Background

Currently, China is one of the countries with the highest prevalence of diabetes in Asia and has the largest absolute disease burden of diabetes worldwide. According to the most recent epidemiological survey, the overall prevalence of diabetes was estimated to be 11.6% in Chinese adults [1]. Diabetes-related vascular complications such as cardiovascular disease and stroke are major causes of morbidity and mortality for type 2 diabetic patients (T2DM) [2]. Inflammatory processes have been found to play a role in the pathogenesis of both diabetes and atherosclerosis, and may offer a biological link between the two diseases [3]. Various circulating markers of inflammation have been extensively evaluated for their role as risk predictors of diabetes and atherosclerosis [4–6]. Among these markers, lipoprotein-associated phospholipase A2 (Lp-PLA2) and secretory phospholipase A2 (sPLA2) have gained considerable interest in the last decade. Many prospective studies have also indicated that Lp-PLA2 and sPLA2 are independent predictors of coronary heart disease and carotid stenosis, and they are more closely associated with atherosclerosis as compared to classic inflammatory markers such as C-reactive protein [7, 8].

Lp-PLA2 and sPLA2 are two subtypes of the PLA2 superfamily, a family of enzymes that propagate inflammation by catalyzing the hydrolysis of glycerophospholipids, thereby producing non-esterified fatty acids (FA) such as arachidonic acid. Lp-PLA2, also known as platelet-activating factor acetylhydrolase, is a 50-kD Ca2+-independent phospholipase and hydrolyzes phospholipids to produce free FAs (FFAs) and lyso-phosphatidycholine, thus promoting inflammation and atherosclerosis. Another important extracellular enzyme of the PLA2 superfamily, sPLA2, is a 14-kD Ca2+-dependent phospholipase and is normally expressed in arterial walls. Its expression is readily upregulated by inflammatory stimuli, suggesting a potential role in the early phases of response to vessel injury. Most studies have attributed an inflammatory role to Lp-PLA2 and sPLA2 [9, 10]. In addition, studies suggest that Lp-PLA2 is positively associated with insulin resistance and predicts the incidence of T2DM [11]. However, few study has assessed the effect of glycemic status of the patients on sPLA2. Moreover, other studies showed that both of Lp-PLA2 and sPLA2 play critical roles in the development of atherosclerosis and its clinical sequelae [7]. Lp-PLA2 and sPLA2 are upregulated in atherosclerotic plaques and are strongly expressed in macrophages within the fibrous cap of rupture-prone lesions [7, 12].

Short-term intensive insulin therapy can induce effective glycemic control and improve islet function and insulin sensitivity in patients with newly diagnosed T2DM [13]. In the United Kingdom Prospective Diabetes Study (UKPDS), after a 10-year follow-up, patients with newly diagnosed T2DM who received intensive glucose therapy were found to have a reduced risk of microvascular complications, coronary events, and death from any cause [14]. However, the mechanism through which these patients benefit from intensive insulin therapy is poorly understood. To our knowledge, no study has thus far specifically assessed the effect of intensive insulin therapy on Lp-PLA2 and sPLA2 levels in patients with newly diagnosed T2DM. Therefore, this study aimed to evaluate the change in Lp-PLA2 and sPLA2 levels after intensive insulin treatment and the correlation of these two enzymes with insulin resistance and pancreatic β-cell function in patients with newly diagnosed T2DM.

Methods

Subjects

Patients with newly diagnosed T2DM were enrolled in this study at the Sun Yat-sen Memorial Hospital of Sun Yat-sen University from October 2010 to March 2012. Diabetes was diagnosed in accordance with the diagnostic criteria of American Diabetes Association (ADA, 2010). All patients were 30–60 years of age, negative for pancreatic islet autoantibody, and had never taken hypoglycemic drugs. Patients with acute hyperglycemic complications, severe hepatic disease, renal dysfunction, cardiac dysfunction, and acute or chronic infections were excluded.

