Efficacy and Safety of Tanshinone for Chronic Kidney Disease: A Meta-Analysis

Objective To systematically evaluate the efficacy and safety of tanshinone for chronic kidney disease (CKD). Methods Randomized controlled trials (RCTs) on the treatment of CKD using tanshinone were searched using 4 Chinese databases (China National Knowledge Infrastructure (CNKI), Value In Paper (VIP), Wanfang, and Chinese Biology Medicine (CBM)) and 3 English databases (PubMed, Cochrane Library, and Excerpta Medica Database (Embase)). The results included data on blood urine nitrogen (BUN), serum creatinine (Scr), glomerular filtration rate (GFR), 24 h urine protein, microalbuminuria (mALB), β2-macroglobulin (β2-MG), cystatin C (CysC), and safety events. The data were analyzed using Revman 5.3 and Stata 12.0 software. Results Twenty-one studies were entered into this meta-analysis, which involved 1857 patients including 954 cases from the tanshinone treatment group and 903 cases from the control group. BUN levels in the tanshinone treatment group were significantly reduced compared with the control (standardized mean difference (SMD) = −0.65, 95% confidence interval (CI): −0.81 to −0.49, p < 0.01). In addition, subgroup analysis indicated that tanshinone had a significant effect in reducing Scr levels at 14, 21, and 28 days. Scr levels in the tanshinone treatment group were significantly reduced compared with the control group (SMD = −1.40, 95% CI: −2.09 to −0.71, p < 0.01); subgroup analysis based on treatment time also yielded the same results. GFR in the tanshinone treatment group was better than that in the control group (SMD = 0.83, 95% CI: 0.59 to 1.07, p < 0.01). In terms of urine protein levels, 24 h urine protein level, mALB, and β2-MG of CKD patients were reduced to some degree compared with controls, and CysC levels in the tanshinone treatment group were also significantly reduced compared with the control group (SMD = −0.24, 95% CI: −0.44 to −0.03, p < 0.05). Safety in the tanshinone treatment group did not differ significantly from that of the control group (risk ratio (RR) = 7.78, 95% CI: 0.99 to 61.05, p > 0.05). Conclusion This meta-analysis showed that tanshinone could control urine protein level in CKD patients, improve kidney function, and delay the evolution of CKD without significant side effects. However, the results were limited and should be interpreted with caution because of the low quality of the included studies. In the future, more rigorous clinical trials need to be conducted to provide sufficient and accurate evidence.


Introduction
Chronic kidney disease (CKD) is a condition of functional and structural dysfunction of the kidney with impairment longer than 3 months, caused by various factors. It is characterized by normal or abnormal glomerular filtration rate (GFR) from pathological glomerular injury, abnormal renal function indices in urine or blood, abnormal imaging findings, or decreased GFR for more than 3 months of unknown etiology (<60 mL/min·1.73 m 2 ) [1]. With increasing life expectancy and parallel increases in the standard of living, the incidence of metabolic diseases such as diabetes, hyperlipidemia, and hypertension is consistently on the rise, leading to annual increases in the incidence of CKD [2].
However, there are presently no effective drugs for treating CKD. CKD is treated mainly with hormone therapy and immune suppressors in addition to control urine protein, blood pressure, and blood glucose levels and maintaining homeostasis. Although these therapies can reduce kidney impairment to a certain degree, they are unable to radically delay the evolution of this disease. Furthermore, hormone and immune suppressors are only effective in part of the pathogenesis of kidney injury and lead to high rates of relapse. eir long-term application can also lead to severe infection and osteonecrosis of the femoral head, among other side effects [3,4]. erefore, safe and efficacious drugs are needed for CKD treatment. Tanshinone is the main active ingredient of the frequently used traditional Chinese medicine Danshen. is agent functions to protect body tissues by inducing dilation of the cardiocerebral-renal vasculature, increasing blood flow to ischemic myocardial tissue, scavenging reactive oxygen species (ROS) resulting from lipid peroxidation during organ ischemia/reperfusion injury, and inhibiting intracellular calcium overload. Tanshinone has been widely administered in patients with angina pectoris and myocardial infarction [5,6]. However, research on CKD revealed that tanshinone can also protect the kidney and delay diabetes-related renal fibrosis [7]. Clinically, a number of studies have shown that tanshinone administration can restore a certain degree of kidney function in CKD patients, but this conclusion has yet to be confirmed.
In the present study, published randomized controlled studies (RCTs) on CKD treatment with tanshinone were found, and the efficacy and safety of tanshinone for CKD were systematically evaluated. e results of this metaanalysis provide some evidence for the clinical application of tanshinone.

