Dermoscopic Features of Nail Psoriasis: Revisited

Psoriasis is an immune-mediated inflammatory disease with a global prevalence of 1–2%. Almost half of the patients suffer from nail involvement, with a reported risk of up to 90% in lifetime. Nail involvement is a potential cause of significant morbidity in terms of functional limitations, pain, and cosmetic concerns. Moreover, nail psoriasis is a negative predictive factor for the development of both arthritis and severe cutaneous psoriasis. Nail psoriasis represents a major clinical burden independent from the severity of cutaneous disease [1-6].

Nail dermoscopy is a fundamental method in differentiating nail psoriasis from other nail disorders when clinical findings are not apparent [7, 8]. A few studies [9-16], including ours [17], have described both classical and novel dermoscopic features of nail psoriasis. The diagnosis of nail psoriasis can easily be made when characteristic dermoscopic features are detected [7]. On the other hand, there is an ongoing confusion on the assessment of typical dermoscopic features of nail psoriasis [17, 18]. Thus, we aimed at assessing the characteristic dermoscopic features of nail psoriasis by re-evaluating both classical and novel findings, whilst exploring the relationship between these features and disease severity, as well as with psoriatic arthritis.

We prospectively enrolled 35 patients (26 men and 9 women; mean age, 45.5 ± 13.9 years [range: 24–76 years]) with severe nail psoriasis over a period of 6 months. The study was conducted in accordance with the Declaration of Helsinki and approved by the Local Ethics Committee. We included patients over 18 years of age with severe nail psoriasis (Nail Psoriasis Severity Index [NAPSI] score of more than 50), previously treated/treatment-naïve patients [2]. For all patients, the diagnosis of psoriasis was confirmed with a histopathological examination. Patients were classified into four groups according to age: less than 35 years old (Group 1), 35–54 years old (Group 2), 55–64 (Group 3), and over 65 years old (Group 4). A detailed history regarding the demographic profile, medical history, the onset and duration of psoriasis was obtained and documented. A complete clinical evaluation, including musculoskeletal assessment and dermatological examination of all nails, was carried out for each patient. Patients with onychomycosis and nail trauma were excluded from our study. Nail psoriasis severity was graded according to the NAPSI score, which is the sum of scores given for the presence of manifestations of psoriasis in each of four imaginary quadrants of each nail [19]. Dermoscopic evaluation was performed using a digital dermoscope (Fotofinder dermoscope, FotoFinder Systems GmbH, Bad Birnbach, Germany) for all nail plate surfaces, periungual folds, distal-free edges, and hyponychial regions. The 20-fold magnification power was preferred, unless otherwise. Ultrasound gel was used as an interface medium during the examination.

Analysis of data was carried out using SPSS software, Version 20 (SPSS Inc.) Descriptive frequencies were calculated for variables related to demographic and clinical characteristics and were presented as means (standard deviations) or medians (interquartile range [IQR]) for continuous variables and numbers (percentages) for categorical variables. Prior to comparisons, the variables were tested using the Kolmogorov-Smirnov test for skewness and kurtosis. Histograms and probability plots were used to describe the distribution of data. Comparisons and associations were performed using the χ² test for categorical variables and Mann-Whitney U test for continuous variables. The correlation between duration of psoriasis and nail psoriasis severity was analysed using Spearman correlation. To identify the dermoscopic features associated with psoriatic arthritis, univariate and multivariate logistic regression analysis was performed. Dermoscopic features with a p value <0.2 in univariate analysis were included in multivariate logistic regression analysis and odds ratios with 95% confidence intervals were used to report the risk of psoriatic arthritis. A p value <0.05 was considered statistically significant.

From our analysis, 22.9% of patients (n = 8) were in Group 1, 48.6% (n = 17) in Group 2, 17.1% (n = 6) in Group 3, and 11.4% (n = 4) in Group 4 in terms of age. The median duration of psoriasis was 120 months (IQR: 36–240 months). Arthritis was observed in 14 (40%) patients. Fingernails were affected in all patients, and the median of affected fingernails was 10 (IQR: 9–10 nails). Toenails were affected in 23 (65.7%) patients, and the median of affected toenails was 2 (IQR: 0–10 nails). The median of the total number of affected nails was 12 (IQR: 9–20 nails). The median NAPSI score of the study population was 64 (IQR: 51–116).

