To the Editor: In a recent issue of Diabetologia, Cederberg et al report a dose-dependent increased risk of type 2 diabetes with long-term simvastatin or atorvastatin treatment [1]. Impaired insulin sensitivity and reduced pancreatic insulin secretion may mediate this diabetogenic effect [1]. Some comments may be of interest.

  1. 1.

    Dose-response and 3-hydroxy-3-methylglutaryl-CoA reductase inhibition We previously showed that rosuvastatin increased insulin resistance in patients with impaired fasting glucose in a dose-dependent manner [2]. It was suggested that statins reduce insulin sensitivity in parallel with their 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) reductase inhibitory capacity [3]. This effect might be most prominent for the potent statin rosuvastatin.

  2. 2.

    Lipophilic vs hydrophilic statins Impaired insulin secretion induced by the lipophilic simvastatin and atorvastatin may not be relevant for the hydrophilic rosuvastatin [3, 4]. Post-treatment changes in HOMA of beta cell function (HOMA-B) [5] as a surrogate of pancreatic beta cell function were calculated using insulin and glucose values from a previous dataset of rosuvastatin-treated patients [2]. Rosuvastatin at daily doses of 10, 20 and 40 mg dose-dependently increased HOMA-B by 14.2% (p = not significant vs baseline), 25.4% (p < 0.05 vs baseline; p < 0.05 vs 10 mg) and 46.1% (p < 0.05 vs baseline; p < 0.05 vs 10 and 20 mg), respectively. The resultant compensatory hyperinsulinaemia may prevent hyperglycaemia in the short-term [2]. In this context, it is relevant to prospectively compare the time-dependent diabetogenic effects of lipophilic and hydrophilic statins.