1887

Abstract

Background

Worldwide, cerebrovascular disease (CVD) is the leading disease-related cause of chronic disability; the second most common cause of death and dementia; and a significant burden to patients, caregivers, and healthcare systems. Major risk factors for stroke include diabetes, hypertension, smoking and dyslipidemia which are highly prevalent in the gulf region. Indeed in Qatar, the incidence of cerebrovascular disease is increasing and most patients have major vascular risk factors. In data from the HMC Stroke Registry ∼71% of patients presenting with stroke had diabetes and indeed diabetes was undiagnosed at the time of presentation in 35% of patients. Current imaging techniques to detect early sub-clinical cerebral damage and hence those at risk of stroke includes computerized tomography (CT) and magnetic resonance imaging (MRI). MRI in particular can detect features of small vessel disease such as accumulating silent infarcts, cerebral microbleeds [CMB], periventricular white matter hyperintensity and perivascular spaces which all predict a higher risk of stroke. However, these abnormalities provide a measure of vascular and not neuronal integrity and it is not feasible or practical to undertake MRI simply to identify those at risk of stroke. There is therefore an unmet need for a rapid, non-invasive, sensitive, cost-effective, reproducible and easily imaging biomarker for neuronal damage in subjects at risk of stroke. We have pioneered the technique of corneal confocal microscopy (CCM). Our studies (NIH (5RO1-NS46259-03, R01DK077903-01A1 NINDS), JDRF (17-2008-1031, 8-2008-362)) have shown that corneal confocal microscopy (CCM) can quantify early axonal damage in diabetic neuropathy (1), reliably (2), with high sensitivity and specificity (3). Furthermore, we have shown that subjects with impaired glucose tolerance (IGT) and other vascular risk factors also show corneal nerve loss using CCM (4). Furthermore, these abnormalities improve with an improvement in blood pressure, lipids and HbA1c, all risk factors for stroke (5) and with change in glucose tolerance (6). We have recently published a large normative reference database (7) and developed an automated image analysis algorithm to enable rapid and objective quantification of corneal nerve morphology (8). We therefore hypothesized that CCM may identify corneal nerve loss driven by the common vascular risk factors which may lead to stroke and thereby provide a surrogate for cerebral neuronal loss. Hence, CCM may allow us to identify subjects who may be at high risk of developing stroke and enable risk stratification and targeted risk factor intervention in these individuals.

Aim

We undertook CCM in a cohort of patients admitted to HMC with a clinical and radiological diagnosis of stroke.

Methods

17 stroke patients (age: 53.13 ± 9.09, years) and 15 age-matched healthy control participants (age: 55.30 ± 8.98, years) underwent corneal nerve assessment using CCM (Heidelberg HRT III) to quantify corneal nerve fiber density (CNFD)(no./mm2), corneal nerve fiber branch density (CNFBD)(no./mm2), corneal nerve fiber nerve length (CNFL)(mm/mm2) and corneal nerve fiber tortuosity (TC) and metabolic testing. Stroke patients underwent detailed neurological assessment to define the severity of the stroke using a National Institute of Health Stroke Scale (NIHSS), 3D Carotid Doppler and 3 Tesla MRI. Stroke patients were stratified into different groups for analysis: Normoglycemia (HbA1c <  5.7) (age: 57 ± 8.12, years, n = 5) v dysglycemia (HbA1c ≥  5.7), n = 12); neurological disability: NIHSS <  4 (n = 9) v NIHSS ≥  4 (n = 8), Pathology on MRI: <  10 silent infarcts n = 5 v > 10 silent infarcts (n = 4), patients with periventricular white matter hyperintensity (n = 11) v no periventricular white matter hyperintensity (n = 4).