Research design

This was a self-control study of patients with newly diagnosed T2DM. All the enrolled diabetic patients were admitted to the hospital. The same researcher measured the height, weight, waist circumference, and blood pressure of the subjects on an empty stomach using consolidated tools. Body mass index (BMI) was calculated as weight divided by the square of the height (kg/m2). After an overnight fast, venous blood samples were drawn for measuring blood glucose, hemoglobin A1c (HbA1c), lipid profiles (including total cholesterol [TC], triglyceride [TG], high-density lipoprotein cholesterol [HDL-C], and low-density lipoprotein cholesterol [LDL-C]), traditional inflammatory markers (including white blood cells [WBC], FFA, and high-sensitivity C-reactive protein [hsCRP]), Lp-PLA2, and sPLA2. Subsequently, the patients underwent an intravenous glucose-tolerance test (IVGTT) with 50 mL of 50% glucose solution using the standard protocol. Blood samples were collected before and 3, 5, 7, and 10 min after the injection to assess the glycemic and insulin status. The next day, patients underwent the oral glucose-tolerance test (OGTT) with 75-g glucose load. To determine the glycemic and insulin status, blood samples were obtained before and 30, 60, 120, and 180 min after oral administration.

Thereafter, the patients received insulin lyspro (Humalog Lilly, USA) via an insulin-infusion pump (MiniMed 712) for approximately 2 weeks. The initial insulin doses were 0.4–0.6 IU/kg/day. Total daily doses were divided into 50% basal and 50% bolus injection. The doses of bolus and basal insulin infusions were adjusted according to the capillary blood glucose levels. Patients were asked to monitor their glycemic levels before and 2 h after each meal, and at bedtime. The target blood glucose level was defined as fasting plasma glucose (FPG) level < 6.1 mmol/L and 2-h postprandial plasma glucose (2hPG) level < 7.8 mmol/L. A severe hypoglycemic episode (PG <2.8 mmol/L) was defined as an event requiring the assistance of another person to actively administer carbohydrate, glucagon, or other resuscitative treatments. No oral hypoglycemic agents, antiplatelet medicine, or lipid drugs were administered during the study. Antihypertensive drugs affecting the renin-angiotensin system were also avoided in patients with complications of hypertension. Twenty-four hours after completion of the treatment, all the measurements mentioned above were reviewed.

Laboratory analyses

For plasma samples, blood was collected in EDTA vials and plasma was separated by centrifugation at 1000 × g for 10 min. The plasma samples were stored at −80°C and analyzed later for the mass of Lp-PLA2 and sPLA2. Concentrations of Lp-PLA2 and sPLA2 were assayed using a double-antibody sandwich enzyme-linked immunosorbent assay kit (Uscn Life Science, Peking, China). Three repetitive experiments showed that the average of coefficient of variation within-run was below 5%.

The plasma glucose level was determined by the glucose-oxidase method using a semi-automatic biochemistry analyzer (GF-D200, Shandong, China). The HbA1c concentration was measured using high-pressure liquid chromatography (BIO-RAD, California, USA). The direct chemiluminescence method was used to measure serum insulin levels with an automatic biochemistry analyzer (Immulite 2000, DPC, USA). The lipid profiles were measured by enzymatic colorimetry using a semi-automatic biochemistry analyzer (GF-D200, Shandong, China). Homeostasis model assessment (HOMA) was used to estimate insulin resistance (HOMA-IR) and β-cell function (HOMA-β) as follows: HOMA-IR = FPG × fasting insulin (FINS)/22.5 and HOMA-β = 20 × FINS/(FPG - 3.5). Other indicators used to evaluate β-cell function were as follows. The acute insulin response (AIR) was used to assess the first-phase insulin secretion, including the average increase in plasma insulin levels between the third or fifth minute and the baseline (AIR3–5) during IVGTT. The area under the curve of insulin (AUCIns) was calculated as the incremental trapezoidal area during the first 10 min as follows: IVGTT AUCIns = (3 × FINS + 5 × INS3 + 4 × INS5 + 5 × INS7 + 3 × INS10)/2. The early phase of insulin secretion was expressed as ∆INS30/∆G30 = (INS30 - FINS)/(G30 - FPG). The second-phase of insulin release was indicated as AUCIns during OGTT and was calculated as OGTT AUCIns = (FINS + 2 × INS30 + 3 × INS60 + 4 × INS120 + 2 × INS180)/4. Modified β-cell function index (MBCI) was calculated as MBCI = (FPG × FINS)/(G120 + G60 - 7) and was used for comprehensive assessment of β-cell function. Moreover, three other indicators were adopted to estimate insulin sensitivity: IAI, (calculated as 1/[FINS × FPG]), QUICKI (calculated as 1/[lgFPG + lgFINS]), and ISIced (calculated as MCR/LgMSI, MCR = M/MG, MG = [G0min + G30 min + G60 min + G120 min]/4, MSI = [I0 min + I30 min + I60 min + I120 min]/4, M = 75000/120 + [G0 min - G120 min] × 1.15 × 180 × 0.19 × body weight/120). In addition, all the indicators for β-cell function were divided by HOMA-IR to exclude interference of insulin resistance [15]. Furthermore, ultrasonography and electrocardiography were performed to screen atherosclerosis in the lower limbs (whether femoral artery or posterior tibial artery had definite plaques), cervical vessels (whether carotid intima-media thickness was more than 1.2 mm), and coronary artery (whether ECG had changes of ST-T which indicated myocardial ischemia).

