Nail Fold Capillaroscopy for Diagnosis of Onychodystrophies: A Prospective Cross-Sectional Study

Abstract

Background

Nail diseases are often associated with significant physical and psychosocial burden, but diagnosis is challenging due to non-specific clinical and histological findings. Nail fold capillaroscopy has been studied for diagnosis of systemic diseases, but studies on nail diseases are lacking.

Objective

Our objectives were to characterize and compare capillary changes in a set of nail conditions vs. controls, between nail groups, and based on demographic/clinical criteria.

Methods

This was a prospective cross-sectional study of patients with nail psoriasis, onychomycosis, idiopathic onycholysis, brittle nail syndrome, nail lichen planus, retronychia, other nail conditions, and no nail findings (controls) undergoing capillaroscopy imaging/analysis.

Results

Nail psoriasis vs. control patients demonstrated decreased capillary length/density and increased abnormal capillaries, with higher frequency in older, male patients. Onychomycosis was associated with increased meandering capillaries compared to controls, nail psoriasis, and nail lichen planus. Retronychia is associated with increased disorganized polymorphic capillaries compared to controls and onychomycosis.

Limitations

Limitations include small sample size for certain nail conditions and small numbers of nail psoriasis patient with psoriatic arthritis.

Conclusion

Our findings highlight nailfold capillaroscopy as a potentially quick, cost-effective, and non-invasive imaging modality, as an adjunct for diagnosis and treatment initiation for patients with onychodystrophies.

Capsule Summary

• How does this article integrate into what was already known?

  • Nail fold capillaroscopy can help visualize specific capillary changes in psoriasis patients, but these changes have not been characterized for patients with nail diseases.
• How does it change practice?

  • We characterize capillary changes amongst a set of nail diseases, which can be used as an adjunct for diagnosis and treatment initiation.

Introduction

Nail fold capillaroscopy (NC) is a non-invasive clinical imaging technique that has been used to assess nail capillary changes in rheumatic disease patients with cutaneous manifestations.¹ Specifically, NC has been utilized for detection of the characteristic “scleroderma pattern,” defined as dilated capillaries, avascularity, microhemorrhages, and capillary architecture distortion, which helps to differentiate between scleroderma spectrum disorder and primary Raynaud’s phenomenon.¹

For psoriasis and other dermatological conditions, utility of NC as a diagnostic tool has been far less well studied compared to connective tissue diseases. Prior capillaroscopy studies in psoriasis patients have demonstrated lower capillary density, increased morphologically abnormal capillaries, and increased avascular areas compared to controls, but have only included small patient cohorts.² Likewise, studies assessing NC in nail psoriasis (NP) patients specifically are lacking, especially in patients with isolated NP, defined as psoriatic changes in the nails only with no or limited skin involvement (5–10% body surface area, 5–10% of all psoriatic patients).^3–4

Diagnosis of nail conditions is notoriously difficult due to overlapping clinical and histological findings between different diseases, often resulting in misdiagnosis and treatment delays.^3–7 Impact of these diseases extends beyond aesthetics, with patients having trouble performing daily activities and having negative effects on quality of life.⁴ NP specifically is associated with increased psoriatic arthritis (PsA) risk.^7,8 Therefore, additional tools are needed to aid in diagnosis. However, current knowledge on utility of NC, and specific capillary features of different nail diseases, is lacking. Thus, our objectives were to characterize and compare nail fold capillary dimension and morphological changes amongst common nail diseases compared to healthy control patients, as well as between patients with different nail conditions.

Methods

A single-center, observational cross-sectional pilot study at the Weill Cornell Medicine (WCM) Department of Dermatology was conducted 1/1/2022–5/30/2023 after WCM Institutional Review Board (#21–07023732) approval. Patients ages ≥7 years were screened and enrolled based on clinical diagnosis of one of eight groups: NP, onychomycosis, idiopathic/traumatic onycholysis, brittle nail syndrome (BNS), nail lichen planus (NLP), retronychia, other nail conditions (onychopapilloma, eczema, etc.), and patients with no nail findings (controls). Only patients with newly diagnosed nail conditions, without prior treatment, and with no other concurrent nail conditions were included. Patients with scleroderma or CREST syndrome, Raynaud’s phenomenon, systemic lupus erythematous, hypertension, diabetes mellitus, as well as pregnant individuals, were excluded. All patients with clinical diagnosis of NP and onychomycosis had nail clippings performed with histopathology to confirm the diagnosis.

