Therapeutic effects of Salvia miltiorrhiza injection combined with telmisartan in patients with diabetic nephropathy by influencing collagen IV and fibronectin: A case-control study.

Involvement of collagen IV (ColIV) and fibronectin (FN) in the occurrence and development of diabetic nephropathy (DN) and the effects of telmisartan and Salvia miltiorrhiza injection in the treatment of the patients were investigated. Two hundred and fifty-eight patients with stage IV DN were selected as the case group, and another 110 normal healthy subjects were incorporated as the control group. Involved patients were subdivided into different groups according to different treatment therapies; patients in the telmisartan group (T group) were given oral telmisartan; patients in the Salvia miltiorrhiza injection + telmisartan (S + T group) were administered with Salvia miltiorrhiza injection combined with telmisartan treatment, and there was a group of patients who received no intervention as the placebo group. After intervention, levels of glycemic indexes and renal damage indexes indicated downwards trends both in the T group and the S + T group when compared to the placebo group; besides, levels in the S + T group were much lower than those in the T group (all P<0.05). Additionally, in comparison among the above three intervention groups, differences in the fasting blood glucose, 2 h post-prandial blood glucose, glycosylated hemoglobin, blood urea nitrogen, serum creatinine and urinary albumin excretion rate were significant after treatment (all P<0.05). Further, before intervention, both Co1IV and FN in the urine were increased in the case group compared to the control group (all P<0.05). After intervention, both levels were apparently decreased. There were remarkable differences of Co1IV and FN levels in the urine when compared among three different intervention groups after treatment (P<0.05). Increased ColIV and FN levels may be partially responsible for the development of DN. Salvia miltiorrhiza injection with telmisartan have beneficial synergistic effects for DN patients through attenuating the increase in ColIV and FN, reversing hyperglycemia state and postponing ultrastructure changes of glomerular basement membrane.

RCT Entities:   Population Interventions Outcomes

Introduction

Diabetes is a group of metabolic diseases characterized by a sustained increase of high blood glucose levels over a prolonged period, acting as one of the leading causes of morbidity and mortality worldwide; type 1 diabetes (T1D) and type 2 diabetes (T2D) are the two major types (1). Diabetic nephropathy (DN) is estimated to be one of the major serious chronic complications in patients with diabetes, affecting approximately one third of people with T1D or T2D (2). The rising tide of DN may be greatly associated with the dramatically increasing rate of diabetes population currently, with concomitant increase in associated cardiovascular mortality and end-stage renal disease, producing significant social and economic ramifications in the developing world particularly (3,4). According to its pathophysiological changes, DN can be divided into the following five stages: glomerular filtration and renal hypertrophy stage (stage I), normal albumin urine stage (stage II), early stage of DN (stage III), clinical stage of DN (stage IV), and end stage renal failure (stage V). Main known risk factors affecting progression of DN include baseline albumin excretion, senior age, glucose/lipids/protein metabolism disorders, blood pressure, oxidative stress, hemodynamic changes and the involvement of varieties of cytokines/inflammatory factors, but the above factors fail to explain all of the interindividual variability for the rates of developing DN (5).

Collagen IV (ColIV or Col4) is an exclusively distributed collagen within the basement membrane zone, of which the basement membrane is an extracellular matrix (ECM) separating connective tissue from epithelia, endothelia, muscle fibers and the entire nervous system (6,7). Mutations of the gene coding for ColIV may cause thinning and splitting of the glomerular basement membrane, which present as isolated hematuria, sensorineural hearing loss, and ocular disturbances (8,9). Clinically, the deposition of ColIV in the liver may induce liver fibrosis and cirrhosis, whereas decreased levels may be found in serum or hepatic tissues following successful treatment (10). In addition, increased glomerular and mesangial deposition of ColIV may be positively correlated with the extent of renal injury, and thereby partially responsible for the development of DN (11). On the other hand, fibronectin (FN) is a high-molecular weight glycoprotein of the ECM binding to membrane-spanning receptor proteins called integrin (12). Similar to integrin, FN can bind ECM components such as collagen, fibrin, and heparan sulfate proteoglycans (13). Furthermore, FN plays a major role in cell adhesion, growth, migration, and differentiation, and it is important for wound healing and embryonic development (14–16). Altered FN expression, degradation, and/or organization have been associated with a number of pathological changes, including fibrosis and cancer (17,18).