The following dermoscopic findings were observed in the study population: splinter haemorrhages (SH) in 94.3% (n = 33), distal onycholysis (DO) in 88.6% (n = 31), pitting in 74.3% (n = 26), salmon patches (SP) in 74.3% (n = 26), thickened white-yellow nail plates (TYP) in 68.6% (n = 24), fuzzy lunula (FL) in 68.6% (n = 24), longitudinal erythema of the nail bed in (LE) 57.1% (n = 20), dilated hyponychial capillaries (DHC) in 51.4% (n = 18), subungual hyperkeratosis (SUH) in 45.7% (n = 16), dilated nail bed capillaries (DBC) in 45.7% (n = 16), nail plate crumbling (NPC) in 45.7% (n = 16), mottled lunula (ML) in 45.7% (n = 16), leukonychia in 28.6% (n = 10), and red spots in the lunula (RSL) in 22.9% (n = 8).

All patients with DBC had DHC (n = 16, p < 0.001), and all patients with leukonychia had pitting (n = 10, p = 0.029). Patients with LE were significantly more likely to exhibit DBC (65% vs. 35%, p = 0.010) and DHC (75% vs. 25%, p = 0.002). Moreover, 81.2% of patients with DBC had LE (n = 13), whilst 83.3% of patients with DHC had LE (n = 15). In addition, patients with DHC were more likely to exhibit SP (88.9% vs. 11.1%, p = 0.049). A significant relationship was observed between the presence of RSL and the presence of ML (p = 0.010). Likewise, patients with RSL were more likely to exhibit NPC (87.5% vs. 12.5%, p = 0.010).

Significant relationships were observed between the presence of ML and the presence FL (p = 0.030), as well as the presence of FL and the presence of TYP (p = 0.006). Patients with ML were more likely to exhibit FL (87.5% vs. 12.5%) and those with FL were more likely to exhibit TYP (54.2% vs. 45.8%). Furthermore, 58.3% of patients with FL had ML (n = 14), and 54.2% of patients with TYP had FL (n = 13). The presence of SUH was inversely related to the presence of LE (p = 0.034). Patients with SUH were less likely to exhibit LE (37.5% vs. 62.5%). However, those with NPC were more likely to exhibit DO (75% vs. 25%, p = 0.035) and TYP (87.5% vs. 12.5%, p = 0.030). Moreover, 58.3% of the patients with TYP had NPC (n = 14), and 38.7% of the patients with DO had NPC (n = 12).

We performed a statistical analysis to detect any association between clinical characteristics of the study group and the presence of any dermoscopic feature. A significant correlation was observed between the presence of toenail involvement and the presence of SUH (p = 0.001). SUH was observed in 65.2% of patients with toenail involvement, and in 8.3% of the patients without toenail involvement. Besides, in patients with SUH, the median of affected toenails was 6 (IQR: 2–10 nails), and number of affected nails was 16 (IQR: 2–10 nails). These values were higher than in patients without SUH (p = 0.011, p = 0.020, respectively). Patients with toenail involvement were less likely to exhibit ML (p = 0.012). ML was observed in 30.4% of patients with toenail involvement, and in 75% of patients without toenail involvement. In patients with ML, the median of affected toenail was 2 (IQR: 0–4 nails), and number of affected nails was 10 (IQR: 7.25–14 nails). These values were lower than those obtained in patients without ML (p = 0.007, p = 0.039, respectively).

A significant correlation was observed between the presence of LE and toenail involvement (p = 0.024). LE was observed in 43.5% of patients with toenail involvement, and in 83.3% of patients without toenail involvement. Although statistically insignificant, the median of the affected toenail was lower in patients with LE compared to those without LE (1 vs. 5, p = 0.073). When compared with patients without leukonychia, patients with leukonychia had more affected fingernails (10 vs. 8, p = 0.032) and whole nails (17 vs. 11, p = 0.042). Dermoscopic features did not reveal any significant associations between sex, age groups, and duration of psoriasis (p > 0.05, for all). In addition, the presence of arthritis was not associated with sex (p = 1.000), age groups (p = 0.685, p = 0.1000, p = 0.664, p = 0.133, group 1–4 respectively), and duration of psoriasis (p = 0.661).

Statistical analysis did not reveal any significant correlation between the duration of psoriasis (rho = −0.058, p = 0.739), sex (p = 0.088), age groups (p = 0.180, p = 0.778, p = 0.323, p = 0.876, group 1–4, respectively), presence of arthritis (p = 0.091) and NAPSI scores. However, significant associations were observed between the presence of SUH and ML. Patients with SUH were more likely to have severe nail psoriasis (p = 0.041). The median NAPSI score was 86.5 (IQR: 53–132.75) in patients with SUH and 61 (IQR: 50–75) in patients without SUH. Nail psoriasis severity was significantly less in patients with ML (p = 0.006). The median of NAPSI was 56.50 (IQR: 50.25–63.75) in patients with ML, whilst it was 86 (IQR: 56–123) in patients without ML. Statistical analysis revealed that patients with toenail involvement were more likely to have severe nail psoriasis (p = 0.004). The median NAPSI score was 86 (IQR: 56–123) in patients with toenail involvement, and 53.5 (IQR: 50–63.75) in patients without toenail involvement.

