Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Apr 4.
Published in final edited form as: Clin Exp Optom. 2022 Aug 17;106(7):752–758. doi: 10.1080/08164622.2022.2111204

Changes in tear glucose and insulin concentrations following an oral glucose tolerance test

Cecilia Chao a,b, Carolina Lema a, Rachel Redfern a, Kathryn Richdale a
PMCID: PMC11970258  NIHMSID: NIHMS2065921  PMID: 35977531

Abstract

Clinical relevance:

Tear glucose and insulin are responsible for the health of the ocular surface; thus, it is important for clinicians to detect the tear glucose and insulin using point-of-care methods.

Aim:

To determine if changes in blood glucose and insulin levels following an oral glucose tolerance test are reflected in the tears and to test the association between gene expression and tear insulin and glucose.

Methods:

Twenty healthy young adults were enrolled. Basal tears and peripheral blood samples were collected to assess glucose and insulin using a point-of-care glucometer and ELISA assays in fasted subjects, and 1.5 and 3 h after an oral glucose challenge. Conjunctival impression cytology was collected to determine gene expression of insulin receptor (INSR) and glucose transporters (GLUT1 and GLUT4). Changes were examined using non-parametric one-way ANOVA. Spearman tests were conducted to examine associations between variables.

Results:

Glucose and insulin levels increased 1.5 h after oral glucose in both blood (P < 0.001) and tears (P < 0.049) and returned to near baseline values after 3 h. There was a positive correlation between glucose levels in the blood and tears (rho = 0.57, P < 0.001), but not between blood and tear insulin levels (P = 0.18). Glucose and insulin levels in tears were correlated (rho = 0.32, P = 0.048). Tear glucose concentration at 1.5 h after oral glucose was associated with INSR expression (rho = 0.49, P = 0.03), and there was a trend with GLUT1 (P = 0.06) but not GLUT4.

Conclusion:

Tear glucose reflected blood glucose levels but this correspondence was not observed for insulin. Further studies are required to determine the role of glucose and insulin on the ocular surface in both health and diabetes.

Keywords: Conjunctival glucose transporter, conjunctival insulin receptor, ocular surface, oral glucose tolerance test, tear glucose, tear insulin

Introduction

Over 34 million US adults have type 2 diabetes mellitus, a condition with chronically high blood glucose levels due to the body’s inability to utilise pancreatic insulin.1 New research has identified multiple clinical and subclinical changes in the anterior eye, which occur with type 2 diabetes, which could serve as one of the biomarkers of pre-diabetes or undiagnosed diabetes2–4 (reviewed in Richdale5) in optometric settings. Therefore, it is important to understand how glucose and insulin change during the disease state and how these fluctuations may affect the ocular surface. Tears are a readily accessible body fluid; however, compared to blood, there is limited understanding of normal insulin and glucose levels in the tears.6,7

Type 2 diabetes can cause alteration on the ocular surface and poor wound healing.2,8 Tear glucose has been measured using laboratory methodologies, including infrared spectroscopy, glucose oxidase, glucose dehydratase and other methods (reviewed in Baca et al.7). These techniques do not provide point-of-care results and therefore do not support detection and management during the eye care visits. Hyperglycaemia in tears has also been detected using novel contact lens devices,9,10 but these are not yet commercially available. Recently, Lee et al. developed and validated a modified point-of-care glucose monitor and reported tear glucose levels ranging from 0.34 to 1.57 mg/dL,11,12 which are approximately 50-fold lower than plasma glucose concentrations.13

Glucose transporters (GLUT) are the proteins responsible for glucose transportation across cell membranes.14,15 GLUT1 and GLUT4 are some of the major proteins involved in glucose haemostasis and transportation into cells. The function and expression of GLUT1 is independent and GLUT4 is dependent on insulin expression.15–17 The association between expression of these proteins and tear glucose was not previously investigated.

The main role of insulin is to modulate glucose metabolism and facilitate cell survival.18 It was suggested that corneal epithelial cells did not require insulin for glucose uptake19,20; although topical insulin has been successfully used to improve corneal wound healing.21–23 Only one study to date has quantified endogenous insulin in human tears using radioimmunoassay24 and found that tear insulin increased 1–2 h after a meal (11.5 ± 2.0 μIU/mL) compared to a fasted state (4.3 ± 2.0 μIU/mL) in healthy subjects.24 Corneal and conjunctival epithelial cells have insulin receptors, suggesting insulin in tears may modulate glucose uptake at the ocular surface.24 The ocular surface expression of insulin receptors has not been quantified nor has it been evaluated in the context of glucose uptake.

