Dosing accuracy of two insulin pre-filled pens

This study was designed to determine the dose accuracy of two commonly available insulin pre-filled pens for use in diabetes - FlexPen (Novo Nordisk A/S, Bagsvaerd, Denmark) (FP) and SoloStar (Sanofi-Aventis, Paris, France) (SS).

Dosing accuracy was tested at 5 U, 10 U and 30 U doses - three previously unused pens of both pre-filled pens were used for each dose. At the 5 U dose each pen was tested 42 times, at the 10 U dose each pen was tested 25 times and at the 30 U dose each pen was tested 9 times. The pre-filled pens were used strictly according to manufacturers' instructions and measurements were made on a sensitive balance and corrected for the specific density of the insulin formulations used. Specified limits were based on ISO standards (+/-1 U for the 5 U and 10 U doses and +/-1.5 U for the 30 U dose).

FP was more accurate for injecting 5 U, 10 U and 30 U doses (absolute mean 4.95 +/- 0.19 U, 9.61 +/- 0.27 U and 29.70 +/- 0.34 U, respectively) than SS (absolute mean 4.86 +/- 0.39 U, 9.27 +/- 0.52 U and 28.73 +/- 0.47 U, respectively). No doses were outside specified limits for 5 U and 30 U with FP and 1.3% of doses were outside these limits at the 10 U dose. For SS; 1.6%, 29.3% and 33.3% of doses were below the pre-specified threshold for the 5 U, 10 U and 30 U doses, respectively.

This non-blinded comparison indicates that FP is a more accurate insulin pre-filled pen than SS at three different insulin doses. In this single-user study, FP delivered consistent and accurate doses of insulin but SS had a high frequency of under dosing. No assessment was made of user variability in this study.