Diabetes is one of the most common systemic diseases afflicting Australians, placing them at risk of premature mortality, macrovascular complications such as cardiovascular disease, and microvascular complications including nephropathy leading to kidney failure, potentially blinding diabetic retinopathy, and diabetic peripheral neuropathy.1 The American Diabetes Association estimates that 25.8 million children and adults, approximately 8.3% of the US population, have the condition,2 placing financial strain on the health system.3 Complications resulting from diabetes impose a significant burden on patients, their families, and their communities because of higher utilization of health services.3 A fundamental goal is earlier detection or prevention of complications such as neuropathy because this not only improves the health of the individual but also brings significant economic benefits to the broader community.
Diabetic peripheral neuropathy is estimated to affect 60–70% of people with diabetes2 and has been defined by the Toronto Diabetic Neuropathy Expert Group as a chronic, symmetrical, length-dependent diabetic sensorimotor polyneuropathy.4 Painful neuropathy can have a devastating effect on the quality of life of a patient,4 affecting 25% of people with diabetes.5 Importantly, foot ulceration occurs in 7% of patients with diabetic neuropathy compared with 1% of individuals with diabetes but without neuropathy,6 with lower limb amputation required in advanced cases.7 In addition to the quality-of-life costs associated with diabetic peripheral neuropathy, the financial burden is also significant. Ramsey and colleagues found that attributable costs for a middle-aged diabetic male to be approximately US $28,000 two years after a new foot ulcer.8
There is currently no Food and Drug Administration–approved treatment that is capable of preventing or reversing diabetic neuropathy. Attention is therefore targeted at its detection in order to identify those at risk. Examination of nerve morphology via skin punch biopsy9 allows for objective and sensitive assessment of small nerve fiber damage. This is, however, invasive and non-repeatable.10 The current gold standard for diagnosis of diabetic peripheral neuropathy is nerve conduction using biothesiometry,11 which measures large nerve fiber function. The lack of a sensitive, non-invasive, and repeatable endpoint to measure changes in small nerve fibers is a major factor holding back clinical trials for the treatment of diabetic peripheral neuropathy.12 Unlike nerves elsewhere in the body, corneal nerves can be imaged in vivo and non-invasively with in vivo confocal microscopy.13 Recent studies have shown that morphological changes in the corneal sub-basal nerve plexus correlate with changes in the peripheral nerves and may therefore be a good surrogate measure for diabetic peripheral neuropathy.6,14,15 Significantly, a 2015 study reported that in vivo confocal microscopy of corneal nerve fiber density (CNFD) can predict the onset of diabetic peripheral neuropathy in individuals with type 1 diabetes.16
Both the cornea and the tear film are known to be affected in diabetes.17,18 Substance P is a neuropeptide present in the tear film, which is released from trigeminal sensory nerve endings in the cornea, conjunctiva, and lacrimal gland into the tear film.19 Substance P has a role in wound healing, but also maintains corneal integrity as it promotes migration, proliferation, and differentiation of epithelial cells. In the trigeminal nerve, substance P has been found to be decreased in concentration in diabetes relative to healthy controls.20 Although altered corneal nerve structure in people with diabetes has been reported,10 the contribution of neuropeptides such as substance P in the tear film relative to diabetic corneal neuropathy has not been explored. This study therefore aims to investigate the impact of diabetes on the concentrations of substance P in the tear film relative to CNFD.
MATERIALS AND METHODS
This was a cross-sectional, single-visit case–control study approved by the Institutional Ethics Committee of the University of New South Wales (Approval number: HC15255). The tenets of the Declaration of Helsinki were followed. Informed consent was obtained from nine participants with diabetes (one female, eight male, mean 63.5 ± 14.5 years) with all but one participant having type II diabetes. Informed consent was also obtained from 17 participants without diabetes (9 female, 8 male, mean 49.2 ± 20.1 years). The sample size was based on a statistically significant difference at 95% confidence and with 80% power, of 136 ± 75 pg/mL of substance P.21 Participants were recruited through the general university population and through approved email notices from Diabetes New South Wales website, the Centre for Eye Health, and local newspapers. Inclusion into the study required participants to be over 18 years of age, have an adequate understanding of the English language, no allergies to eye drops, eye injuries, recent or current infection, and not be taking any corticosteroids, doxycycline, or prostaglandin analogues. Additionally, a requirement of the study was that the participants were either non-contact lens wearers or non-habitual lens wearers who had not worn contact lenses in the month preceding the study. Pregnant or lactating women were excluded from the study.
