Introduction

Nail fold capillaroscopy (NFC) is a simple, non-invasive, painless in vivo technique to assess microvasculature. It is an established method to assess microcirculation in connective tissue diseases (CTDs) and helps in the early diagnosis and monitoring of Systemic sclerosis (SSc). The proximal nail fold, being an acral area, is affected early by disorders that affect microvasculature and has the advantage of easy accessibility for repeated examination.1 Diabetic retinopathy (DR) is the most common microvascular complication of diabetes and remains a major cause of preventable blindness.2 As compared to studies evaluating the retinal microvascular changes in diabetics, very little is known regarding the NFC features of these patients.3 The objectives of the present study were to assess the nail fold capillaries in diabetic retinopathy to evaluate the micro-vascular involvement, to identify if there is any correlation between nail fold capillaroscopic findings and the severity of DR and determine whether NFC changes have a relationship with the duration of diabetes and glycaemic index.

Materials and Methods

This cross-sectional observational study enrolled 200 patients that included 100 cases of DR and 100 healthy age and sex-matched controls and was conducted in the Department of Dermatology and Ophthalmology of a tertiary care institute over 1.5 years. Sample size calculation was done using convenience sampling. The Institutional Ethical Committee approved the study protocol (IEC No-3385/D-26/2020 Batch). Adults (>18 years) diagnosed with T2DM, based on the American Diabetes Association criteria (ADA)4 and DR as per Diabetic Retinopathy Disease Severity Scale5 and consenting to the study protocol, were included. Patients who did not give consent, those with type 1 diabetes mellitus, nail unit infections, trauma, patients with known CTD or on drugs affecting peripheral circulation, patients with aesthetic treatments in the last two weeks before the study, and pregnant and lactating females were excluded. Informed written consent was taken from all the patients, before their inclusion in the study [Figure 1].

Figure 1: The protocol of the study conducted.

Export to PPT

A detailed NFC examination was done for all 10 fingernails, using a dermatoscope (AM 7515MZT Dino-Lite Edge Dermatoscope with 20x–220x magnification). The patients were made to sit for 15–20 min at an ambient temperature (around 20–22°C), with their hands at the level of the heart. Before the procedure nail plate and folds were cleaned with spirit and ultrasound gel was applied to the PNF to improve capillary visibility. NFC was performed, first at a low magnification for global visualisation of capillaries in the entire PNF and then at a higher magnification to assess individual capillary architectural change. Later the NFC changes were evaluated by the supervisor and recorded.

Qualitative morphological alterations in the capillary loops were visualised in all the fingers. The presence of dilated and giant capillaries (width of the capillary >2 times or >10 times the normal respectively); tortuous (capillary limb curled but not crossing over), neo-angiogenesis/bushy capillaries (small, multiple buds originating from the distal loop); meandering (limbs crossed upon themselves several times); micro-haemorrhages, capillary dropouts; avascular areas (absence of two or more adjacent capillaries from the distal-most row); and bizarre capillaries (atypical morphology not conforming to the predefined morphologies) were recorded.6,7 Receding capillaries were seen as individual capillaries maintaining their position in the distal-most row but present slightly proximal to the distal-most loops. They represent that the capillary had receded backwards but had not dropped out yet. A special type of tortuous capillary reported was angulated capillary where one or both the arms of the distal capillary loop were seen to bend at an angle.3 The visibility of the subpapillary plexus was also recorded [Figures 2–8].

Figure 2: Nail fold capillaroscopy showing normal capillaries and dilated capillaries. (black arrow). [195.1 X] (DL0 – first dermoscopic calibration, DL1 – second dermoscopic calibration, H – represents the line used during the calibration.)

Export to PPT

Figure 3: Nail fold capillaroscopy showing bushy capillaries and tortuous capillaries. (black arrow). [203.1x]

Export to PPT

Figure 4: Nail fold capillaroscopy showing cross-linked (black arrow, blue circle), meandering capillaries (black arrow, blue circle) and avascular areas (blue arrow). [203.6 X]

Export to PPT

Figure 5: Nail fold capillaroscopy showing dilated and giant (black arrow) capillaries. [71.4x]. DL0: first dermoscopic calibration, DL1: second dermoscopic calibration, DL2: third dermoscopic calibration, H (red lines) signify the line used during calibration.

