Background

Microvascular complications (neuropathy, nephropathy and retinopathy) affect hundreds of millions of patients with type 2 diabetes mellitus (T2DM) [1,2,3]. The progression of these complications can lead to loss of visual, renal, and neurologic functions, impaired mobility and cognition, poor quality of life, limitations for employment and productivity, and increased costs for the patient and society [4, 5]. If left uncontrolled or untreated, they lead to irreversible damage and even death. Thus, it is of great importance to investigate the risk factors and carry early intervention [6].

Arterial calcification in the lower extremity, particularly in the setting of T2DM, doubles cardiovascular mortality and quadruples the risk of amputation [7]. It is classified as two forms: intimal calcification is a marker of atherosclerotic disease and is associated with arterial stenosis; while T2DM patients usually have medial arterial calcification (MAC) [8]. MAC is a condition that leads to stiffening of the elastic layer of the arterial wall, resulting in a series of cardiovascular events [9]. Ultrasonography is widely used to assess arterial intimal abnormalities such as intima-medial thickness [10] and endothelial dysfunction [11]. Compared with conventional radiography, ultrasound can directly aid visualization of the arterial lumen of the peripheral arterial tree and has its advantages in detecting MAC accurately [12]. Previous studies mostly focused on the association between MAC in lower limb arteries and various diabetic macrovascular complications including peripheral arterial diseases (PAD) [13, 14]. Lower limb arteries mainly consist of above-knee arteries [superficial femoral arteries (SFA)], and below-knee arteries [anterior tibial arteries (ATA) and posterior tibial arteries (PTA)]. However, it remains a debate on whether below- or above-knee calcification plays a role in T2DM-related microvascular complications. Guzman RJ et al. showed that tibial artery calcification was associated with foot ulcer in T2DM [15]. By contrast, the results from a recent study indicated an association between SFA calcification and T2DM-related microvascular complications [16]. These studies were limited for two reasons. On one hand, they used computed tomography measurement, which may underestimate MAC severity; on the other hand, they just measured one segment of lower limb arteries and did not explore the related risk factors either.

Thus, in this present study, we improved these flaws and adapted a sensitive method (ultrasonography) to evaluate MAC in the whole lower limb artery branches (SFA, ATA and PTA) of T2DM patients. We aimed to assess which lower limb artery segment calcification was associated with diabetes-related complications and to evaluate its risk factors.

Methods

Study Participants

This was a retrospective, single-center study consisting of T2DM patients admitted in our unit between March 2015 and December 2017. The demographic, biomedical and US profiles were extracted from medical database. The main inclusion criteria were based on the T2DM diagnostic guideline on the 1999 World Health Organization [17]. The main exclusion criteria were (1) type 1 diabetes mellitus; (2) recent infection inflammatory disorders, serious cardiovascular diseases, renal dysfunction or hepatic diseases; (3) recent receiving treatments of several agents affecting mineral metabolism such as steroid hormones and anti-osteoporosis drugs within three months; (4) malignancy or disability to complete required measurement. This study protocol conformed to the ethical guidelines of the Declaration of Helsinki as reflected in a priori approval by the Ethics Committee of Sun Yat-sen University.

Measurement of Blood Pressure and Body Mass Index (BMI)

Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured three times using digital sphygmomanometers (OMRON Healthcare, Hoofddorp, The Netherlands) and the mean values of measurements were used as SBP and DBP in the analysis. BMI was calculated by weight/height2 (kg/m2).

Biochemical Measurements

Blood samples were collected after an overnight fast for at least 8 h. Fasting plasma glucose (FPG), hemoglobin A1c (HbA1c), serum total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C), creatinine, calcium and phosphorus were measured on a standardized and certified TBA-120 auto-analyzer (Toshiba Medical Systems, Japan) in the central laboratory of our unit. Notably, the triglyceride and glucose (TyG) index (Ln [TG (mg/mL) × FPG (mg/mL)/2]) and estimated glomerular filtration rate (eGFR) were calculated with established formulas: 141 × min (Scr/κ, 1)α × max (Scr/κ, 1)–1.209 × 0.993Age × 1.018[if female]_1.159[if black], where κ is 0.7 for females and 0.9 for males, α is − 0.329 for females and − 0.411 for males, min indicates the minimum of Scr/κ or 1, and max indicates the maximum of Scr/κ or 1 [18].