Participants were asked to complete a questionnaire on their age, ethnicity, type of diabetes, control of diabetes, most recent HbA1c, blood glucose levels, and comorbidities. Visual acuity using computerized letter charts was measured and slit-lamp biomicroscopy was performed to exclude any pre-existing conditions. Bulbar and limbal redness, palpebral redness and roughness, and corneal staining were assessed using the CCLRU Grading scales.22 The tests were ordered so as to minimize the impact on the ocular surface before tear collection. Visual acuity and evaluation of the cornea and conjunctiva without fluorescein were therefore conducted before tear collection whereas fluorescein evaluation and lid eversion were conducted after tear collection.
In Vivo Confocal Microscopy
Laser scanning in vivo confocal microscopy was conducted to determine CNFD at the central corneal location. The Heidelberg Retinal Tomograph II with the Rostock Corneal Module (Heidelberg Engineering GmbH, Heidelberg, Germany) was used and set up according to standard technique, as previously described.23 Participants were instructed to fixate on a meandering white spot on an LCD screen placed 12.5 cm perpendicular to their contralateral eye, while the confocal microscope captured 100 high-quality contiguous images of the central corneal sub-basal nerve plexus, as previously described.24
Corneal Nerve Analysis
The scanned images for each subject were arranged into a montage, where eight high-contrast images that did not overlap by more than 20% were selected to represent the central corneal sub-basal nerve plexus, as suggested by Vagenas et al.25 These images were then analyzed using a fully automated nerve fiber image analysis program (ACCMetrics; M.A. Dabbah, Imaging Science and Biomedical Engineering, Manchester, UK)26 to determine the CNFD (mm/mm2) for each participant.
Flush tears as described previously27 were collected from both eyes using a 10 μL microcapillary tube (Blaubrand intraMark, Wertheim, Germany). Right and left eyes were pooled. Flush tears were collected by instilling a 60 μL drop of non-preserved, unit dose saline (sodium chloride injection 0.9%; AstraZeneca, Clifford Hallam Pharmaceuticals North Ryde, Australia) into the inferior palpebral fold using an Eppendorf pipette. The eyes were then closed gently. Participants tilted their head toward the side of collection and the microcapillary tube rested in the lateral tear meniscus to collect tears. A limit of 1 minute per eye was imposed to avoid reflex tearing.28 Tear collection rate was monitored for all samples by noting the time and volume collected. The sample was expelled from the capillary tube into a siliconized polypropylene microcentrifuge tube of 0.65 mL capacity (Sigma-Aldrich, Steinheim, Germany) and placed on ice until processing.
Samples were centrifuged at 1145g force for 20 minutes at 4 °C to remove cellular debris. The supernatants were collected and stored in siliconized polypropylene microcentrifuge tubes (Sigma-Aldrich, Steinheim, Germany) at −80 °C in four aliquots, one for each analysis.
Total Protein Content Analysis
Total protein content was determined using the FluoroProfile Protein Quantification Kit (Sigma-Aldrich, St Louis, USA) and Nunc flat-bottom 96-well microplates (Thermo Fisher Scientific, Rochester, NY) as described previously.29 Briefly, a standard curve was constructed using serial dilutions of bovine serum albumin. Tears were added to MilliQ water in a 1:400 dilution. A 50 μL volume of both the standard and the samples were added to each well in triplicates. To this 50 μL of the FluoroProfile Fluorescent reagent was added. The plate was incubated in the dark for 30 minutes and then imaged using a Tecan Safire2 microplate reader (Tecan, Switzerland) at 510 nm excitation wavelength and emission wavelength of 620 nm. The total protein content of each sample was calculated from the standard curve.
