Table 1.
Characteristics of studies included in the primary analysis
| Study name, location | Baseline year(s), (years of follow-upa) | Study size, n (% female) | Mean age, in years | Diabetes, n (% female) | Ascertainment of diabetes | Incident PAD, n (% female) | Method of PAD/PVD ascertainment | Maximum adjustment available |
|---|---|---|---|---|---|---|---|---|
| Alzamora et al. [23], Spain | 2011–2012, (5) | 2256 (59.0%) | 63 | 289 (48.8%) | Self-report and clinical history | 95 (51.6%) | ABI value < 0.9 | Age, atherogenic dyslipidemia, BMI, central obesity, education level, high LDL, high triglycerides, hypercholesterolemia, hypertension, low LDL, physical activity, sex, smoking |
| Emanuelsson et al. [24], Denmark | 1976-1978 (Copenhagen City Heart Study) or 2003-2018 (Copenhagen General Population Study) (9b) | 117,193 (55.1%) | 58c | 2437 (47.5%) | ICD-8 (249, 250) or ICD-10 (E10, E11, E14) diagnosis of type 1 or type 2 diabetes | 3615 (%) | ICD-8 (249.04, 249.05, 250.04, 250.05, 440-441, 443.99, 445) or ICD-10 (E10.5, E11.5, E14.5, 170–172, 173.9) diagnosis of PAD | Birth year, current smoking, pack-years smoked, BMI, hypertension, LDL cholesterol, time since last meal, and menopausal status (in women) |
|
Kennedy et al. [25], USA |
1989–1990; 1992–1993, (6) | 3126 (64.0%) | 74 | 378 | Use of antidiabetic medications or by the 1997 American Diabetes Association criteria | 251 (59.8%) | 1.4 ≥ ABI > 0.9 at baseline, with a decline in ABI of > 0.15 and to ABI ≤ 0.9 at follow-up; OR hospitalization (s) with ICD codes 440.2 or 443.9. | Age, cigarette smoking, fibrinogen, history of myocardial infarction, HDL, history of stroke, hypertension, lipid-lowering drug use, LDL, race, triglycerides |
| Krause et al. [26], Germany | 2001, (7) | 5735 (59.1%) | 72 | 1328 (52.6%) | Previous clinical diagnosis or HbA1c ≥ 6.5% or use of antidiabetic medications | 740 (58.2%) |
Any of the following symptom(s): history of peripheral revascularization, necrosis/gangrene, and/or peripheral amputation OR ABI ≤ 0.9, as assessed by linear regression modelling of multiple follow-up points |
Age, antihypertensive medication, BMI, CVD co-morbidity, education, GFR, homocysteine, LDL, sCRP, smoking, statin use, systolic BP, vitamin D |
| Shah et al. [27], England | People who were (or turned) 20 years or older between Jan 1, 2009 to March 25, 2010, (6b) | 1921,260 (49.7%) | 45 | 34,198 (46.2%) | Coded diagnoses recorded in CPRD or hospital episode statistics. Type 1 diabetes cases excluded. | 11,066 |
Coded diagnoses and procedures in primary care, secondary care and death certificates, including for, but not limited to, intermittent claudication, limb ischemia or gangrene due to atherosclerotic disease in the arteries of the legs. *Patient follow-up ended upon death or initial presentation of any cardiovascular disease. |
Age, antihypertensive medication, BMI, HDL, smoking status, socioeconomic status, statin, systolic blood pressure, total cholesterol |
| Turnstall-Pedoe et al. [28], Scotland | 1984–1995, (20) | 15,737 (52.0%) | 49 | 236 (48.6%) | Measured | 499 (41.7%) | At baseline: self-report and documented hospital discharge diagnosis. At endline: hospital diagnoses (ICD 9 = 440.2, 443.9, and/or 250.6; ICD-10 = I70.2, I73.9, E10.5, E11.5, E12.5, E13.5, OR E14.5) | Tobacco smoker, family history of CHD, age, hsC-reactive protein, systolic BP, expired carbon monoxide, cotinine, SIMD score, Lipoprotein (a), R-250 HD (adj), NT-pro-BNP, Glucose, triglycerides, cystatin-C |
| Weiss et al. [29], USA | 2000–2002, (13b) | 5953 (50.0%) | 703 (46.2%) | Fasting glucose > 125 mg/dL or use of antidiabetic medication | 168 (47.0%) | Self-reported diagnosis, hospital records review, or Centers for Medicare and Medicaid Services records | Age, race/ethnicity, smoking, hypertension, BMI |
a Reported as mean, unless otherwise specified
b Follow-up time reported as median
c Age reported as median