From 2012 to 2022, the number of literature on traditional Chinese non-pharmacological therapies in the treatment of KOA showed an overall fluctuating trend, with 375 literature published, or an average of 34 literature per year. After 2019, the number of publications increased significantly. In the same year, China promulgated the “opinions on promoting the inheritance and innovative Development of Traditional Chinese Medicine,” clearly pointing out that to promote the inheritance, openness and innovative development of TCM (Huang, 2019), the number of literature on KOA and other TCM treatments was expected to continue to grow in the future (Figure 1).

There were five types of literature in total, of which, 268 “Articles” were the largest type of literature (71.4%), indicating that a large number of clinical and basic trials have been conducted in the past decade to verify the efficacy of traditional Chinese non-pharmacological therapies in the treatment of KOA. This was followed by “Review” with 82 articles, indicating that retrospective research was also valued by researchers, while 25 articles (6.5%) were of other types (Table 2).

According to the national analysis, a total of 32 countries involved in the research of traditional Chinese non-pharmacological therapies on KOA from 2012 to 2022. China published 252 articles and was the most active country, the United States was the second (71), England (Wang et al., 2021) and South Korea (Vickers et al., 2012). Among them, China and the United States had the strongest correlation, indicating close cooperation among the regions. From the perspective of centrality, the United States ranked first (0.68), followed by Australia (0.21) and China (0.19). The top three most-cited countries were the United States (2363), China (1688), and England (1664; Table 3). Regionally, Asia and Europe were the research centers, with a clear clustering phenomenon. For example, China, Japan, South Korea, and Singapore were the major contributors to Asian research, while Spain, England, and Germany were the main drivers of European research (Figure 2).

According to the analysis of 557 institutions participating in studies of traditional Chinese non-pharmacological therapies on KOA over the past decade, 36 institutions published more than five articles (Figure 3), with Beijing University of Chinese Medicine (Tu et al., 2021) published the most, followed by Shanghai University of Chinese Medicine (National Institute for Health and Care Excellence, 2014), and Capital Medical University (White A. et al., 2012), which noted that traditional Chinese non-pharmacological therapies helped to improve pain and restore the knee function in KOA patients (Wang M. et al., 2020; Zhang Y. et al., 2020; Tu et al., 2021), suggesting that this effect may be achieved through the changes of inflammatory cytokines TNG-α, IL-1β, and IL-13 (Shi et al., 2020), which have received more attention in recent years. In terms of centrality, Beijing University of Chinese Medicine (0.1) ranked the top one, followed by Capital Medical University (0.1), and Korea Medical College of Oriental Medicine (0.08). The top three most frequently cited universities were University of York (1314), University of Southampton (1237), and Memorial Sloan-Kettering Cancer Center (1185, Table 4). Through the close collaboration with Klee University and Charite, their study indicated that the effect of acupuncture on KOA was related to the amount and duration of acupuncture (MacPherson et al., 2013; Vickers et al., 2018), and the effect was long-lasting (MacPherson et al., 2017), suggesting that acupuncture was a recommended therapy for KOA (Vickers et al., 2012).

Over the past decade, 2,021 authors participated in the study of traditional Chinese non-pharmacological therapies for KOA, among which, 27 authors have published more than five articles. The three authors with the most published literature were Cun-Zhi Liu (Li et al., 2021), Jian-Feng Tu (Li et al., 2021), and Li-Qiong Wang (Zhang X. et al., 2012), who concluded that acupuncture was an effective method by observing specific changes in the brain function in patients with KOA (Zhang N. et al., 2020). They also found that intestinal flora was a possible target (Wang et al., 2021), and improving the frequency of treatment may lead to better outcomes (Lin et al., 2020). The three authors with the highest centrality were Wang X (0.11), Li X (0.08), and Li Y (0.08). The most frequently cited author was Lewith G (1153; Table 5), who draws a negative conclusion on traditional Chinese non-pharmacological therapies, suggesting that acupuncture and Qi Gong were not particularly effective compared to controls and that patient’s confidence in the physician and treatment method had a significant effect on the outcome (Macfarlane et al., 2012; White P. et al., 2012). Macpherson H (1128) and Vickers A (1066) also had high citation frequency (Table 5). The authors with the purple outer circles indicate a high between centrality and were core researchers in the research field. The authors with the red mark represent a prominent frequency over a period of time. For example, Macpherson H, Cun-Zhi Liu, and Li-Qiong Wang were the most active authors in the last decade (Figure 4).