The purpose of this study was to quantify insulin and glucose changes in tears following a glucose tolerance test, and to compare tear to blood levels in healthy young adults. Also, the relationship between conjunctival insulin receptor (ISNR), GLUT1 and GLUT4 mRNA expression and tear glucose and insulin concentration was examined.

Methods

This was a two-day, single site prospective study conducted at the University of Houston (UH) College of Optometry. This study followed the tenets of the Declaration of Helsinki, with ethics approval from the UH Institutional Review Board. After explanation of the nature of the study, written informed consent was obtained from all subjects before participation. The first and second study visits were conducted at least 1 h after awakening and after at least 8 h of fasting. On the second visit, subjects were given 75 g of oral glucose (Trutol®, ThermoFisher Scientific, Waltham, MA). Outcome measures were performed at the first visit (fasted) and at the second visit 1.5 and 3 h ± 10 min after the ingestion of glucose. The 1.5-h interval was selected because testing 80–120 min after glucose challenge was reported to be associated with the highest morbidity rates in diabetic.25 Additional studies report there is a lag of up to 15–25 min before glucose levels equilibrate in the tears after eating.9,26

The sample size was calculated based on the difference in tear insulin concentration before and after meals as determined by radioimmunoassay by Rocha et al. (0.5 ± 0.2 ng/ml vs 0.3 ± 0.1 ng/ml).24 A sample size of 20 was estimated using G*Power 3.1 (Universitat Düsseldorf, Germany) to provide sufficient power between visits (1 – β error = 0.8, α = 0.05, effect size = 0.80) to detect differences in tear insulin.

All participants were aged between 18 and 30 years, with a normal body mass index (BMI: 18.5–24.9 kg/m2). Exclusion criteria included autoimmune and metabolic diseases, such as type 1 and 2 diabetes and Sjögren’s syndrome, ocular malformation, corneal scars, rigid contact lens wear, ocular surgery or injury, chronic use of any anti-inflammatory medications and/or eyedrops, such as steroid, non-steroidal anti-inflammatory, or anti-histamine, active eye inflammation and infection, smoking, alcoholism and other comorbid conditions, such as heart and kidney diseases and asthma, or pregnancy/breastfeeding during the study, which may affect the insulin level. Participants who were soft contact lens wearers were asked to not wear their contact lenses at least 8 h prior to and during the study visits.

Height, weight, BMI and body fat percentage (impedance biometer, Tanita Body Composition Analyser, Arlington Heights, IN, USA) were collected. An anterior segment slitlamp examination was conducted to confirm the health of the ocular surface. After tear collection, sodium fluorescein (GloStrips, Amcon, St Louis, MO, USA) and Lissamine Green (BioGlo, Alta Loma, CA, USA) dyes were instilled to assess the ocular surface integrity using a modified Oxford scale.

Up to 15 μl of basal tears were collected from a randomly selected eye using disposable glass capillary tubes (Blaubrand intraMARK; BRAND GMBH, Wertheim, Germany) at the first visit. In order to reduce the chance of stimulating tears, the alterative eye was used for the second sample. The same eye as the first visit was used for the last visit. All tears were expelled into an extended capacity microcentrifuge tube as previously described.27 Blood was collected via finger prick. Both tears and plasma samples were stored in a −80°C freezer until insulin analyses.

Based on the validated methodology from Cha et al. and Lee et al.,11,12 a commercially available glucometer with a test strip (Accu-Chek Aviva, Roche Diagnostics, Switzerland) was attached to a potentiostat (DY2113, Digi-Ivy, USA).11,12 After applying 150 mV, the output current obtained was measured using the amperometric i–t curve mode.12 The potentiostat was connected to a desktop computer (Dell, OptoPlex 7050, Round Rock, TX) in order to visualise the reading in nanoamperes (nA). Increasing concentrations of glucose solution (0, 0.01, 0.05, 0.1, 0.2, 0.4 and 0.8 mM – 0, 0.2, 0.9, 1.8, 3.6, 7.2 and 14.2 mg/dl) were prepared in phosphate-buffered saline (PBS) at pH 7.6. The 0 mM (blank) reading was subtracted from the sample readings for background correction. Six technical replicates were run for each glucose solution (1 μl) on 6 different days to create a standard curve at a temperature of 20°C and humidity between 42% and 52% (Figure 1).

Figure 1.

Figure 1.

