Greater corneal nerve loss at the inferior whorl is related to the presence of diabetic neuropathy and painful diabetic neuropathy

We assessed whether a measure of more distal corneal nerve fibre loss at the inferior whorl(IW) region is better than proximal measures of central corneal nerve damage in relation to the diagnosis of diabetic peripheral neuropathy(DPN), painful DPN and quality of life(QoL). Participants underwent detailed assessment of neuropathy, QoL using the SF36 questionnaire, pain visual analogue score(VAS), and corneal confocal microscopy(CCM). Corneal nerve fibre density (CNFD), branch density (CNBD) and length (CNFL) at the central cornea and inferior whorl length (IWL) and average(ANFL) and total(TNFL) nerve fibre length were compared in patients with and without DPN and between patients with and without painful DPN and in relation to QoL. All CCM parameters were significantly reduced, but IWL was reduced ~three-fold greater than CNFL in patients with and without DPN compared to controls. IWL(p = 0.001), ANFL(p = 0.01) and TNFL(p = 0.02) were significantly lower in patients with painful compared to painless DPN. The VAS score correlated with IWL(r = −0.36, P = 0.004), ANFL(r = −0.32, P = 0.01) and TNFL(r = −0.32, P = 0.01) and QoL correlated with CNFL(r = 0.35, P = 0.01) and IWL(r = 0.4, P = 0.004). Corneal nerve fibre damage is more prominent at the IW, lower in patients with painful compared to painless neuropathy and relates to their QoL. IWL may provide additional clinical utility for CCM in patients with DPN.

combination of corneal nerve fibre length at the centre and inferior whorl may increase the sensitivity of CCM in detecting DPN 11 .
Painful neuropathy affects approximately 21% of patients with diabetes 17,18 and can significantly reduce the quality of life 19,20 . A small study has demonstrated a greater reduction in central corneal nerve fibre length in patients with painful diabetic neuropathy 3 . In the present study, we have compared corneal nerve pathology in the central and inferior whorl regions in relation to the severity and presence of DPN and painful DPN and quality of life.

Methodology
Study subjects. 116 patients with type 1 or type 2 diabetes and 22 age-matched healthy controls underwent assessment of peripheral neuropathy and corneal confocal microscopy (CCM). Patients were excluded if they had a history of malignancy, neuropathy of non-diabetic cause, current or active diabetic foot ulceration, deficiency of B12 or folate, chronic renal impairment or liver failure, connective tissue or systemic disease, corneal trauma, systemic disease that involves the cornea and cystic corneal disorders. Before participation informed consent was obtained from each participant. This research adhered to the tenets of Declaration of Helsinki and was approved by Greater Manchester east Research Ethics Committee.
Clinical and peripheral neuropathy assessment. All participants underwent assessment of body mass index (BMI) and blood pressure. Neurological deficits were assessed using the Neuropathy Disability Score (NDS), which includes an assessment of vibration perception, pinprick, temperature sensation and presence or absence of ankle reflexes 21 . Based on the NDS score patients were divided into two groups: without neuropathy (DN−) (NDS < 3) (n = 47) and with neuropathy (DN+) (NDS ≥ 3) (n = 69). VPT was assessed using a Horwell Neurothesiometer (Scientific Laboratory Supplies, Wilfrod, Nottingham, UK). Cold (CT) and warm (WT) perception thresholds, were assessed on the dorsolateral aspect of the non-dominant foot (S1) using a TSA-II NeuroSensory Analyser (Medoc, Ltd., Ramat-Yishai, Israel). Electrodiagnostic studies were undertaken using a Dantec ''Keypoint" system (Dantec Dynamics Ltd, Bristol, UK), equipped with a DISA temperature regulator to keep the limb temperature constantly at 32-35 °C. Sural sensory nerve amplitude (SSNA), sural sensory nerve conduction velocity (SSNCV), peroneal motor nerve amplitude (PMNA), and peroneal motor nerve conduction velocity (PMNCV) were assessed by a consultant neurophysiologist.
Quality of life was assessed using The Short Form 36 Health Survey (SF-36) questionnaire [15]. This survey consists of 36 questions, measures eight different dimensions, which are scored from 0 to 100. We used SF-36 total score, an average of eight dimensions as a single measure of quality of life.
A diagnosis of painful diabetic neuropathy was based on the presence of typical symmetrical neuropathic symptoms such as burning, aching pain in the feet and a visual analogue scale (VAS) > 4 mm in patients with abnormal distal sensation or decreased/absent distal reflexes (NDS ≥ 3) 22 .
Corneal Confocal Microscopy. All participants underwent CCM examination using a laser scanning corneal confocal microscope HRT III (Heidelberg Retinal Tomograph III Rostock Cornea Module, Heidelberg Engineering, Heidelberg, Germany) for both eyes using our established protocol for the central and inferior whorl area of the cornea 11,23 . Six images from the central cornea and four images from inferior whorl at the level of sub-basal nerve plexus were selected and manually quantified using CCMetrics (The University of Manchester, Manchester, UK). Images were selected based on their quality and variability using our established protocol 11,24 . Four corneal parameters were quantified: corneal nerve fibre density (CNFD -total number of main nerves per square millimetre (no./mm 2 ), corneal nerve branch density (CNBD -number of nerve branches per square millimetre), corneal nerve fibre length (CNFL -total length of main nerves and nerve branches per square millimetre) (mm/mm 2 ), inferior whorl length (IWL -total length of nerves per square millimetre at the IW region) (mm/mm 2 ), average nerve fibre length (ANFL) − (CNFL + IWL/2 (mm/mm 2 )) and total nerve fibre length (TNFL) − (CNFL + IWL (mm/mm 2 )).