Literature Search.
is search was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement for the conduct of meta-analyses of intervention studies [8]. Four Chinese databases (China National Knowledge Infrastructure (CNKI), Value In Paper (VIP), Wanfang, and Chinese Biology Medicine (CBM)) and 3 English databases (PubMed, Cochrane Library, and Excerpta Medica Database (Embase)) were used for the literature search. Conference papers and academic dissertations were searched manually. In cases where the information was incomplete, the corresponding authors of those papers were contacted and the information necessary for the meta-analysis was obtained. e literature search was concluded on June 1, 2019, and therefore included papers published before this date.

Study Subjects.
e subjects in the RCTs were patients diagnosed with CKD. Diagnostic criteria were based on the relevant guidelines for CKD diagnosis and treatment.

Preventative Measures.
e treatment group comprised patients treated with tanshinone, and treatment with other medications plus tanshinone was also included. e control group was treated with placebo or conventional western medicine, and any form of tanshinone treatment was excluded.

Criteria for Literature Exclusion.
e exclusion criteria for literature were as follows: (1) cases not definitively diagnosed with CKD, (2) repeatedly published articles, (3) too little information reported to be useful, and (4) the number of cases and treatments not accurately provided.

Data
Extraction. Data were extracted collaboratively by two researchers. ey independently extracted data consistent with the study requirements, recorded the data sequentially in a prepared information sheet, and crosschecked the data. Disagreement was resolved by having a third party to make the decision.

Assessment of Literature Quality.
e literature included in this meta-analysis was subject to quality assessment according to the Cochrane criteria [9,10].
e assessed 2 Evidence-Based Complementary and Alternative Medicine content included random design, blind concealment, blinding, completeness of the final data, selectively reported research results, and publication bias. e risk of each content was evaluated. e risk assessment table was produced using Revman 5.3 software.
2.5. Statistical Analysis. Stata 12.0 and Revman 5.3 software were both used to analyze the data. For consecutive variables, the mean value ± standard deviation (SD) and 95% confidence interval (CI) represented the efficacy of treatment. For binary outcome data, risk ratio (RR) was used to calculate the summary. Heterogeneity was tested by the chi-squared test and I 2 test. A random-effects model was used when a chisquared test (p < 0.10) or I 2 > 50% revealed significant evidence of homogeneity across the studies, or else a fixed-effects model was used. Publication bias was evaluated using Begg's test. p values less than 0.05 indicated statistical significance. If the number of included trials was sufficient, sensitivity analysis was used to assess the stability of the results.

Results of Our Literature Search.
Computer searches and manual searching resulted in 297 publications, of which 127 potentially eligible reports were identified. First, 69 articles were excluded based on identification of titles and abstracts. Next, 2 single-arm studies, 18 animal studies, 3 repeated publications, and 14 articles with uncertain diagnosis or preventative measures were excluded after reading the entire article. Finally, 21 studies were included in the meta-analysis ( Figure 1).

Basic Characteristics of the Included Studies.
In total, 21 studies were included in the present meta-analysis, which involved 1875 patients comprising 954 patients from the tanshinone treatment group and 903 patients from the control group. Authors, year of publication, treatment plan, sample size, course of treatment, and index of evaluation of each study are shown in Table 1.

Risk of Bias Assessment of the Literature Included in the Study.
e 21 studies included in this analysis were all RCTs, but only 3 had detailed descriptions of the methods. One study had a double-blind design. None of the 21 studies had allocation concealment. We could not confirm any other forms of bias because the available information was insufficient. Assessment of bias risk of the studies in the metaanalysis is shown in Figures 2(a) and 2(b).

BUN.
A total of 8 studies compared BUN levels between the tanshinone treatment group and control group. We classified these studies into different subgroups based on the time of tanshinone treatment, namely, 14, 21, and 28 days. As shown in Figure 3(a), there was low heterogeneity (I 2 � 32.7%, p > 0.1), so a fixed-effects model was used to analyze the data. BUN levels in the tanshinone treatment group were significantly reduced compared with the control group (SMD � − 0.65, 95% CI: − 0.81 to − 0.49, p < 0.01). e results of subgroup analysis indicated that tanshinone had a significant effect in reducing BUN levels at 14, 21, and 28 days. As shown in Figure 3(b), Begg's test showed that there was no publication bias (p > 0.05).

Scr.
A total of 9 studies compared Scr levels between the tanshinone treatment group and control group. We classified these studies into different subgroups based on time of tanshinone treatment, namely, 14, 21, and 28 days. As shown in Figure 4(a), the random-effects model was used to analyze the data because of the high heterogeneity. Scr levels in the tanshinone treatment group were significantly reduced compared with the control group (SMD � − 1.40, 95% CI: − 2.09 to − 0.71, p < 0.01). Subgroup analysis indicated that tanshinone had a significant effect on reducing Scr levels for 14, 21, and 28 days. As shown in Figure 4(a), Begg's test showed that there was no publication bias (p > 0.05).

GFR.
A total of 4 studies compared GFR between the tanshinone treatment and control groups. As shown in Figure 5(a), there was low heterogeneity (I 2 � 0%, p > 0.1), so the fixed-effects model was used to analyze the data. e tanshinone treatment group had better GFR than the control group (SMD � 0.83, 95% CI: 0.59 to 1.07, p < 0.01). Begg's test showed that there was no publication bias (p > 0.05; Figure 5(b)).  Evidence-Based Complementary and Alternative Medicine

Quantification of 24 h Urine Protein.
A total of 6 studies quantitatively compared the 24 h urine protein between the tanshinone treatment group and control group. As shown in Figure 6(a), the fixed-effects model was used to analyze the data because of low heterogeneity (I 2 � 35.7%, p > 0.1). Tanshinone significantly reduced 24 h urine protein level compared with the control group (SMD � − 0.94, 95% CI: Begg's test indicated that there was no publication bias (p > 0.05; Figure 6(b)).

mALB.
Two studies compared mALB between the tanshinone treatment group and control group. As shown in Figure 7(a), the fixed-effects model was used to analyze the

β2-MG.
A total of 3 studies compared β2-MG between the tanshinone treatment and control groups. As shown in Figure 8(a), the fixed-effects model was used to analyze the data because of low heterogeneity (I 2 � 8.9%, p > 0.1). Tanshinone significantly reduced β2-MG levels compared with the control group (SMD � − 1.21, 95% CI: − 1.52 to − 0.89, p < 0.01). Begg's test also suggested that there was no publication bias (p > 0.05; Figure 8(b)).

CysC.
A total of 3 studies compared CysC levels between the tanshinone treatment and control groups. As shown in Figure 8(a), the fixed-effects model was used to analyze the data because of low heterogeneity (I 2 � 21.1%, p > 0.1). CysC content in the tanshinone treatment group was significantly lower than that of the control group (SMD � − 0.24, 95% CI: − 0.44 to -0.03, p < 0.05). No publication bias was observed according to Begg's test (p > 0.05; Figure 9(b)).
3.11. Safety. Two publications addressed the safety of tanshinone in treating CKD. In terms of side effects, 1 article reported mild stomach upset in some patients, and in another study patients complained of slight dizziness, with facial flushing that did not affect treatment in another article. As shown in Figure 10(a), the fixed-effects model was used to analyze the data because of low heterogeneity (I 2 � 0%, p > 0.1). Safety in the tanshinone treatment group did not differ significantly from that of the control group (RR � 7.78, 95% CI: 0.99 to 61.05, p > 0.05). is suggested that no significant side effects occurred after the addition of tanshinone. Likewise, no publication bias was observed according to Begg's test (p > 0.05; Figure 9(b)).

Discussion
Danshen is a perennial herbaceous plant belonging to the mint family Lamiaceae [31]. Its root is used as a traditional Chinese medicine that has been widely used clinically for the treatment of cardiovascular diseases. Tanshinone is the active ingredient of Danshen and contains more than 50 compounds such as liposoluble compound tanshinone I, tanshinone IIA, and tanshinone IIB as well as the watersoluble compounds salvianolic acid A, salvianolic acid B, and tanshinol [32], of which tanshinone IIB is the most active ingredient, and tanshinone IIA-sodium sulfonate injection is a water-soluble compound derived from the sulfonation of tanshinone, a diterpene quinone compound isolated from Danshen. Many studies have demonstrated its anti-inflammatory, antioxidative, and ROS scavenging activities [33,34].
CKD is a clinically prevalent, chronic, noninfectious, progressive, and irreversible disease characterized mainly by decreasing GFR. Epidemiological evidence suggests that CKD is mainly the result of glomerular disease and vascular disease caused by diabetes among other reasons [35].
In the present meta-analysis, the effects of tanshinone on CKD were compared with those of the control groups in terms of the three indicators, BUN, Scr, and GFR. Taken together, the results showed that BUN levels significantly decreased in the Gu et al. [30] Liao et al. [29] Wang [22] Yan et al. [20] 0 Overall (I−squared = 35.7%, p = 0.169) Wang [26] Gao and Gao et al. [19] Cao et al. [14] ID Tuxunguli [15] Deng [21] Li and Gao [13]  tanshinone treatment group compared with the control group, and subgroup analysis based on treatment time yielded the same results. For all the CKD patients included in this metaanalysis, Scr levels decreased significantly, while GFR markedly increased. ese results suggest that tanshinone protects the kidney in CKD patients and improves GFR.
A common clinical feature of CKD patients is persistent proteinuria. We first compared 24 h urine protein in the tanshinone treatment and control groups. e comparison revealed that tanshinone treatment significantly reduced urine protein, while Begg's test showed that there was no publication bias, indicating that tanshinone effectively reduced urine protein level in CKD patients. To further confirm this, we reviewed literature on mALB, and the combined results showed that tanshinone effectively reduced the mALB levels in CKD patients. Next, we selected 2 publications on the effects of tanshinone treatment on β2-MG levels in CKD patients, and the combined data showed that tanshinone treatment significantly reduced β2-MG levels in CKD patients.
ese results indicate that tanshinone has a role in controlling proteinuria in patients with CKD.
Presently, CysC is considered an endogenous marker of GFR, reflecting to a certain extent the filtration capacity of the glomerulus [36]. CysC could be an index of renal function. ree studies on CysC were included in the present meta-analysis. e results showed that tanshinone significantly reduced CysC levels in CKD patients compared with those of the control group. Meanwhile, the results of Begg's test did not indicate publication bias, so the results can be taken as credible. e results of the aforementioned studies suggest that tanshinone could effectively reduce CysC levels in CKD patients and thus protect the kidney.
Only 2 studies evaluated the safety of tanshinone. e combined results showed that there was no significant increase in the side effects seen after the addition of tanshinone, suggesting that tanshinone treatment is relatively safe and has few side effects.
Although the effects of tanshinone on CKD were evaluated using a multiple dimensional meta-analysis, there were some limitations. e quality of the literature included in this present meta-analysis was relatively low, which may have affected the reliability of the results. Although the included literature claimed to have a randomized controlled design, only 3 studies clearly explained their methods, and 1 study used the double-blinded method. When combining the Scr data, the subgroup was highly heterogeneous, which may have impacted the reliability of the results. Limited by language barriers, only Chinese and English databases were searched, and only Chinese studies were included in the meta-analysis, which may have affected the final results to a certain degree.

Conclusion
Tanshinone controls urine protein level in CKD patients, improves kidney function, and delays the progression of CKD. Meanwhile, no major adverse effects have been reported. However, this conclusion should be interpreted with caution because of the poor quality of the included trials. More rigorous clinical trials including double-blind, randomized, placebo-controlled designs are required to provide more accurate evidence for the future.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Disclosure
Yao Zhou and Shi-min Jiang are co-first authors.

Conflicts of Interest
e authors declare that they have no conflicts of interest.