Open in a new tab

A standard curve was generated from known concentrations of glucose (0–0.8 mm) and current (nA) measurements made using a modified glucometer. Plot represents the mean ± SD of six technical replicates. Note: 0.01 nm is not visible on the x-axis due to its proximity with the origin.

The tear glucometer was calibrated with 0 mM control solution each day of a study visit to standardise the potentiostat’s measurements. One μl of collected basal tears was then loaded by pipette into the strip and read from the potentiostat. The tear glucose measurement was repeated twice for each subject at each visit. The standard curve was used to convert the readout in nA to mg/dl (glucose concentration) and the average was obtained for analysis.

Blood glucose concentration was quantified with an unmodified glucometer and test strips (Accu-Chek Aviva, Roche Diagnostics, Switzerland).

Additional blood was collected in an EDTA tube and centrifuged at 250 rpm for 10 min at room temperature to extract the plasma and stored at −80°C along with the collected tears until insulin analysis. Tear and blood insulin ELISA were run separately, 1 day apart. Concentrations of insulin in tears (dilution 1:8) and blood (dilution 1:2) were determined using the Insulin Human ELISA kit (ab100578, Abcam, Boston, MA) according to manufacturer instructions but with a smaller sample volume and a higher concentration (2×) of horseradish peroxidase (HRP) conjugation agent as suggested by the manufacture to improve the assay sensitivity. Blood samples were diluted in sample diluent A and tear samples were diluted in sample diluent B. Insulin concentration was calculated from the standard curve and multiplied by the respective dilution factors.

Impression cytology samples were collected from the temporal conjunctiva, approximately 1 mm away from the corneal limbus, on both eyes on the first visit. Prior to collection, one to two drops of 0.5% proparacaine was instilled into the lower conjunctival sac. Superficial layers of conjunctival cells were collected using a Millicell® membrane (Merck Millipore, Ireland).28 A Millicell membrane was placed on the conjunctiva using sterile forceps and removed carefully after 5 s. It was placed in an Eppendorf tube with RLT lysis buffer (Qiagen, Germantown, MD) and β-mercaptoethanol and stored in a −80°C freezer until analysis.

Messenger RNA (mRNA) expression of human ISNR, GLUT 1 and GLUT4 on the ocular surface was determined using quantitative real-time PCR (qRT-PCR). Frozen conjunctival impression cytology samples were thawed at room temperature and RNA was extracted using the RNeasy Mini RNA extraction kit (Qiagen Sciences, Germantown, MD) following the manufacturer instruction as previously described.29 Total RNA concentration was measured using a microvolume spectrophotometer (DeNovix Inc., Wilmington, DE). Complementary DNA (cDNA) was obtained by reverse transcription using 250 ng of total RNA via iScript™ Reverse Transcription Supermix (Bio-Rad, Hercules, CA). Real-time PCR amplification of cDNA was performed with SsoAdvanced™ SYBR® Green Supermix (Bio-Rad) and PrimePCR™ SYBR® Green primers for human insulin receptor (INSR unique assay ID: qHsaCID0017132, Bio-Rad), glucose transporter 1 (GLUT1: qHsaCID002223, Bio-Rad), glucose transporter 4 (GLUT4: qHsaCID001334, Bio-Rad) and housekeeping RPL27 (qHsaCID0023846, Bio-Rad) using a CFX96™ Real-Time System.29 No reverse transcriptase (Bio-Rad) and water were used as negative controls. Samples were tested in triplicated, and all qRT-PCR reactions were carried out in 20 μl containing 10ng of cDNA. Quantification cycle (Cq), the number of PCR cycles required for the fluorescent signal to be detected/quantified, was determined and averaged. Delta Cq values were then calculated by the difference in Cq value between INSR (target gene) and RPL27 (housekeeping gene). Higher delta Cq values indicate lower mRNA expression of the target genes.

Statistical analysis

Data were analysed using SPSS 25 (IBM, Armonk, NY). Demographic information was reported as counts, mean and standard deviation or median and interquartile range (IQR), as appropriate. Non-parametric one-way ANOVA was carried out to compare the changes in glucose and insulin in tears and blood samples and were presented as estimated mean ± standard errors. Post-hoc paired t-test were used to compare changes between visits. Spearman (rho) and partial correlation (r) tests were conducted to examine associations between variables for fasted versus the non-fasted (1.5 h post glucose) states. Significance was considered for a P value less than 0.05 based on a two-tailed analysis.

