Autonomic neuropathy predisposes to rosiglitazone-induced vascular leakage in insulin-treated patients with type 2 diabetes: a randomised, controlled trial on thiazolidinedione-induced vascular leakage.

AIMS/HYPOTHESIS: The mechanism of fluid-related complications caused by thiazolidinedione derivatives is unclear. One potential mechanism is thiazolidinedione-induced arterial vasodilatation, which results in vascular leakage and a fall in blood pressure, normally counterbalanced by sympathetic activation and subsequent renal fluid retention. We hypothesised that thiazolidinedione-induced vascular leakage will be particularly prominent in patients with autonomic neuropathy. METHODS: We conducted a randomised, double-blind, placebo-controlled, parallel study in 40 patients with type 2 diabetes on insulin treatment recruited from a university medical centre. The randomisation was performed by a central office using a randomisation schedule. Both treatment groups, placebo (n = 21) and rosiglitazone (n = 19), were stratified for sex and level of autonomic neuropathy as assessed by Ewing score (or=2.5). We investigated the effects of 16 weeks of treatment with rosiglitazone 4 mg twice daily on vascular leakage (transcapillary escape rate of albumin, TERalb), body weight, extracellular volume and plasma volume. RESULTS: Thirty-nine patients were included in the analysis. In patients with high Ewing scores (n = 16), rosiglitazone increased TERalb significantly (DeltaTERalb: rosiglitazone +2.43 +/- 0.45%/h, placebo -0.11 +/- 0.15%/h, p = 0.002), while rosiglitazone had no effect in the patients with low Ewing scores (n = 23). Rosiglitazone-induced increases in TERalb and Ewing score at baseline were correlated (r = 0.65, p = 0.02). There was no correlation between Ewing score and rosiglitazone-induced changes in fluid variables. One subject was withdrawn from the study because of atrial fibrillation. CONCLUSIONS/INTERPRETATION: Rosiglitazone may increase vascular leakage in insulin-treated patients with type 2 diabetes with autonomic neuropathy. Autonomic neuropathy did not exaggerate rosiglitazone-induced fluid retention. Therefore, autonomic neuropathy should be considered as a risk factor for thiazolidinedione-induced oedema, not for thiazolidinedione-induced fluid retention. TRIAL REGISTRATION: ClinicalTrials.gov NCT00422955. FUNDING: GlaxoSmithKline