Several devices for self-monitoring of blood glucose (SMBG) (e.g., AtLast, Amira; OneTouch Ultra, LifeScan; FreeStyle, TheraSense, Alameda, CA; Glucometer-Elite XL+Microlet-Vaculance, Bayer; and Sof-Tact, Abbott) recently received Food and Drug Administration approval for alternative site monitoring of capillary blood glucose. These alternatives are marketed with considerable efforts under the assumption that capillary blood glucose measurements, e.g., those taken at the forearm, do not differ from the results obtained by classic finger pricking. Diabetic patients using different devices for SMBG reported discrepancies between clinical symptoms of hypoglycemia and normoglycemic SMBG values at the forearm. Neither standardized quality control assessments of technical performance of such SMBG devices (1,2) nor patient device handling resulted in any obvious explanation of the reported discrepancies. Because of this, we examined whether or not fast blood glucose changes over a larger range of blood glucose concentrations could result in clinically relevant blood glucose differences between forearm and fingertip.
Capillary blood glucose samples were taken from the fingertip and the forearm of six male type 1 diabetic patients on intensified insulin treatment (age 26–54 years, diabetes duration 0.1–25 years) using the FreeStyle system (TheraSense) because it required the smallest blood glucose amount: 0.3 μl/sample. To avoid any disturbance of the normal regional blood flow, the forearm skin was not rubbed before blood glucose sampling, as recommended by the manufacturer. The following protocol was applied: after an overnight fast the usual prebreakfast insulin was omitted and the breakfast was replaced by oral Dextro O.G.T. (Roche, Mannheim, Germany), equivalent to 75 g glucose, in order to achieve blood glucose values of 300–400 mg/dl. Then the patient’s usual short-acting insulin was given intravenously at an individual dose (6–15 U/injection). The blood glucose decrease was followed every 5–15 min until either steady state or hypoglycemia (<60 mg/dl) was reached. Hypoglycemia was compensated by oral glucose. For control purposes, additional blood glucose samples from the fingertip were analyzed by the Gluco-quant method (Roche, Mannheim, Germany).
The capillary blood glucose decrease (mean ± SD) at the forearm (208 ± 38 mg/dl) was significantly smaller than at the fingertip (295 ± 16 mg/dl) (Student’s paired t test: P < 0.01) within 111 ± 26 min for all patients. An example is shown in Fig. 1. For the two patients with hypoglycemic unawareness, the first asymptomatic hypoglycemic values at the fingertip (51 and 53 mg/dl) were accompanied by normoglycemic values at the forearm (142 and 159 mg/dl). Compared with the fingertip, it took an additional 27–34 min until the capillary blood glucose levels at the forearm reached hypoglycemic values.
Despite the preliminary state of our investigation, the consistency of clinically relevant delays of blood glucose changes at the forearm prompted us to draw attention to a potentially very dangerous situation. Our results raise the possibility that the delayed glucose concentration changes at the forearm occur physiologically. To our knowledge, this has not been fully recognized as a potential problem by the certifying administrations in the U.S. or Europe.
Even a few delays of hypoglycemia detection could unnecessarily endanger the life of diabetic patients. Because of this, we strongly recommend providing sufficient evidence that the suggested use of SMBG at the forearm and other alternative sites does not result in a risky delay of hypoglycemia detection. Meanwhile, SMBG at the forearm should only be used when ongoing fast blood glucose changes can be excluded.
Address correspondence to Theodor Koschinsky, MD, Auf’m Hennekamp 65, D-40225 Duesseldorf, Germany. E-mail: email@example.com.