After screening and verification of eligibility criteria, enrolled patients underwent NC imaging using the Dino-Lite Edge USB digital handheld microscope. A magnification of 210–220x was used to visualize and photograph the nail fold capillaries along the proximal nail fold of the most severely affected fingernail(s) or toenail(s), seen clinically. DinoXcope computer-based software was utilized to analyze capillary parameters in each captured image, including capillary loop density (number of capillaries per millimeter), capillary loop length (in micrometers), normal/minor capillary changes (crossing, tortuous, enlarged), abnormal/major capillary changes (branching, meandering, bushy/ramified, bizarre), and other capillary parameters (disorganized polymorphic capillaries, presence of hemorrhages, avascular areas, and presence of subpapillary plexus) (Supplementary Figures 1–2, available via Mendeley at https://data.mendeley.com/datasets/k7pwbpttds/3).⁹ Assessment of nail fold capillaries was performed in concordance with standardized capillaroscopy guidelines.^9,10

For data analysis, capillary features were compared between patients with nail conditions vs. controls, as well as between nail groups. Fingernail conditions and toenail conditions were compared separately, against fingernail or toenail controls, respectively. Within each nail group, capillary parameters were compared between patient groups based on demographic characteristics (patient age, gender, race/ethnicity). For the NP subgroup, additional clinical variables, including disease duration, presence of concomitant cutaneous psoriasis (PsO) or PsA, and signs of nail matrix (pitting, crumbling, leukonychia, red lunula) vs. nail bed involvement (onycholysis, subungual hyperkeratosis, salmon patches/oil spots, splinter hemorrhages) were analyzed. Mean scores were analyzed using student t-tests, and categorial variables with chi-squared tests (α<0.05).

Results

One hundred and twenty-eight patients were included, with average age 46.1 years (range 8–84) and 52.3% being female (Supplementary Table 1). Age, gender, and racial distributions were similar across all nail groups (including controls), with the exception of the BNS group (average age 68.8 years, 100% female, 100% non-Hispanic white). Amongst control groups with no nail conditions (n=11 for fingernails, n=7 for toenails), there were no significant differences in capillary parameters based on age (40+ vs. <40 years), gender (male vs. female), or race/ethnicity (white vs. non-white).

The NP, onychomycosis, NLP, onychopapilloma, and retronychia groups differed in terms of measured capillary parameters compared to the control group (Figures 1–4, Table 1, Supplementary Figures 3–6, Supplementary Table 2). The control groups (n=11 for fingernails, n=7 for toenails), on average, demonstrated findings consistent with established normal capillary parameters, with an average 9.45 capillaries/mm (normal 7–12), average 185.94μm capillary length, and presence of normal/minor morphological changes, including crossed, tortuous, and enlarged capillaries (Figure 1).

Figure 1.

Nail fold capillaroscopy imaging findings for healthy fingernail controls: 7–12 capillaries/millimeter, normal morphological changes including crossing, tortuous, and/or enlarged capillaries (field: 1 millimeter)

Figure 4.

Nail fold capillaroscopy imaging findings for nail lichen planus (NLP): decreased capillary length (field: 1 millimeter)

Table 1.

