Article Text

Download PDFPDF

Concise report
‘Deep Koebner’ phenomenon of the flexor tendon-associated accessory pulleys as a novel factor in tenosynovitis and dactylitis in psoriatic arthritis
  1. Ilaria Tinazzi1,
  2. Dennis McGonagle2,
  3. Sibel Zehra Aydin3,
  4. Donatella Chessa4,
  5. Antonio Marchetta1,
  6. Pierluigi Macchioni5
  1. 1 Unit of Rheumatology, Sacro Cuore Don Calabria Hospital, Negrar-Verona, VR, Italy
  2. 2 NIHR Leeds Musculoskeletal Biomedical Research Unit, Leeds Teaching Hospitals Trust and The University of Leeds, Leeds, UK
  3. 3 Department of Rheumatology, University of Ottawa, The Ottawa Hospital Research Institute, Ottawa, Ontario, Canada
  4. 4 Medicina Interna, Ospedale Paolo Dettori, Tempio Pausania, OT, Italy
  5. 5 Rheumatology Department, Ospedale S.Maria Nuova, Reggio Emilia, Italy
  1. Correspondence to Dr Ilaria Tinazzi, Unit of Rheumatology, Ospedale Sacro Cuore Don Calabria, Negrar VR 37024, Italy; ilariatinazzi{at}


Objectives Skin and joint involvement in psoriasis (PsO) and psoriatic arthritis (PsA) are thought to relate to the so-called Koebner response. Given that dactylitis is non-randomly distributed in the digits, this study tested the hypothesis that the accessory pulleys linked to the flexor tendons were thickened in PsA and thus exhibited koebnerisation.

Methods Ninety-six subjects (27 PsA, 27 rheumatoid arthritis (RA), 23 PsO and 19 healthy controls (HCs)) were enrolled. The A1, A2 and A4 pulley thickness was measured using a high-resolution probe (22 MHz). All patients were in remission or low disease activity with current dactylitis being excluded.

Results Within 864 pulleys investigated, patients with PsA had thicker pulleys in every digit compared with both RA (P<0.001 and P=0.003) and HCs (P<0.001). RA and PsO groups had some pulleys in some digits thicker than HCs whereas some others were comparable. The second digit A1 pulley thickness was higher in patients with PsA with previous dactylitis (P=0.020). More pulleys were thickened in the PsA group (165/243, 68%) than RA (41/243, 17%; P<0.001) and HCs (13/171, 7.6%; P<0.001).

Conclusions In established PsA, the accessory pulleys are thickened compared with RA, PsO or HCs and especially in subjects with a history of dactylitis. These findings implicate the involvement of pulleys in PsA-related tenosynovitis and dactylitis supporting the idea of deep koebnerisation in dactylitis and sites of high physical stress.

  • psoriatic arthritis
  • rheumatoid arthritis
  • ultrasonography
View Full Text

Statistics from


Dactylitis is a hallmark of psoriatic arthritis (PsA) occurring in around 40% of cases at some point in the disease course.1 The original MRI studies of dactylitis suggested that the main cause of this pathology was flexor tenosynovitis in addition to soft-tissue oedema, joint synovitis and osteitis.2 3 It has been suggested that enthesitis is the primary lesion in spondyloarthritis (SpA) including PsA.4 In support for a role for the enthesis in dactylitis pathogenesis, high-resolution MRI studies showed that enthesitis was common in distal interphalangeal joints PsA involvement pointing towards a link between small joint swelling and dactylitis.5 Also, recent MRI studies showed entheseal changes in the digital pulleys at flexor tendons in dactylitis.6

The digital pulleys have constituent fibrocartilage and have the role of tracking the tendon accurately maintaining the apposition of tendon and bone across the joint to restrict ‘bowstringing’ of tendon during flexion.6 Ultrasonography (US) has increasingly being used for its ability and sensitivity to visualise soft tissues in the early stages of inflammatory arthritis.7 8 The A1 pulley can be visualised using high-frequency US,9 a structure that was demonstrated to be abnormal in PsA dactylitis.6 Also, the pulleys are frequently thickened as determined by US imaging of trigger fingers.10 11