Results

Twenty participants, including 14 females, with a mean age of 24 ± 1 years, were enrolled. Seven were habitual soft contact lens wearers. The average BMI was 22.1 kg/m2 (19.7–23.1 kg/m2) and body fat was 20.8% (15.3–28.3%), indicating a healthy population. Subjects had no significant ocular health findings based on slitlamp examination and ocular surface staining (grade 1 or lower for all parameters).

Tear and blood glucose levels increased 1.5 h after oral glucose ingestion and returned to near baseline levels by 3 h (Figure 2A). Specifically, fasted tear glucose concentration was 1.3 ± 0.3 mg/dl, increased to 3.0 ± 0.3 mg/dl at 1.5 h after oral glucose (F = 10.3, df = 57, P < 0.001) and decreased to 1.6 ± 0.29 mg/dl at 3 h (P = 0.45 vs fasted). Similarly, fasted blood glucose concentration was 93.3 ± 3.0 mg/dl, increased to 122.7 ± 3.0 mg/dl at 1.5 h (F = 51.7, df = 57, P < 0.001) and returned to 80.4 ± 3.1 mg/dl at 3 h (P = 0.005 vs fasted).

Figure 2.

Figure 2.

Open in a new tab

Absolute change in blood (B) and tear (T) concentration of glucose (A) and insulin (B) at the fasted state and 1.5 and 3 h after oral glucose ingestion. Box plots represent the median and interquartile ranges for blood (solid circles and red box) and tear (empty circles and blue box) samples (n = 20 each).

Blood (F = 17.9, df = 57, P < 0.001) and tear insulin concentrations (F = 3.2, df = 57, P = 0.049) also increased after oral glucose and then returned to fasted levels (Figure 2B). Fasted blood insulin increased from 4.3 ± 2.3 μIU/ml to 21.1 ± 2.3 μIU/ml 1.5 h after oral glucose (P < 0.001) and returned to fasted levels at 3 h (4.6 ± 2.4 μIU/ml, P = 0.94 vs fasted). Tear insulin concentration increased from fasted (15.5 ± 2.0 μIU/ml) to 1.5 h after oral glucose (22.1 ± 2.0 μIU/ml, P = 0.02) and returned to baseline at 3 h (16.7 ± 2.0 μIU/ml, P = 0.68 vs fasted).

The largest fold-increases in concentration were for blood insulin (average from fasted to 1.5 h after glucose: 5.8, IQR 2.4–10.4), while the smallest fold change was blood glucose (1.3, 1.2–1.8, Figure 3). The fold change in tear glucose was 2.0 (1.4–3.2) and insulin was 1.3 (1.2–2.0) after oral glucose test.

Figure 3.

Figure 3.

Open in a new tab

Fold change of blood and tear glucose and insulin levels before and 1.5 h after oral glucose tolerance test. Box plots represent the median and interquartile ranges for blood (solid circles and red box) and tear (empty circles and blue box) samples (n = 20 each).

Blood glucose was highly correlated with blood insulin (rho = 0.61, P < 0.001, Figure 4A) while tear glucose was weakly correlated with tear insulin (rho = 0.32, P = 0.048, Figure 4B). Blood and tear glucose concentrations were also highly correlated (rho = 0.58, P < 0.001, Figure 4C), but there was no significant association between tear and blood insulin concentrations in this study (P = 0.18, Figure 4D).

Figure 4.

Figure 4.

Open in a new tab

Correlations between blood glucose and insulin (A), tear glucose and insulin (B), blood and tear glucose (C) and blood and tear insulin (D) concentrations. Rho: Spearman correlation coefficient.

Average INSR, GLUT1 and GLUT4 mRNA values for the conjunctiva were 1.9 ± 0.4, 4.6 ± 0.5 and 10.7 ± 0.9 delta Cq, respectively. There was an association between INSR and GLUT1 mRNA expression (rho = 0.82, P < 0.001, Figure 5A) but not between INSR and GLUT4 (P = 0.29, Figure 5B) nor between GLUT1 and GLUT4 (P = 0.38). At 1.5 h after oral glucose ingestion (the maximum glucose levels recorded), tear glucose levels showed a positive correlation with INSR mRNA expression (rho = 0.49, P = 0.03, Figure 5C) and there was a trend with GLUT1 mRNA expression (rho = 0.44, P = 0.06, Figure 5D). However, no correlation was found between tear insulin levels and mRNA expression (Figure 5E–F).

Figure 5.

Figure 5.