Nail fold capillaroscopy measurements (n=128) for various nail conditions compared to healthy controls

Diagnosis	n	Findings	Clinical Implications
Fingernail Conditions (Controls: n=11)
NP	34	Decreased capillary density vs. controls, OM, and LP Decreased capillary length vs. controls and OL Increased frequency of crossed/tortuous capillaries vs. controls Increased frequency of abnormal morphology vs. controls	Look for decreased capillary density/length and increased abnormal morphology Differentiate from OM and LP based on decreased capillary density
OM	8	Increased frequency of abnormal morphology (meandering capillaries) vs. controls, NP, and LP	Look for increased abnormal morphology (especially meandering capillaries) Differentiate form NP and LP based on increased meandering capillaries
LP	11	Decreased capillary length vs. controls	Look for decreased capillary length
OP	6	Decreased capillary length vs. controls
BN	13	No significant differences vs. controls
OL	7
E	6
Toenail Conditions (Controls: n=7)
OM	13	Increased frequency of abnormal morphology (meandering capillaries) vs. controls	Look for increased abnormal morphology (especially meandering capillaries)
RN	12	Decreased capillary length vs. controls Increased frequency of disorganized polymorphic capillaries vs. controls	Look for disorganized polymorphic capillaries

NP: nail psoriasis, OM: onychomycosis, LP: nail lichen planus, OP: onychopapilloma, BN: brittle nail syndrome, OL: idiopathic/traumatic onycholysis, E: eczema, RN: retronychia

Compared to controls, the NP (n=34) cohort had both decreased average capillary density and decreased length (P<0.001), increased frequency of crossed and tortuous capillaries (P<0.02), as well as increased frequency of any abnormal morphology (P=0.03) (Figure 2, Supplementary Figure 3). Specific abnormal morphologies seen within the NP cohort included branching (26.5%), meandering (26.5%), bizarre (17.6%), and bushy (8.8%) capillaries. Amongst the NP group, there was also a trend towards a greater proportion of patients having avascular areas (n=10, 27.8%), defined as an area of absence of >2 successive capillaries, compared to the control group (n=1, 9.1%), but did not reach statistical significance (P=0.2). Presence of other capillary changes, including presence of hemorrhages and presence of subpapillary plexus were also noted in a few patients (n=3, 8.8% each).

Figure 2.

Nail fold capillaroscopy imaging findings for nail psoriasis (NP): decreased capillary density/length, increased abnormal morphology including branching, meandering, bushy/ramified, and/or bizarre capillaries (field: 1 millimeter)

Both fingernail onychomycosis (n=8) and toenail onychomycosis (n=13) cohorts showed increased frequency of abnormal morphology (P<0.02), as well as increased frequency of meandering capillaries (75.0% for fingernails, 76.9% for toenails, P<0.01) compared to controls (Figure 3, Supplementary Figure 4). Within the onychomycosis cohort, other abnormal morphologies were much less frequently seen than meandering capillaries, including branching (0%−7.7%) and bizarre (12.5%−30.8%) capillaries. Presence of hemorrhages and subpapillary plexus were only noted in one case each (4.8%).

Figure 3.

Nail fold capillaroscopy imaging findings for onychomycosis (OM): increased frequency of meandering capillaries (field: 1 millimeter)

NLP (n=11, P<0.01) and onychopapilloma groups (n=6, P=0.01) both showed decreases in average capillary length compared to controls (Figure 4, Supplementary Figure 5). The retronychia (n=12) cohort also showed decreased average capillary length (P=0.03), as well as increased frequency of disorganized polymorphic capillaries compared to controls (P=0.02) (Supplementary Figure 6). The only other nail conditions that also showed disorganized polymorphic capillaries included NLP (18.2%) and NP (5.9%), but much less often than that of retronychia (50.0%). The BNS (n=13), idiopathic/traumatic onycholysis (n=7), and eczema (n=6) groups showed no differences in any capillary parameters compared to healthy controls (P>0.05).

There were also differences in NC parameters between nail groups (Supplementary Table 3). The NP group showed decreased average capillary density compared to both onychomycosis (P=0.0002) and LP (P=0.03) cohorts, as well as decreased average capillary length compared to the idiopathic/traumatic onycholysis cohort (P=0.01). Onychomycosis patients had increased frequency of meandering capillaries compared to both NP (P=0.01) and NLP (P=0.04) patients.