Because of the pathogenic complexity of DN, new therapeutic interventions targeting primary mechanisms contributing to renal damage are critical for the future treatment of this disease. In clinical care, according to the objective and intuitive analysis of the human body, as well as corresponding etiological analysis based on anatomy and physiology, western medical treatment focuses on the control of blood glucose and blood pressure, swelling reduction, control of protein uria and protection of renal function, significantly different from traditional Chinese medicine, strengthening a balance of yin and yang on the basis of the yin and yang theory (19,20). Telmisartan is an angiotensin II receptor antagonist [angiotensin receptor blocker (ARB)] which is mainly used in treating hypertension (21). At present, angiotensin converting enzyme inhibitors (ACEI) and ARB have attracted extensive attention in the treatment of DN. Studies have shown that ACEI and ARB can significantly reduce the progression of proteinuria and maintain the stability of serum creatinine (Scr) mainly through hemodynamic and nonhemodynamic mechanisms (22,23). However, the exact mechanism of the latter is unclear. Furthermore, Salvia miltiorrhiza injection is isolated and extracted from the root of Salvia miltiorrhiza known as Danshen (24). Danshen is a traditional Chinese medicine that can activate the anti-oxidization system in the body to eliminate free-radicals and to prevent calcium overload (25). The injection of Salvia miltiorrhiza have been used commonly and successfully clinically for treating cardiovascular diseases (26,27). For example, such injection has been suggested to scavenge the oxygen free radicals generated from ischemia-reperfusion injury in the myocardium as effectively as SOD (28). Yet related application and potential mechanism for its therapeutic basis is poorly understood for DN patients.

In this study, we aimed to investigate the roles of ColIV and FN in kidney of DN patients and the curative effects of telmisartan and Salvia miltiorrhiza injection, and to explore the mechanism of ColIV and FN taking part in the occurrence and development of DN and the possible nonhemodynamic mechanisms of telmisartan and Salvia miltiorrhiza injection to protect the kidneys. Our study might provide theoretical basis for the pathogenesis, prevention and treatment of DN.

Materials and methods

Ethics statement

All procedures performed in studies involving human participants were approved and permitted by the Ethics Committee of the Third Affiliated Hospital of Guangzhou Medical University (Guangzhou, China), in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. All the patients signed the informed consents before the investigation.

Subjects' selection

A total of 258 patients diagnosed with stage IV DN were selected in the Third Affiliated Hospital of Guangzhou Medical University from January 2014 to January 2015 as the research subjects (Clinical registration number of ChiCTR-IPR-17011581). The diagnosis of diabetes mellitus was confirmed according to the definition, diagnosis and classification of diabetes mellitus proposed by the World Health Organisation (1999), with fasting blood glucose (FBG) ≥7.0 mmol/l (126 mg/dl) or OGTT 2 h plasma glucose ≥11.1 mmol/l or symptoms of diabetes + the levels of blood glucose detected at any time ≥11.1 mmol/l (200 mg/dl). Stage IV DN was diagnosed and staged in accordance with the internationally recognized Mogensen staging criteria (29) namely the urinary albumin excretion rate (UAER) >300 mg/day at least 2 times within 6 months.