Statistical analysis

Statistical analysis was performed using SPSS 13.0 software for windows. Values of variables are expressed as mean ± standard deviation (SD) if their data fit the normal distribution and as median (interquartile range) if their data did not fit the normal distribution. Change in the variables after treatment was assessed using a paired t-test or Wilcoxon rank sum test. Pearson or Spearman’s rank correlation analysis was used to assess the association between variables. Multiple linear regression analysis was used with HOMA-IR as a dependent variable. Variables that had skewed distributions were logarithmically transformed (lg10). A two-sided value of P < 0.05 was considered statistically significant in all analyses.

Results

Baseline characteristics

A total of 90 patients were enrolled in our study. The mean age of the patients was 48.11 ± 9.21 years, and the sex ratio (M/F) was 66:24. The FPG, 2hPG, and HbA1c levels were 11.79 ± 3.57 mmol/L, 24.28 ± 5.90 mmol/L, and 11.70% ± 2.40%, respectively. Waist circumference was 87.49 ± 8.26 cm (normal range in Chinese: <85 cm for men and <80 cm for women), and BMI was 24.03 ± 2.93 kg/m2 (normal range in Chinese: 18.5 ~ 23.9 kg/m2). Blood pressure was (129 ± 14)/(83 ± 10) mmHg. Lipid profiles were as follows: TC was 5.82 ± 1.21 mmol/L, TG was 1.76(1.26, 3.05)mmol/L, HDL-C was 1.18 ± 0.29 mmol/L, and LDL-C was 3.73 ± 0.98 mmol/L. Traditional inflammatory markers like hsCRP, WBC and FFA were 1.84(1.08,3.67)mg/L, 7.62 ± 2.02 × 109/L, 607.48 ± 225.00 umol/L, respectively. Tests for diabetic angiopathy showed that 14, 9, and 6 patients had atherosclerosis in the lower limbs, cervical vessels, and coronary artery, respectively (including two patients with atherosclerosis both in the lower limbs and cervical vessels and one patient with atherosclerosis both in the lower limbs and coronary artery).

Changes in the clinical features of diabetic patients after intensive insulin treatment

Ideal glucose control was attained in 4.1 ± 2.0 days and was maintained for 10.0 ± 2.4 days. The maximum dosage of insulin administered was 50.32 ± 13.23 U/day (0.72 [0.64, 0.89] U/kg/day). The frequency of hypoglycemia (≤3.9 mmol/L) during CSII treatment was 2 (1, 3) times. There were no severe hypoglycemic (<2.8 mmol/L) episodes during the short-term intensive interventions. After treatment, the measures of waist circumference, blood pressure, HbA1c, AUC of glucose (AUCGlu) of IVGTT and OGTT, TC, TG, LDL-C, and inflammatory markers (including WBC and FFA, but not hsCRP) decreased significantly (P < 0.05), whereas the levels of HDL-C and AUCIns of IVGTT and OGTT increased significantly after CSII (P < 0.05) (Table 1).