Substance P Tear Analysis
Substance P concentration was determined using a competitive enzyme-linked immunosorbent assay with the Abcam human substance P kit (Melbourne, Australia) following the instructions of the manufacturer. The plate was pre-coated with a goat anti-rabbit IgG antibody. The samples were added in duplicate in a 1:20 dilution and standards added according to the instructions of the manufacturer along with an alkaline phosphate conjugated substance P antigen and a polyclonal rabbit antibody specific to substance P. Standard concentration ranged from 10,000 pg/mL to 9.76 pg/mL. The plate was incubated for 1 hour at 37 °C. Excess reagents were washed three times with wash buffer and substrate was added. After a 30-minute incubation at 37 °C, stop solution was added and the optical density of each well and was read at 405 nm with correction at 580 nm. Where the reading was below the detection range, zero was allocated. A standard curve was generated from the standard wells, with the average absorbance of the blank wells subtracted from each sample well to determine the final absorbance.
Data was analyzed using PASW version 18.0 GP (Chicago, USA). Normality was tested using the Shapiro-Wilk test and the data for substance P and total protein content were not normally distributed. A two-tailed independent t-test was administered to analyze CNFD and substance P between the two study groups. Pearson’s correlation and a two-tailed 0.05 level of significance calculation were administered to determine the correlation between groups. Results are reported as mean ± standard deviation.
Table 1 illustrates the demographics of the participants in this study. There was a statistically significant difference between participants with diabetes and control participants for age (P = .015).
Corneal Nerve Fiber Density
There was a significant difference in the CNFD between the participants with diabetes and control subjects (16.1 ± 5.7 vs. 21.5 ± 7.0 mm/mm2, P = .04, Table 2).
Tear Collection Flow Rate
Flow rate was 7.3 ± 2.2 μL/min in participants with diabetes and 9.1 ± 4.0 μL/min in controls. This was not significantly different between groups (P = .16).
Total Protein Content
Table 2 lists the concentrations of the proteins measured. There was no significant difference in total protein content between the two study groups (P = .26).
The concentration of substance P was significantly greater in the tear film of the control group compared to the diabetes group (P = .047, Table 2 and Fig. 1).
Table 3 illustrates the associations between tear film variables and CNFD. There was a statistically significant, moderate correlation between substance P and CFND (r = 0.48, P = .01, Fig. 2). There was no statistical correlation between total protein content and CFND (r = 0.24, P = .25). There was a significant correlation between age and both CFND (r = −0.46, P = .02) and substance P (r = −0.62, P = .001).
Previous studies have found CNFD to be significantly reduced in people with diabetes compared to healthy controls,15,30–33 a finding supported by our data. Corneal nerves are responsible for the production of neuropeptides, including substance P, insulin-like growth factor-1 (IGF-1), calcitonin gene-related peptide (CGRP), neuropeptide Y, and vasoactive-intestinal peptide (VIP).34 Given the reduction in corneal nerve density, it would be anticipated that these proteins are also reduced. This hypothesis is supported by our data which indicate a reduction in tear film substance P in the diabetes group compared to the control group.
In isolation and in combination with other constituents, substance P plays a role in the maintenance and nutrition of the cornea, by promoting migration and proliferation of corneal epithelial cells.35 This reduction in substance P levels may contribute to the poorer wound healing and increased susceptibility to corneal neurotrophic ulcers that is seen in diabetes.36,37 With reduction in substance P, corneal epithelial migration may decrease and neurotrophic keratopathy may develop, resulting in further corneal complications.38
In support of our findings, substance P levels in the central and peripheral nervous system have been found to be significantly decreased in type I diabetes, particularly in those with peripheral neuropathy.39 However, the literature is conflicted with regard to levels of tear film substance P in diabetes. A study by Yamada et al.38 found the concentration of tear film substance P to be reduced in diabetes. In agreement with this, the same group found that eyes with corneal hypoesthesia have decreased concentrations of substance P compared to healthy control eyes.21 The data shown in Fig. 1 demonstrate a large standard deviation for the concentrations of substance P. We have previously reported on tear film substance P in healthy controls using the same tear collection technique, showing similar variability in the data.40 The larger standard deviations found in the current study compared to Yamada et al.21 may be a result of the different tear collection technique implemented, where Yamada et al. collected basal tears whereas the current study collected flush tears. Marfurt and Echtenkamp measured the level of substance P and other neuropeptides, namely CGRP and VIP in the corneas and irides of diabetic mice.41 Although no significant difference was found in the corneas of the diabetic and control mice, a significant difference was found in the iridal levels of CGRP and VIP.41 These authors argue that the absence of changes in corneal neuropeptides suggests that trophic peptides in corneal sensory nerves are not related to the keratopathy seen in diabetics.41 Clearly, more needs to be done to understand the change in tear film neuropeptides in diabetes and what impact this has on corneal wound healing.