Stage IV DN patients were incorporated and patients with other causes of renal damage were excluded from this study, in combination with the following inclusion criteria: i) patients with clear history of diabetes; ii) UAER >20 g/min (30 mg/l), or positive urinary protein or 24 h urine quantitative protein >0.5 g determined by conventional detection method; iii) patients with mild renal function (Scr <265 umol/l); iv) patients who met the above diagnosis criteria; v) patients with age ranged from 18–70 years. Exclusion criteria: i) patients who were clinically confirmed with nephritis, urinary tract infection, renal vascular disease and urinary tract obstruction diseases that could cause proteinuria; ii) pregnant or lactating women; iii) patients who had grievous heart, liver, renal dysfunction; iv) patients with thyroid disease, Cushing disease and other diseases of the endocrine system; v) patients of allergic constitution or allergic to the present medicine; vi) psychiatric patients and who were unable to cooperate properly; vii) patients younger than 18 years and aged more than 70 years. Furthermore, patients who had incomplete data during the experiment were also excluded from this study. According to the random number table method, enrolled patients were equally divided into three groups, 86 cases in each group; patients in two groups received drugs treatment, and patients in another group received placebo therapy. In addition, a total of 110 age and sex-matched normal healthy subjects receiving health examination in the same hospital were incorporated in this study during the same period, without diabetes history in the kidney examination and imaging detection. Baseline characteristics of all the subjects are listed in Table I, including sex, age, disease duration, blood pressure, total cholesterol (TC), triglycide (TG), high density lipoprotein cholesterol (HDL-C), low density lipoprotein cholesterol (LDL-C), very low density lipoprotein cholesterol (VLDL-C), FBG, 2 h post-prandial blood glucose (2h PBG), and glycosylated hemoglobin (HbA1c), blood urea nitrogen (BUN), Scr and UAER.

Table I.

Comparison of baseline characteristics between the case group and the control group.

Variables	Case group (n=258)	Control group (n=110)
Sex (Male/Female)	140/118	52/58
Age (years)	52.00±3.39	50.00 ±2.87
Disease duration (years)	5.70±1.23
SBP (mmHg)	140.00±15.00	134.00±12.00
DBP (mmHg)	85.00±10.00	82.00±9.00
TC (mmol/l)	4.44±0.41	4.23±0.61
TG (mmol/l)	1.16±0.13	1.01±0.12
HDL-C (mmol/l)	1.42±0.13	1.14±0.14
LDL-C (mmol/l)	2.42±0.46	2.22±0.38
VLDL-C (mmol/l)	0.96±0.18	0.85±0.14
FBG	8.04±0.72	6.29±1.02
2h PBG	12.09±1.40	8.27±0.79
HbA1c	10.49±1.53	8.22±1.46
BUN (mmol/l)	9.83±2.99	9.31±1.90
Scr (µmol/l)	109.74±8.77	79.50±3.46
UAER	126.73±11.14	20.05±1.31

SBP, systolic blood pressure; DBP, diastolic blood pressure; TC, total cholesterol; TG, triglycide; HDL-C, high density lipoprotein cholesterol; LDL-C, low density lipoprotein cholesterol; VLDL-C, very low density lipoprotein cholesterol; FBG, fasting blood glucose; 2h PBG, 2 h post-prandial blood glucose; HbA1c, glycosylated hemoglobin; BUN, blood urea nitrogen; Scr, serum creatinine; UAER, urinary albumin excretion rate.

Sample collection

Both the blood samples and urine samples were collected from all the included subjects. Venous blood samples (5 ml) were obtained from each subject early in the morning after overnight fasting. Supernatants were harvested by centrifugation (2,500 × g, 4°C) for 10 min and the samples were frozen at −80°C. Further, another 2-hours postprandial (after meal) venous blood sample (2.0 ml) was taken and centrifuged as above. Urine samples (10–15 ml) of each patient were collected with plugged tubes and stored at 2–8°C for further usage.