Ultrasonography Detection of MAC

All the T2DM patients underwent ultrasonography examination of bilateral superficial femoral arteries for the presence of MAC. An ultrasonography unit (LOGIQ E9, GE, USA) was used with an L5-12 MHz or L9 MHz transducer. The examinations were performed and interpreted by two sonographers, who were blinded to the clinical data. The SFA, ATA and PTA of both thighs were scanned in the cross-sectional plane from their origins at the groin to their distal parts at the entrance of the adductor canal to have an overall view of the blood vessels (Fig. 1). The MAC was assigned a score from 0 to 4 according to severity of the calcification within a 4-cm scanned area (which was usually the approximate size of the ultrasonography transducer). The definitions of unilateral artery were as follows: score 0, no MAC; score 1, linear MAC with a length of less than 1 cm; score 2, linear MAC with a length between 1 and 2 cm; score 3, more extensive non-stenotic MAC with a length between 2 and 3 cm; and score 4, diffuse linear MAC with a length more than 3 cm (Fig. 1) [12].

Fig. 1
figure 1

Representative ultrasound images from T2DM population with different MAC severities in the lower limb arteries. a MAC with score 1 (linear calcification length of PTA is less than 1 cm in this scanning area). b MAC with score 2 (linear calcification length of ATA occupies approximately 1–2 cm in this scanning area). c MAC with score 3 (more extensive non-stenotic calcifications of PTA with approximately 2–3 cm in this scanning area). d MAC with score 4 (diffuse linear calcification of SFA with extent more than 3 cm in this scanning area). The white arrows indicate MAC. ATA, anterior tibial arteries; MAC, medial arterial calcification; PTA, posterior tibial arteries; SFA, superficial femoral arteries; T2DM, type 2 diabetes

Clinical Assessment of Diabetic Complications

The included patients underwent comprehensive clinical evaluations. The definitions of various diabetic complications were consistent with recent guidelines [19,20,21]. The definition of diabetic neuropathy was as follows: if T2DM patients had clinical symptoms like pain, numbness, abnormal sensation, etc., accompanied with any one abnormality in examinations (ankle reflex, acupuncture pain, vibration, pressure and temperature sensation); or in the absence of clinical symptoms, 2 out of 5 tests were abnormal [19]. Diabetic nephropathy was defined as overt albuminuria (urinary albumin excretion rate > 300 mg/24 h) and eGFR < 60 mL/min per 1.73 m2 (no end-stage renal disease or kidney transplantation) in T2DM, without any clinical or laboratory evidence of other kidney diseases [20]. Diabetic retinopathy was defined as the presence of dot and blot hemorrhages, hard exudates, cotton wool spots, neovascularization, laser scars, or a history of vitrectomy in T2DM [21].

Statistical Analysis

Continuous variables with a normal distribution were reported as mean ± standard deviation; skewed data as median (25–75th quartiles). Categorical variables were presented as numbers (percentages). Baseline variables among patients with different severity of SFA calcification groups were compared using variance or Kruskal–Wallis test followed by a LSD comparison test or Pearson chi-square according to the data types. Receiver-operating characteristics curve (ROC) analysis was used to quantify the assessing value of different lower limb artery calcification for diabetic complications. Univariate and multivariate logistic analysis were used to investigate the risk factors for SFA calcification, as well as in further subgroup analysis. Data were analyzed with SPSS version 20 (SPSS, Inc, Chicago, IL), and two-sided P values < 0.05 were considered statistically significant.

Results

Characteristics of Recruited T2DM Patients

A total of 359 consecutive patients with T2DM were finally included in this study. Baseline characteristics of all the T2DM patients were summarized in Table 1. The mean age was 58.6 ± 11.6 years and 162 (45.1%) T2DM patients were male. The mean SBP and DBP were 134.8 ± 21.3 and 78.5 ± 11.6 mmHg, respectively. The median duration of T2DM was 5.0 (0–12.0) years. Moreover, the mean serum calcium, phosphorus, eGFR and TyG index in these patients were 2.24 ± 0.12 mmol/L, 1.17 ± 0.21 mmol/L, 75.83 ± 16.13 mL/min per 1.73 m2 and 9.27 ± 0.75, respectively. The prevalence of calcification in different lower limb artery branches were: SFA calcification 210 (58.5%), ATA calcification 140 (39.0%) and PTA calcification 135 (37.6%). As regard to diabetic microvascular complications, 227 (63.2%) T2DM patients had diabetic neuropathy, 105 (29.2%) T2DM patients had diabetic nephropathy, and 103 (28.7%) T2DM patients had diabetic retinopathy.