The presence of the following dermoscopic features was not significantly different in patients with and without arthritis: DO (p = 0.13), SH (p = 0.23), SP (p = 0.75), DBC (p = 0.27), DHC (p = 0.21), LE (p = 0.99), pitting (p = 0.75), leukonychia (p = 0.45), TYP (p = 0.77), RSL (p = 0.51), ML (p = 0.78), and FL (p = 0.77). SUH and NPC were observed in 64.5% of patients with arthritis, and in 33.3% of patients without arthritis. Although the difference was clinically important, a significant difference was not observed (p = 0.072, for both). On the other hand, further statistical analysis revealed that patients with arthritis were more likely to exhibit a combination of SUH and NPC compared with patients without arthritis (42.9% vs. 4.8%, p = 0.006). Moreover, the presence of a combination of SUH and NPC led to a 15-fold increase in the risk of developing an arthritis (95% confidence interval: 1.55–145.2, p = 0.019).

Typical dermoscopic features of nail psoriasis include SH, DO, SP, SUH and pitting, which are also seen with a naked eye. Dermoscopy allows for the recognition of characteristic features of nail psoriasis that are difficult to differentiate with a naked eye, such as erythematous border around the onycholytic area [20] and ML [21]. In the present study, we assessed both classical and novel dermoscopic features of nail psoriasis. FL is a new dermoscopic feature of nail psoriasis described by Chauhan et al. [15] and was defined as irregular wide white lunula. We detected FL in 68.6% of patients. Although we did not support the description of FL as wide as the lunula of some of the patients were not wide (Fig. 1d, e), we agreed on the specificity of FL for psoriasis. Chauhan et al. [15] did not comment on the possible underlying pathogenic mechanism of FL. One explanation might be related to the involvement of the nail matrix in the disease process. It is well-known that the lunula represents the visible portion of the distal nail matrix [22]. The jagged characteristics of FL may indicate the pathological process in the distal nail matrix (Fig. 1).

In our opinion, the pathogenesis of FL is related to both the nail matrix and the nail bed. Moreover, all psoriasis-related findings in the lunula (FL, ML, RSL) are interconnected with each other, and all are linked with the vascular changes seen in psoriasis. In Figure 2, FL, LE, DBC, SH, DO, DHC, and pitting are seen (Fig. 2), representing the most characteristic features of nail psoriasis. They include the jagged border of FL (red circles), several LE (blue rectangle), and DBC (black circles) (Fig. 2). The jagged border of FL occurs due to the presence of multiple LE, which runs longitudinally through the lunula. These LE are dilated vascular channels on the epidermal-dermal ridges, which end with DBC distally (Fig. 2 [black circles]). In the lunula, the longitudinal fashion of LE is preserved causing the jagged border and the marbled appearance of the lunula. In Figure 3, LE with some parallel SH (Fig. 3a) and a prominent dilated capillary in between two LE (Fig. 3b [red circle]) are demonstrated.

Some authors prefer to call the RSL as ML [21]. We only designate RSL, when red spots are seen in the lunula (Fig. 4). Otherwise, we prefer to use the term ML, when different shades of white, pink and red occupy the lunula, giving a marbled appearance (Fig. 1a–d, f, Fig. 4a–c, e). In 2016, Peña-Romero et al. [21] described the presence of ML in three patients with psoriasis, one of whose histopathological examination of nail biopsy specimens demonstrated dilated capillaries in the superficial papillary dermis and thinning of the suprapapillary plate. It was postulated that RSL was the clinical representation of dilated blood vessels in the lunula [21]. Lichen planus is another dermatological disorder with a known cause of RSL. Kaur and Jakhar [23] described intraoperative dermoscopic features of patients with nail lichen planus presenting with RSL. They demonstrated that multiple linear lunular bands were the evident dermoscopic feature corresponding to the increased, dilated, and tortuous vasculature in the nail bed and matrix [23]. The linear lunular bands were identical to the LE described here. It is obvious that ML is composed of multiple linear bands with different shades of white, pink, and red (Fig. 1a–d, f, Fig. 4a–c, e), some of which are intermingled with red spots (Fig. 4a–c, e).

