Association of fructosamine with capillary blood glucose in patients with diabetes

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: SUN 807-838-Diabetes - Diagnosis, Complications & Outcomes
Sunday, June 16, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board SUN-813
Kathryn Elizabeth Berkseth*1, Dace Lilliana Trence2 and Irl B Hirsch3
1University of Washington, Seattle, WA, 2Univ of WA Medical Ctr, Bellevue, WA, 3Univ of Washington Med Ctr, Seattle, WA

Hemoglobin A1c (A1C) has become the standard measure for monitoring glycemic control in patients with diabetes.  However, there are clear limitations to the use of A1C in many subgroups of patients.  In these cases, fructosamine (a measure of glycated proteins) may be helpful as an alternative or adjunct measurement tool.  Fructosamine is thought to reflect glycemic control over the previous 2-3 weeks with higher values indicating higher average serum glucose.  There are currently no published data that directly compare elevated fructosamine values with capillary blood glucose data in patients with diabetes.


The primary objective of the current study is to identify the relationship between fructosamine and average blood glucose in subjects with diabetes. 

Study design:

Results of all fructosamine assays reported by the University of Washington laboratory from August 2011 to February 2012 were obtained.  Retrospective chart review was performed on all 160 identified charts.  Inclusion criteria were: available capillary blood glucose data for 4 weeks prior to fructosamine level and capillary blood glucose levels measured on average at least 3 times daily over that period.  Final sample size was 30 results from 28 patients.  Subset analysis was performed on 13 patients with known normal serum albumin (>3.5mg/dl).  Fructosamine and average capillary blood glucose levels were compared using simple linear regression.


The mean patient age was 53.5 years (range 20-75 years), and 16 subjects were female.  Hematologic measurements (mean, range) included: A1C (7.2%, 5.4-10.1%), fructosamine (314 𝜇mol/L, 209-386 𝜇mol/L), and average blood glucose for the preceding 4 weeks (165 mg/dl, 121-235 mg/dl). On average, patients measured capillary blood glucose 4.3 times per day (range: 3-7), and two patients used continuous glucose monitoring.  A significant association was found between average capillary blood glucose and fructosamine in the primary analysis group (p= 0.007, R2=0.232) as well as in the subset of patients with normal albumin (p=0.004, R2= 0.5457). The following equations summarize the predicted average capillary blood glucose (BG) over the past four weeks based on measured fructosamine for each data set: 

All patients: Average BG (mg/dL) = 38 + 0.41*fructosamine (𝜇mol/L)

Patients with known normal albumin: Average BG (mg/dL) = -59 + 0.78*fructosamine (𝜇mol/L)

Assessment of prediction intervals for the above models and graphical analysis of regression results was also performed.


This study can be used as an initial estimation for interpretation of fructosamine values; however, the prediction intervals for average blood glucose are large.  Additional data are required before fructosamine results can be interpreted with confidence as a surrogate measure of glycemic control in patients with diabetes.

Nothing to Disclose: KEB, DLT, IBH

*Please take note of The Endocrine Society's News Embargo Policy at