The difference between groups in tear film substance P indicates that this may be a therapeutic target in the case of persistent ulcers. Topical treatment with substance P derivatives and insulin growth factor-1 has been shown to have good efficacy in healing epithelial defects and restoring ocular surface integrity in neurotrophic keratopathy, including diabetic keratopathy.42,43 In contrast, eye drops such as diclofenac have been found to reduce substance P concentrations in tears.44 Diclofenac is a non-steroidal anti-inflammatory drug that acts to inhibit the COX-2 pathway of the inflammatory process, ultimately blocking prostaglandin release.45 This has implications in patients with diabetic corneal neuropathy and therefore hypoesthesia. With reduction in substance P as a result of diclofenac intake, corneal epithelial migration would decrease further and neurotrophic keratopathy may develop, resulting in further corneal complications.38
A limitation of our study was that diagnosis of diabetes, its duration, and HbA1c measurements were reported by the study participant, which may have been inaccurate. Additionally, there was a significant age difference between groups, with the diabetic group being significantly older. Although age has been shown to reduce CFND,46–48 the effect of age on tear film substance P has not been explored. Dehgani et al. have reported a linear decrease of 0.05 mm/mm2 in corneal nerve fiber length with increasing age.46 The average age difference between groups in this study was 14.3 years (Table 1), and it would be anticipated that based on a yearly decrease of 0.05 mm/mm2, there would be a difference of 0.7 mm/mm2 between groups based on age alone. As the difference between groups (Table 2 and Fig. 1) exceeded this, it is unlikely that age significantly confounded the CNFD data. However, given the association shown between CNFD and tear film substance P (Fig. 2), any age-related effect on CNFD may have also contributed to the reduction in substance P shown here. We were additionally limited by the small sample size in this feasibility study. Future work should also look at expanding the sample size and including the assessment of neuropathic status to further explore the relationship between tear film substance P and peripheral neuropathy.
In conclusion, this study has shown that there is a substantial difference in the concentration of substance P in the tear film between healthy controls and those with diabetes. The positive correlation between substance P and corneal nerve density indicates that substance P may be a potential biomarker for corneal nerve health.
School of Optometry and Vision Science
Level 3, North Wing, Rupert Myers Building
Gate 14, Barker Street
The University of New South Wales
Sydney NSW 2052
The authors would like to thank Diabetes New South Wales and the Centre for Eye Health for their assistance with the study recruitment.
This work was supported by UNSW start-up funds (MM). JK was supported by an Australian Research Council (ARC) Future Fellowship.
There are no conflicts of interest for any of the authors that could have influenced the results of this work. This work has been presented in part at the Association for Research in Vision and Ophthalmology Conference in Seattle, WA, on May 5, 2016.
Received January 22, 2017; accepted May 5, 2017.
1. Cade WT. Diabetes
-related microvascular and macrovascular diseases in the physical therapy setting. Phys Ther 2008;88:1322–35.
2. Charnogursky G, Lee H, Lopez N. Diabetic neuropathy. Handb Clin Neurol 2014;120:773–85.
3. Magliano DJ, Shaw JE, Shortreed SM, et al. Lifetime risk and projected population prevalence of diabetes
. Diabetologia 2008;51:2179–86.
4. Brod M, Pohlman B, Blum SI, et al. Burden of illness of diabetic peripheral neuropathic pain: a qualitative study. Patient 2015;8:339–48.
5. Snyder MJ, Gibbs LM, Lindsay TJ. Treating painful diabetic peripheral neuropathy: an update. Am Fam Physician 2016;94:227–34.