Export to PPT

Figure 6: Nail fold capillaroscopy showing angulated capillaries (blue arrow). [149.8x]

Export to PPT

Figure 7: Nail fold capillaroscopy showing reduced capillary density and bushy capillaries (red arrow). [155.5x]. DL0: dermoscopic calibration used. H (red line) signify the line used during calibration.

Export to PPT

Figure 8: Nail fold capillaroscopy showing reduced capillary density, microhaemorrhages (green arrow) and avascular areas (blue arrow). [155.5x]. DL0: dermoscopic calibration used. H (red line) signify the line used during calibration.

Export to PPT

Results were tabulated and analysed objectively at the end of the study statistically. Categorical variables were presented in number and percentage, and continuous variables were presented as mean ± SD. Quantitative variables were compared using an unpaired t-test between two groups. Qualitative variables were correlated using the Chi-square test. A ‘p’ value of <0.05 was considered statistically significant. The data were analysed using Statistical Package for Social Sciences (SPSS) version 22.0.

Results

A total of 100 patients with diabetic retinopathy and 100 healthy controls were evaluated and the baseline characteristics are summarised in Table 1.

Table 1: Baseline characteristics and comparative analysis of qualitative nail fold capillaroscopy features observed between cases (type 2 diabetics with retinopathy) and healthy controls

Baseline characteristics Cases -Diabetic Retinopathy (n = 100), n (%) Healthy controls (n = 100), n (%) p-value Mean age ±SD (years) 58.32 ± 9.00 56.43 ± 7.62 0.926 Male: Female ratio 48:52 59:41 0.156 Mean disease duration ±SD (years) 12.24 ± 5.121 - PDR 14.52 ± 4.229 - Severe NPDR 11.92 ± 3.752 - Moderate NPDR 9.75 ± 5.149 - Mild NPDR 6.20 ± 4.658 - BMI 25.86 ± 3.77 23.97 ± 3.50 Mean fasting blood sugar ±SD (mg/dl) 177.01 ± 48.40 94.91 ± 3.75 Mean HbA1c±SD (%) 8.80 ± 1.60 5.43 ± 0.21 NFC parameters Presence of dilated capillaries 55 (55%) 6 (6%) 0.001* Evidence of capillary dropouts 4 (4%) 7 (7%) 0.352 Presence of avascular areas 71 (71%) 3 (3%) 0.001* Presence of micro-haemorrhages 27 (27%) 5 (5%) 0.001* Bushy capillaries 42 (42%) 3 (3%) 0.001* Meandering capillaries 33 (33%) 7 (7%) 0.001* Tortuosity (>5% of capillaries) 82 (82%) 19 (19%) 0.001* Reduced capillary density 61 (61%) 9 (9%) 0.001* Subpapillary plexus visibility 10 (10%) 4 (4%) 0.096 Receding capillaries 19 (19%) 7 (7%) 0.012* Angulated capillaries 31 (31%) 0 0.001* Normal capillary distribution 9 (9%) 89 (89%) 0.001*

The qualitative capillary changes in healthy controls and patients with DR are depicted in Table 1. The difference in the NFC changes between healthy controls and DR patients was statistically significant (p = 0.001).

Patients with DR were further subdivided into those with proliferative DR (PDR) and non-proliferative DR (NPDR) (mild, moderate and severe) as per Diabetic Retinopathy Disease Severity Scale5 and their nail fold changes were observed and analysed [Table 2]. Statistically significant morphological capillary alterations were observed in PDR as compared to NPDR [Figures 9 and 10].

Table 2: Comparative analysis of nail fold capillaroscopy features in diabetic retinopathy patients (According to the severity of Diabetic Retinopathy (DR)).

NFC parameters NPDR (n = 50), n (%) PDR (n = 50), n (%) p value Mild (n = 5), n (%) Moderate (n = 32), n (%) Severe (n = 13), n (%) Presence of dilated capillaries 4 (80%) 16 (50%) 10 (76.92%) 25 (50%) 0.202 Evidence of capillary dropouts 0 0 0 4 (8%) 0.244 Presence of avascular areas 0 13 (40.63%) 11 (84.62%) 47 (94%) 0.001* Presence of micro-haemorrhages 1 (20%) 1 (3.13%) 4 (30.77%) 21 (42%) 0.002* Bushy capillaries 0 3 (9.38%) 4 (30.77%) 35 (70%) 0.001* Meandering capillaries 0 0 2 (15.38%) 31 (62%) 0.001* Tortuosity (>5% of capillaries) 2 (40%) 19 (59.38%) 11 (84.62%) 50 (100%) 0.001* Reduced capillary density 0 3 (9.38%) 10 (76.92%) 48 (96%) 0.001* Subpapillary plexus visibility 0 1 (3.13%) 1 (7.69%) 8 (16%) 0.230 Receding capillaries 0 0 0 19 (38%) 0.001* Angulated capillaries 0 0 3 (23.08) 28 (56%) 0.001* Figure 9: Fundus examination showing findings suggestive of proliferative diabetic retinopathy with clinically significant macular oedema (CSME). Black arrows suggesting dot and blot haemorrhages, flame shaped haemorrhages and hard exudates.

