The Efficacy and Safety of Traditional Chinese Medicine Tonifying-Shen (Kidney) Principle for Primary Osteoporosis: A Systematic Review and Meta-Analysis of Randomized Controlled Trials

Objective This study aimed to appraise the efficacy and safety of the tonifying-Shen (kidney) principle (TS (TK) principle) for primary osteoporosis (POP). Methods Randomized controlled clinical trials (RCTs) using the TS (TK) principle for POP were searched from eight electronic databases to search for relevant literature that was published from the initiation to September 2019. Two reviewers performed study selection, data extraction, data synthesis, and quality assessment independently. Review Manager 5.3 software was used to assess the risk of bias and conduct the data synthesis. We assessed the quality of evidence for outcomes by using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system. Results Thirty-six studies with 3617 participants were included. Meta-analysis showed a consistently superior effect of the TS (TK) principle combined with conventional Western medicine (CWM) in terms of total effectiveness rates (RR = 1.28; 95% CI (1.23, 1.33); P < 0.00001), BMD of the lumbar spine (SMD = 0.71; 95% CI (0.47, 0.95); P < 0.00001) and proximal femur (SMD = 0.94; 95% CI (0.49, 1.38); P < 0.00001), TCM symptom integral (SMD = −1.23; 95% CI (−1.43, −1.02); P < 0.00001), and VAS scores (SMD = −3.88; 95% CI (−5.29, −2.46); P < 0.00001), when compared to using CWM alone and with significant differences. Besides, in respect of adverse effects, it showed no significant statistical difference between the experimental and control groups, RR = 0.99 and 95% CI (0.65, 1.51), P=0.97. Conclusion Our meta-analysis provides promising evidence to suggest that using the TS (TK) principle combined with CWM for POP is more effective than using CWM alone. Also, both of them are safe and reliable for POP.


Introduction
Primary osteoporosis (POP), also called "a silent disease," is one of the pandemic public health issues that seriously threaten people's health all over the world [1]. eoretically, POP and secondary osteoporosis belong to the field of osteoporosis (OP). POP includes two major types: postmenopausal osteoporosis (PMOP) and age-related or senile osteoporosis [2]. POP is characterized by decreased bone mass and degenerated bone microstructure, which contributes to a high risk of bone fragility and fracture [3,4]. It is estimated that the number of individuals aged 50 years or over at high risk of osteoporotic fracture worldwide was at 158 million in 2010 and is set to double by 2040 [5]. In China, because of the largest population and the increasing proportion of elderly people, osteoporosis has become a severe challenge to the Chinese family, society, and government [6,7]. erefore, an effective prevention and treatment method is urgently needed for POP. Management of POP includes nonpharmacologic treatment and pharmacologic treatment. Conventional Western medicines (CWM) including antiresorptive or anabolic are widely used in the treatment of POP [8][9][10]. However, due to adverse effects and risk of cancer, the use of CWM as long-term treatment is limited [11].
It is well documented that traditional Chinese medicine (TCM) is commonly combined with pharmacotherapy for POP in China [12][13][14]. In the theory of TCM, there is no particular disease named POP. According to the clinical symptoms, POP belongs to the TCM category of "Guwei (flaccidity of bones)," "Guku (dryness of bones)," "Guji (disease of bones)," "Gukong (depletion of bones), and Gubi (impediment of bones)" [15]. "Shen (kidney) dominating the bone" is the most fundamental theory in TCM. Shen essence (kidney essence) is closely related to bone physiology and pathology. e sufficient Shen essence can keep the bone healthy and strong. On the contrary, the deficiency of Shen essence will lead to a series of bone-related symptoms. Besides, "liver controlling tendon," is involved in bone health. Also, the spleen and stomach are considered as the postnatal foundations of organisms and are the material sources of bone growth. erefore, the tonifying-Shen (kidney) principle (TS (TK) principle) mainly includes bushen huoxue, bushen zhuanggu, bushen jianpi, and buyi ganshen, which are widely used in combination with other TCM therapies or CWM for the treatment of POP. And, this significant principle has its essential meaning in curing POP [16].
In recent years, numerous meta-analyses were carried out to investigate the efficacy of acupuncture or Chinese herbal medicine for POP [17]. However, there was still no sufficient evidence to draw definitive conclusions as most studies were not comparative analysis aiming at the effect of the TS (TK) principle for POP. Besides, most of the metaanalyses did not explore the safety of particular interventions.
erefore, the purpose of this study was to systematically identify available randomized clinical trials (RCTs) using the TS (TK) principle combined with CWM for POP to appraise its efficacy and safety.

Study Registration.
e study has been registered in PROSPERO (registration number: CRD42020151768). e review reporting was conducted in compliance with the preferred reporting items for systematic reviews and metaanalyses (PRISMA) statement guidelines.

Inclusion Criteria
(1) Types of Participants and Interventions. All RCTs that were reporting the application of the TS (TK) principle combined with CWM for POP were included. e interventions of these studies must include the TS (TK) principle in the experimental group. Studies must be published in English or Chinese language. TS (TK) principle mainly includes bushen huoxue, bushen zhuanggu, bushen jianpi, and buyi ganshen. Specific treatment methods include traditional Chinese herbal medicine, acupuncture and moxibustion combined with traditional Chinese herbal medicine, acupoint catgut embedding, and acupoint injection. POP patients were included. ere is no restriction on gender, race, ethnicity, or nation. Patients in the treatment group were given the TS (TK) principle combined with CWM, while patients in the control group were given CWM alone.
e dosages and courses were not limited in our studies.
(2) Types of Outcome Measures. e primary outcomes included effectiveness rate and bone mineral density (BMD) of the lumbar spine and proximal femur (femoral neck or total hip). e secondary outcomes consisted of VAS scores and TCM symptom integral. Besides, adverse events were also assessed as a safety measurement.

Exclusion Criteria.
Studies with the following characteristics were excluded: irrelevant to TS (TK) principle studies; studies without consistent diagnostic criteria or relevant outcome indicators; non-English or Chinese-language articles; duplicate reports or the data cannot be extracted; case reports, animal experiences, qualitative studies, comments, or review articles.

Literature Search Strategy.
We searched four international electronic databases (PubMed, Cochrane Library, EMBASE, and Web of Science) and four Chinese electronic databases (CNKI, VIP, Wanfang, and CBM) from their initiation to September 2019 to collect for relevant literature.
e literature search was constructed around search terms for TS (TK) principle, POP, and randomized controlled trials and adapted for each database as necessary. e references of the included studies were also screened for further material for inclusion. e detailed search strategy for PubMed is in Table 1. Search strategies were also used for other electronic databases.

Study Selection and Data Extraction.
As a first step in the data handling process, titles and abstracts of all studies retrieved by the search strategies were screened for relevance, and all those that were clearly irrelevant have been discarded.
As a second step, two review team members (Junquan Liang and Fengyi Wang) independently assessed the eligibility of the studies by using the predefined inclusion and exclusion criteria. Besides, for the studies that meet the inclusion criteria, the whole article was read by reviewers to ensure that the entire study met the criteria and was prepared to extract relevant information. e disagreements on whether including a specific study or not were resolved by discussion between the reviewers. e lacking information was requested by contacting the writer of the original article. e information extracted by the two independent review team members included the following: study setting, population study, participant demographics and baseline characteristics, details of the intervention and control conditions, study methodology, outcomes and treatment periods, information for the assessment of the risk of bias. e discrepancies were identified and resolved through discussion (with a third author where necessary). Missing data were requested from the study authors.

Risk of Bias Assessment.
ere were two reviewers involved in the quality assessment process, and any major disagreements were resolved by discussion to define the final set of included studies.
Two independent reviewers assessed the risk of bias by considering the following characteristics: randomization sequence generation, treatment allocation concealment, blinding method, completeness of outcome data, selective outcome reporting, and other sources of bias. Besides, the Cochrane Collaboration's risk of bias assessment tool was used to assess the quality of the individual included studies.
2.6. Data Synthesis. Review Manager 5.3 software was used to carry out the quantitative synthesis. Mean difference (MD) or standardized mean difference (SMD) was used for continuous data. Risk ratio (RR) was used for the analysis of dichotomous data. Both were given a 95% confidence interval (CI). In the case of homogeneous data (I 2 ≤ 50%, P > 0.10), the fixed-effect model was adopted for the metaanalysis. Otherwise, the sources of heterogeneity were further analyzed. After excluding the influence of marked clinical heterogeneity, a random-effect model was adopted to perform the meta-analysis. Sensitivity and bias risk analyses were also performed.
2.6.1. Analysis of Subgroups. We performed some planned subgroup analysis: different specific therapies (bushen huoxue, bushen zhuanggu, bushen jianpi, and buyi ganshen) included in the TS (TK) principle, different kinds of treatment methods (traditional Chinese herbal medicine, acupuncture and moxibustion combined with traditional Chinese herbal medicine, acupoint catgut embedding, and acupoint injection), different parts of BMD examination (lumbar spine and proximal femur (femoral neck or total hip)), and different treatment periods of the TS (TK) principle (≤3 months, 3-6 months, and >6 months).

Sensitivity Analysis and Reporting Bias Analysis.
Sensitivity analysis was carried out to identify the robustness and stability of pooled outcome results by removing the lowquality studies. We have performed a funnel plot of the primary outcome (effectiveness rates of different treatment methods included in the tonifying-Shen (kidney) principle) to evaluate the reporting bias.

Quality of Evidence.
We assessed the quality of evidence for outcomes by using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system [18].

Risk of Bias Assessment.
We used the Cochrane Collaboration's risk of bias assessment tool to assess the quality of the included studies. Firstly, all studies reported the method of randomization, and 30 studies described the method of generating a randomization number table [19-30, 32, 34, 35, 37-42, 44, 46, 48-54].
irdly, 3 studies were assessed as appropriate doubleblinding of participants and provided detailed information for double-blinding during treatment as well as an outcome assessment [19,20,43]. None of the studies reported any incomplete outcome data (Table 3 and Figure 2).
is result suggests that TCM herbal medicine combined with CWM in the treatment of POP can significantly improve clinical efficacy when compared with using CWM alone. ree studies reported effectiveness rates of acupuncture and moxibustion combined with TCM herbal medicine [42,44,45]. e result showed that there was no statistical heterogeneity among studies (chi 2 � 3.94, P � 0.14; I 2 � 49%), so we adopted a fixed-effect model to calculate the combined RR and 95% CI as 1.34 (1.19, 1.51), P < 0.00001, indicating a statistically significant difference between the experimental group and the control group. is result suggests that, in respect of effectiveness rates, using acupuncture and moxibustion combined with TCM herbal medicine plus CWM for POP was better than using CWM alone. ere are 6 studies which reported effectiveness rates of acupoint catgut embedding [25,40,[47][48][49][50]. e heterogeneity was not detected among studies (chi 2 � 4.53, P � 0.48; I 2 � 0%), so a fixed-effect model was used to calculate the combined RR and 95% CI as 1.42 (1.27, 1.58), P < 0.00001, indicating a statistically significant difference between acupoint catgut embedding combined with the CWM group and the CWM alone group. is result shows that the effectiveness rates of catgut embedding combined with CWM for POP were better than using CWM alone. ere is only 1 study which reported effectiveness rates of acupoint injection [53]. e combined RR and 95% CI was 1.25 (1.05, 1.48), P < 0.00001, indicating the difference between acupoint injection combined with the CWM group and the CWM alone group. Owing to the small sample size, this result would show that the effectiveness rates of acupoint injection combined with CWM for POP were better than using CWM alone. All in all, the pooled data showed that different treatment methods included in the TS (TK) principle combined with CWM were more effective than using CWM alone in improving effectiveness rates, with significant differences (RR � 1.28; 95% CI (1.23, 1.33); P < 0.00001) (Figure 3(a)).

Effectiveness Rates of Different Specific erapies Included in TCM Herbal Medicine.
ere were 16 studies which reported effectiveness rates of different specific therapies included in TCM herbal medicine [21-23, 25, 27, 28, 30, 31, 33, 34, 36-41]. Five studies reported the effectiveness rates of bushen huoxue therapy [23,[36][37][38][39], three studies reported the effectiveness rates of bushen zhuanggu therapy [27,31,33], three studies reported the effectiveness rates of bushen jianpi therapy [22,28,30], and five studies reported the effectiveness rates of buyi ganshen therapy [   ; P < 0.00001, respectively, indicating a statistically significant difference between the experimental group and the control group. is result suggests that bushen huoxue, bushen zhuanggu, bushen jianpi, and buyi ganshen TCM herbal medicine combined with CWM in the treatment of POP can significantly improve clinical efficacy when compared with using CWM alone. e combined data showed that different specific therapies included in TCM herbal medicine combined with CWM were more effective than using CWM alone in improving effectiveness rates, with significant differences (RR � 1.23; 95% CI (1.18, 1.29); P < 0.00001) (Figure 3(b)).

Effectiveness Rates of Different Treatment Periods.
Twenty-three literature studies reported effectiveness rates of different treatment periods [21-23, 26-28, 30, 31, 33, 34, 37-42, 44, 45, 47-50, 53]. Eleven literature studies reported treatment periods of less than 3 months [23,27,28,37,38,41,44,45,[48][49][50]. However, there was high statistical heterogeneity among studies (chi 2 � 22.88, P � 0.01; I 2 � 56%). A sensitivity analysis was performed to identify the source of heterogeneity. By removing one trial [38], no heterogeneity was detected (chi 2 � 4.88, P � 0.84; I 2 � 0%). We confirmed the accuracy of the data without publication bias after contacting the author. So, we adopted a random-effect model to calculate the combined RR and 95% CI as 1.35 (1.27, 1.43), P < 0.00001, indicating a statistically significant difference between the TS (TK) principle combined with the CWM group and the CWM alone group. It is suggested that using the TS (TK) principle combined with CWM for POP was better than using CWM alone in improving effectiveness rates when the treatment periods were less than 3 months. Twelve literature studies reported treatment periods of three to six months [21,22,25,30,31,34,39,40,42,44,47,53]. ere was no statistical heterogeneity among studies (chi 2 � 7.92, P � 0.85; I 2 � 0%). erefore, the fixed-effect model was applied to calculate the combined RR and 95% CI as 1.25 (1.18, 1.32), P < 0.00001, indicating a statistically significant  Evidence-Based Complementary and Alternative Medicine difference between the TS (TK) principle combined with the CWM group and the CWM alone group. It is suggested that using the TS (TK) principle combined with CWM for POP was better than using CWM alone in improving effectiveness rates when the treatment periods were three to six months. e pooled data showed that different treatment periods combined with CWM were more effective than using CWM alone in improving effectiveness rates, with significant differences (RR � 1.29; 95% CI (1.24, 1.35); P < 0.00001) (Figure 3(c)).

TCM Symptom
Integral. e TCM symptom integral was established according to the Clinical Research Guidance of New Chinese Herbal Medicine [55]. Four studies reported TCM symptom integral [26,40,46,54], and there was low statistical heterogeneity among studies (chi 2 � 7.49,     erefore, the fixed-effect model was applied. e meta-analysis showed that there was a statistically significant difference between the experimental group and the control group (SMD � −1.23; 95% CI (−1.43, −1.02); P < 0.00001). is result suggests that the TS (TK) principle combined with CWM in the treatment of POP can significantly improve TCM symptom integral when compared with using CWM alone (Figure 3(g)).

Publication Bias.
e funnel plots were generated for studies with data on the effectiveness rates of different treatment methods included in the TS (TK) principle. e results showed that most of the points in the funnel plots were symmetrical. However, two points were outside the 95% CIs, which indicates that there may have been publication bias in our studies and that might influence the results of our analysis, as can be seen in Figure 4.

Quality of Evidence.
ere were 8 results for levels of evidence in our study. e detailed GRADE evidence profile of results is shown in Figure 5.

