May 2005
Volume 46, Issue 13
ARVO Annual Meeting Abstract  |   May 2005
Diabetes Determination via Tear Glucose Values
Author Affiliations & Notes
  • J. Lane
    CIBA Vision, Duluth, GA
  • D. Krumholz
    State College of Optometry, State University of New York, New York, NY
  • R. Sack
    State College of Optometry, State University of New York, New York, NY
  • C. Morris
    Southern Cross University, New South Wales, Australia
  • Footnotes
    Commercial Relationships  J. Lane, CIBA Vision F; D. Krumholz, None; R. Sack, None; C. Morris, CIBA Vision F.
  • Footnotes
    Support  CIBA Vision
Investigative Ophthalmology & Visual Science May 2005, Vol.46, 355. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      J. Lane, D. Krumholz, R. Sack, C. Morris; Diabetes Determination via Tear Glucose Values . Invest. Ophthalmol. Vis. Sci. 2005;46(13):355.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Abstract: : Purpose:To determine if tear glucose values can discriminate between diabetic and normal patients. Methods: 121 diabetic and normal patients were recruited and randomized into fasting or nonfasting groups. Patients in the fasting group were asked to complete an overnight fast prior to the testing day. All diabetic patients returned on 2 subsequent days to be retested. The resulting groups were fasting normals (n=50), nonfasting normals (n=23), fasting diabetics (n=33), and nonfasting diabetics (n=15).Upon arrival at the testing site, baseline capillary blood and tear samples were taken from each patient. Capillary blood was tested immediately via a home blood glucometer. Tear samples (1–5 µL) were taken via a micropipette and stored at –80°C until analysis via HPLC–PAD. After baseline samples were taken, a standard glucose load was ingested by each patient. Tear and capillary blood samples were collected at 15, 30, 60, and 90 minutes post–load. Results: Values of 2.0 and 2.5 mg/dL were selected as potential diabetic cutoff values in tear glucose (TG). A standard value of 126 mg/dL was used as the diabetic cutoff value for capillary blood glucose (CBG). Sensitivity, specificity, and the predictive value of positive and negative tests for TG were calculated at both TG cutoff values. These values were compared to the same values calculated for CBG at 126 mg/dL. At all time points, sensitivity and specificity of the TG values were lower than for the CBG value. In some instances, the predictive value of the positive test was higher with the TG than with the CBG. Conclusions: Although TG values do not have the sensitivity and specificity of CBG values for predicting the presence of diabetes mellitus, TG levels above 2.0 mg/dL may indicate the need for a patient to have further testing by their primary care physician. As one–third of all diabetics in the United States (5.2 million) are unaware of their disease, an increase in testing of the general population may help identify these individuals and prevent health issues from prolonged hyperglycemia such as diabetic retinopathy, neuropathy, and nephropathy.

Keywords: diabetes • clinical laboratory testing • detection 

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.