Export to PPT

Figure 10: Fundus examination showing findings (black arrows suggesting dot and blot haemorrhages, cotton wool spots and hard exudates) suggestive of moderate nonproliferative diabetic retinopathy.

Export to PPT

The NFC parameters were compared in DR patients according to the duration of diabetes [Table 3]. The findings were found to be statistically significant (p < 0.03). Patients with DR were analysed based on their haemoglobin A1c (HbA1c) levels (<7%, 7–9%, 9.1–11%, >11%) reflecting their glycaemic control and the comparison is depicted in Table 4. Morphological capillary changes were more commonly seen in those with poor glycaemic control and were details are given in Table 4.

Table 3: Comparative analysis of nail fold capillaroscopy features in diabetic retinopathy patients (According to the duration of diabetes mellitus)

NFC Parameters Duration of diabetes (in years) P-value ≤5 (n = 9), n (%) 6–10 (n = 27), n (%) 11–15 (n = 47), n (%) 16–20 (n = 13), n (%) >20 (n = 4), n (%) Presence of dilated capillaries 6 (66.67%) 14 (51.85%) 24 (51.06%) 8 (61.54%) 3 (75%) 0.778 Evidence of capillary dropouts 0 0 3 (6.38%) 1 (7.69%) 0 0.588 Presence of avascular areas 1 (11.11%) 16 (59.26%) 38 (80.85%) 12 (92.31%) 4 (100%) 0.001* Presence of micro-haemorrhages 0 5 (18.52%) 17 (36.17%) 3 (23.08%) 2 (50%) 0.112 Bushy capillaries 0 6 (22.22%) 27 (57.45) 7 (53.85%) 2 (50%) 0.003* Meandering capillaries 0 4 (14.81%) 18 (38.30%) 8 (61.54%) 3 (75%) 0.03* Tortuosity (>5% of capillaries) 3 (33.33%) 20 (74.07) 42 (89.36%) 13 (100%) 4 (100%) 0.001* Reduced capillary density 0 10 (37.04%) 36 (76.60%) 12 (92.31%) 3 (75%) 0.001* Subpapillary plexus visibility 0 0 8 (17.02%) 1 (7.69%) 1 (25%) 0.105 Receding capillaries 0 3 (11.11%) 12 (25.53%) 3 (23.08%) 1 (25%) 0.315 Angulated capillaries 0 4 (14.81%) 17 (36.17%) 7 (53.85%) 3 (75%) 0.005*

Table 4: Comparative analysis of nail fold capillaroscopy features in diabetic retinopathy patients -according to the glycaemic index (HbA1c levels)

NFC parameters HbA1c levels P-value <7.0 (n = 15), n (%) 7.0–9.0 (n = 42), n (%) 9.1–11.0 (n = 36), n (%) >11.0 (n = 7), n (%) Presence of dilated capillaries 6 (40%) 23 (54.76%) 24 (66.67%) 2 (28.57%) 0.150 Evidence of Capillary drop outs 0 0 3 (8.33%) 1 (14.29%) 0.109 Presence of avascular areas 8 (53.33%) 24 (57.14%) 32 (88.89%) 7 (100%) 0.002* Presence of microhaemorrhagesC 2 (13.33%) 11 (26.19%) 12 (33.33%) 2 (28.57%) 0.537 Bushy capillaries 3 (20%) 12 (28.57%) 22 (61.11%) 5 (71.43%) 0.003* Meandering capillaries 3 (20%) 5 (11.90%) 21 (58.33%) 4 (57.14%) 0.001* Tortuosity (>5% of capillaries) 10 (66.67%) 30 (71.43%) 35 (97.22%) 7 (100%) 0.005* Reduced capillary density 4 (26.67%) 17 (40.48%) 33 (91.67%) 7 (100%) 0.001* Subpapillary plexus visibility 1 (6.67%) 4 (9.52%) 4 (11.11%) 1 (14.29%) 0.943 Receding capillaries 1 (6.67%) 4 (9.52%) 12 (33.33%) 2 (28.57%) 0.027* Angulated capillaries 1 (6.67%) 8 (19.05%) 17 (47.22%) 5 (71.43%) 0.001* Discussion