Therapeutic regimens

Two groups of patients were given routine diabetic diet management (to ensure balance and optimal nutrition of carbohydrates, protein and fat for the daily needs of human body) and oral hypoglycemic agents and/or insulin to control blood glucose within the range of ideal level (FBG <8.0 mmol/l, postprandial blood glucose <10.0 mmol/l). Subjects in the T group (telmisartan group) were given oral telmisartan (Zhejiang Tailison Pharmaceutical Co. Ltd., Jiaxing, China; batch no. H20140032, specifications: 40 mg/tablet) at a dose of 40 mg daily for 4 weeks. Patients in the S + T group (Salvia miltiorrhiza injection + telmisartan) were administered with preprandial Salvia miltiorrhiza intravenous injection by the patients themselves combined with the administration of telmisartan, and the use of telmisartan was the same as in the T group, at a dose of 40 mg per day for 4 weeks. Salvia miltiorrhiza injection was offered by a local pharmaceutical company (China Pharmaceutical University Pharmaceutical Co. Ltd., Nanjing, China; batch no. Z32020678, specifications: 10 ml/branch). Patients in the S + T group were treated by intravenous drip of 20 ml Salvia miltiorrhiza injection with 250 ml physiological saline injection once daily, in a total of 4 weeks of continuous medication.

Main outcome measures

Glucose level in the serum of incorporated patients were compared in the 1st week, 2nd week and 4th week after treatment. The measurement of FBG, 2h PBG, and HbA1c was conducted with a biochemical analyzer (Beckman-700, Beckman Coulter, Inc., Fullerton, CA, USA). Blood pressure (systolic blood pressure and diastolic blood pressure) monitoring was based on a standard mercury sphygmomanometer on the right arm of each patient. Measurements were usually taken by the same investigator of the same item. Furthermore, a comparison of renal function indexes was performed before and after treatment (1st week, 2nd week and 4th week). Scr and BUN levels were examined by using full automatic biochemical analyzer (Beckman-700, Beckman Coulter, Inc.). The levels of UAER was detected in 24 h urine samples by radioimmunoassay.

In addition, the levels of ColIV and FN in urine samples of patients were examined before treatment and in the 1st week, 2nd week and 4th week after treatment. Urine ColIV and FN levels were detected by enzyme-linked immunosorbent assay (ELISA) kit. Due to the low concentration of ColIV in urine, collected samples was concentrated and then tested. The concentration processing was as follows: 10 g/l bovine γ-globulin (0.8 ml) and 450 g/l PEG-4000 (5 ml) were added in the 10 ml urine specimens in turn, and the prepared specimens were mixed subsequently. Followed by centrifugation at 1,500 × g for 30 min at 4°C, the supernatant was removed, and samples were dissolved in 1 ml of physiological saline to achieve 10-fold concentration. The absorbance (A) [optical density (OD)] value was determined in dual wavelength mode (450 nm) with a calibration wavelength of 570 nm by full wavelength spectrophotometer (Hitachi, Ltd., Tokyo, Japan). Using OD value as a vertical coordinate, the standard product concentration as the horizontal coordinate, the standard curve was drawn and the urine ColIV and FN levels were calculated accordingly. The test was repeated three times to ensure the repeatability of the experiment.

Adverse events

Adverse drug reaction was defined as adverse consequences of drug therapy in the course of treatment (30). Adverse events of patients in each drug intervention groups of T group and S + T group were observed in an observation period of 6 months, and the incidence rate of adverse reaction was recorded simultaneously. Observed adverse drug reaction included dyslipidemia, nausea and headache, dizziness, fatigue, ankle oedema, constipation, gastrointestinal reaction, hyperphosphatemia, hypercalcemia and hypotension (31,32). Various adverse events unrelated to the purpose of the treatment were carefully recorded in the target patients, including the extent, duration, requirement for special handling and the clinical outcome. All the adverse events and therapeutic regimens were analyzed according to the following five items, namely, ‘affirmation’, ‘probably’, ‘may’, ‘suspicious’ and ‘impossible’ (33) and the first three items were predefined as drug adverse reactions. Adverse drug reaction and corresponding rates were then compared between groups.

