.07) | 1.03 (1.00–1.05) | 1.03 (1.00–1.06) | |
|
0.004 | 0.067 | 0.063 |
Model A: crude model; model B: adjusted for age, region, and occupation; model C: adjusted for age, region, occupation, smoking status, BMI, WC, and T2DM duration.
When alcohol consumption was analysed as a continuous outcome, models A to C showed that increased alcohol consumption was associated with an increased risk of LEAD (all
In this study, we observed a significant association between alcohol consumption and LEAD in patients with T2DM. We used a standard and universal measure of alcohol consumption; in China, individuals usually drink white spirits distilled from sorghum or maize or beer, and “liang” (50 g) is usually used as the measurement for the amount of alcohol consumed [
As shown in a study by Mukamal et al. [
This study had several limitations. As the information on alcohol consumption was based on recall, recall bias could not be completely excluded; however, the information was confirmed by patients and their relatives to ensure accuracy. Second, our sample may not be completely representative of patients with T2DM in China because our hospital is one of the best hospitals in Zibo, and the inpatients here have higher proportions of diabetic complications. However, the representativeness of our sample should not substantially affect the internal validity of this study. Third, the cholesterol, blood pressure, and prevalent cardiovascular disease data were missing for 229 participants, and thus, we did not include these three variables as confounders for PSM. Furthermore, we did not collect information about physical activity or homocysteine levels, which are also risk factors for LEAD [
In summary, our study observed a dose-response relationship between alcohol consumption and LEAD among inpatients with T2DM. We used a standard and universal measurement of alcohol consumption and increased the comparability of the two groups using the PSM method. Alcohol consumption may be a risk factor for LEAD in patients with T2DM; however, further cohort studies should be conducted to verify the causal relationship. Based on our findings, patients with T2DM should be advised to stop drinking, or at least to avoid heavy drinking, to prevent the onset of LEAD.
Ankle-brachial index
Body mass index
95% confidence interval
Diabetes mellitus
Lower extremity arterial disease
Odds ratio
Peripheral artery disease
Propensity score matching
Type 2 diabetes mellitus
Toe-brachial index
Waist circumference.
The authors declare that they have no conflicts of interest.
Dr. Shanshan Yang and Professor Zhengguo Yang conceived and designed the study; Shuang Wang, Bo Yang, Jinliang Zheng, and Yuping Cai performed the experiments; Shanshan Yang and Shuang Wang analysed the data; and Shanshan Yang, Shuang Wang, and Zhengguo Yang contributed materials/analysis tools and wrote the paper. Shanshan Yang and Shuang Wang contributed equally to this work. This manuscript has been read and approved by all the authors, the requirements for authorship have been met, and each author believes that the manuscript represents honest work.
The authors would like to thank Lei Xu, Qian Li, Wei Jia, Xinli Cai, Lihui Liu, Ying Zhang, Jinjuan Zhao, and Xinai Yan from the 148th PLA Hospital for assisting with the research described in this study.