Results

There was significant difference in HbA1c (8.51 ± 2.35; 5.72 ± 0.36, %, p <  0.001) between stroke patients with dysglycemia and healthy control participants and patients with dysglycemia compared to normoglycemia (8.51 ± 2.35; 5.18 ± 0.35, %, p <  0.009). There was no difference in the total cholesterol (4.96 ± 1.62, 5.34 ± 0.80, mmol/L, p <  0.457), triglycerides (1.32 ± 0.35, 1.64 ± 0.48, mmol/L, p <  0.118), HDL (1.73 ± 1.57, 1.44 ± 0.47, mmol/L, p <  0.527), LDL (2.92 ± 1.42, 3.16 ± 0.69, mmol/L, p <  0.593), hypertension (hypertensive/normal tension, 1/11, 0/15, p <  0.255), height (172.00 ± 4.24, 169.24 ± 9.85, cm, p <  0.708) weight (88.50 ± 13.44, 81.51 ± 13.37, kg, p <  0.500) or BMI (30.00 ± 2.83, 28.47 ± 4.44, Kg/m2, p <  0.649) between stroke patients and control subjects. There was a significant reduction in CNFD (30.47 ± 5.76; 39.50 ± 8.49; p <  0.043) but no change in CNBD (117.50 ± 25.99; 100.54 ± 37.30; p <  0.365), CNFL (29.49 ± 2.65; 27.45 ± 5.56; p <  0.447) or CNFT (19.27 ± 5.07; 15.32 ± 3.94; p <  0.091) in the stroke patients with NGT as compared to the healthy control participants. There was a significant reduction in CNFD (30.17 ± 5.68; 39.06 ± 8.36; p <  0.004) and increase in CNBD (127.71 ± 34.09; 97.80 ± 37.49; p <  0.042) and CNFT (18.57 ± 4.36; 14.93 ± 4.08; p <  0.035) but no change in CNFL (27.06 ± 4.76; 26.97 ± 5.67; p <  0.965) in stroke patients with dysglycemia as compared to the healthy control participants. There was no difference in CNFD (30.17 ± 5.68; 30.47 ± 5.76; p <  0.921), CNBD (127.71 ± 34.09; 117.50 ± 25.99; p <  0.559), CNFT (18.57 ± 4.36; 19.27 ± 5.07; p <  0.777) or CNFL (27.06 ± 4.76; 29.49 ± 2.65; p <  0.306) between stroke patients with and without dysglycemia. There was no significant difference in CNFD (30.36 ± 5.26; 30.14 ± 6.17; p <  0.941), CNBD (123.46 ± 32.58; 126.12 ± 32.37; p <  0.869), CNFT (19.26 ± 5.30; 18.22 ± 3.48; p <  0.645) or CNFL (27.05 ± 4.38; 28.60 ± 4.38; p <  0.479) in stroke patients with NIHSS <  4 compared to NIHSS ≥  4. There was no significant difference in CNFD (30.52 ± 5.27; 32.16 ± 7.73; p <  0.715), CNBD (120.00 ± 39.76; 134.96 ± 43.98; p <  0.609), CNFT (15.71 ± 2.55; 17.61 ± 4.59; p <  0.453) or CNFL (26.54 ± 5.43; 28.43 ± 4.90; p <  0.606) in stroke patients with silent infarcts <  10 vs ≥  10. There was no significant difference in CNFD (30.18 ± 5.15; 30.08 ± 6.00; p <  0.973), CNBD (127.20 ± 28.08; 121.35 ± 45.78; p <  0.766), CNFT (20.36 ± 4.53; 15.15 ± 2.56; p <  0.051) or CNFL (28.91 ± 2.95; 25.50 ± 5.67; p <  0.144) in stroke patients with white matter ischemia compared to stroke patients without white matter ischemia.

Conclusion

Corneal confocal microscopy identifies greater corneal nerve fibre abnormalities in patients admitted with stroke, which is present in patients with and without dysglycemia. It may therefore reflect the consequence of vascular risk factors independent of glycemic status. However, the extent of corneal nerve fibre pathology does not differ in relation to the severity of neurological disability or extent of pathology on MRI. Larger, longitudinal follow up studies are required to determine the prognostic ability and utility of CCM as a surrogate imaging biomarker for stratifying patients at risk of stroke.

References

1. Petropoulos IN, Alam U, Fadavi H, Asghar O, Green P, Ponirakis G, et al. Corneal nerve loss detected with corneal confocal microscopy is symmetrical and related to the severity of diabetic polyneuropathy. Diabetes care. 2013;36(11):3646–51.

2. Petropoulos IN, Manzoor T, Morgan P, Fadavi H, Asghar O, Alam U, et al. Repeatability of in vivo corneal confocal microscopy to quantify corneal nerve morphology. Cornea. 2013;32(5):e83–e9.

3. Petropoulos IN, Alam U, Fadavi H, Marshall A, Asghar O, Dabbah MA, et al. Rapid automated diagnosis of diabetic peripheral neuropathy with in vivo corneal confocal microscopy. Investigative ophthalmology & visual science. 2014;55(4):2071–8.

4. Asghar O, Petropoulos IN, Alam U, Jones W, Jeziorska M, Marshall A, et al. Corneal confocal microscopy detects neuropathy in subjects with impaired glucose tolerance. Diabetes care. 2014;37(9):2643–6.

5. Tavakoli M, Kallinikos P, Iqbal A, Herbert A, Fadavi H, Efron N, et al. Corneal confocal microscopy detects improvement in corneal nerve morphology with an improvement in risk factors for diabetic neuropathy. Diabetic medicine: a journal of the British Diabetic Association. 2011;28(10):1261–7.

6. Azmi S, Ferdousi M, Petropoulos IN, Ponirakis G, Alam U, Fadavi H, et al. Corneal Confocal Microscopy Identifies Small-Fiber Neuropathy in Subjects With Impaired Glucose Tolerance Who Develop Type 2 Diabetes. Diabetes care. 2015;38(8):1502–8.

7. Tavakoli M, Ferdousi M, Petropoulos IN, Morris J, Pritchard N, Zhivov A, et al. Normative values for corneal nerve morphology assessed using corneal confocal microscopy: A multinational normative data set. Diabetes care. 2015;38(5):838–43.

8. Dabbah M, Graham J, Petropoulos I, Tavakoli M, Malik R. Automatic analysis of diabetic peripheral neuropathy using multi-scale quantitative morphology of nerve fibres in corneal confocal microscopy imaging. Medical image analysis. 2011;15(5):738–47.

Loading

Article metrics loading...

/content/papers/10.5339/qfarc.2016.HBPP1589
2016-03-21
2020-09-27
Loading full text...

Full text loading...

http://instance.metastore.ingenta.com/content/papers/10.5339/qfarc.2016.HBPP1589
Loading
This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error