Statistical analysis

Using SPSS 22.0 statistical software (IBM Corp., Armonk, NY, USA), measurement data were presented as mean ± standard deviation. To compare continuous variables with normal distribution, we used student's t-test or analysis of variance (ANOVA), followed by post hoc test (Least Significant Difference), for multiple group comparisons. Bilateral P<0.05 was statistically significant

Results

Basic information comparison among groups

As shown in Table I, there were 258 cases included in the case group (male/female: 140/118), all of whom were diagnosed with stage IV DN, with a mean age of 52.00±3.39 years. Further, in the control group, there were 110 cases of subjects, including 52 males and 58 females, with a mean age of 50.00±2.87 years. No differences in age and sex were found between groups (both P>0.05), suggesting comparability between groups. In addition, duration of disease in the case group was 5.70±1.23 years; further, in the case group, after being divided into the three case groups (the placebo group, the T group, and the S + T group) randomly and equally, there were 86 cases in each group.

Comparison of glycemic indexes and renal damage indexes

Table II shows the comparison results of glycemic indexes and renal damage indexes between the control group and the case group, as well as within different intervention groups. Firstly, in the comparison between the case group and the control group, parameters all showed significantly increased trends in the case groups, with statistically differences (all P<0.05).

Table II.

Comparison results of glycemic indexes and renal damage indexes between the control group and the case group, within different intervention groups.

		Case group
Variables	Control group (n=110)	Placebo group (n=86)	T group (n=86)	S + T group (n=86)
FBG	6.18±1.98	7.80±1.49[a]	8.36±2.64[a]	7.96±2.20[a]
2 h PBG	8.56±2.04	13.17±1.10[a]	11.75±2.45[a]	11.63±1.15[a]
HbA1c	7.56±1.76	11.00±2.35[a]	10.93±2.96[a]	11.62±1.91[a]
BUN (mmol/l)	9.25±1.88	10.12±4.02[a]	11.81 ± 2.92[a]	11.83±3.78[a,b]
Scr (µmol/l)	81.05±8.49	117.21±7.61[a]	113.65 ± 6.25[a]	119.64±7.17[a]
UAER	20.04±2.67	120.00±19.94[a]	129.16±15.17[a]	127.46±17.38[a,b]
ColIV	165.40±10.59	193.61±28.16[a]	186.04±24.37[a]	195.44±23.65[a,b]
FN	69.04±9.64	102.55±13.67[a]	95.95±12.12[a,b]	95.56±18.58[a,b]

FBG, fasting blood glucose; 2h PBG, 2 h post-prandial blood glucose; HbA1c, glycosylated hemoglobin; BUN, blood urea nitrogen; Scr, serum creatinine; UAER, urinary albumin excretion rate; ColIV, collagen IV; FN, fibronectin.

P<0.05, comparing target parameters between the control group and the case group, including the placebo group (with no drugs intervention, just placebo therapy), the T group (telmisartan group, given oral telmisartan administration), the S + T group (Salvia miltiorrhiza injection + telmisartan, administered with Salvia miltiorrhiza injection combined with telmisartan treatment)

P<0.05, comparing target parameters between the placebo group and the T group and the S + T group, respectively.

Furthermore, as presented in Table II, comparison within the intervention groups revealed that there were statistical differences with respect to correlated levels after intervention (all P<0.05), all indicated downwards trends both in the T group and in the S + T group when compared to the placebo group; besides, levels in the S + T group were much lower than those in the T group. In comparison, there were statistical significance regarding levels of BUN and UAER between the placebo group and the S + T group. Furthermore, as shown in Table III, difference in the FBG, 2h PBG, HbA1c, BUN, Scr and UAER were shown to be significant at different time-points after treatment when compared to the levels before treatment in the T group. There were signifcant differences in the comparison of the above-mentioned glycemic indexes at different time-points after treatment than those before treatment for S + T group, which are presented in Table IV. Furthermore, the majority of the detected parameters were significantly decreased in the comparison of the 1st week after treatment with the 2nd and 4th week, and in the comparison between the 2nd week and the 4th week, showing statistical difference (all P<0.05).

Table III.

Comparison of results of glycemic indexes and renal damage indexes within telmisartan group with oral telmisartan administration at different time-points after treatment.