6. Pritchard N, Edwards K, Shahidi AM, et al. Corneal markers of diabetic neuropathy. Ocul Surf 2011;9:17–28.
7. Davies M, Brophy S, Williams R, et al. The prevalence, severity, and impact of painful diabetic peripheral neuropathy in type 2 diabetes
8. Ramsey SD, Newton K, Blough D, et al. Incidence, outcomes, and cost of foot ulcers in patients with diabetes
9. Quattrini C, Tavakoli M, Jeziorska M, et al. Surrogate markers of small fiber damage in human diabetic neuropathy. Diabetes
10. Edwards K, Pritchard N, Vagenas D, et al. Utility of corneal confocal microscopy for assessing mild diabetic neuropathy: baseline findings of the landmark study. Clin Exp Optom 2012;95:348–54.
11. Albers JW, Brown MB, Sima AA, et al. Nerve conduction measures in mild diabetic neuropathy in the Early Diabetes
Intervention Trial: the effects of age, sex, type of diabetes
, disease duration, and anthropometric factors. Tolrestat Study Group for the Early Diabetes
Intervention Trial. Neurology 1996;46:85–91.
12. Tavakoli M, Quattrini C, Abbott C, et al. Corneal confocal microscopy: a novel noninvasive test to diagnose and stratify the severity of human diabetic neuropathy. Diabetes
13. Oliveira-Soto L, Efron N. Morphology of corneal nerves using confocal microscopy. Cornea 2001;20:374–84.
14. Mehra S, Tavakoli M, Kallinikos PA, et al. Corneal confocal microscopy detects early nerve regeneration after pancreas transplantation in patients with type 1 diabetes
15. Dehghani C, Pritchard N, Edwards K, et al. Natural history of corneal nerve morphology in mild neuropathy associated with type 1 diabetes
: development of a potential measure of diabetic peripheral neuropathy. Invest Ophthalmol Vis Sci 2014;55:7982–90.
16. Pritchard N, Edwards K, Russell AW, et al. Corneal confocal microscopy predicts 4-year incident peripheral neuropathy in type 1 diabetes
17. Misra SL, Braatvedt GD, Patel DV. Impact of diabetes
mellitus on the ocular surface: a review. Clin Exp Ophthalmol 2016;44:278–88.
18. Misra SL, Patel DV, McGhee CN, et al. Peripheral neuropathy and tear film
dysfunction in type 1 diabetes
mellitus. J Diabetes
19. Davidson HJ, Kuonen VJ. The tear film
and ocular mucins. Vet Ophthalmol 2004;7:71–7.
20. Troger J, Humpel C, Kremser B, et al. The effect of streptozotocin-induced diabetes
mellitus on substance P
and calcitonin gene-related peptide expression in the rat trigeminal ganglion. Brain Res 1999;842:84–91.
21. Yamada M, Ogata M, Kawai M, et al. Decreased substance P
concentrations in tears from patients with corneal hypesthesia. Am J Ophthalmol 2000;129:671–2.
22. Cornea and Contact Lens Research Unit. CCLRU Grading Scales University of New South Wales, Sydney; 1996.
23. Zhivov A, Stave J, Vollmar B, et al. In vivo confocal microscopic evaluation of Langerhans cell density and distribution in the normal human corneal epithelium. Graefes Arch Clin Exp Ophthalmol 2005;243:1056–61.
24. Lum E, Golebiowski B, Swarbrick HA. Mapping the corneal sub-basal nerve plexus in orthokeratology lens wear using in vivo laser scanning confocal microscopy. Invest Ophthalmol Vis Sci 2012;53:1803–9.
25. Vagenas D, Pritchard N, Edwards K, et al. Optimal image sample size for corneal nerve morphometry. Optom Vis Sci 2012;89:812–7.
26. Dabbah MA, Graham J, Petropoulos IN, et al. Automatic analysis of diabetic peripheral neuropathy using multi-scale quantitative morphology of nerve fibres in corneal confocal microscopy imaging. Med Image Anal 2011;15:738–47.
