Peer Review in Medical Journals: A Bad Example

Abstract

To the Editor: Dr. Robin (Chest 1987; 91:252-55) has maintained his stimulating and provocative tradition by challenging the conventional peer review system used by many distinguished medical journals for evaluating manuscripts for publication.1Robin ED Burke CM Peer review in medical journals.Chest. 1987; 91: 252-255Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar One of his major concerns, and rightly so, is that manuscripts containing errors in research methodology and/or interpretations of findings, if published, may result in harm to patients. In support of his position, he cites two specific examples, one relating to the treatment of breast cancer, about which I am not qualified to comment, and one relating to the possible protective effect of barbiturates in the treatment of global cerebral ischemia,2Bleyaert AL Nemoto EM Safer P Stezosld SW Mickell JJ Moossy J et al.Thiopental amelioration of brain damage after global ischemia in monkeys.Anesthesiology. 1978; 49: 390-398Crossref PubMed Scopus (144) Google Scholar about which I am qualified to comment since I was Editor-in-Chief of the journal at the time the article was published. Dr. Robin is critical of the latter publication because it suggested that barbiturates may protect the brain from global ischemia; as a consequence, innumerable patients were treated with a potentially harmful therapy until the authors did another study, published eight years later, which contravened their earlier work. In attempting to prove his point, he has chosen the wrong example. What he failed to recognize and acknowledge is that the article was accompanied by an editorial which was written by two noted authorities in the field of barbiturates and brain protection.3Rockoff MA Shapiro HM Barbiturates following cardiac arrest: possible benefit or Pandora's box?.Anesthesiology. 1978; 49: 385-387Crossref PubMed Scopus (15) Google Scholar In their editorial, Drs. Rockoff and Shapiro articulated in a clear and comprehensive way the methodologic flaws of the study, and concluded by stating that additional laboratory studies and controlled human studies must be done before high-dose barbiturate therapy has clinical application in the treatment of cerebral ischemia. In contrast to Dr. Robin's conclusion, this represents peer review at its best Both prospectively and retrospectively, the peer review process worked to the benefit of patients. First, the topic was and is important because global ischemia is a devastating event for which there is no specific, assured, ameliorating treatment. Second, the Bleyaert study was flawed, a criticism incidentally that is applicable to every study, but the flaws were well defined for the reader in the article and accompanying editorial. Third, the editorial provided appropriate, conservative guidelines for both the investigators and readers as to how they might use this information. And finally, neither Dr. Robin nor I have any data proving that any patient was harmed by publication of this article. Progress in research requires publication of controversial, inconclusive studies, which the Bleyaert study was. When done in a responsible manner, it ultimately leads to improvement in patient care—which it did.