Psoriasis is an inflammatory disease with a pathogenesis intricately linked to angiogenesis. In psoriatic lesions, changes in microvasculature (such as an increased number of dilated, tortuous, and leaky vessels) are typical histopathological features [24]. It has been shown that angiogenic mediators including vascular endothelial growth factor significantly increase in psoriatic lesions giving rise to increased vascular permeability, thereby leading to cutaneous infiltration of the inflammatory cells. Vascular network expansion is a well-established histopathological hallmark of psoriasis, which occurs before epidermal changes and prevails even after the improvement of psoriatic lesions [25]. We can depict that FL, RSL, ML, LE, DBC, and DHC share the same pathogenesis: an increased dermal vascularity.

One of the important findings of our study was the observation of the nail bed red spots (Fig. 5), which is another proof of the involvement of nail bed in the pathogenesis of red spots. In addition, we observed that some red spots, whether they were on the lunula (Fig. 4a) or on the nail bed (Fig. 5), had a surrounding white halo identical to that around the DBC (Fig. 6, Fig. 7a, b, d, Fig. 8d). Likewise, in Figure 7, peripheral white halos were observed around SP (Fig. 7a–c), some of which are accompanied by a peripheral erythema (Fig. 7b). The peripheral white halo is also related to vascular remodelling and morphological abnormalities of microvessels seen in psoriasis [26]. SP are yellowish-brownish spots with a red margin. Peripheral erythema (Fig. 7b) and DBC (Fig. 7d) around SP are the dermoscopic representations of red margins seen clinically, which are presumably linked with vascular alterations.

Psoriatic DO also has a reddish margin that differentiates it from other causes of DO (Fig. 2, 6–8, 10). In the present study, we detected further findings that distinguish psoriatic DO from onychomycotic onycholysis: one of the most common causes of DO. Typical psoriatic DO has a silvery-whitish hue, which is more pronounced on the lateral edges of the nail plates (Fig. 2, Fig. 6b, Fig. 9e), shows how the dermoscope can capture images with a very slight angle due to the convex nature of the plates. Psoriatic DO occurs because of separation of parakeratotic areas [1], which creates individual parallel splits. These splits are parallel in both the longitudinal plane (Fig. 2, Fig. 6b, Fig. 7f, Fig. 8c–f, Fig. 10a) and the horizontal plane (Fig. 6 [red circle]). Moreover, each split has a round linear proximal edge (Fig. 2, Fig. 6b, Fig. 7f, Fig. 8c–f, Fig. 10a). In onychomycotic onycholysis, however, a jagged proximal edge with spikes is the most characteristic feature. The spikes correspond to the onset of fungal invasion, where pathogens occupy the longitudinal rete ridges from the hyponychium through the lunula. In addition, yellowish white is the typical colour of spikes, instead of silvery-white colour of splits seen in psoriatic DO [27].

Previously, we described pseudo-fibre sign as a novel dermoscopic feature of nail psoriasis [17], which was defined as the presence of filamentous structures located along the cuticle or underneath the distal-free edge or exposed areas where the nail plate had detached. Although it was impossible to differentiate these structures from real fibres, we assumed these structures to be organic, rather than real fibres. Nevertheless, in the present study, we demonstrated the solid uniform nature of these structures (Fig. 10e, f). Thus, we no longer support the terminology used above and we do not approve the originality of these structures as organic. These structures were real fibres caught on the dystrophic areas of the nail, where NPC was evident. They were also seen in far deep areas of the hyponychium, where they were easily trapped. Presumably, accompanying dotted structures were minute artefacts, some of which might have been caught perpendicularly to the surface. The unique red and black colour of these structures may represent just a mere coincidence.

In the present study, we also detected other well-known dermoscopic features of nail psoriasis, including pitting, SH, TYP, DHC, SUH, NPC, and leukonychia. Since we included patients with severe nail psoriasis [2], the frequencies of dermoscopic findings in the study population were high. SH was the most common dermoscopic finding, found in 94.3% of the patients. The second most common dermoscopic feature was DO followed by pitting. Our findings revealed that the presence of a nail bed-related or matrix-related finding increases the likelihood of the presence of other nail bed-related or matrix-related findings. For instance, all patients with DBC had DHC and leukonychia had pitting. On the other hand, nail bed-related findings had more significant relationships. LE, DHC, DBC, ML, FL, and SP were significantly associated with each other. NPC was the most common matrix-related finding that had significant relationships, one of which was with TYP, another matrix-related finding. The only inverse relationship detected was between SUH and LE, both of which are nail bed-related findings. Patients with SUH were less likely to exhibit LE. This could be explained by the fact that when SUH occurs, it covers the nail bed, where LE is normally seen. According to our results, other than NPC, nail bed-related findings had more relevant associations in defining nail psoriasis. Thus, nail bed-related findings play a key role in differentiating nail psoriasis from other nail disorders.