Table 1 Baseline characteristics of included T2DM patients

SFA Calcification Is a Good Marker for Predicting Diabetic Complications

To explore the roles of different lower limb artery root calcification for predicting diabetic microvascular complications, we performed ROC curve analysis. As shown in Fig. 2, compared with ATA, PTA calcification, the area under the curve for SFA calcification was the largest [SFA vs. ATA vs. PTA: 0.679 (0.619, 0.739) vs. 0.636 (0.577, 0.696) vs. 0.641 (0.582, 0.700); all P < 0.001]. SFA calcification with a threshold value of 1.5 provided 63.1% sensitivity and 69.2% specificity. In addition, we divided the SFA calcification into three levels (Fig. 3): no SFA calcification (score 0); mild SFA calcification (score 1–4); and severe SFA calcification (score 5–8). As shown in Fig. 4, with the severity SFA calcification elevated, the prevalence of diabetic microvascular complications was significantly increased [diabetic neuropathy: 73 (49.0%) vs. 70 (70.0%) vs. 84 (76.3%); diabetic nephropathy: 27 (18.1%) vs. 28 (28.0%) vs. 50 (45.5%); diabetic retinopathy: 22 (14.8%) vs. 33 (33.0%) vs. 48 (43.6%); all P < 0.05].

Fig. 2
figure 2

Receiver-operating characteristic curve for SFA, ATA and PTA calcification in assessing T2DM-related microvascular complications, respectively. ATA anterior tibial arteries, PTA posterior tibial arteries, SFA superficial femoral arteries, T2DM type 2 diabetes

Fig. 3
figure 3

Representative ultrasound images of 3 levels of SFA calcification in T2DM patients. No SFA calcification (score 0): score 0 in the left SFA (1L) and score 0 in the right SFA (1R) from a 52-year-old man; Mild SFA calcification (score 2): score 1 in the left SFA (1L) and score 1 in the right SFA (1R) from a 54-year-old woman. Severe SFA calcification (score 8): score 4 in the left SFA (1L) and score 4 in the right SFA (1R) from a 64-year-old man. The white arrows indicate MAC. MAC medial arterial calcification, SFA superficial femoral arteries, T2DM type 2 diabetes

Fig. 4
figure 4

The prevalence of T2DM-related microvascular complications among different SFA calcification severity groups. SFA superficial femoral arteries, T2DM type 2 diabetes. *P < 0.05 vs. no SFA calcification group; #P < 0.05 vs. mild SFA calcification group

Characteristics of T2DM Patients According to Different SFA Calcification Severity

Then, we compared the characteristics among T2DM patients with different SFA calcification severity. As shown in Table 2, age, duration of T2DM, and SBP were significantly elevated with the severity of SFA calcification increased [53.0 ± 10.9 vs. 61.2 ± 9.3 vs. 63.8 ± 11.3 years; 0 (0, 8.0) vs. 6.0 (0, 12.8) vs.10.0 (4.0, 15.3) years; 129.7 ± 17.7 vs. 136.2 ± 22.3 vs. 140.6 ± 23.4 mmHg; all P < 0.05]. As regard to the biomedical characteristics, T2DM patients with severe SFA calcification had significantly lower FPG, TyG index and eGFR than those with mild or no SFA calcification [7.8 ± 2.6 vs. 8.8 ± 3.9 vs. 9.1 ± 3.4 mmol/L; 9.16 ± 0.73 vs. 9.22 ± 0.77 vs. 9.39 ± 0.74; 64.77 ± 14.14 vs. 76.10 ± 12.55 vs. 83.81 ± 14.89 mL/min per 1.73 m2; all P < 0.05]. There were no significant differences of sex, smoking, SBP, DBP, BMI, WC, serum lipids, calcium and phosphorus among three groups (all P > 0.05).

Table 2 Characteristics of T2DM patients among different SFA calcification severity groups

Reduced eGFR Is an Independent Risk Factor of SFA Calcification

We further performed logistic analysis to explore the potential risk factors of SFA calcification. On the univariate logistic analysis, age (OR 1.090; 95% CI 1.064–1.117; P < 0.001), duration of T2DM (OR 1.090; 95% CI 1.064–1.117; P < 0.001), SBP (OR 1.021; 95% CI 1.010–1.032; P < 0.001), TyG index (OR 0.688; 95% CI 0.517–0.917; P = 0.011) and eGFR (OR 0.925; 95% CI 0.905–0.945; P < 0.001) were significantly associated with the presence of SFA calcification. However, when we put these factors into the multivariate logistic analysis, only age (OR 1.053; 95% CI 1.025–1.082; P < 0.001) and eGFR (OR 0.953; 95% CI 0.931–0.976; P < 0.001) were independent risk factors of SFA calcification (Table 3).

