Article Text

PDF

Tear secretion and tear film function in insulin dependent diabetics
  1. MURAT DOGRU
  1. Uludag University School of Medicine, Department of Ophthalmology, Gorukle Bursa 16059, Turkey
    1. M J GOEBBELS
    1. Dozent der Med Fakultät der Univ Bonn, Belegarzt des Krankenhauses Düren, Oberstrasse 45, 52349 Düren, Germany

      Statistics from Altmetric.com

      Editor,—I read the article by Goebbels1 with interest and would like to share a few opinions with you.2 Goebbels stated in his paper that BUT results did not differ between diabetics and controls and also pointed out that BUT is a very rough test for the detection of tear film stability. He found low Schirmer test values and conjunctival squamous metaplasia in diabetics compared with controls. He hypothesised that a decrease in reflex tearing inducing conjunctival surface damage, disturbance of the trophic function of the tear film, or metabolic alterations might be responsible. First of all, we believe with many others that BUT is an invaluable and direct test of tear film stability when performed carefully.2-4 In addition, a keen observation of the breaking tear film provides a lot of information on the minute changes on the ocular surfaces. We found in our study that BUT scores, Schirmer test values, central corneal sensitivity, and goblet cell density were significantly lower in NIDDM patients compared with controls. NIDDM patients also had significantly higher squamous metaplasia grades. We showed that tear film function and impression cytology variables significantly fared poorly in those patients with diabetes with peripheral neuropathy, decreased corneal sensitivity, and poor metabolic control without any correlation with duration of diabetes and status of retinopathy. We believe that the ocular surface disease in diabetes is characterised by squamous metaplasia and goblet cell loss which seems to evolve in close proximity to the status of metabolic control and peripheral neuropathy. Corneal and conjunctival epithelial damage caused by disruption of tear quantity and quality and diabetic neuropathy may be important determinants of diabetic ocular surface disease.

      Our final comment and request to all researchers who carry out impression cytology studies with devotion is that methodology of the procedure should be reported in each paper with photographs of the samples so that we can compare and refine our own procedures despite variability in cytology techniques and difficulties in comparing impression cytology studies with one another. Besides, no impression cytology study should be without information on figures of squamous metaplasia grade and goblet cell densities. Absence or presence of mucin pick up of filter papers must be mentioned without fail since such observations prove noteworthy; mucin being one of the major components in increasing the tear film stability and the wettability of the ocular surface.5

      References

      Reply

      Editor,—Dogru found decreased Schirmer test values and significantly higher grades of squamous metaplasia in NIDDM patients, thus confirming the data we obtained in our study on insulin dependent diabetics. Furthermore, he made two points with regard to break up time (BUT) and impression cytology (IC). Undoubtedly, the determination of the BUT is a helpful tool for the clinical assessment of tear film disorders, especially when showing significantly reduced BUT values. However, there is some question as to whether or not the measurement of the BUT is a sensitive technique for the quantitative and reproducible determination of tear film stability. Even when performed properly, BUT values are often characterised by significant intraindividual variability. In our study, BUT values did not differ between diabetics and controls. Thus, either the technique is not able to detect a difference between groups, or there isn't one. We obtained a total of 1332 IC specimens.

      The degree of squamous metaplasia was evaluated in a masked fashion according to a scale described by Tseng as mentioned in the paper. In our opinion, the demonstration of two or three figures of IC samples would provide poor additional information.

      View Abstract

      Request permissions

      If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.