Variables	Before treatment	1 week after treatment	2 weeks after treatment	4 weeks after treatment
FBG	8.36±2.64	7.96±1.58	8.41±1.11[a,b]	7.49±0.97[a,b,c]
2 h PBG	11.75±2.45	9.41±1.69[a]	10.75±1.74[a,b]	8.03±1.64[a,b,c]
HbA1c	10.93±2.96	10.68±1.48	11.26±2.33[a]	8.91±0.94[a,b,c]
BUN (mmol/l)	11.81±2.92	10.64±1.63[a]	9.06±1.27[a,b]	8.43±2.13[a,b,c]
Scr (µmol/l)	113.65±6.25	103.69±9.88[a]	89.62±6.95[a,b]	79.99±6.62[a,b,c]
UAER	129.16±15.17	117.41±15.89[a]	71.92±7.63[a,b]	103.36±8.47[a,b,c]
ColIV	186.04±24.37	168.86±15.47[a]	160.44±18.58[a,b]	156.69±10.39[a,b]
FN	95.95±12.12	91.78±7.79[a]	83.56±7.73[a,b]	77.03±7.73[a,b,c]

FBG, fasting blood glucose; 2h PBG, 2 h post-prandial blood glucose; HbA1c, glycosylated hemoglobin; BUN, blood urea nitrogen; Scr, serum creatinine; UAER, urinary albumin excretion rate; ColIV, collagen IV; FN, fibronectin.

P<0.05, comparing target parameters at different time-points before treatment and after treatment

P<0.05, comparing target parameters at different time-points after treatment, including the comparison between the first week and the second/fourth week, respectively

P<0.05, comparing target parameters at different time-points between the second week and the fourth week.

Table IV.

Comparison of results of glycemic indexes and renal damage indexes within Salvia miltiorrhiza injection + telmisartan group administered with Salvia miltiorrhiza injection combined with telmisartan treatment at different time-points after treatment.

Variables	Before treatment	1 week after treatment	2 weeks after treatment	4 weeks after treatment
FBG	7.96±2.20	7.39±1.62[a]	6.19±1.36[a,b]	6.22±1.12[a,b]
2 h PBG	11.63±1.15	11.73±2.74	8.55±1.58[a,b]	7.98±1.15[a,b,c]
HbA1c	11.62±1.91	10.10±2.56	10.35±2.47	8.34±1.14[a,b,c]
BUN (mmol/l)	11.83±3.78	9.97±2.11	9.64±2.46	7.76±3.05[a,b,c]
Scr (µmol/l)	119.64±7.17	112.13±8.89	99.05±7.49[a,b]	81.94±5.37[a,b,c]
UAER	127.46±17.38	108.15±10.43[a]	93.28±7.84[a,b]	66.38±10.02[a,b,c]
ColIV	195.44±23.65	171.21±24.28[a]	177.40±29.07[a]	164.44±21.01[a,c]
FN	95.56±18.58	79.71±17.24[a]	72.09±11.19[a,b]	61.16±10.54[a,b,c]

FBG, fasting blood glucose; 2h PBG, 2 h post-prandial blood glucose; HbA1c, glycosylated hemoglobin; BUN, blood urea nitrogen; Scr, serum creatinine; UAER, urinary albumin excretion rate; ColIV, collagen IV; FN, fibronectin.

P<0.05, comparing target parameters at different time-points before treatment and after treatment

P<0.05, comparing target parameters at different time-points after treatment, including the comparison between the first week and the second/fourth week, respectively

P<0.05, comparing target parameters at different time-points between the second week and the fourth week.

ColIV and FN levels at different time periods after treatment

As listed in Table II, it was detected that before intervention, both the levels of Co1IV and FN in the urine were all increased evidently in the case groups compared to the control group, showing statistical difference (both P<0.05). Besides, within the case groups, urine FN level was decreased in the T group and the S + T group when compared with the placebo group, showing statistical differences (all P<0.05); yet none obvious statistical difference was found in Co1IV that requires further identification with larger sample size.