Open in a new tab

Correlations between insulin receptor (INSR) and glucose transporter (GLUT) 1 (A) and 4 (B) mRNA expressions from conjunctival impression cytology. Correlations between INSR (C, E) or GLUT1 (D, F) mRNA expression and non-fasted tear glucose and insulin concentration. Rho: Spearman correlation coefficient.

Contact lens wearers had a lower concentration of tear insulin than non-lens wearers (14.1 ± 7.4 μIU/ml vs 20.4 ± 9.5 μIU/ml, df = 54, P = 0.008) but there were no differences in tear glucose between contact lens and non-contact lens wearers (P = 0.26, data not shown). Age, sex, BMI, and body fat were not correlated with measures of tear or blood glucose or insulin levels in this healthy population (all P > 0.06, data not shown).

Discussion

This was the first study to quantify both tear and blood glucose and insulin concentrations in a healthy young adult population before and after an oral glucose tolerance test. Human tear insulin had not previously been quantified after an oral glucose tolerance test. The findings confirmed that systemic glucose levels can regulate glucose and insulin on the ocular surface. Blood glucose and insulin concentrations were also measured, and their changes were consistent with the literature.25

Tear glucose concentration at fasted and after oral glucose ingestion in this study are in accordance to values previously reported for healthy population (0.58 and 2.88 mg/dl, respectively).7,10,11,30 Consistent with blood glucose, tear glucose levels in this study increased by 1.5 h and returned to near baseline levels by 3 h after oral glucose. Aihara et al.31 showed that tear glucose concentrations increased from 0.36 to 1.3–1.4 mg/dl between 30 and 120 min after an oral glucose tolerance test using a high-pressure liquid chromatography method. The higher concentration reported in the current study could be due to differences in race, age, study settings (outpatient vs inpatient population) or different analysis methodologies. The fasted glucose concentration in the current study is consistent with Cha et al. (average of 1.1 mg/dl),11 who reported excellent repeatability using the same point-of-care method and also found that it was comparable with mass spectrometry.11 The main advantage of mass spectrometry is that only 1 μl of tears is required to quantitate glucose7,10,32; however, this method is expensive, time-consuming and has poor batch-to-batch reproducibility.33 The point-of-care tear glucose measure using a modified commercial glucose metre and standard test strips is relatively non-invasive, inexpensive and requires a tear volume of only 0.5–1 μl to determine the glucose level in tears.11,12

There was an association between tear and blood glucose, which was consistent with the changes reported in a study after daily meal intake using enzymatic detection methods10,13 and with oral glucose tolerance test using high-pressure liquid chromatography.31 Based on the blood glucose concentration measured using a standard glucose metre, the concentrations at the fasted state (<95 mg/dl) and after oral glucose tolerance test (<155 mg/dl at 1.5 h and <140 mg/dl at 3 h) were within expected ranges for healthy participants, confirming the recruited participants did not have diabetes. This result further supports that tear glucose concentrations can be a reliable method to monitor blood glucose levels, although there is an expected 15–25 min delay between blood and tear equilibration.9,26 Higher blood and tear glucose concentrations have been reported in diabetic patients compared to healthy subjects in both fasted and non-fasted states.30,31 But assessment of tear glucose concentrations in diabetic subjects using the current point-of-care method is required to confirm its sensitivity and reliability in a disease state.

Tear and blood insulin concentrations also increased significantly at 1.5 h and returned to baseline by 3 h after an oral glucose tolerance test. This finding was consistent with a previous study that measured tear insulin fasted (4.3 μIU/mL) and 1–2 h after a meal (11.5 μIU/mL) in healthy subjects.24 The concentrations reported in the current study were 2–3 times higher, which could be due to the type of collected tears (stimulated vs non-stimulated), full meal intake (vs glucose drink) or quantitation methodologies. The collection of stimulated tears may cause the release of additional insulin from the conjunctival tissue and blood vessels, which requires confirmation in a future study. The increase in blood insulin in this study was 5.8 times higher 1–2 h after oral glucose, which is consistent with a previous study using radioimmunoassay.25

Tear and blood insulin concentrations were not significantly correlated. This is in agreement with findings in a rat model conducted by Cunha et al.20 The findings in this study suggested that increased glucose can enhance the secretion of insulin from the lacrimal gland; thus, after glucose challenge, the level of tear insulin may be slightly higher than blood insulin,20 which is consistent with findings in this study.