Statistical analysis.
Analysis was carried out using SPSS (Version 22.0 for Macintosh, IBM Corporation, New York, NY, USA). Independent T-test (Mann Witney U test for nonparametric variables) was used to assess the estimates between two groups. The One-Way Anova (post hoc LSD) was used to compare means among groups. The analysis of covariance (Ancova) (post hoc LSD) was used to compare variables between groups, while statistically controlling for the effects of age. Pearson's correlation coefficient (Spearman's for non-parametric) was calculated to assess the associations among different variables. All the data are expressed as mean ± standard error (SE). P < 0.05 was considered as significant. Graphpad Prism (Version 7.0c for Macintosh, Graphpad Software, La Jolla California, USA) was used for plotting the graphs. Data availability. The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Discussion
CCM is a rapid, reproducible ophthalmic technique that can quantify corneal nerve fibre abnormalities in diabetic and a number of other peripheral neuropathies 3,25-28 . Two recent studies have shown that CCM has a diagnostic ability, which is comparable to the current gold standard technique of skin biopsy 9,29 . We confirm that CCM can detect sub-clinical corneal nerve damage in patients without evidence of DPN 11 .
DPN is length-dependent neuropathy and will therefore affect the most distal nerve fibres first 30 i.e. the nerves at the inferior whorl which are more distal to the nerves in the central cornea. Corneal nerve loss has been reported to be comparable at the inferior whorl and central cornea in diabetic patients with and without neuropathy 11,15,16 . However, in the present study, we demonstrate a ~2-3-fold greater reduction in corneal nerve fibre length at the IW  able 3. CCM parameters in controls, patients with painless and painful neuropathy. £ Represents values adjusted for age using ANCOVA; *P < 0.05 compared to controls, $ P < 0.05 compared to painful neuropathy; ^P < 0.01 compared to controls; II P < 0.01 compared to painful neuropathy.
SCIENTIFIC REPORtS | (2018) 8:3283 | DOI:10.1038/s41598-018-21643-z compared to the central cornea in diabetic patients with and without DPN compared to controls. We also show that whilst 30% of patients with a reduction in inferior whorl length had a normal corneal nerve fibre length, only 13.5% of patients with a normal inferior whorl length had an abnormal corneal nerve fibre length; confirming more prominent distal corneal nerve loss. Given this difference we postulated that a combination of both corneal nerve fibre length and inferior whorl length, either as a total or average length may be more useful than either measure alone. Indeed, inferior whorl length, average and total nerve fibre length rather than the standard parameters of corneal nerve fibre density, branch density and length correlate with the quality of life and the presence and severity of painful diabetic neuropathy. Given that new therapies for diabetic neuropathy should induce nerve repair in the most distal part of the nerve, we propose that quantification of inferior whorl length and the average and total nerve fibre length may provide a more powerful means to assess a therapeutic response than has already been demonstrated by quantifying central corneal nerve morphology [31][32][33] . An additional advantage is the unique pattern of the IW, which facilitates its use as a landmark in longitudinal and interventional studies.
To our knowledge the association between CCM parameters and QoL and presence and severity of neuropathic pain has not been previously reported. QoL is an important patient outcome and we used the validated SF-36 questionnaire 34 to assess the QoL in patients in this study. We show a more significant correlation between QoL and IWL compared to corneal nerve fibre length. Furthermore, inferior whorl length, average and total nerve fibre length differentiate patients with and without painful neuropathy and correlate with the severity of painful neuropathy, whereas corneal nerve fibre density, branch density and length, do not.
A potential limitation of our study is that the diagnosis of painful diabetic neuropathy was principally based on the VAS and NDS scores, but this is in accord with the IASP definition for neuropathic pain as adopted by a recent large painful neuropathy phenotyping study 20 . We believe future studies should better stratify cases in relation to the severity and sub-types of diabetic neuropathy.
In conclusion we have shown that there is more prominent distal corneal nerve fibre damage at the inferior whorl in patients with diabetic neuropathy. It also provides an objective measure of small nerve fibre damage, which relates to the presence of painful diabetic neuropathy and the quality of life. Further studies deploying these novel measures of more distal corneal nerve damage and repair are required in longitudinal studies and in clinical trials of new therapies to define their clinical utility.