In addition, after intervention by Salvia miltiorrhiza injection and telmisartan, both Co1IV and FN levels were apparently decreased, especially in the S + T group, with statistical difference (P<0.05). Differences in the levels of Co1IV and FN were shown to be significant at different time-points after treatment in the T group and in the S + T group (Tables III and IV, respectively). To be specific, there were statistical differences in that both urine levels of Co1IV and FN were significantly decreased comparing data before and after treatment at different time-points of the 1st week, 2nd week and 4th week (all P<0.05). Statistical differences were also found in levels of Co1IV and FN in the urine indicated by decreased tendency at the 2nd and 4th week when compared to the 1st week (all P<0.05). Importantly, results indicated that there were significantly positive correlations of Co1IV and FN with FBG, 2h PBG, HbA1c, BUN, Scr and UAER, corresponding differences were significant (all P<0.05).

Comparison of adverse drug reaction

During the observation period of 6 months, there were 5 cases of dyslipidemia, 7 cases of fatigue, 4 cases of nausea and headache, 2 cases of constipation and dizziness, respectively in the T group, and the rate of adverse drug reaction was 20.93%. Furthermore, in the S + T group, there were 3 cases of dyslipidemia, 4 cases of fatigue, 2 cases of constipation, 2 cases of nausea and headache, and corresponding adverse drug reaction rate was 12.79%. No other adverse drug reaction was found in either group. The statistical difference of adverse drug reaction was much lower in the S + T group compared to that in the T group (P<0.05).

Discussion

DN is one of the common and severe microvascular complications of diabetes, and end-stage renal disease caused by DN has become a major cause in chronic renal failure (34). At present, the pathogenesis of DN has not been fully elucidated, and previous studies focused on the exploration of lipid metabolism, oxidative stress, hemodynamics, vasoactive substances and genetic susceptibility (35–37). In general, DN is characterized by glomerular and tubular structural changes, including the following pathological features, such as the thickening of glomerular and tubular basement membrane, the expansion of mesangial matrix, and the occurrence of renal interstitial fibrosis. Besides, typical clinical pathological features are the accumulation of glomerular basement membrane and renal tubular basement membrane, as well as the ECM accumulation in mesangial cells, which eventually contribute to the occurrence of diffuse and nodular glomerulosclerosis (38,39).

Statistical analysis of the present study indicated that urine levels of ColIV and FN were significantly higher in DN patients compared to normal controls; further results suggested that there were obvious positive correlations of ColIV and FN with the levels of FBG, 2h PBG, HbA1c, BUN, Scr and UAER. The change of Scr concentration is mainly determined by the glomerular filtration rate, and the increased concentration of Scr usually indicates a decreased filtration capacity and thus correlated with a damage in the kidney accurately. Besides, both BUN and UAER are common and accurate parameters for the evaluation of renal impairment. Those results indicated that increased levels of ColIV and FN in the urine might indicate the progression of glomerular and tubular dysfunction, and correlated obviously with blood glucose levels changes. Higher PBG, HbA1c levels indicated the more severe and longer glucose toxicity on the kidney, suggesting the importance and necessity of 2h PBG and HbA1c regulation in the prevention and treatment of DN.

