Reply to: Tocolytic magnesium sulfate: Are the epidemiologic data reassuring?

Abstract

To the Editors:It is true that the acknowledged standard for decision making in clinical medicine is the well-designed, well-conducted, randomized controlled trial. It is also true that a poorly conducted clinical trial—one that is too small, does not take into account major determinants of the outcome, and does not ascertain outcome fully in both arms of the trial—can be profoundly misleading. Some of these limitations may be present in the small clinical trial to which Mittendorf and Pryde allude, as suggested by the letters that followed its publication.1Benichou J Zupan V Fernandez H Marpeau L Marret S Tocolytic magnesium sulphate and paediatric mortality [letter].Lancet. 1998; 351: 5-6Abstract Full Text Full Text PDF Google Scholar, 2Crowther C Hiller J Doyle L Lumley J Carlin J Tocolytic magnesium sulphate and paediatric mortality [letter].Lancet. 1998; 351: 6-7Abstract Full Text Full Text PDF Scopus (13) Google Scholar, 3Leveno KJ Tocolytic magnesium sulphate and paediatric mortality [letter].Lancet. 1998; 351: 7-8Abstract Full Text Full Text PDF PubMed Google ScholarSeveral large randomized controlled trials are now in progress to evaluate the relationship of antenatal treatment with magnesium to perinatal and infant mortality and neurologic outcome. These trials are likely to be informative and possibly conclusive. In the meantime, obstetric management must rely on available evidence and, in doing so, must judge the quality of that evidence. Case-control studies have played an important role in the history of medical science and disease reduction. Unless Mittendorf and Pryde can identify some specific source of bias in our study or particular flaws in its analyses, it should not be summarily dismissed because of its case-control design. To the Editors:It is true that the acknowledged standard for decision making in clinical medicine is the well-designed, well-conducted, randomized controlled trial. It is also true that a poorly conducted clinical trial—one that is too small, does not take into account major determinants of the outcome, and does not ascertain outcome fully in both arms of the trial—can be profoundly misleading. Some of these limitations may be present in the small clinical trial to which Mittendorf and Pryde allude, as suggested by the letters that followed its publication.1Benichou J Zupan V Fernandez H Marpeau L Marret S Tocolytic magnesium sulphate and paediatric mortality [letter].Lancet. 1998; 351: 5-6Abstract Full Text Full Text PDF Google Scholar, 2Crowther C Hiller J Doyle L Lumley J Carlin J Tocolytic magnesium sulphate and paediatric mortality [letter].Lancet. 1998; 351: 6-7Abstract Full Text Full Text PDF Scopus (13) Google Scholar, 3Leveno KJ Tocolytic magnesium sulphate and paediatric mortality [letter].Lancet. 1998; 351: 7-8Abstract Full Text Full Text PDF PubMed Google ScholarSeveral large randomized controlled trials are now in progress to evaluate the relationship of antenatal treatment with magnesium to perinatal and infant mortality and neurologic outcome. These trials are likely to be informative and possibly conclusive. In the meantime, obstetric management must rely on available evidence and, in doing so, must judge the quality of that evidence. Case-control studies have played an important role in the history of medical science and disease reduction. Unless Mittendorf and Pryde can identify some specific source of bias in our study or particular flaws in its analyses, it should not be summarily dismissed because of its case-control design. It is true that the acknowledged standard for decision making in clinical medicine is the well-designed, well-conducted, randomized controlled trial. It is also true that a poorly conducted clinical trial—one that is too small, does not take into account major determinants of the outcome, and does not ascertain outcome fully in both arms of the trial—can be profoundly misleading. Some of these limitations may be present in the small clinical trial to which Mittendorf and Pryde allude, as suggested by the letters that followed its publication.1Benichou J Zupan V Fernandez H Marpeau L Marret S Tocolytic magnesium sulphate and paediatric mortality [letter].Lancet. 1998; 351: 5-6Abstract Full Text Full Text PDF Google Scholar, 2Crowther C Hiller J Doyle L Lumley J Carlin J Tocolytic magnesium sulphate and paediatric mortality [letter].Lancet. 1998; 351: 6-7Abstract Full Text Full Text PDF Scopus (13) Google Scholar, 3Leveno KJ Tocolytic magnesium sulphate and paediatric mortality [letter].Lancet. 1998; 351: 7-8Abstract Full Text Full Text PDF PubMed Google Scholar Several large randomized controlled trials are now in progress to evaluate the relationship of antenatal treatment with magnesium to perinatal and infant mortality and neurologic outcome. These trials are likely to be informative and possibly conclusive. In the meantime, obstetric management must rely on available evidence and, in doing so, must judge the quality of that evidence. Case-control studies have played an important role in the history of medical science and disease reduction. Unless Mittendorf and Pryde can identify some specific source of bias in our study or particular flaws in its analyses, it should not be summarily dismissed because of its case-control design.