NFC can be used to assess T2DM-related complications in the proximal nail fold capillaries.8 The most common qualitative capillary changes in healthy controls were tortuous, meandering, capillary dilation and micro-haemorrhages as observed in previous studies.3,9,10 In our study, the NFC changes were significantly higher (p = 0.001) in DR patients. Tortuosity, avascular areas, reduced capillary density, micro-haemorrhages, and dilated, bushy, receding and angulated capillaries were common NFC findings in DR patients. The typical findings in diabetes may be due to the advanced glycation end-products (AGEs), which are known to be associated with macro- and microvascular complications.11 One of the pathomechanisms is the apoptosis of retinal capillary pericytes and endothelial cells.12 Contrarily, the well-known NFC features of CTD may be due to interaction between VEGF, antiangiogenic factors (angiostatin, endostatin, tumstatin, canstatin) and elevated levels of hypoxia-induced factor, which affects the capillaries and the surrounding connective tissue matrix.13 Thus, NFC features are regressive in advanced disease (capillary dropouts, avascular areas) and predominantly proliferative in early stages (capillary dilation, giant, bushy capillaries).

Maldonado et al. demonstrated that NFC changes in diabetic patients collaborated with ophthalmological findings in 18% of patients.14 A study done by Kaminska–Winciorek reported that the presence of mega capillaries, abnormal loops and altered capillary densities were seen more in DM of longer duration and high HbA1C.15 Various other studies have also recorded reduced capillary density, tortuosity, neoangiogenesis, haemorrhages, and avascular zones as the common findings in diabetics, which significantly correlated with the presence and degree of DR, duration and control of diabetes.16–20

The findings of our study were consistent with those of Mohanty et al., which reported tortuosity in 100% of patients with PDR and 81.8% with NPDR.21 In a study by Chang et al., including 35 diabetic patients and 20 healthy controls, it was found that tortuosity was the most common finding in PDR (68%).22

We studied the NFC features in DR patients according to the duration of diabetes. The findings were consistent with the study by Uyar et al. in which it was reported that diabetes years of patients having tortuosity, aneurysm, bizarre and micro-haemorrhage were significantly longer in patients with DR than without DR.8 Positive correlation of capillaroscopic findings and diabetes duration was also stated in Chang et al. study.22 The findings in our study are not consistent with the study done by Barchetta et al. in which NFC findings were independent of the duration of diabetes.16 Although these findings do not predict when the microvascular changes formed, there is a significant correlation between diabetes years and capillaroscopic findings in our study. It can be assumed that early detection of tortuosity, reduced capillary density, and dilated and microhaemorrhages may be a precursor of DR.

We studied the correlation of nail fold capillaries with the glycaemic index of the patient. The findings were consistent with the studies done by Hsu et al,18 Uyar et al 8 and, Rajaei et al 23 s which identified mega capillaries, increased tortuosity and neoangiogenesis as frequent changes in diabetic patients with longer duration and high HbA1c.

In a study done by El-Khalik et al, there was a significantly longer disease duration and significantly higher HbA1c in patients with DR than with non-DR.24 Also, it was seen that the frequencies of tortuosity and precapillary oedema were increased significantly in patients with HbA1c≥7% compared to patients with HbA1c <7% and most of these patients were with diabetic retinopathy. This could be attributed to some studies showing HbA1c as a predictor of diabetic retinopathy. A study revealed that when HbA1c was ≥6.8%, the odds ratio for diabetic retinopathy increased significantly.25

Limitations

A larger sample size study with different demographic populations could have provided a broader picture of NFC changes in type 2 DM.

Conclusion

Changes in NFC in diabetics had a significant association with the severity of DR, disease duration, and glycaemic control. These changes perhaps run parallel to the changes in the retina. Hence, nail fold capillaroscopy can be used as a valuable non-invasive tool not only for the diagnosis of microvascular complications but also for the prognosis of the disease and identifying patients at high risk of DR, thereby helping in early diagnosis and improving the quality of life of patients.