Table 1 Clinical features of 90 newly diagnosed type 2 diabetic patients at baseline and after CSII
Full size table

With regard to the insulin-secretory function of islets, none of the diabetic patients showed first-phase insulin secretion, and the peak of second-phase secretion appeared later than it normally does (120 min vs. 60 min, respectively). After 2 weeks of CSII, the levels of insulin at each time point in both IVGTT and OGTT were significantly elevated (P < 0.001) and the first-phase insulin secretion was recovered, whereas the peak of second-phase secretion was delayed (Fig. 1). Indicators of β-cell function including AIR3–5, ΔINS30/ΔG30, HOMA-β, and MBCI significantly improved after treatment (P < 0.05). The results remained the same when they were divided by HOMA-IR to exclude the influence of insulin resistance (P < 0.001) (Table 1).

Fig. 1
figure 1

Concentrations of blood glucose and insulin during IVGTT and OGTT before and after CSII (n = 90). a Means ± SD for glucose concentrations during IVGTT before (■) and after CSII (▲). The AUCGlu in IVGTT decreased significantly (89.43 ± 17.38 mmol · min vs. 67.01 ± 11.74 mmol · min, P < 0.01) after CSII. b Median (interquartile range) for insulin concentrations during IVGTT before (■) and after CSII (▲). The AUCIns in IVGTT was significantly elevated (11.91 [8.00, 29.05] mU · min vs. 25.64 [10.32, 46.79] mU · min, P < 0.01) after CSII. c Means ± SD for glucose concentrations during OGTT before (■) and after CSII (▲). AUCGlu in OGTT decreased significantly (79.57 ± 18.88 mmol · min vs. 46.49 ± 9.28 mmol · min, P < 0.01) after CSII. d Median (interquartile range) for insulin concentrations during OGTT before (■) and after CSII (▲). AUCIns in OGTT was significantly elevated (27.60 [12.20, 64.37] mU · min vs. 75.45 [46.37, 123.53] mU · min, P < 0.01) after CSII. Values of P < 0.05 are statistically significant. CSII, continuous subcutaneous insulin infusion; IVGTT, intravenous glucose-tolerance test; OGTT, oral glucose-tolerance test; SD, standard deviation; AUCGlu, area under the curve of glucose; AUCIns, area under the curve of insulin

Full size image

With respect to insulin resistance, the HOMA-IR decreased significantly after treatment, but the IAI, QUICKI, and ISIced significantly improved after treatment (P < 0.001) (Table 1).

Plasma levels of Lp-PLA2 and sPLA2 in diabetic patients and their changes after CSII

The plasma levels of Lp-PLA2 and sPLA2 were 102.98 (76.34, 134.31) ng/mL and 219.33 (130.03, 337.30) ng/mL, respectively, in diabetic patients before CSII. In addition, the plasma levels of Lp-PLA2 and sPLA2 in diabetic patients with evidence of atherosclerosis were significantly higher than those in diabetic patients without evidence of atherosclerosis (Lp-PLA2 levels: 133.43 [111.54, 145.17] ng/mL vs. 99.11 [63.02, 130.85] ng/mL; sPLA2 levels: 235.73 [180.48, 416.46] ng/mL vs. 182.97 [90.08, 280.79] ng/mL, respectively, P < 0.001) (Fig. 2a). The differences within and among product batches were <10 and <12%, respectively.

