Very tight vs. tight control: what should be the criteria for pharmacologic therapy dose adjustment in diabetes in pregnancy? Evidence from randomized controlled trials

IntroductionThere is inconclusive evidence from randomized controlled trials (RCTs) to support any specific criteria for pharmacologic therapy dose adjustment in diabetes in pregnancy. Our objective was to analyze the criteria for dose adjustment of pharmacologic treatment for diabetes mellitus (DM) in pregnancy. Material and methodsData sources: MEDLINE, OVID and Cochrane Library were searched from their inception to September 2017. Selection criteria included all trials of DM in pregnancy managed by oral hypoglycemic agents or insulin reporting criteria for pharmacologic therapy dose adjustment. RCTs in women with pregestational DM and gestational DM (GDM) were included. For each trial, data regarding glucose values used for pharmacologic therapy dose adjustment were extracted and carefully reviewed. ResultsOf 51 RCTs on therapy for GDM or pregestational DM, 17 (4230 women) were included as they reported criteria for pharmacologic therapy dose adjustment. Most of them (88%, 15/17) included women with GDM only. For RCTs including women with GDM, 12/16 (75%) used the two-step approach, three (19%) the one-step approach and one (6%) either the one- or two-step approach. Regarding the type of initial therapy, 13 (77%) RCTs used different types and doses of insulin; nine (53%) used metformin; five (30%) used glyburide; and one (6%) used placebo. In most RCTs, glucose monitoring was assessed four times daily, i.e. fasting (all RCTs) and two hours (15 RCTs, 88%) after each of the three main meals - breakfast, lunch, and dinner. For fasting glucose target, all used a value <105mg/dL; nine (53%) used 95mg/dL as target. Of the 15 RCTs using a two-hour postprandial value as target, 11 (73%) had 120mg/dL as cutoff. Regarding the criteria for pharmacologic therapy dose adjustment, we found six different criteria. The majority of RCTs (9/17, 53%) used either one or two values per week higher than the target values, of which two-thirds used only one value (35% of total), and one-third (18% of total) two values. Five RCTs (29%) used >50%, one (6%) >30%, and one (6%) >20% of the values higher than the target value; one (6%) used the appearance of glycosuria. ConclusionsWhen evaluating RCTs which included criteria for pharmacologic GDM therapy dose adjustment, the most common criterion for diagnosis was the two-step test, and the most common used therapies were insulin and metformin. Regarding glucose monitoring, the most common frequency was four times per day, fasting and two hours after each main meal, using as target glucose values 95 and 120mg/dL, respectively. Importantly, we found six different criteria for pharmacologic GDM therapy dose adjustment, with the majority using very tight criteria of either one or two values per week higher than the target values, of which two-thirds used only one value, and one-third used two values.