Comparison of glyburide and insulin for the management of gestational diabetes in a large managed care organization

Abstract

To the Editors: The study by Jacobson et al1Jacobson G.F. Ramos G.A. Ching J.Y. Kirby R.S. Ferrara A. Field D.R. Comparison of glyburide and insulin for the management of gestational diabetes in a large managed care organization.Am J Obstet Gynecol. 2005; 193: 118-124Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar comparing the use of glyburide and insulin for the management of gestational diabetes was extremely interesting, and we congratulate the authors on their clever study design using a “natural experiment” with a historical control group. However, the authors of the study, as well as the accompanying editorial,2Durnwald C. Landon M.B. Glyburide: the new alternative for treating gestational diabetes?.Am J Obstet Gynecol. 2005; 193: 1-2Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar concluded that glyburide appeared to be a safe and effective alternative to insulin for the treatment of gestational diabetes mellitus in appropriately selected patients. We would like to point out 2 potential problems with this conclusion. First, because the authors are primarily highlighting their negative findings, it would be important to know how likely it may have been that they could have a type II error for the neonatal outcomes they are reporting. Was a power analysis, even post hoc, performed to address this issue? For example, we note that they report a 1% rate of birth injuries in the insulin group versus a 3% rate in the glyburide group (P = .08); certainly this finding, among others, is not particularly reassuring with regard to the safety of glyburide management in those with A2 gestational diabetes. In the most cited study on this same issue, Langer et al3Langer O. Conway D.L. Berkus M.D. Xenakis E.M. Gonzalez O. A comparison of glyburide and insulin in women with gestational diabetes mellitus.N Engl J Med. 2000; 343: 1134-1138Crossref PubMed Scopus (666) Google Scholar are similarly underpowered to examine neonatal outcomes. We would request that the authors submit an estimate of power in this regard. Second, the authors did find several worse outcomes in patients treated with glyburide, such as an increased need for neonatal phototherapy (despite similar rates of hyperbilirubinemia in both groups), and an increased incidence of pre-eclampsia. This latter finding is particularly concerning in light of recent findings by Crowther et al4Crowther C.A. Hiller J.E. Moss J.R. McPhee A.J. Jeffries W.S. Robinson J.S. Effect of treatment of gestational diabetes mellitus on pregnancy outcomes.N Engl J Med. 2005; 352: 2477-2486Crossref PubMed Scopus (2337) Google Scholar that pre-eclampsia is more prevalent among patients with untreated gestational diabetics. We are left wondering whether in fact glyburide is less effective in treating gestational diabetes, leading to a higher rate of preeclampsia. Given these 2 issues, we suggest that an excellent study such as this one not seek to capsulate its findings into an overarching reassuring conclusion, when one should not be reasonably reached. Rather, such a study should encourage the field to conduct further research on this topic and to incite clinicians to exert caution when adopting new modalities of treatment that may, in the end, lead to worse outcomes for our patients. ReplyAmerican Journal of Obstetrics & GynecologyVol. 195Issue 2PreviewTo the Editors: We thank Drs Land and Caughey for their comments. In designing our observational study, glycemic control and birth weight were treated as primary outcome variables. We felt these outcomes were clinically relevant and could be realistically powered to find significant differences. Although we collected data on a number of other outcomes of interest, we anticipated many would be uncommon as well as unpredictable, and it was not feasible to power a study to assess them all. Our power for detecting the given difference for birth injury was only 41%. Full-Text PDF