In type 2 diabetes, hyperglycaemia occurs when the body is unable to use the insulin secreted for glucose uptake.1 It can also be observed on the eye and the ocular surface.34 For example, increased glucose concentration in the cornea induces the formation of advanced glycation end products (AGEs) which leads to increased corneal thickness, stiffness2–4 and auto-fluorescence.35 It also reduces the activity of Na+/K+ ATPase pump in corneal endothelial cells and decreases transport regulation3 and increases oxidative damage.3 GLUT are responsible for glucose transfer through the cell membrane with and without insulin. GLUT1 induction is greatest when glucose levels are low, and GLUT2 expression is more pronounced when glucose levels are high.36,37 GLUT4 is the glucose carrier residing in the cytoplasmic vesicles and only translocate into the cell membrane with increased insulin for glucose consumption.17 These indicate that their roles in glucose transport are different. In this study, the qPCR data revealed the presence of GLUT1 and GLUT4 in the conjunctival epithelium although GLUT4 expression was quite low. Previous studies reported the expression of GLUT1 on the ocular surface20 but was not able to detect GLUT4 in human eyes.38 This study found low GLUT4 expression, which might imply diminished ability of conjunctival epithelia cells to respond to tear insulin20 and confirms that insulin secretion on the ocular surface may not be required for glucose uptake but enhances glucose metabolism.20

GLUT1, but not GLUT4, expression positively correlated with INSR, which is unexpected given that GLUT1 expression is independent of insulin while GLUT4 is dependent on insulin.15–17 In addition, there was a trend of GLUT1, but not GLUT4, expression and tear glucose concentration. This could be due to the low levels of GLUT4 expression in the conjunctiva. This finding is consistent with the functions of GLUT1 based on the concentration gradient of the glucose.15 Therefore, the findings may indicate that both INSR and GLUT1 expression are mutually regulated to favour glucose homoeostasis via insulin binding to INSR in the conjunctiva, and by facilitated glucose transport across the cell membrane through GLUT1.

This study showed that INSR presents on the ocular surface, further suggesting that endogenous tear insulin may be important in maintaining ocular surface homoeostasis.39 Lower INSR expression (high delta Cq) on the conjunctiva was associated with a higher tear glucose after the oral glucose test. Future study of INSR expression on the ocular surface and associations with tear glucose in diabetes has not been investigated but would add further understanding of the impact of metabolic diseases on the ocular surface.

Topical insulin has been shown to speed corneal wound healing in patients with diabetes.21–23 However, wound healing using topical insulin (Actrapid HM 100 U/mL, Denmark = 1000 U = 40 mg of insulin40 did not show a dose-dependency, and only the 0.5 unit (125 U = 5 mg) dose of topical insulin produced a significant improvement.22,23 Insulin can act as an anti-inflammatory mediator and downregulate the immune response.41 The use of 6 nM (35 mg) of insulin enhances glucose metabolism in a corneal cell culture.20 To date, no studies have quantified tear insulin before or after topical insulin treatment.

There are limitations to this study. Conjunctival impression cytology samples were only collected at the first visit. Even though INSR expression was not expected to change within 2–3 days future studies could assess repeatability. Repeat sample collection may likely need to be extended beyond 2–3 days to allow the conjunctival tissue to fully heal.

Tear glucose and insulin levels may provide direct insight into ocular surface health. It is important to understand the typical endogenous insulin concentration prior to determining the concentration of topical insulin to promote healing. Future studies should aim to quantify baseline and post-treatment levels, as well as the expression of INSR and GLUTs, to better understand the ideal tear insulin levels to promote corneal health and modulate ocular immune responses and the association with ocular surface disease.

Acknowledgements

We would like to thank Prof. Young Bin Choy (Seoul National University, College of Medicine) and Mr. Alexander Schill (UHCO, NIH-NEI P30 EY007551) for helping with the construction of the tear glucometer. Also, we would like to acknowledge Prof. Alan Burns for helpful comments.

Footnotes

Disclosure statement

No potential conflict of interest was reported by the author(s).

