One in three or four patients with psoriasis will have psoriatic arthritis. If nail psoriasis is present, there is a 3-fold increased risk of developing psoriatic arthritis in patients with psoriasis, particularly if onycholysis occurs, and nail involvement is also associated with distal interphalangeal arthritis [28].
Unlike rheumatoid arthritis, in which synovitis is the main unit of involvement, in psoriatic arthritis, the main process is enthesitic and periarticular and is not necessarily associated with synovitis. Therefore, the physical evaluation in spondyloarthritis is more nonspecific because we may be faced with active psoriatic arthritis without florid arthritis. There is difficulty in the clinical evaluation of entheses, especially when the symptoms are less severe, which may delay the diagnosis, as enthesitis may have mechanical causes, and it may take time to identify that there is an inflammatory cause. Thus, technologies have emerged to assist in the management of these patients, among which ultrasound is particularly favorable.
The changes in the morphology of the nail plates were classified according to one of the pioneer descriptions of nails with psoriasis provided by Dr. Ximena Wortmsan from Chile in 2004, who defined four types of morphological changes in nails with psoriasis according to GS imaging, producing the most frequently used system in subsequent studies.
Most studies that evaluated the distance (or thickening) of the nail bed found a significant difference between patients with psoriasis/psoriatic arthritis and controls [12, 15, 16, 18, 23]. According to Aydin et al. [13], there was no association between nail plate thickening and distal interphalangeal involvement. Aydin et al. [13] and Kasman et al. [12] identified greater total nail thickening in psoriatic patients (psoriasis and psoriatic arthritis) than in controls, and Kasman et al. [12] confirmed this finding for patients with psoriatic arthritis, while De Rossi et al. [14] did not find such a difference between the psoriatic groups and controls. Aydin et al. [13] and Mondal et al. [15] identified a higher mean matrix in patients with psoriatic arthritis than in controls, and Acquitter et al. [29] found greater matrix thickening in patients with clinically present nail psoriasis.
Gutierrez-Manjarrez et al. [19] identified normal measurements for the nail bed up to 2.5-3 mm after comparing patients with psoriatic arthritis and controls, while Sandobal et al. [16] found that a 2 mm nail bed could discriminate patients with psoriasis or psoriatic arthritis from healthy controls or controls with rheumatoid arthritis.
Wortsman et al. [18] obtained measurements such as the interplate distance (IPD) and the distance between the ventral plate and the bone edge of the distal phalanx (nail bed thickness), as shown in Fig. 1, and identified a significant difference between control and psoriatic participants, the latter of whom had higher values. This difference was also found by Sandobal et al. [16] between patients with psoriasis or psoriatic arthritis and healthy controls or those with rheumatoid arthritis, as well as in a comparison of patients with psoriatic arthritis and controls in subsequent studies [12, 15, 23]. In the present study, we did not find such a difference in nail bed thickness between patients with psoriatic arthritis and controls (1.79 versus 1.67 mm; p = 0.073), in agreement with the findings of De Rossi et al. [14].
Nail ultrasound studies vary regarding the number and which fingers are examined, and thus demonstrate different results. Regarding the number of digits examined, studies have reported values ranging from 2 digits—corresponding to the nail most clinically affected and that of the contralateral finger [13, 18] —or 3 fingers of the dominant hand plus fingers with clinical onychopathy or pain in the DIJs [16], or second and third fingers bilaterally[14], up to 10 fingers [15] or 12 digits (the 10 fingers of the hands and the 2 halluces) in the ultrasound evaluations [12]. Given the variability of the results in the literature, we chose to evaluate all the digits of the hands and verify if there were differences between the fingers for this study population according to the parameters evaluated. The finger most affected was the right thumb (44.2%), followed by the left thumb, both also identified by Wortsman et al. [18], with regard to nail plate changes and the mean score (which considers nail plate changes and the presence of synovitis, paratendinitis and enthesitis), where the right thumb was also the most affected, followed by the left thumb, second and third fingers of the right hand. According to Mondal et al. [15], who also evaluated the 10 fingers of the hands, the right thumb was also the finger that most frequently presented with Wortsman’s alterations (93.33%). Acer Kasman et al. [12], who evaluated 12 digits (including the 2 halluces), found that the most discriminative fingers varied according to the parameter, each of which was evaluated individually: the left thumb for the nail plate index, the fifth finger of the left hand for nail bed and total thickening, and the third finger of the right hand for Doppler activity; in contrast, Mendonça et al. [23] found that the second and third fingers had the greatest changes on spectral Doppler and the lowest resistivity index, respectively. Undoubtedly, when evaluating all the fingers, the examination time is longer, which can reduce the number of people willing to participate in the study. Nevertheless, we were able to show that the fingers most subjected to microtrauma are the ones most often affected, which corroborates the Koebner phenomenon for psoriatic arthritis.
