Reply to Retnakaran R, Shah BR. Abnormal screening glucose challenge test in pregnancy and future risk of diabetes in young women. Diabet Med 2009; 26: 474-477 I am not convinced from the study by Renakaran and Shah [1] claiming that an abnormal gestational diabetes mellitus (GDM) test during pregnancy followed by a normal oral glucose tolerance test (OGTT) provides prognostic information that could not be obtained rather more simply. The authors created a matched cohort based upon age, region, socio-economic status and year of delivery. However, they did not match for body mass index (BMI), nor for non-pregnant abdominal circumference. I wonder if Renakaren and Shah could explore their database to determine if the women who developed Type 2 diabetes mellitus (T2DM) during the study were more overweight and had a greater abdominal circumference, and whether these risk factors were similar in the cohort and the matched population? As the 'number needed to test' is 303 to identify one patient who will go on to develop diabetes; one wonders if information such as abdominal circumference and BMI would be practical indicators of risk. Competing interestsNothing to declare. J. Andrews Exenatide and renal failureExenatide is a new agent, approved in 2005 and indicated for the treatment of Type 2 diabetes. It belongs to a class of drugs known as incretin mimetics. As with the endogenous incretin hormone glucagon-like peptide 1 (GLP-1), exenatide effectively lowers blood glucose levels by stimulating insulin release, inhibiting glucagon secretion and delaying gastric emptying. Nausea and vomiting are the most common adverse reactions, which lead to discontinuation in as many as 4-14% of patients in some studies (1,2). We report one patient (female; age 58 years) with exenatideassociated renal failure who presented with nausea, vomiting and reduction of blood pressure. She had Type 2 diabetes with mild chronic nephropathy [creatinine 97.2 lmol ⁄ l, creatinine clearance 65 ml ⁄ min], microalbuminuria and hypertension. Exenatide was initiated and maintained at a dosage of 5 lg twice daily. The patient had been taking stable doses of metformin, repaglinide, angiotensin-converting enzyme inhibitors and diuretics for two years. She was not using nonsteroidal anti-inflammatory drugs. The time between starting exenatide therapy and diagnosis of renal failure was three months (creatinine 227.2 lmol ⁄ l; creatinine clearance 39 ml ⁄ min) and the drug was then discontinued. She did not require hospitalization. Fifteen days after withdrawal of exenatide, metformin and secretagogues, renal function had recovered (creatinine 96.3 lmol ⁄ l; creatinine clearance 68 ml ⁄ min).