Eight randomised controlled trials (763 participants, range 40 to 227) were included in the review. Jadad scores showed four trials to be good quality (one trial scored 5 and three scored 3) and four trials to be poor quality (three trials scored 2 and one scored 1). Study duration ranged from three to 12 months.
Statistically significant benefits on fasting plasma glucose were observed with treatment with telmisartan compared to other angiotensin receptor blockers (WMD -8.63mg/dL, 95% CI -12.29 to -4.98; Ι²=66%, eight trials).
Subgroup analyses according to telmisartan dose showed statistically significant benefits with 40mg telmisartan (WMD -7.00mg/dL, 95% CI -11.14 to -2.85; Ι²=70%; six trials) and 80mg telmisartan (WMD -14.46mg/dL, 95% CI -20.01 to -8.91; Ι²=0%; two trials). Significantly greater benefit was seen with 80mg than with 40mg.
There were no statistically significant differences observed between telmisartan and other angiotensin receptor blockers in fasting plasma insulin (seven trials; Ι²=89%) and insulin resistance measured by homeostasis model assessments of insulin resistance (seven trials; Ι²=65%).
Subgroup analyses showed statistically significant decreases with telmisartan doses of 80mg in both fasting plasma insulin (WMD -6.06mg/dL, 95% CI -9.27 to -2.84; Ι²=33%; two trials) and insulin resistance (MD -1.60, 95% CI -3.18 to -0.02; one trial).
Statistically significant increases in adiponectin levels were observed with telmisartan compared with other angiotensin receptor blockers (WMD 0.93ug/dL, 95% CI 0.28 to 1.59; Ι²=4%; six trials) and with telmisartan administered at a dose of 40mg (WMD 1.03ug/dL, 95% CI 0.25 to 1.81; Ι²=20%; five trials). One study of telmisartan administered at 80mg found no differences in adiponectin levels compared to irbesartan.
Funnel plots and Egger's test revealed no evidence of publication bias.