Table 3 Logistic analysis for the risk factors of SFA calcification in T2DM

Identifying the T2DM Patients with High Risk of SFA Calcification

Since aging is an unmodifiable fact OR we then further investigated the association between eGFR and SFA stratified by traditional cardiovascular risk factors, including age, sex, smoking, SBP, BMI and HbA1c, respectively. As shown in Fig. 5, we found that the negative association between eGFR and SFA calcification was significant in PA patients with ≤ 60 years of age (OR 0.925; 95% CI 0.889–0.964; P < 0.001), HbA1c > 7.0 (OR 0.947; 95% CI 0.920–0.975; P < 0.001). However, the association between eGFR and SFA calcification became nonsignificant in T2DM patients with > 60 years of age and HbA1c ≤ 7.0 subgroups (all P > 0.05). Moreover, this association remained significant among sex, smoking and BMI subgroups, suggesting it was not influenced by sex, smoking and BMI difference (all P < 0.05).

Fig. 5
figure 5

Subgroup analysis of assessing the association between eGFR and SFA calcification in T2DM. Multivariate logistic analysis after adjustment for age, duration of T2DM, BMI, SBP, and TyG index was performed in subgroup according to age (≤ 60 or > 60 years), sex (male or female), systolic blood pressure (SBP; ≤ 140 or > 140 mmHg), and BMI (20–25 or > 25 kg/m2). Data were expressed as the odds ratio and 95% confidence interval (CI). *P < 0.05. BMI body mass index, eGFR estimated glomerular filtration rate, SBP systolic blood pressure, SFA superficial femoral arteries, T2DM type 2 diabetes, TyG index triglyceride and glucose index

Discussion

Major findings from our study demonstrated that (1) SFA calcification had a strongest correlation with microvascular complications in patients with T2DM; (2) reduced eGFR was an independent risk factor for SFA calcification, especially in T2DM patients ≤ 60 years old with HbA1c > 7.0.

Microvascular complications are great threats to patients with T2DM, which reduces their life quality and increases the hospitalized duration [22]. MAC is a common condition in T2DM patients, which was shown to associate with various diabetic macrovascular complications [12, 13]. Chowdhury et al. showed that CT-based lower limb arterial calcification was strongly associated with cardiovascular events [23]. However, these studies were limited by radiology technique, which was inferior to US in evaluating MAC [12, 13]. In addition, it is under debate whether below- or above-knee calcification plays a role in T2DM-related microvascular complications. Thus, we adapted ultrasonography measurement and compared the roles of MAC in above-knee arteries (SFA), and below-knee arteries (ATA and PTA) in predicting microvascular complications in T2DM. It was emphasized that compared with ATA and PTA, SFA lesion had a more important role in predicting prognosis. McDermott MM et al. showed that lipid-rich necrotic core in the SFA was associated with higher rates of clinical PAD events [24]. Another study demonstrated that greater plaque quantity and smaller lumen area in the proximal SFA were associated with higher mobility loss in people with PAD [25]. Consistently, our results showed that SFA calcification had the strongest association with various diabetic microvascular complications than ATA and PTA calcification. Thus, early diagnosis and appropriate control of the risk factors of SFA calcification was possibly useful to reduce T2DM-related complications.

Instead of a passive product of calcium and phosphate precipitation, vascular calcification is now regarded as an actively regulated process involving many factors [26]. Although there were a series of studies investigating the risk factors of larger-sized lumen vascular calcification like coronary artery calcification, few have been known about lower limb arterial calcification in T2DM. Insulin resistance and hyperglycemia are major characteristics of T2DM [27]. It was demonstrated that insulin resistance was associated with coronary artery calcification [28]. As a reliable indicator of insulin resistance, high-TyG index was reported to be independently associated with the presence of coronary artery calcification in healthy Korean adults [29]. Although our data showed that neither high-TyG index nor FPG was an independent risk factor of SFA calcification, it indicated that under impaired kidney function condition, hyperglycemia may accelerate the development of SFA calcification. It was supported by the fact that the induction of in vitro T2DM calcification model required the high phosphate microenvironment (mimicked the reduced eGFR) [30,31,32]. Vascular calcification was thought to be a manifestation of vascular aging [33]. We consistently showed that older age was a strong risk factor of SFA calcification. Interestingly, the results of our subgroup analysis suggested that in T2DM patients with bad controlled hyperglycemia, downregulation of eGFR might trigger SFA calcification in the early stage. And this may explain why improving eGFR showed little effect on SFA calcification in older T2DM patients even with better control of blood glucose. However, further investigation is still required to verify our results.

This study had several limitations that should be highlighted. First, the number of T2DM patients enrolled in the study was relatively small. Therefore, further studies with larger sample size are required. Second, because of the cross-sectional design of the present study, causality between reduced eGFR and SFA calcification extent cannot be established. Third, some residual factors may confound our results despite careful adjustment, further studies are needed to verify our findings.

In summary, we provided clinical evidence that SFA calcification was a good predictor of microvascular complications in T2DM patients. Reduced eGFR was probably an important risk factor of SFA calcification in T2DM patients, especially in those young with bad controlled hyperglycemia.