27. Markoulli M, Papas E, Petznick A, et al. Validation of the flush method as an alternative to basal or reflex tear collection. Curr Eye Res 2011;36:198–207.
28. Sack RA, Tan KO, Tan A. Diurnal tear cycle: evidence for a nocturnal inflammatory constitutive tear fluid. Invest Ophthalmol Vis Sci 1992;33:626–40.
29. You JJ, Hodge C, Wen L, et al. Tear levels of SFRP1 are significantly reduced in keratoconus patients. Mol Vis 2013;19:509–xxx.
30. Ziegler D, Papanas N, Zhivov A, et al. Early detection of nerve fiber loss by corneal confocal microscopy and skin biopsy in recently diagnosed type 2 diabetes
31. Zhivov A, Winter K, Hovakimyan M, et al. Imaging and quantification of subbasal nerve plexus in healthy volunteers and diabetic patients with or without retinopathy. PLoS One 2013;8:–.
32. Chen X, Graham J, Dabbah MA, et al. Small nerve fiber quantification in the diagnosis of diabetic sensorimotor polyneuropathy: comparing corneal confocal microscopy with intraepidermal nerve fiber density. Diabetes
33. Petropoulos IN, Alam U, Fadavi H, et al. Corneal nerve loss detected with corneal confocal microscopy is symmetrical and related to the severity of diabetic polyneuropathy. Diabetes
34. Lambiase A, Micera A, Sacchetti M, et al. Alterations of tear neuromediators in dry eye disease. Arch Ophthalmol 2011;129:981–6.
35. Nishida T, Nakamura M, Ofuji K, et al. Synergistic effects of substance P
with insulin-like growth factor-1 on epithelial migration of the cornea. J Cell Physiol 1996;169:159–66.
36. Oyibo SO, Jude EB, Tarawneh I, et al. The effects of ulcer size and site, patient’s age, sex and type and duration of diabetes
on the outcome of diabetic foot ulcers. Diabet Med 2001;18:133–8.
37. Hyndiuk RA, Kazarian EL, Schultz RO, et al. Neurotrophic corneal ulcers in diabetes
mellitus. Arch Ophthalmol 1977;95:2193–6.
38. Yamada M, Ogata M, Kawai M, et al. Substance P
in human tears. Cornea 2003;22:S48–54.
39. Kunt T, Forst T, Schmidt S, et al. Serum levels of substance P
are decreased in patients with type 1 diabetes
. Exp Clin Endocrinol Diabetes
40. Markoulli M, Gokhale M, You J. Substance P
in flush tears and Schirmer strips of healthy participants. Optom Vis Sci 2017;94:527–33.
41. Marfurt CF, Echtenkamp SF. The effect of diabetes
on neuropeptide content in the rat cornea and iris. Invest Ophthalmol Vis Sci 1995;36:1100–6.
42. Shaheen BS, Bakir M, Jain S. Corneal nerves in health and disease. Surv Ophthalmol 2014;59:263–85.
43. Abdelkader H, Patel DV, McGhee C, et al. New therapeutic approaches in the treatment of diabetic keratopathy: a review. Clin Exp Ophthalmol 2011;39:259–70.
44. Yamada M, Ogata M, Kawai M, et al. Topical diclofenac sodium decreases the substance P
content of tears. Arch Ophthalmol 2002;120:51–4.
45. Cashman JN. The mechanisms of action of NSAIDs in analgesia. Drugs 1996;52(Suppl. 5):13–23.
46. Dehghani C, Pritchard N, Edwards K, et al. Morphometric stability of the corneal subbasal nerve plexus in healthy individuals: a 3-year longitudinal study using corneal confocal microscopy. Invest Ophthalmol Vis Sci 2014;55:3195–9.
47. Niederer RL, Perumal D, Sherwin T, et al. Age-related differences in the normal human cornea: a laser scanning in vivo confocal microscopy study. Br J Ophthalmol 2007;91:1165–9.
48. Parissi M, Karanis G, Randjelovic S, et al. Standardized baseline human corneal subbasal nerve density for clinical investigations with laser-scanning in vivo confocal microscopy. Invest Ophthalmol Vis Sci 2013;54:7091–102.