A Network Meta-analysis Of Traditional Chinese Medicine For The Treatment Of Knee Osteoarthritis Based On GRADE Rating

Background:The clinical manifestations of knee osteoarthritis(KOA)are joint pain,dysfunction,and joint stiffness,joint swelling,muscle atrophy in some severe cases,which seriously affect the quality of life of the patients and are one of the main causes of lower limb dysfunction in the elderly people.At present,most of the domestic and international studies on the treatment of KOA with TCM external treatment methods are RCTs or systematic evaluations or Meta-analyses of single TCM external treatment methods compared with Western treatment methods,and there is a lack of cross-sectional comparisons between multiple TCM external treatment methods.Network Meta-analysis(NMA)can quantitatively compare different interventions for similar diseases,combine direct and indirect comparative evidence,and rank them according to the superiority of an outcome indicator to select the best option.In this study,we used the clinical literature to identify several TCM interventions commonly used in modern clinics,and retrieved relevant RCTs that used single and combined TCM interventions for KOA,developed inclusion and exclusion criteria,performed risk of bias assessment,network Meta-analysis,and quality of evidence for outcome indicators:and selected the most appropriate treatment protocol for KOA.Objective:To investigate the effectiveness and safety of acupuncture,Tui-na,fire needle,moxibustion,electro-acupuncture,warm acupuncture,small needle knife,thermosensitive moxibustion,external application of traditional Chinese medicine,and Tai chi in the treatment of KOA by using systematic evaluation in evidence-based medicine and based on the network Meta-analysis method and to establish an evidence-based individualized TCM treatment protocol for KOA.The results obtained can provide certain evidence-based medical evidence for clinical KOA guidelines,which will provide a scientific basis for the standardization and standardization of TCM external treatment methods,and also lay the foundation for the promotion and popularization of TCM traditional therapies to the international market.Methods:A computer search of PubMed,EMbase,The Cochrane Library,VIP,CNKI,and WanFang Data databases was conducted to collect domestic and international literature,including clinical randomized controlled trials(RCTs)using single TCM external treatments and combinations of TCM external treatments,with the search period from the date of creation to February 29,2020.The search period was from the establishment of the database to February 29,2020.After two evaluators independently screened the literature,extracted information,and evaluated the risk of bias of the included studies,the risk of bias was summarized using RevMan 5.3 regarding the risk of bias assessment method provided in Cochrane Handbook 5.1.0.For the network Meta-analysis section,data analysis and inconsistency testing were performed using STATA 15.0 software,and network evidence plots,forest plots,intervention SUCRA treatment effect ranking plots,step plots,and funnel plots were drawn,and the quality of evidence was rated using the GRADEpro system,and the assessment results were presented in SoF tables.Results:A total of 14,696 papers were retrieved,and 58 clinical studies were finally included after checking and screening to exclude irrelevant literature,involving a total of 18 types of TCM external treatments including general acupuncture,Tui-na,moxibustion,fire needle,moxibustion,electro-acupuncture,warm acupuncture,small needle knife,thermosensitive moxibustion,external application of Chinese herbs,Tai chi,and combined application between the above treatments.For the assessment of the risk of bias in the included literature,the results showed that the risk of selective bias,implementation bias,and reporting bias exceeded 75%,and the bias caused by other confounding factors exceeded 50%,and the overall literature risk of bias assessment scale was more moderate and high risk.The results of the network Meta-analysis showed that:(1)in the outcome of clinical treatment efficiency,the best treatment option probably was:warm acupuncture+herbal topical application(SUCRA 87.9%),followed by:Tui-na+herbal external application(SUCRA 85%),Tui-na+warm acupuncture(SUCRA 83.3%),and Tui-na+small needle knife(SUCRA 81.8%).But there was no statistically significant difference in the indirect comparison results between the four interventions with each other;(2)in the outcome of VAS scores,the best treatment option probably was:Tui-na+small needle knife(SUCRA 90.9%),followed by:Tui-na+warm acupuncture(SUCRA 87.7%),fire needle(SUCRA 81.9%),and warm acupuncture+herbal external application(SUCRA 75%).The effect of Tui-na+small needle knife compared with warm acupuncture+herbal external treatments was statistically significant(MD=-0.66,95%CI[-2.22,0.89]);(3)In the outcome of WOMAC-stiffness score,the treatment options that were probably the best were:fire needle(SUCRA 78.8%),followed by:warm acupuncture(SUCRA 78%),acupuncture(SUCRA 50.9%),and Tai Chi(SUCRA 42.8%),of which warm acupuncture vs acupuncture was statistically significant in improving this outcome(MD=-0.78,95%CI[-2.46,0.91]);(4)in reducing the WOMAC-Functional Disorders score,the treatment option that may be the best treatment option was:moxibustion(SUCRA 88.8%),followed by Tai Chi(SUCRA 75.2%),electro-acupuncture(SUCRA 69.8%),and fire needle(SUCRA 62.2%),but there were no statistically significant differences yet in the indirect comparison between the four interventions.The results of the safety analysis of different interventions,among the 58 included papers,54 studies did not report adverse reactions,and 4 studies reported a total of 21 cases of adverse reactions,accounting for 21/5416 of the total sample.specifically,10 cases of subcutaneous ecchymosis after acupuncture(2 cases in the small needle knife group,4 cases in the electroacupuncture group,and 4 cases in the general acupuncture group),10 cases of moxibustion blisters after moxibustion,and 1 case of skin burns with warm acupuncture.No infection or poor prognosis occurred after common treatmentThe results of GRADE evidence quality evaluation showed that(1)in terms of treatment efficiency,a total of 8 groups of interventions were statistically significant for comparison,of which Tui Na vs.topical Chinese medicine,Tui Na vs.electroacupuncture,warm acupuncture vs.electroacupuncture,thermal moxibustion vs.warm acupuncture,and fire needle vs.warm acupuncture were rated as low quality of evidence,and the rest of the comparisons were very low quality of evidence;(2)in terms of improvement of VAS pain scores,a total of 94 groups of interventions with statistically significant comparisons,including Tui-na+small needle knife vs.small needle knife,Tui-na+small needle knife vs.Tui-na,fire needle vs.electric acupuncture,warm acupuncture+herbal topical vs.acupuncture,electric acupuncture+warm acupuncture vs.acupuncture,electric acupuncture vs.small needle knife,electric acupuncture vs.Tui-na,electric acupuncture vs.acupuncture,acupuncture+herbal topical vs.acupuncture,small needle knife vs.acupuncture,warm acupuncture vs.acupuncture,and warm acupuncture vs.acupuncture+moxibustion which were low quality of evidence in 13 groups,and the rest were all very low quality of evidence;(3)in the outcome of improving WOMAC stiffness scores,there were a total of 3 groups of interventions with statistically significant comparisons,but all were rated as very low quality of evidence;(4)in the outcome of improving WOMAC dysfunction scores,there were a total of 12 groups of interventions with statistically significant comparisons.The quality of evidence for fire needle vs acupuncture and warm acupuncture vs acupuncture was low quality,and the remaining comparisons were all very low qualityConclusions:A network Meta-analysis revealed that the combination of two external treatments was superior to a single external treatment in terms of efficiency in treating KOA,with warm acupuncture+herbal topical application probably having the most significant efficacy,and among the monotherapy intercomparisons,the efficacy of fire needle was relatively significant The ranking in terms of improvement of knee pain VAS scores showed that the combination of Tui Na+needle knife and Tui Na+warm acupuncture may have the highest efficacy.For the outcome index WOMAC knee stiffness score,the efficacy of fire needle was relatively significant,and moxibustion has the potential to be the best efficacious intervention in terms of improving the WOMAC knee dysfunction score.However,because of the generally low quality of the literature included in this study,the lack of multicenter,large sample size,and long-term follow-up studies,the cumbersome variety of interventions,and the low variability in ranking probabilities among them,it is still expected that more multicenter,large-scale,and high-quality randomized controlled trials can be conducted subsequently to minimize the occurrence of bias and to be able to comprehensively evaluate the high evidence treatment of KOA with external Chinese medicine therapies.