Portal de Periódicos da CAPES

Medical Literature Ahead: Proceed With Caution

2007; Elsevier BV; Volume: 23; Issue: 12 Linguagem: Inglês

10.1016/j.arthro.2007.10.001

1526-3231

James H. Lubowitz, Gary G. Poehling,

Sports injuries and prevention

“Based on past experience, it is likely that half of what we today believe to be correct arthroscopic knowledge or treatment is wrong. The problem is…I don’t know which half.”Donald H. Johnson, M.D.Arthroscopy Journal Board of Trustees We won’t go so far as to say don’t believe what you read, but we do issue a warning: proceed with caution. As we have said before, clinically relevant articles of high levels of evidence are required to change surgical practice, and a single publication must be interpreted in the context of the complete body of medical literature.1Lubowitz J.H. Poehling G.G. Clinically relevant articles of high levels of evidence are required to change surgical practice.Arthroscopy. 2007; 23 (Editorial): 803Abstract Full Text Full Text PDF PubMed Scopus (21) Google Scholar The conclusion of a scientific article depends on the results, the results are determined by the methods, and the methods must address the purpose of a study. (We work backward from the conclusion to remind readers of the logical simplicity of the scientific method that forms the framework of our complex research endeavors.) Thus, when scientific reports reach contradictory conclusions, it does not follow that one of the articles is wrong. Rather, it may be that the studies used different methods that led to different results. In such a case, additional research may ultimately clarify our understanding. In this context, we introduce the meta-analysis of the incidence of anterior cruciate ligament tears as a function of gender, sport, and knee injury reduction regimen by our friend and Editorial Board member, Chadwick C. Prodromos and his colleagues.2Prodromos C.C. Han Y. Rogowski J. Joyce B.T. Shi K. A meta-analysis of the incidence of anterior cruciate ligament tears as a function of gender, sport and a knee injury–reduction regimen.Arthroscopy. 2007; 23: 1320-1325Abstract Full Text Full Text PDF PubMed Scopus (646) Google Scholar Because their meta-analysis was systematically conducted, we were initially puzzled by the findings. In particular, the conclusions regarding skiers contradict those that we published in a previous review.3Rossi M.J. Lubowitz J.H. Guttmann D. The skier’s knee.Arthroscopy. 2003; 19: 75-84PubMed Scopus (18) Google Scholar We found injury rates nearly double in female alpine skiers whereas Prodromos found no difference. Yet further investigation clarifies. Rossi et al.3Rossi M.J. Lubowitz J.H. Guttmann D. The skier’s knee.Arthroscopy. 2003; 19: 75-84PubMed Scopus (18) Google Scholar reviewed 98 published reports in preparation for their Current Concepts review of “The Skier’s Knee,” whereas the meta-analytic methods of Prodromos et al.2Prodromos C.C. Han Y. Rogowski J. Joyce B.T. Shi K. A meta-analysis of the incidence of anterior cruciate ligament tears as a function of gender, sport and a knee injury–reduction regimen.Arthroscopy. 2007; 23: 1320-1325Abstract Full Text Full Text PDF PubMed Scopus (646) Google Scholar resulted in the inclusion of only four publications regarding skiers. Finally, two of the four studies included by Prodromos et al. are evaluations of ski area employee Workers’ Compensation data. In fact, when at work, it is hypothesized that ski area employees often ski on terrain or in a manner that diminishes their risk of injury.3Rossi M.J. Lubowitz J.H. Guttmann D. The skier’s knee.Arthroscopy. 2003; 19: 75-84PubMed Scopus (18) Google Scholar Meta-analytic methods do not adjust for this fact. This is not to diminish the work of Prodromos and his group, who have recently published, in addition to the cited article, two other meta-analyses of ACL injury.4Prodromos C. Joyce B. Shi K. A meta-analysis of stability of autografts compared to allografts after anterior cruciate ligament reconstruction.Knee Surg Sports Traumatol Arthrosc. 2007; 15: 851-856Crossref PubMed Scopus (161) Google Scholar, 5Prodromos C.C. Joyce B.T. Shi K. Keller B.L. A meta-analysis of stability after anterior cruciate ligament reconstruction as a function of hamstring versus patellar tendon graft and fixation type.Arthroscopy. 2005; 21: 1202Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar Rather, we mean to elucidate that, as with original research articles, the conclusions reached by meta-analyses depend highly on the methods. Different meta-analytic methods will result in different studies being included or excluded from an investigation. In addition, different statistical methods may result in meta-analyses of the same studies reaching different conclusions. We choose to emphasize the importance of meta-analytic methods for a variety of reasons. To begin, this is not the first time that the results of a meta-analysis have contradicted previous publications in the Journal. 5Prodromos C.C. Joyce B.T. Shi K. Keller B.L. A meta-analysis of stability after anterior cruciate ligament reconstruction as a function of hamstring versus patellar tendon graft and fixation type.Arthroscopy. 2005; 21: 1202Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar, 6Goldblatt J.P. Fitzsimmons S.E. Balk E. Richmond J.C. Reconstruction of the anterior cruciate ligament: Meta-analysis of patellar tendon versus hamstring tendon autograft.Arthroscopy. 2005; 21: 791-803Abstract Full Text Full Text PDF PubMed Scopus (269) Google Scholar, 7Yunes M. Richmond J.C. Engels E.A. Pinczewski L.A. Patellar versus hamstring tendons in anterior cruciate ligament reconstructions: A meta-analysis.Arthroscopy. 2001; 17: 248-257Abstract Full Text Full Text PDF PubMed Scopus (246) Google Scholar Second, the level of evidence of a meta-analysis is only as high as that of the articles included. The meta-analysis of Prodromos et al.2Prodromos C.C. Han Y. Rogowski J. Joyce B.T. Shi K. A meta-analysis of the incidence of anterior cruciate ligament tears as a function of gender, sport and a knee injury–reduction regimen.Arthroscopy. 2007; 23: 1320-1325Abstract Full Text Full Text PDF PubMed Scopus (646) Google Scholar in this issue of Arthroscopy, for example, represents Level IV evidence as a result of the low levels of evidence of the previously published literature available for review. Third, depending on the methods, meta-analyses may vary widely in quality, and meta-analytic quality may be assessed using an objective scoring system.8Bhandari M. Morrow F. Kulkarni A.V. Tornetta III, P. Meta-analyses in orthopaedic surgery A systematic review of their methodologies.J Bone Joint Surg Am. 2001; 83: 15-24Crossref PubMed Scopus (234) Google Scholar Finally, as we have recently predicted,9Lubowitz J.H. Poehling G.G. Arthroscopy: The future.Arthroscopy. 2007; 23: 453-454Abstract Full Text Full Text PDF Scopus (13) Google Scholar concomitant with the evidence-based medicine movement, publication of meta-analyses is becoming more frequent. Readers, and we Editors, must carefully scrutinize the methods when we consider the conclusions of a meta-analysis or a systematic review.