Fig. 2
figure 2

Comparison of the Lp-PLA2 and sPLA2 levels between diabetic patients with macroangiopathy and those without macroangiopathy (133.43 [111.54, 145.17] ng/mL vs. 99.11 [63.02, 130.85] ng/mL, 235.73 [180.48, 416.46] ng/mL vs. 182.97 [90.08, 280.79] ng/mL, respectively) (a); changes of the Lp-PLA2 and sPLA2 levels after continuous subcutaneous insulin infusion (CSII) in all newly diagnosed type 2 diabetes (102.98 [76.34, 134.31] ng/mL vs. 88.35 [76.74, 125.18] ng/mL, 219.33 [130.03, 337.30] ng/mL vs. 173.78 [80.95, 278.09] ng/mL, respectively, n = 90) (b); changes of the Lp-PLA2 and sPLA2 levels after CSII in diabetic patients with or without macroangiopathy (Lp-PLA2: 133.43 [111.54, 145.17] ng/mL vs 92.16 [80.26, 128.78] ng/mL, P <0.05; 99.11 [63.02, 130.85] ng/mL vs. 85.94 [60.53, 119.76] ng/mL, P =0.102; sPLA2: 235.73 [180.48, 416.46] ng/mL vs. 183.76 [93.88, 271.54] ng/mL, P < 0.01; 182.97 [90.08, 280.79] ng/mL vs. 147.88 [76.58, 265.66] ng/mL, P < 0.05) (c). *P < 0.05, ** P < 0.01. Values of P < 0.05 are statistically significant

Full size image

The plasma sPLA2 level was significantly reduced after CSII in diabetic patients (from 219.33 [130.03, 337.30] ng/mL to 173.78 [80.95, 278.09] ng/mL; P < 0.05) (Table 1, Fig. 2b), regardless of whether they had macroangiopathy or not (from 235.73 [180.48, 416.46] ng/mL to 183.76 [93.88, 271.54] ng/mL, P < 0.01; and from 182.97 [90.08, 280.79] ng/mL to 147.88 [76.58, 265.66] ng/mL, P < 0.05) (Fig. 2c). However, we did not find any statistical difference in the plasma Lp-PLA2 level after treatment in all diabetic patients (from 102.98 [76.34, 134.31] ng/mL to 88.35 [76.74, 125.18] ng/mL, P = 0.087) (Table 1, Fig. 2b). Further analysis showed that the decrease of Lp-PLA2 level after treatment in diabetic patients with macroangiopathy was statistically significant (from 133.43 [111.54, 145.17] ng/mL to 92.16 [80.26, 128.78] ng/mL, P <0.05), while it was not the case in those without macroangiopathy (from 99.11 [63.02, 130.85] ng/mL to 85.94 [60.53, 119.76] ng/mL, P >0.05) (Fig. 2c).

Correlation analysis

In patients with newly diagnosed T2DM before CSII, the Lp-PLA2 level was positively correlated with the HbA1c concentration (r = 0.283, P = 0.020), LDL-C level (r = 0.269, P = 0.029), and HOMA-IR (r = 0.309, P = 0.037). Correlation analysis for sPLA2 and clinical parameters showed a significant correlation of sPLA2 with waist circumference (r = 0.243, P = 0.043) and HOMA-IR (r = 0.506, P <0.01). No significant correlation was observed between the Lp-PLA2 or sPLA2 levels and BMI, blood pressure, TC, TG, HDL-C, all indicators used to evaluate β-cell function, and other inflammatory factors (WBC, FFA, and hsCRP) (Table 2).

Table 2 Correlations between Lp-PLA2, sPLA2 and clinical parameters in newly diagnosed type 2 diabetes
Full size table

After CSII, the decrease in the Lp-PLA2 (ΔLp-PLA2) and sPLA2 (ΔsPLA2) levels were both positively correlated with the reduction in HOMA-IR (ΔHOMA-IR) (r = 0.590, P < 0.05 and r = 0.476, P < 0.05, respectively) (Fig. 3a, 3b). Further, multiple linear regression analysis showed that both Lp-PLA2 and sPLA2 levels were independent factors that were correlated with HOMA-IR (β = 0.372 and β = 0.560 respectively, P < 0.05).

