To the Editor: Nathan and colleagues recently published a review of diabetes prevention trials, specifically looking at their long-term effects on reducing the diabetes-related vascular complications [1]. Data from nine diabetes prevention studies were thoroughly and comprehensively analysed to investigate the association of diabetes prevention with microvascular or cardiovascular complications. The majority of participants enrolled in the nine studies had either impaired glucose tolerance (IGT) or impaired fasting glucose (IFG) or both. We found that in Table 1 of their review [1], Nathan et al included details of the specific diagnostic criterion for IFG, which was a fasting plasma glucose (FPG) level of 5.3–6.9 mmol/l in the Diabetes Prevention Program (DPP) trial, but no details were given for other trials. In fact, the definitions given for IFG or IGT in terms of the range of FPG were inconsistent across studies.

Table 1 Description of major diabetes prevention studies reporting long-term complications
Full size table

We have adjusted the table to provide the FPG ranges of the definitions (Table 1). The definition of IGT in terms of 2 h plasma glucose (7.8–11.0 mmol/l) was the same in nine studies, but the diagnostic criterion for IFG or IGT in term of FPG was different across studies. In particular, in the Da Qing Diabetes Prevention Study (DQDPS), Finnish Diabetes Prevention Study (FDPS) and the Study to Prevent Non-Insulin-Dependent Diabetes Mellitus (STOP-NIDDM), some of the original participants who were enrolled decades ago with an FPG of 7.0–7.7 mmol/l should have been diagnosed with diabetes, rather than prediabetes, according to the current diabetes diagnostic criterion as set out in 1999 [2]. Reviewing these diabetes prevention cohorts without a concrete interpretation of the diagnostic criteria may, to some extent, lead to an insufficient understanding of their original results. These minor differences in the IFG or IGT diagnostic criterion according to FPG may at least partly explain the discrepant results between these diabetes prevention studies. Therefore, we have added the information regarding the criteria in Table 1 as a supplement to the work by Nathan et al.