References

  • 1.American Diabetes A 2. Classification and diagnosis of diabetes. Diabetes Care. 2017; 40: S11–S24. [DOI] [PubMed] [Google Scholar]
  • 2.Calvo-Maroto AM, Cervino A, Perez-Cambrodi RJ, et al. Quantitative corneal anatomy: evaluation of the effect of diabetes duration on the endothelial cell density and corneal thickness. Ophthalmic Physiol Opt. 2015; 35: 293–298. [DOI] [PubMed] [Google Scholar]
  • 3.McNamara NA, Brand RJ, Polse KA, et al. Corneal function during normal and high serum glucose levels in diabetes. Invest Ophthalmol Vis Sci. 1998; 39: 3–17. [PubMed] [Google Scholar]
  • 4.Bao F, Deng M, Zheng X, et al. Effects of diabetes mellitus on biomechanical properties of the rabbit cornea. Exp Eye Res. 2017; 161: 82–88. [DOI] [PubMed] [Google Scholar]
  • 5.Richdale K, Chao C, Hamilton M. Eye care providers’ emerging roles in early detection of diabetes and management of diabetic changes to the ocular surface: a review. BMJ Open Diabetes Res Care. 2020; 8: e001094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Zhang J, Hodge W, Hutnick C, et al. Noninvasive diagnostic devices for diabetes through measuring tear glucose. J Diabetes Sci Technol. 2011; 5: 166–172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Baca JT, Finegold DN, Asher SA. Tear glucose analysis for the noninvasive detection and monitoring of diabetes mellitus. Ocul Surf. 2007; 5: 280–293. [DOI] [PubMed] [Google Scholar]
  • 8.Dagdeviren M, Altay M, Yildiz Z, et al. The relationship between obesity, insulin resistance, and conjunctival impression cytology. Turk Patoloji Derg. 2019; 35: 119–127. [DOI] [PubMed] [Google Scholar]
  • 9.Lane JD, Krumholz DM, Sack RA, et al. Tear glucose dynamics in diabetes mellitus. Curr Eye Res. 2006; 31: 895–901. [DOI] [PubMed] [Google Scholar]
  • 10.Sen DK, Sarin GS. Tear glucose levels in normal people and in diabetic patients. Br J Ophthalmol. 1980; 64: 693–695. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Cha KH, Jensen GC, Balijepalli AS, et al. Evaluation of commercial glucometer test strips for potential measurement of glucose in tears. Anal Chem. 2014; 86: 1902–1908. [DOI] [PubMed] [Google Scholar]
  • 12.Lee SH, Cho YC, Bin Choy Y. Noninvasive self-diagnostic device for tear collection and glucose measurement. Sci Rep. 2019; 9: 4747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Daum KM, Hill RM. Human tear glucose. Invest Ophthalmol Vis Sci. 1982; 22: 509–514. [PubMed] [Google Scholar]
  • 14.Chen Y, Lippincott-Schwartz J. Selective visualization of GLUT4 storage vesicles and associated rab proteins using IRAP-pHluorin. Methods Mol Biol. 2015; 1298: 173–179. [DOI] [PubMed] [Google Scholar]
  • 15.Yaribeygi H, Farrokhi FR, Butler AE, et al. Insulin resistance: review of the underlying molecular mechanisms. J Cell Physiol. 2019; 234: 8152–8161. [DOI] [PubMed] [Google Scholar]
  • 16.Moraes-Vieira PM, Saghatelian A, Kahn BB. GLUT4 expression in adipocytes regulates de novo lipogenesis and levels of a novel class of lipids with antidiabetic and anti-inflammatory effects. Diabetes. 2016; 65: 1808–1815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Huang S, Czech MP. The GLUT4 glucose transporter. Cell Metab. 2007; 5: 237–252. [DOI] [PubMed] [Google Scholar]
  • 18.Carter DA, Wobken JD, Dixit PK, et al. Immunoreactive insulin in rat salivary glands and its dependence on age and serum insulin levels. Proc Soc Exp Biol Med. 1995; 209: 245–250. [DOI] [PubMed] [Google Scholar]
  • 19.Friend J, Snip RC, Kiorpes TC, et al. Insulin sensitivity and sorbitol production of the normal rabbit corneal epithelium in vitro. Invest Ophthalmol Vis Sci. 1980; 19: 913–919. [PubMed] [Google Scholar]
  • 20.Cunha DA, Carneiro EM, Alves Mde C, et al. Insulin secretion by rat lacrimal glands: effects of systemic and local variables. Am J Physiol Endocrinol Metab. 2005; 289: E768–775. [DOI] [PubMed] [Google Scholar]
  • 21.Bastion ML, Ling KP. Topical insulin for healing of diabetic epithelial defects?: a retrospective review of corneal debridement during vitreoretinal surgery in Malaysian patients. Med J Malaysia. 2013; 68: 208–216. [PubMed] [Google Scholar]