Summary of Main Results.
ere were 36 included RCTs with 3617 participants in our research. Even though most of the trials had small sample sizes and poor methodological quality, our meta-analysis reached the following results: (1) analysis of the pooled data showed a consistently superior effect of the TS (TK) principle combined with CWM in terms of total effectiveness rates, BMD of the lumbar spine and proximal femur (femoral neck or total hip), TCM symptom integral, and VAS scores when compared to using CWM alone; (2) in terms of adverse effects, the same safety was obtained for the TS (TK) principle combined with CWM or using CWM alone for POP; (3) different treatment methods included in the TS (TK) principle combined with CWM were more effective than using CWM alone in improving effectiveness rates. It should be noted that since the sample size of acupoint injection effectiveness rates was small, the combined RR and 95% CI was reported from the original study, not from the meta-analysis results; (4) in respect of TCM herbal medicine, different specific therapies combined with CWM for POP were more effective than using CWM alone; (5) compared with using CWM alone, the TS (TK) principle combined with CWM was more effective for POP in the aspect of different treatment periods; (6) in the three-to six-month treatment period, the TS (TK) principle combined with CWM for POP in terms of BMD of the lumbar spine and proximal femur (femoral neck or total hip) was better than using CWM alone; (7) according to the guideline of GRADE, the effectiveness rates of different treatment methods included in the TS (TK) principle, effectiveness rates of different specific therapies included in TCM herbal medicine, effectiveness rates of different treatment periods, BMD (lumbar spine) of different treatment periods, BMD (proximal femur (femoral neck or total hip)) of different treatment periods, TCM

Analysis of TS (TK) Principle. TCM has been used in a range of medical management and health interventions in
China and any other Asian countries for over 2500 years. POP patients are usually seeking TCM treatment, when the therapeutic effect of CWM is unsatisfactory. According to TCM theory, the establishment of a therapeutic principle is based on TCM syndromes, not symptoms. e clinical diagnosis of the TCM syndrome relies on the gathering of clinical information through inspection, auscultation and olfaction, inquiry, and palpation [56]. For POP, the fundamental physiopathological changes of the bone depend on whether Shen (kidney) essence is sufficient or not. erefore, the TS (TK) principle is the key point to prevent and treat Shen (kidney)-deficiency syndrome of POP [57][58][59][60]. On the one hand, the effective mechanisms of pharmaceutical treatment included in the TS (TK) principle on POP have been demonstrated in the voluminous literature. Icariin (ICA), similar to estrogen, has a definite antiosteoporotic effect [61,62]. Besides, oleanolic acid (OA) and psoralen have been reported to prevent bone loss by inhibiting osteoclast formation [63][64][65]. On the other hand, in respect of nonpharmaceutical treatment included in the TS (TK) principle, experiments have demonstrated that acupuncture could alleviate osteoporosis by regulating the expression of members in OPG/RANKL, Wnt/β-catenin, and MAPK pathways [66]. Acupoint catgut embedding could regulate the hypothalamic-pituitary-ovarian axis to raise the serum E 2 level which would be significant in preventing osteoporosis [67]. Even experiments have shown that acupoint catgut embedding ameliorated the ovariectomization-(OVX-) caused metabonomic changes more effectively than hormone replacement therapy (HRT) with nilestriol [68]. Also, nonpharmaceutical treatment could alleviate related symptoms of osteoporosis and improve the quality of life [43]. erefore, the TS (TK) principle would be a promising approach for POP, and it can not only improve physiological and biochemical indicators but also alleviate the TCM syndromes [16].

Limitations of Research.
However, some limitations in our meta-analysis should be mentioned. (1) Although all of our included studies were RCTs, the methodological quality of them was generally improvable. Most of them failed to describe the blinding methods in detail, allocation, and concealment methods. (2) Among 36 studies, only 8 studies reported follow-up. e longest follow-up period was 120 months, and no further follow-up data were collected. e long-term effect of the TS (TK) principle for POP should be further studied. (3) For POP, fracture incidence should be the most patient-important outcomes. However, no fracture incidence data were collected in our studies. It is necessary to focus on fracture incidence as a patientimportant outcome in further studies. (4) e studies included in this analysis were insufficient, especially in terms of subgroup analysis.
us, potential publication bias probably exists. (5) e high heterogeneity among studies may be related to the different treatment methods, treatment periods, and even the skill level of the practitioners. It is a common problem in the research of TCM therapy. (6) e study for different specific therapies of TCM herbal medicine was insufficient, and further data mining should be carried out.

Conclusion
In summary, our meta-analysis suggests that using the TS (TK) principle combined with CWM for POP is effective and safe. However, the limitation in the quality and quantity of the included RCTs might weaken the overall reliability of this conclusion.
erefore, large-volume, well-designed RCTs with extensive follow-up are awaited to confirm and update the findings of this analysis.