Amongst the NP cohort (n=34), average age was 39.2 years (range 8–72), with equal distribution of male/female patients and average disease duration of 3.5 years (Supplementary Table 4). Eleven patients (32.3%) had concomitant PsO and two (5.9%) had concomitant PsA. Average capillary density was 7.56±1.57 capillaries per millimeter and average capillary length was 130.26±45.54 micrometers, with no differences based on age (40+ vs. <40 years), gender (male vs. female), or race/ethnicity (white vs. non-white). Nineteen patients (55.9%) had abnormal capillary morphology, which was more common in NP patients 40+ years of age vs. <40 years (80.0% vs. 36.8%, P=0.01), and in male vs. female NP patients (76.5% vs. 35.3%, P=0.02). There were no differences in capillary dimensions or capillary morphology changes based on any clinical variable (disease duration, concomitant PsO, matrix/bed involvement; all P>0.05).

Discussion

Our study demonstrates that NC imaging is potentially a valuable, non-invasive, in vivo technique for studying microvascular abnormalities in nail diseases. We identified new capillaroscopic characteristics for eight dermatological conditions, which provides a framework for future studies assessing NC as a tool to aid in clinical diagnosis.

We found that NP patients demonstrated decreased capillary density/length, increased crossing/tortuous capillaries, as well as increased abnormal capillaries compared to controls. Notably, previous psoriasis studies have not examined NC changes in isolated NP patients, which comprised the majority (67.6%, n=23) of our NP cohort. Our NP findings were similar to that of a systematic review of 22 PsO/PsA NC studies (with or without NP), demonstrating reduced density/length and increased abnormal morphology vs. healthy patients.² The decreased density and length in psoriasis patients might be due to reduced angiogenesis and blood flow to the nail fold triggered by psoriatic inflammatory responses and subsequent thickening of vessel walls.^11,12 Furthermore, 56% of patients in our NP cohort demonstrated abnormal capillary morphology, similar to the 63% frequency identified in a prior study of 19 NP patients.¹³ Notably in our study, although there was a statistically significant increase in any abnormal morphology, there was no correlation of NP with any specific singular type (branching, meandering, bushy/ramified, vs. bizarre). Prior studies have also shown an increase in avascular areas amongst PsO/PsA patients compared to controls, which did not reach statistical significance in our NP dataset.^14–16

We identified important novel findings for capillary measurements amongst NP patients. Overall, NP patients had increased frequency of abnormal capillaries compared to controls, which was statistically higher in male vs. female patients, and in older (40+) vs. younger (<40) patients. We found no correlation of capillary changes based on NP disease duration (3+ years vs. <3 years), and no correlation based on nail matrix vs. bed disease. It was hypothesized that NC was more important for diagnosis of nail matrix vs. bed disease.^17,18 However, we show that regardless of disease duration or nail area involvement, NC may serve as a valuable diagnostic tool with identifiable differences between NP patients and controls.

Overall, findings in NP patients demonstrate that NC may have significant value as an adjunct in early diagnosis and treatment initiation of affected patients, especially in those with isolated NP. A prior study of 87 patients with isolated NP demonstrated an average 2.9 years delay in diagnosis and 11.5% presence of concurrent PsA (most often diagnosed concurrently with NP), emphasizing the importance of earlier clinical recognition.⁷ The same study also demonstrated that histological features of NP were often non-specific and inconsistent (79% of NP patients with parakeratosis, 48% with neutrophilic infiltration, 32% with incomplete keratinization), highlighting the need for additional diagnostic tools to aid in NP diagnosis.⁷ NC may thus serve as a quick and useful clinical imaging resource to address this gap, and could complement histopathological findings. In further characterizing these NC findings, future studies can help determine whether capillary changes correlate with NP severity and if capillaroscopic findings change with treatment.

Our study also delineated capillary parameters for a variety of other common nail diseases. Notably, onychomycosis patients showed not only an increased proportion of patients with any abnormal capillary morphology, but also specifically an increase in the meandering type compared to controls, NP, and NLP patients, which was similar for both fingernail (75%) and toenail onychomycosis (77%). Further studies can help to verify this correlation, and corroborate if high frequency of meandering capillaries can aid in differentiating non-specific onychodystrophy of onychomycosis vs. NP or NLP. Likewise, both NLP and onychopapilloma patients demonstrated decreased capillary length, compared to controls.