Fig. 3
figure 3

Correlation between ΔLp-PLA2 and ΔHOMA-IR in diabetic patients after CSII (r = 0.590, P < 0.05) (a); correlation between ΔsPLA2 and ΔHOMA-IR in diabetic patients after CSII (r = 0.476, P < 0.05) (b). Values of P < 0.05 are statistically significant. ΔLp-PLA2, change in Lp-PLA2 levels; ΔsPLA2, change in sPLA2 levels; ΔHOMA-IR, change in homeostatic model assessment for insulin resistance

Full size image

Discussion

Early intensive insulin therapy could effectively reduce glucotoxicity and lipotoxicity, protect islet function, and alleviate insulin resistance in patients with newly diagnosed T2DM [13]. Using comprehensive indicators, our study showed that early intensive insulin treatment could not only reduce waist circumference, blood pressure, blood glucose (FPG, 2-h postprandial PG, HbA1c concentration, AUCGlu during both IVGTT and OGTT), blood lipid levels (TG, TC, and LDL-C), levels of traditional inflammatory factors (WBC and FFA), and insulin resistance (HOMA-IR), but also distinctly improve the HDL-C levels, insulin sensitivity (IAI, QUICKI, and ISIced) and β-cell function (HOMA-β, MBCI, AIR3–5, ∆Ins30/∆G30, and AUCIns during IVGTT and OGTT) in patients with newly diagnosed T2DM.

In the present study, the plasma concentration of Lp-PLA2 was positively correlated with the HbA1c concentration in patients with T2DM during the early stages of disease, which may indicate an important role of Lp-PLA2 in predicting T2DM [11]. The relationship between sPLA2 and the risk of diabetes is still unclear, and studies with a larger sample size are needed to confirm this relationship. In addition, Lp-PLA2 and sPLA2 levels were significantly increased in diabetic patients with macroangiopathy as compared to those without macroangiopathy. This finding is consistent with the view that Lp-PLA2 and sPLA2 are important risk predictors for atherosclerosis [7, 16]. However, it is uncertain whether plasma Lp-PLA2 and sPLA2 could be markers of subclinical atherosclerosis for diabetic patients [17, 18]. In our study, we found positive correlations between these two enzymes and several atherogenic factors such as LDL-C, waist circumference, and IR in patients with newly diagnosed diabetes. Therefore, we propose that Lp-PLA2 and sPLA2 may affect the incidence of atherosclerosis in diabetic patients. A recent study suggested that the relative distribution of Lp-PLA2 between LDL and HDL determines its pro- or anti-inflammatory action. According to this assumption, Lp-PLA2 is anti-inflammatory when bound to HDL, but pro-inflammatory when bound to LDL [19]. Moreover, an increase in the binding between Lp-PLA2 and LDL was observed in diabetic patients [20], which could support the pro-atherosclerotic role of Lp-PLA2 in diabetes. However, studies on sPLA2 in diabetic patients are lacking in the literature.

Currently, animal experiments and clinical trials involving an Lp-PLA2 inhibitor (darapladib) are underway and have been shown to reduce the progression of atherosclerosis in animals [21, 22] but not in humans [23] with or without diabetes. On the other hand, studies on a sPLA2 inhibitor (varespladib) showed that administration of 500 mg daily for 8 weeks could significantly reduce the post-ACS (acute coronary syndrome) inflammatory response, especially in diabetic patients [24]. In our study, we aimed to determine whether intensive insulin treatment could reduce the levels of Lp-PLA2 and sPLA2 in patients with newly diagnosed T2DM. This issue has not been studied previously and may help in resolving the atherosclerotic complications resulting from intensive insulin therapy in these patients. To our knowledge, this study is the first to provide evidence that the plasma concentration of sPLA2 is significantly decreased after short-term intensive insulin therapy in patients with newly diagnosed T2DM. The more sensitive change in sPLA2 levels could be attributed to its role as an acute-phase protein [25]. Although result of changes in Lp-PLA2 after treatment in all diabetic patients was negative—which may be due to the small sample size and short time of therapy or observation—the level of Lp-PLA2 was significantly decreased in diabetic patients with macroangiopathy. Considering the close relationship between Lp-PLA2 or sPLA2 and atherosclerosis, we believe that intensive insulin treatment could alleviate inflammation in diabetic patients by decreasing the levels of Lp-PLA2 and sPLA2, and such treatment may further prevent or postpone the occurrence and progression of diabetic atherosclerotic complications.