After intervention treatment, the majority of blood glucose correlated parameters and renal damage indexes all showed obvious decreased trends along with the time of treatment when compared to the DN patients without intervention, suggesting that both the treatment of Salvia miltiorrhiza injection and telmisartan might have positive effect in decreasing blood glucose and retarding kidney damage. Furthermore, such positive effect was much more obvious in the subgroup managed by Salvia miltiorrhiza injection with telmisartan than treated by telmisartan alone, highlighting that there might be certain synergistic effect of both drugs. It should be noted that both levels of ColIV and FN in urine indicated evidently decreased tendency in the targeted patients, we therefore speculated that Salvia miltiorrhiza injection with telmisartan have beneficial treatment effect for patients with DN through attenuating the increase in ColIV and FN, inversing the increased levels of FBG, 2h PBG, HbA1c, BUN, Scr and UAER, and thus postponing or reversing the changes of the ultrastructure of glomerular basement membrane. Meanwhile, in our study, relevant adverse drug reactions were reported in patients under intervention, such as dyslipidemia, fatigue, nausea and headache, constipation and dizziness, and the observed rate of adverse drug reaction was much lower in the S + T group than compared to that in the T group, which might suggested that the combination of Salvia miltiorrhiza injection and telmisartan would be superior to single application of telmisartan for DN patients, combined with relatively lower adverse drug reaction. Importantly, all recorded adverse drug reactions were of mild nature that did not influence the continuity of treatment in the study, suggesting the efficacy and feasibility of both therapies in the treatment of DN.

Generally speaking, the glomerular is the filtering device of the kidney, and mesangial cells and ECM support the capillary network of glomerular (40). DN is found to be caused by excessive accumulation of ECM, especially the deposition of ColIV, which is the result of the damage of glomerular function and eventually leading to renal fibrosis (41). Secreted by endothelial cells and epithelial cells of special membrane protein, ColIV is the main chemical composition of the basement membrane; its abnormal synthesis and degradation correlate significantly with the development of DN (42,43). As for FN, it is an important component of ECM, and high expression of FN can be detected in the glomerular mesangial region, as well as in the renal tubule (44,45). Excessive accumulation of FN in glomerular and tubular cells may be the main reason for the declined kidney function (46). In this study, before intervention, there were more positive increases of ColIV and FN in the DN patients than compared to the normal healthy controls. We therefore postulated that under hyperglycemic state, ColIV and FN were both increased in synthesis and decreased in degradation, resulting in excessive accumulation of ECM, critical in the damage of renal function and further renal fibrosis. After telmisartan treatment alone or combined with Salvia miltiorrhiza injection, downward trends of the detected parameters were found in the intervened subjects when compared with those without treatment, and the hyperglycemic state was reversed, indicating that Salvia miltiorrhiza injection with telmisartan might be effective in suppressing the expression and syntheses of ColIV and FN in the kidney. Possible functionary mechanism may be correlated with the reason that the major component of Salvia miltiorrhiza injection is tanshinone, which can improve the microcirculation of the human body and renal blood flow by eliminating free radicals and decreasing lipid peroxidation reaction (47,48). Besides, telmisartan instead of angiotensin II receptor can indicate high affinity binding of AT1 receptor subtype, and the combination action may be persistently effective for lowering blood pressure (49).

However, there were some limitations that should be taken into account in the study. The primary limitation was the relatively small sample size and short observation duration. Insufficient samples might always result in unstable index detection that might affect the overall accuracy, and some results with statistical differences were difficult to reflect. In addition, relatively short observation duration, especially for therapeutic effects evaluation, might result in an underestimation of risk or effect; longer period of observation would be better for more comprehensive assessment of the clinical data. Future studies should be performed on the basis of larger sample size and longer observation duration to evaluate the curative effect and safety of Salvia miltiorrhiza injection and telmisartan. Secondly, further studies are recommended to assess the significance of dose-dependent and time-dependent drugs in the treatment of different stages of DN. Generally, accurate dosage and time control can bring the effect of drugs into full play to reduce adverse effect and prolong effect time. Therefore, further studies in vivo and in vitro are recommended to investigate the value of dose-dependent and time-dependent manner of the present drugs for DN patients in a larger sample size population within longer period of observation.

In conclusion, increased ColIV and FN levels may be partially responsible for the development of DN, associated significantly with the hyperglycemia state and ultrastructure changes of the glomerular basement membranes. Salvia miltiorrhiza injection with telmisartan have beneficial synergistic effect and lower adverse drug reaction for DN patients through attenuating the increase in ColIV and FN, inversing hyperglycemia state and postponing ultrastructure changes of glomerular basement membrane.