Journal analysis showed that a total of 2,994 journals were cited, of which 107 journals were cited more than 20 times. The most frequently cited journal in total was Osteoarthritis Cartilage (560), followed by Pain (409), and Acupuncture in Medicine (365). The top three journals in terms of centrality were Arthritis Rheumatism (0.1), Arthritis Care Research (0.08), and Complementary Therapies in Medicine (0.08). The top three journals in terms of number of publications were Evidence-Based Complementary and Alternative Medicine (Wang et al., 2016), Medicine (Choi et al., 2017), and Trials (Vickers et al., 2012; Table 6). Journals with purple outer rings indicate high between centrality, publish important literature and were key hubs for the entire network. Journals marked in red indicate a high highlighting frequency during a given time period. For example, clinical Journal of Pain, Journal of Pain Research, and Medicine were the journals that have been highlighted most frequently in the last 3 years (Figure 5).

We screened the most representative literature about traditional Chinese non-pharmacological therapies in the treatment of KOA from 2012 to 2022. The yellow line represents literature from 2012 to 2015, and the representative cluster labels were “alternative therapy,” “group acupuncture” and “exercise prescription,” and the green line represents the literature from 2016 to 2022 with the representative cluster labels of “knee osteoarthritis,” “massage analgesia,” “warm acupuncture,” and “non-specific efficacy” (Figure 6). We found that the most frequently cited literature was the 2014 guidelines issued by the International Society for Osteoarthritis Research (OARSI), which listed Tai Chi as a core treatment for KOA (McAlindon et al., 2014). In addition, the American College of Rheumatology guidelines also list Tai Chi as a recommended treatment for KOA (Hochberg et al., 2012). These two guidelines have received extensive attention from researchers, and their recommendations have been adopted in a number of studies (Table 7). Six of the first 10 cited articles discussed the clinical effects of acupuncture on KOA, and the use of which has received a lot attention from researchers, particularly the placebo effect of acupuncture. A randomized controlled clinical trial published by Hinman RS in JAMA showed that acupuncture was no more beneficial than sham acupuncture in treating chronic knee pain (Hinman et al., 2014).

We analyzed literature related to traditional Chinese non-pharmacological therapies on KOA in the past decade and summarized the keywords with the highest co-occurrence frequency and centrality (Table 8). The co-occurrence frequency indicates the frequency of the keywords, and the centrality represents the degree to which keywords are noticed. A node with a centrality greater than 0.1 indicates that the node has greater influence in the study. “Knee osteoarthritis” had the highest co-occurrence frequency (211), “adjunctive therapy” (172) and “pain” (160) were three keywords with the highest co-occurrence. The top three keywords in centrality were acupuncture (0.25), chronic pain (0.19), and alternative medicine (0.18). Figure 7 shows highlighted keywords with red nodes, representing great changes in co-occurrence frequency in a certain period of time and are used to find keywords that are decreasing or increasing. We found that “quality of life,” “older adult,” and “manual therapy” were the keywords that suddenly dominated the most during the period 2020–2022.

Figure 8 shows the keyword change and keyword clustering over time from 2012 to 2022. We found that the traditional Chinese non-pharmacological therapies such as “Tai Chi,” “Qi Gong,” “electro-acupuncture,” and “auricular acupuncture” will always be high-frequency keywords, and the emotion, inflammation and evaluation scales of KOA patients will also have higher co-occurrence frequency. Among the first 10 keyword clusters, “chronic pain” was the largest cluster with a total of 67 members.

We found that many traditional Chinese non-pharmacological therapies such as acupuncture and moxibustion involve the use of acupoints. In order to understand the use of acupoints in the treatment of KOA, we used VOSviewer and SCImago Graphica with co-occurrence frequency as screening condition. The most commonly used acupoints for treating KOA were analyzed. In addition, for other traditional Chinese non-pharmacological therapies, we analyzed the use frequency and corresponding evaluation indexes in the same way, hoping to provide ideas and references for researchers and clinicians.

The results showed that a total of 22 Chinese non-pharmacological interventions were used to treat KOA, and 16 interventions were co-occurring more than five times as often (Figure 9). A total of 49 acupoints have been mentioned in the article, 17 of which had a co-occurrence frequency of more than three times (Figure 10), and a total of 112 evaluation indexes were mentioned. Of these, 19 evaluation indicators had a co-occurrence frequency of more than five times (Figure 11).

The three most frequently used acupoints were ST36 (Zusanli), ST10 (Dubi), and GB34 (Yanglingquan). The three most widely used intervention methods were manual acupuncture, electro-acupuncture, and moxibustion, and the three most widely used evaluation indicators were WOMAC, VAS, and SF-36 (Table 9).