The retronychia cohort demonstrated a higher proportion of patients (50.0%) with disorganized polymorphic capillaries than any other nail condition, a feature that may aid in diagnosis, and in differentiating from onychomycosis and traumatic onychodystrophy. Diagnosis of retronychia is often missed due to its complex clinical nature and lack of established diagnostic criteria other than ultrasonography. Ultrasound diagnostic criteria includes presence of hypoechoic halo surrounding origin of nail plate, decreased distance between origin of nail plate and base of distal phalanx, and proximal nail fold thickening, but such imaging requires specialized expertise to perform and interpret.¹⁹ Nail clippings of retronychia patients may often demonstrate hyphae, but is due to secondary colonization. We demonstrate that presence of disorganized polymorphic capillaries may potentially be one feature supporting diagnosis of retronychia.

Notably, capillary measurements for diagnoses of idiopathic/traumatic onycholysis and BNS yielded no statistically significant differences from controls, which could be especially useful in differentiating onycholysis due to idiopathic/traumatic causes vs. onycholysis from NP, with the former showing a “normal” capillaroscopy picture. Our eczema group (n=6, all with nail findings) also showed no differences from controls in capillary findings, in contrast to a prior study that found decreased capillary density and increased minor morphological changes in eczema (n=2 with nail findings) vs. control patients.⁸ Studies with larger sample sizes can be utilized to clarify these results, and develop algorithms for identifying nail diseases based on these NC measurements.

While these findings require corroboration, our initial results suggest that NC could be particularly useful for diagnosis of NP, onychopapilloma, and retronychia, and as an adjunct for onychomycosis diagnosis along with mycological confirmation. Histopathological findings for NP are often nonspecific, definitive diagnosis of onychopapilloma requires nail biopsy, and there is no diagnostic test for retronychia. Therefore, diagnosis of these conditions with NC could expedite treatment for these patients and circumvent the need for invasive procedures.

Limitations include small sample size for certain nail conditions and being a single physician/center study. Within the NP group, only two patients had PsA, preventing comparison of NC changes in patients with and without PsA. Uneven surface, scaling, onycholysis, and/or thickening made certain capillaroscopy pictures difficult to capture and/or interpret (and thus were excluded), especially with toenails.

Other limitations for adaptation of NC to clinical practice include cost, training, time, and billing. USB digital handheld capillaroscopes are available from manufacturers (Dino-Lite, GOKO Imaging, Italeco, Inspectis), costing $150-$1,200, which is a relatively large initial expenditure.^20–21 However, compared to other diagnostic techniques for nail conditions, including added costs of every nail clipping with histopathology, culture, or polymerase chain reaction testing, or time utilized to perform potassium hydroxide preparation tests with microscopy, NC may be more cost-effective and efficient in the long term. Furthermore, although many imaging techniques often require specialized equipment training or significant time to operate, NC is relatively easy to learn, as well as fast to perform (<1 minute).^1,2,9 Billing practices may also limit wide adaptation of NC use, as there is currently no associated reimbursement. Future advocacy efforts can help to introduce the necessary coding policies to allow for NC billing by dermatologists.

Conclusion

Overall, our pilot study provides valuable knowledge for utilization of NC for diagnosing onychodystrophies, characterizing and comparing NC profiles across eight groups. NP is associated with decreased capillary length/density and increased abnormal capillaries, with higher frequency in older, male patients. Onychomycosis is associated with increased meandering capillaries compared to controls, NP, and NLP. NLP and onychopapilloma are both associated with decreased capillary length. Retronychia is associated with increased disorganized polymorphic capillaries compared to controls and onychomycosis. Our findings highlight NC as a potentially quick, cost-effective, and non-invasive imaging modality as an adjunct for early diagnosis and treatment initiation for patients with onychodystrophies.