Was insulin review biased, or simply a victim of bad timing?

Critical analysis of both protocols provides a plausible explanation. Dr Raskin’s study eliminated insulin secretagogues (sulfonylureas, repaglinide, and nateglinide) and α-glucosidase inhibitors from the treatment regimen; patients were permitted to continue therapy with metformin and pioglitazone. In contrast, in Dr Janka’s study, premixed insulin patients discontinued both secretagogue and metformin therapy; patients were not receiving any prior TZD therapy. It’s possible that the differences in outcomes between the 2 studies would almost certainly be reconciled if both investigators had chosen to continue all oral agents. Furthermore, the incidence of hypoglycemia, which Dr Heise characterizes as “slightly higher,” was actually as follows:

• Janka et al¹⁹ study: 4.07 per patient year with glargine plus oral agents vs 9.87 with premix (P<.0001)
• Raskin et al¹⁴ study: 0.7±2.0 per patient year with glargine vs 3.4±6.6 with premix (P<.05)

I thank Drs Heise and Brunton for their perspective and the points they have raised. Ultimately, each clinician must determine an appropriate and individualized treatment strategy for his or her patients based on the available published information coupled with clinical judgment.

Thomas M. Flood, MD, FACE
Director, Georgia Center for Diabetes, Atlanta