As novel inflammatory markers, both Lp-PLA2 and sPLA2 seem to be related with IR. Studies have suggested that that the concentrations of Lp-PLA2 and sPLA2 are high in patients with metabolic diseases accompanied by IR, such as obesity [26, 27] and polycystic ovary syndrome [28]. Kudolo et al. [29] detected the circulating levels of plasma Lp-PLA2 in nine patients with obesity, six patients with diabetes, and nine healthy controls and for the first time, they showed that the Lp-PLA2 level was significantly correlated with IR. Several studies thereafter observed a similar relation between Lp-PLA2 or sPLA2 and IR, mainly in people with normal blood glucose [11, 30]. Our study indicated the same observations in patients with newly diagnosed T2DM, and the correlation between sPLA2 and IR in diabetic patients was illustrated for the first time. This finding is supported by further results showing that the decrease in both Lp-PLA2 and sPLA2 levels was positively correlated with the reduction in HOMA-IR after intensive insulin treatment. However, the definite relation of the cause and effect between Lp-PLA2 or sPLA2 and IR is still unclear. Numerous studies have shown that low-grade chronic inflammation may induce or aggravate IR [31, 32]. Noto et al. [33] found that hydrolysis of oxidized phospholipids by Lp-PLA2 produces FFA and lysolecithin, both of which stimulate endothelial cells to generate inflammatory cytokines and ultimately induce IR in both muscle and adipose tissue. Furthermore, Iwase [34] and Huang et al. [35] showed that metformin has the potential to improve IR in a rat model of high-fat diet–induced non-alcoholic fatty liver disease. This effect was due to the decrease in sPLA2 levels and inflammation. All the above mentioned studies concluded that Lp-PLA2 or sPLA2 was responsible for the IR. However, several studies have shown contrasting findings. Kudolo et al. [29] believed that the levels of Lp-PLA2 in diabetic patients increased as a result of hyperinsulinemia. In addition, Ramanadham et al. [36] indicated that sPLA2 is present within insulin secretory granules, and stimulation of the Langerhan’s island with insulin secretagogues leads to the co-secretion of insulin and sPLA2. These researchers believe that hyperinsulinemia due to IR increases the Lp-PLA2 or sPLA2 levels. Therefore, the causal relationship between Lp-PLA2 or sPLA2 and IR is still controversial, and further studies are needed to reach a consensus on this issue.

In the current study, we did not observe any correlation between the plasma Lp-PLA2 or sPLA2 levels and β-cell function, irrespective of whether the influence of IR was adjusted for. Mayer et al. found a positive association between Lp-PLA2 activity (not mass) and pancreatic β-cell function [37]. However, studies with a larger sample size are required to establish the association between the plasma Lp-PLA2 or sPLA2 levels and pancreatic β-cell function.

Despite our important findings, there were several limitations to this study. First, the number of patients enrolled in this study was insufficient and the short period of intervention may have resulted in the absence of a significant change in the Lp-PLA2 level after treatment. Second, the Lp-PLA2 and sPLA2 levels in patients with microvascular complications were not discussed because only a few studies have thus far focused on the relationship between microangiopathy and Lp-PLA2 or sPLA2.

Conclusions

In summary, the plasma Lp-PLA2 and sPLA2 levels are increased in patients with newly diagnosed T2DM and macroangiopathy as compared to those with newly diagnosed T2DM but without macroangiopathy. Early intensive insulin therapy can be effectively used to achieve adequate glycemic control, improve β-cell function and IR, and decrease plasma Lp-PLA2 and sPLA2 levels in patients with newly diagnosed T2DM, especially in those with macroangiopathy. The change in both the Lp-PLA2 and sPLA2 levels after CSII were positively correlated with the change in HOMA-IR, and Lp-PLA2 and sPLA2 levels are both independent factors that are positive correlated with IR in patients with T2DM. Therefore, the decrease in Lp-PLA2 and sPLA2 levels by intensive insulin therapy might contribute to the improvement of IR as well as protect from diabetic atherosclerotic complications.