GENERAL COMMENTS

Visboll and colleagues have submitted their trial design for a phase II study of a glucagon-like peptide-1 agonist, liraglutide, on the development of type 2 diabetes in a defined at-risk population (women that had gestational diabetes). The rationale for the study is logical and persuasive, the study is ethically appropriate, randomisation process is valid, allocation concealment is excellent, and the proposed outcomes and analyses are well-founded.

I have only two minor comments.

1. The authors may wish to comment on what they will do with ‘dropouts’. It is conceivable that with a cohort of 100 women of this age, many still within their child-bearing years, that the occasional subject will get pregnant (and need to withdraw) even though they are not planning another pregnancy during the study period.

2. With the sentences (p.5) ‘Patients with T2DM exhibit impaired incretin effect. The cause of this pathophysiological trait is unclear (21-23)’. There has been recent suggestions from Meier and Nauck [1] that the impaired incretin effect observed in patients with type 2 diabetes is just epi-phenomenon related to reduced β-cell function, supported by data from Horowitz and colleagues suggesting that secretion of incretin hormones is unaffected in diabetic patients when the variable of gastric emptying is removed [2]. The authors may wish to identify this when discussing the pathophysiology of the impaired incretin effect in this group.