To assess inflammation of the nail matrix and bed, in addition to measuring the thickness of these structures, local vascularization can also be analyzed using power Doppler imaging according to the grading proposed by Gutierrez-Manjarrez et al. [19] and quantified by measuring the nail vessel resistivity index (RI) by Doppler spectral imaging. Both power Doppler and the RI calculated by spectral Doppler are related to the neovascularization of inflammatory processes. Studies that evaluated the vessel RI in the nail matrix achieved divergent results. Among patients with psoriasis, both greater [30, 31] and sometimes lower values have been obtained with respect to controls [32]; among patients with psoriatic arthritis, the vessel RI is sometimes lower than in control patients [23]; and in a study comparing patients with psoriasis, psoriatic arthritis and controls, no difference was identified between groups [14]. Regarding the resistivity index, assessed in participants with positive power Doppler findings, no patient had an RI < 0.4. The vessel resistance index is still inconclusive in the evaluation of these patients. We believe that because it is measured from microvessels, the difference in Doppler sensitivity of the different machines may interfere with the results.
The power Doppler signal at the nail bed and matrix was more frequent among psoriatic patients (44.2%) than among controls (6%), and the mean score according to the Gutierrez et al. [27] classification was also significantly different - mean power Doppler 0.11 ± 0.05 (0-0.67) for the group with psoriatic arthritis versus 0.01 ± 0 (0-0.10) (p < 0.001) for the control group. This finding is in agreement with what Kasman et al. [12], Arbault et al. [26] and Sandobal et al. [16] found, while De Rossi et al. [14] and Mendonça et al. [23] did not find a difference between the groups.
Regarding the evaluation of the distal interphalangeal joints, the findings for grayscale synovitis were different between the groups, while for power Doppler, no signal was found in any participant. We considered this finding related both to the power Doppler sensitivity of the machine used and to the fact that most patients were on biological medication, circumstances under which the Doppler findings are the most rapidly modified. Positive power Doppler of the distal interphalangeal joint as a discriminative finding is not always investigated because there tends to be greater focus on bed/matrix and nail enthesis, but in studies by Sandobal et al. [16] and Arbault et al. [26], a difference was found between psoriatic and control individuals.
The correlation between ultrasound nail involvement and disease activity varies in the literature, both with regard to the scores compared and to the findings. Nail ultrasound identified more changes in psoriatic patients, even in clinically normal nails [13, 15, 16, 26], although these changes were more frequent when the nails were clinically affected, i.e., there is a good correlation with the clinical nail data verified by the NAPSI or modified NAPSI [13, 15, 29]. We also identified a weak but significant correlation between clinical nail involvement as verified by NAPSI and ultrasound alterations of the nail plate according to the Wortsman et al. [18] classification, both for the presence of this finding and for the mean severity of change.
While Mondal et al. [15] found a moderate correlation between the NAPSI and nail matrix thickness (Spearman r = 0.411, 95% CI: 0.125–0.634, p = 0.005), this analysis was negative in our data. However, we identified a correlation between the mean score of the 10 fingers—which considers abnormal nail plates, the presence of enthesitis of the distal finger extensor, the presence of paratendinitis, grayscale and power Doppler synovitis of the distal interphalangeal joint, and positive power Doppler of the bed or nail matrix–and the NAPSI (Spearman (r = 0.46, p = 0.001).
There is a known positive relationship between nail psoriasis and more severe cutaneous psoriasis with a higher risk of psoriatic arthritis. We found a moderate correlation between the cutaneous involvement as calculated by the PASI and the number of fingers with paratendinitis. This is not the first report of a correlation between the severity of cutaneous disease and ultrasound findings of the nails. Mondal et al. [15] previously identified a correlation between the PASI and nail matrix thickness (Spearman r = 0.302, 95% CI 0.0007–0.553, p = 0.043). No correlation was found between BSA and nail findings on US.
Regarding the composite indices, we did not find an association between the MDA values and the ultrasound findings, but an association was identified with the DAPSA, unlike Mondal et al. [15], who found no such correlation.
When the patients were divided into the different DAPSA classifications (remission, low, moderate and high activity), we did not find an association with the ultrasound findings, but when they were grouped into DAPSA ≤ 14 (remission or low activity) or > 14 (moderate or high activity), we found significant differences; specifically, when DAPSA > 14, patients were more likely to have fingers with greater involvement of the nail plate (grades 3 or 4), greater mean changes in the nail plate and greater mean scores of the 10 fingers. We also identified a weak correlation between DAPSA values and the number of fingers with more severe alterations (grades 3 or 4) of the nail plate (r = 0.30, p = 0.032). We found no association between MDA values and ultrasound findings.
While Arbault et al. [26] found no correlation between disease activity by ASDAS-CRP or CRP alone and any ultrasound parameters in patients with psoriatic arthritis, we found a moderate correlation between the ASDAS-CRP and the number of fingers with nail plate changes, the number of fingers with more severe nail plate changes (3 or 4), the mean of this nail change, and the mean score of the 10 fingers (Spearman: r = 0.57, p = 0.034; r = 0.59, p = 0.026; r = 0.58, p = 0.030; r = 0.66, p = 0.010, respectively), but no correlation was found between BASDAI and the ultrasound findings [1].