  • 22.Fai S, Ahem A, Mustapha M, et al. Randomized controlled trial of topical insulin for healing corneal epithelial defects induced during vitreoretinal surgery in diabetics. Asia Pac J Ophthalmol (Phila). 2017; 6: 418–424. [DOI] [PubMed] [Google Scholar]
  • 23.Aniah Azmi N, Bastion MC. Short-Term results of trial of topical insulin for treatment of dry eyes in diabetics. Eye Contact Lens. 2020; 46 Suppl :S25–S32. [DOI] [PubMed] [Google Scholar]
  • 24.Rocha EM, Cunha DA, Carneiro EM, et al. Identification of insulin in the tear film and insulin receptor and IGF-1 receptor on the human ocular surface. Invest Ophthalmol Vis Sci. 2002; 43: 963–967. [PubMed] [Google Scholar]
  • 25.Metter EJ, Windham BG, Maggio M, et al. Glucose and insulin measurements from the oral glucose tolerance test and mortality prediction. Diabetes Care. 2008; 31: 1026–1030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Geelhoed-Duijvestijn P, Vegelyte D, Kownacka A, et al. Performance of the prototype noviosense noninvasive biosensor for tear glucose in type 1 diabetes. J Diabetes Sci Technol. 2021; 15: 1320–1325. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Chao C, Stapleton F, Willcox MDP, et al. Preinflammatory signs in established reusable and disposable contact lens wearers. Optom Vis Sci. 2017; 94: 1003–1008. [DOI] [PubMed] [Google Scholar]
  • 28.Chao C, Golebiowski B, Stapleton F, et al. Conjunctival MUC5AC+ goblet cell index: relationship with corneal nerves and dry eye. Graefes Arch Clin Exp Ophthalmol. 2018; 256: 2249–2257. [DOI] [PubMed] [Google Scholar]
  • 29.Alven A, Lema C, Redfern RL. Impact of low humidity on damage-associated molecular patterns at the ocular surface during dry eye disease. Optom Vis Sci. 2021; 98: 1231–1238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Michail D, Vancea P, Zolog N. Lacrimal elimination of glucose in diabetic patients. C R Soc Biol Paris. 1937; 125: 194–195. [Google Scholar]
  • 31.Aihara M, Kubota N, Minami T, et al. Association between tear and blood glucose concentrations: random intercept model adjusted with confounders in tear samples negative for occult blood. J Diabetes Investig. 2021; 12: 266–276. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Taormina CR, Baca JT, Asher SA, et al. Analysis of tear glucose concentration with electrospray ionization mass spectrometry. J Am Soc Mass Spectrom. 2007; 18: 332–336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Wehrens R, Hageman JA, van Eeuwijk F, et al. Improved batch correction in untargeted MS-based metabolomics. Metabolomics. 2016; 12: 88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Ljubimov AV. Diabetic complications in the cornea. Vision Res. 2017; 139: 138–152. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Calvo-Maroto AM, Perez-Cambrodi RJ, Garcia-Lazaro S, et al. Ocular autofluorescence in diabetes mellitus. A review. J Diabetes. 2016; 8: 619–628. [DOI] [PubMed] [Google Scholar]
  • 36.Tal M, Liang Y, Najafi H, et al. Expression and function of GLUT-1 and GLUT-2 glucose transporter isoforms in cells of cultured rat pancreatic islets. J Biol Chem. 1992; 267: 17241–17247. [PubMed] [Google Scholar]
  • 37.Yasuda K, Yamada Y, Inagaki N, et al. Expression of GLUT1 and GLUT2 glucose transporter isoforms in rat islets of langerhans and their regulation by glucose. Diabetes. 1992; 41: 76–81. [DOI] [PubMed] [Google Scholar]
  • 38.Mantych GJ, Hageman GS, Devaskar SU. Characterization of glucose transporter isoforms in the adult and developing human eye. Endocrinology. 1993; 133: 600–607. [DOI] [PubMed] [Google Scholar]
  • 39.Cruz-Cazarim ELC, Cazarim MS, Ogunjimi AT, et al. Prospective insulin-based ophthalmic delivery systems for the treatment of dry eye syndrome and corneal injuries. Eur J Pharm Biopharm. 2019; 140: 1–10. [DOI] [PubMed] [Google Scholar]
  • 40.Staten MA, Stern MP, Miller WG, et al. Insulin assay standardization: leading to measures of insulin sensitivity and secretion for practical clinical care. Diabetes Care. 2010; 33: 205–206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Sun Q, Li J, Gao F. New insights into insulin: the anti-inflammatory effect and its clinical relevance. World J Diabetes. 2014; 5: 89–96. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES