An Evidence-based Study On The Repair Effect Of Patch Materials On Rotator Cuff Injury

Background Rotator cuff injury is a common musculoskeletal disease,which can lead to shoulder pain and dysfunction.For patients with a large degree of rotator cuff injury accompanied by tendon retraction and fatty infiltration,patch materials are considered to be the ideal choice for rotator cuff repair.New repair materials must be fully verified in animal models before clinical trials and clinical application.Although a large number of animal experiments have been conducted to explore the repair effect of rotator cuff patch,its efficacy and safety are still controversial.In addition,the scientific establishment of animal models,as well as the scientific design and implementation of animal experiments are the keys to further promote biomedical research and ensure the reproducibility and transferability of results.However,there is no unified standard for the establishment of animal models of rotator cuff injury,and the methodological quality and transparency of reporting standards for animal experiments in this field are still unclear.Objectives 1)The system presents the overall framework of animal experimental research on rotator cuff injury from multiple dimensions,and evaluates the advantages and disadvantages of different animal models;2)To understand the current status of animal experiments on rotator cuff repair with patch materials,and systematically evaluate the safety and effectiveness of rotator cuff repair with patch materials;3)comprehensively evaluate the methods and reporting quality of animal experiments on rotator cuff repair with patch materials,and clarify the current status of animal experimental research in this field;4)To comprehensively understand the current status of Chinese researchers’ cognition on the guidelines for the design and reporting of animal experiments on rotator cuff repair with patch materials.Methods1)Systematic review:PubMed and Ovid Embase databases were searched from the establishment of the databases to November 2020 for animal experimental studies of rotator cuff injury.The characteristics of rotator cuff injury animal models,research objectives,outcome indicators and outcome measurement methods were systematically summarized and analyzed.2)Systematic review and Meta-analysis:PubMed and Ovid Embase databases were searched from the establishment of the databases to November 2020 for animal experimental studies of rotator cuff repair with patch materials.The animal model of rotator cuff repair with patch materials and the histological repair effect of patch materials were systematically reviewed.The Stata software was used to perform a meta-analysis on the biomechanical parameters(maximum failure load and stiffness)after patch repair.3)Literature review:PubMed and Ovid Embase databases were systematically searched from the establishment of the database to November 2020 for animal experimental studies of rotator cuff repair with mesh materials.Based on the SYRCLE bias risk assessment tool and ARRIVE 2019 guidelines,the risk of bias and reporting quality of animal experimental research in this field were evaluated in detail,respectively,and the current situation and existing problems in design methods and reporting transparency were clarified.4)Questionnaire survey:Convenience sampling method was used to investigate researchers from 65 hospitals and universities in 15 provinces/municipalities directly under the Central government based on a self-made electronic questionnaire on the cognition and implementation status of interventional animal experiments on rotator cuff repair with patch materials in the aspects of method design,implementation and reporting guidelines,and to find out the existing problems.The purpose of this study is to standardize and improve the quality of interventional animal experimental research methods,promote the integrity and transparency of animal experimental research reports,and ultimately promote the transformation and utilization of its achievements.Results1)Systematic review of animal models of rotator cuff injury:A total of 409 animal experimental studies were included,using 7 different animal models(including mice,rats,rabbits,dogs,sheep,monkeys,and baboon),of which rats were the most commonly used species(53.6%,219/409),followed by rabbits(25.7%,105/409).In animal models of rotator cuff injury,the most common injury site was supraspinatus tendon(62.1%,254/409),followed by infraspinatus tendon(18.8%,77/409).The most common type and degree of rotator cuff injury was acute fullthickness tear(48.4%,198/409),followed by full-thickness defect(18.8%,77/409).Small animal models(mice and rats)are commonly used to study the pathophysiological changes,gene expression,signaling pathways and other mechanisms after rotator cuff injury.Rabbits and sheep are mostly used to investigate the surgical methods,suture techniques and repair effects of patch after rotator cuff injury.In terms of research objectives,testing the repair effect of patch was the most common research purpose(24.9%,102/409),followed by observing the pathophysiological changes after rotator cuff injury(20.8%,85/409)and testing the effect of drug intervention(11.0%,45/409).The common outcome assessment methods were histological observation(84.6%,346/409)and biomechanical testing(55.0%,225/409),followed by imaging examination(18.6%,76/409)and gene expression analysis(16.4%,67/409).A total of 40.6%(166/409)of studies used two or more types of outcome assessment methods,and the combination of histological assessment and biomechanical testing was the most common combination(49.1%,201/409).2)Effect of rotator cuff patch on rotator cuff repair:Systematic review and metaanalysis based on animal experiments showed that:A total of 74 original animal experimental studies were included,which used four different animal species(rats,rabbits,dogs and sheep).The majority of studies(77.0%,57/74)used small animal models to explore the repair effect of various patch materials.There were five different categories of materials(non-degradable synthetic materials,degradable synthetic materials,autologous tissue,allogeneic tissue,and naturally derived biomaterials),of which naturally derived materials(50.0%,39/78)dominated.Compared with simple suture,the results of meta-analysis showed that:① The biomechanical properties of biodegradable synthetic materials,the maximum failure load:WMD=0.50[95%CI-0.40-1.41],P=0.275(rat model);WMD=-1.63[95%CI11.61-8.34],P=0.748(rabbit model);Neither model was statistically significant.Stiffness:WMD=0.46[95%CI 0.10-0.82],P=0.013(rat model);WMD=0.74[95%CI 0.19-1.30],P=0.008(rabbit model);Both models were statistically significant.②Biomechanical properties of autologous tissue,maximum failure load:WMD=4.21[95%CI 1.35-7.07],P=0.004(rat model);WMD=17.32[95%CI 7.44-27.20],P=0.001(rabbit model);Both models were statistically significant.③ Biomechanical properties of naturally derived materials,maximum failure load:WMD=0.26[95%CI-1.86-2.39],P=0.807(rat model),no statistical significance;WMD=18.68[95%CI 7.71-29.66],P=0.001(rabbit model),which was statistically significant.Stiffness:WMD=-0.02[95%CI-1.21-1.17],P=0.973(rat model),no statistical significance;WMD=1.30[95%CI 0.01-2.60],P=0.048(rabbit model),which was statistically significant.3)The cross-sectional analysis of methods and reporting quality of animal experiments on rotator cuff repair with patch materials showed that 74 animal experimental studies were included.① Quality of original animal experimental methods(based on SYRCLE animal experimental methodological quality assessment tool),our results show that:37.8%(28/74)of the studies were randomized controlled trials(Item 1),but only 4.1%(3/74)of the studies reported the method of randomization,and 1.4%(1/74)of the studies implemented concealed grouping(Item 3).None of the studies reported whether animal breeders and investigators were blinded(Item 5).Only 6.8%(5/74)of the studies used randomized placement of experimental animals(Item 4).21.6%(16/74)of the studies evaluated the results by randomly selecting animals(Item 6);21.6%(16/74)of the studies reported blinding of outcome evaluators(Item 7);The study protocols were not available for all studies(Item 9).② The reporting quality of original animal experiments(based on ARRIVE 2019 guidelines),our results showed that in general,the reporting quality of preclinical animal experiments in the field of rotator cuff injury was not high and the reporting transparency was low.For 10 essential items(22 Sub-items)and 11 recommended items(16 Sub-items),the overall coincidence rates were 46.7%and 53.7%,respectively.The reporting quality of some key items of animal experimental research in this field was not ideal,such as sample size,as high as 91.9%(68/74)of the studies did not explain the calculation method of sample size(Item 2b),89.2%(66/74)of the studies did not report the inclusion and exclusion criteria of animals(Item 3a),and 91.9%(68/74)of the studies did not explain the calculation method of sample size(Item 2b).97.3%(72/74)of the studies did not report the specific randomization method(Item 4a),97.3%(72/74)of the studies did not report the blinding method(Item 5),and all the studies did not report the study registration plan(Item 19).4)The survey results of domestic researchers’ cognition of the guidelines for the design and reporting of animal experiments on rotator cuff repair with patch showed that a total of 126 valid questionnaires were collected,and the survey respondents were 82 clinicians and 44 other researchers.The awareness rate of SYRCLE animal experiment bias risk assessment tool was only 33.3%,and there was no statistically significant difference between the groups(P=0.355).① The cognition of the original animal experimental design method:the researchers’ cognition of all the investigated items was relatively high,ranging from 70.5%to 93.9%.92.7%of clinicians and 75.0%of other researchers believed that it was necessary to take certain measures to keep animal breeders and researchers unaware of the interventions that animals received(Item 5)(P=0.006),and the difference was statistically significant.90.2%of clinicians and 84.1%of other researchers agreed that it was necessary to record and report in detail the data and reasons for animal deaths or exclusions due to inadequate modeling,or any reinclusion in the analysis under each intervention group(compared with the total number of original groups)(Item 8)(P=0.039).②The transparency and consistency of animal experiment research reports:most of the researchers(68.3%)mainly referred to similar published studies,and only 2.4%of the respondents knew and were familiar with the ARRIVE 2010/2019 checklist.There was no significant difference between the two groups(P>0.05).27.8%of the respondents thought that there were insufficient reports in the current published literature on animal experiments,mainly in the aspects of experimental results(14.3%),information about experimental animals(20.0%)and information about blinding(31.4%).Conclusions1)Animal models of rotator cuff injury and repair play an extremely important role in understanding the pathophysiological mechanism of rotator cuff injury,evaluating the repair effect of various patch materials and other related interventions,and implementing new treatment strategies.Before selecting an animal model,the purpose of the study must be clear,and for animal experimental studies,the use of multiple evaluation methods can provide the possibility of observing changes in outcomes from different perspectives.2)The patch has a positive effect on the repair of rotator cuff injury,which can promote the formation of collagen fibers at the tendon-bone interface and improve the biomechanical properties.At present,among the five types of mesh materials(nondegradable synthetic materials,degradable synthetic materials,autologous tissue,allogeneic tissue,and naturally derived biomaterials),naturally derived biomaterials are the most commonly used patch materials.The rat is the most commonly used animal model for rotator cuff repair,but the rabbit model is better for the repair effect of patch material.3)In general,the methodological quality of the published animal experiments on the repair effect of rotator cuff tear is not high and the transparency of reporting is low.In terms of method quality,there are many limitations in animal randomization,hidden grouping and blinding,and all studies have not been registered in advance,which leads to a high risk of selection bias,implementation bias,measurement bias and selective reporting bias in animal experiments in this field.In terms of transparency,there is a lack of transparent,accurate and comprehensive reporting of the key elements of experimental design(sample size and its calculation,randomization,blinding,inclusion and exclusion criteria for animals,etc.)and other elements to avoid bias(registration plan,funding source,and data sharing,etc.).4)Domestic researchers do not have a high awareness of the internationally recognized guidelines for ensuring the internal authenticity and reporting quality of animal experiments,which is the main reason for the unsatisfactory quality and reporting transparency of published animal experiments in this field.Therefore,it is very necessary to popularize and promote the SYRCLE bias risk assessment tool and ARRIVE 2019 guidelines through multiple channels such as industry associations and journals,and introduce "manuscript" in relevant journals at home and abroad,so as to improve the implementation rate of relevant researchers and editors,and promote the improvement of the quality of animal experiment design and reporting in this field.Finally,we will improve the transparency and reproducibility of animal experimental research in this field.