Till date, an erythematous border surrounding the DO, FL, and irregularly distributed, tortuous, elongated DHC, have been described as the unique dermoscopic features of nail psoriasis [7, 11, 15, 20]. Nevertheless, literature is sparse with substantial evidence-based dermoscopic criteria for differentiating nail psoriasis from other nail disorders. With the results of our study, we postulated that a combination of nail bed-related and matrix-related findings could serve as dermoscopic presentation of nail psoriasis, aiding in its diagnosis as shown in Figure 2. In Figure 2, nail bed-related findings (DBC, DHC, SH, DO, FL, LE) and a matrix-related finding (pitting) are demonstrated. As already known, 10 pits in one nail is regarded as proof of psoriasis [1]. Hence, even if nail bed-related findings were not detected, more than 10 pits were seen, making the nail in Figure 2 a typical example of nail psoriasis, in which histopathological confirmation is not required to confirm diagnosis. On the other hand, we agree the absence of enough evidence regarding how many, and which nail bed- and matrix-related findings are necessary to make a definitive diagnosis of nail psoriasis with dermoscopy. Our study only offers preliminary evidence; thus, we recommend further studies to explore the validity of our suggestion about the dermoscopic criteria in the diagnosis of nail psoriasis.

Our primary aim was to clarify the dermoscopic features of nail psoriasis. Moreover, the secondary aim was to assess the relationship between these features and the clinical characteristics of the study group, including the severity of nail psoriasis and psoriatic arthritis. Dermoscopic features did not reveal any significant association between sex, age groups, and duration of psoriasis. The severity of nail psoriasis was not significantly associated with the duration of psoriasis, sex, age groups, and presence of arthritis. When the association between dermoscopic features and severity was investigated, significant associations were observed with the presence of SUH and ML. Patients with SUH and those without ML were more likely to have severe nail psoriasis. Besides, the presence of toenail involvement was significantly associated with the presence of SUH, ML, and LE. Patients with toenail involvement were more likely to exhibit SUH but less likely to exhibit ML and LE. These findings are in line with previous ones, indicating that the presence of SUH and absence of ML reveals severe nail psoriasis. One of the most important findings of the present study was the demonstration of the relationship between arthritis and the presence of SUH and NPC. Our results revealed that both SUH and NPC were observed in higher frequencies in patients with arthritis. Moreover, patients who exhibited a combination of SUH and NPC had a 15-fold increased risk for the occurrence of arthritis. Thus, we emphasise the importance of the initial dermoscopic examination that could provide predictive information about the severity of nail psoriasis and accompanying arthritis.

Dermoscopy is a highly practical and effective tool that waives the need for histopathological confirmation in the diagnosis of a myriad of diseases [28]. Nail psoriasis represents a significant challenge due to its association with uncertainty in its diagnosis, especially in isolated cases [1]. Our study is one of the most recent and comprehensive investigation that assessed characteristic dermoscopic features of nail psoriasis, and would be helpful in differentiating nail psoriasis from other nail disorders. Moreover, significant associations between the dermoscopic features and severity of nail psoriasis, as well as with arthritis were revealed. The findings from this study will provide evidence of the importance of dermoscopy in the diagnosis and evaluation of the patients with nail psoriasis. However, the small sample size of the study population could have limited our power to detect other significant associations. Thus, we recommend further studies with larger sample sizes to assess the value of dermoscopy in the diagnosis and evaluation of patients with nail psoriasis.

The present research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki. This study protocol was reviewed and approved by the Ankara Bilkent City Hospital Medical Ethics Committee with the number of E2-21-470. All subjects have given their written informed consent to publish their case, including publication of images.

The authors have no conflicts of interest to declare.

The authors have received no external funding.

All authors contributed to drafting the work and revising it critically and approved the final version of the manuscript, including conceptualization and writing – review and editing, data curation and writing – original draft, data curation and writing – review and editing, investigation and writing – review and editing, supervision and writing – original draft, methodology and supervision, investigation and writing – review and editing, investigation and writing – review and editing, methodology and supervision, validation and writing – review and editing.

All data generated or analyzed during this study are included in this article. Further enquiries can be directed to the corresponding author.