It has long been considered that disease expression in PsA might relate to abnormal responses to physical stress with Koebner response in the skin and ‘deep Koebner’ response in the joints.12 The small digital pulleys are sites of high physical stress. Therefore, we investigated the digital pulleys in PsA including cases with previous dactylitis and compared this with rheumatoid arthritis (RA) (where flexor tenosynovitis is also common), psoriasis (PsO) subjects13 and healthy controls (HCs). Since our pilot work on the reliability of scanning the pulleys had shown a better agreement for A1, A2 and A4 pulleys, we only focused to those for this study.14


Following local ethical committee approval, written informed consent was obtained from all participants. Consecutive patients with RA (n=27) fulfilling the American College of Rheumatology/European League Against Rheumatism criteria,15 patients with PsA according to CASPAR16 criteria (n=27), PsO without arthritis (n=23) and HCs without any inflammatory rheumatological disease (n=19) were enrolled from two rheumatology and dermatology centres (Verona and Reggio Emilia, Italy).

The inclusion criterion for PsA and RA groups was a history of hand involvement; however, only patients with PsA with minimal disease activity17 for at least 6 months prior to US examination and patients with RA with low disease activity18 were recruited. Patients on biologics, with present active dactylitis, chronic renal failure and diabetes mellitus, were excluded.

Clinical assessments were done by a rheumatologist in each centre (AM, DC). All US was performed blinded to the clinical assessment and diagnosis, by the same experienced investigators all the time (IT, PM) using a MyLabClassC (Esaote, Genova, Italy) equipped with a 10–22 MHz linear transducer using a pre-agreed scanning protocol with reliability testing (details in online supplementary text).

Supplementary file 1

The dominant-hand second to fourth digit volar aspects were imaged in longitudinal and transverse planes. Two measurements (transverse and longitudinal) were done for the A1 pulley at the metacarpophalangeal joint level. The border between the inner rim of the A1 pulley and the flexor tendon was defined by dynamic examination. The A2 and A4 pulley thicknesses were assessed in full extension, in correspondence of the proximal and middle phalanx in both transverse and longitudinal planes measuring them from their thickest site (figure 1A–F).

Figure 1

Transverse (A, B and C) and longitudinal (D, E and F) views of A1, A2 and A4 pulleys, respectively, in a healthy control. The thickened pulleys can be seen in transverse (G, H and I) and longitudinal (J, K, L) views, in the same order of pulleys, in a patient with psoriatic arthritis.

Statistical analysis

The intraclass correlation coefficient values between longitudinal and transverse measurements were analysed. To compare the measurements between the two groups with a continuous scale, the independent two-sample Student t-test or the Mann-Whitney U test were used and dichotomous data were compared using the χ2 test or Fisher exact test, as appropriate. The upper normal value for the thickness of the pulleys was defined as 2 SD above the measurements obtained from HCs. The number of pulleys with increased thickness was compared for patients with PsA and RA. The following possible explanatory variables were entered in a multiple regression analysis with the number of pulleys with increased thickness per patient as the dependent variable: age, sex, body mass index (BMI), duration of disease, clinical diagnosis and manual work. The study was explorative in nature and a power calculation was not performed. Statistical analysis was performed using SPSS V.22 (SPSS, Chicago, Illinois, USA).


The patient characteristics are given in table 1. Patients and HCs were matched for BMI and age. The patients with PsA and PsO were predominantly men and patients with RA and HCs were predominantly women (table 1).

Table 1

Demographic and clinical characteristics of the different groups of patients

The pulleys were identifiable in both planes in all patients and all HCs. A total of 1732 measurements were done in 864 pulleys. The intraclass correlation coefficient values between longitudinal and transverse measurements showed excellent agreement (0.82). As the longitudinal scans permitted detection of the thickest site displaying the whole length of the pulley, the longitudinal scan results are given below (table 2). The transverse measures are also shown in online supplementary table 1.

Supplementary file 2

Table 2

Pulley thickness in different groups of patients and comparisons

Patients with PsA had thicker pulleys in every digit compared with RA and HCs (figure 1G–L). Patients with RA had thicker pulleys than HCs, especially in the A1 pulley, but A2 pulley of the second digit and A4 pulley of the fourth digit were not different. Patients with PsO had thicker pulleys than HCs but not as thick as patients with PsA. In particular, patients with PsA had thicker A1 and A4 pulleys than PsO, whereas A2 pulleys were not statistically different in every digit. More pulleys were thickened in the PsA group (165/243, 68%) than RA (41/243, 17%; P<0.001) (online supplementary table 2). The multiple regression analysis entering age, sex, BMI, duration of disease, diagnosis and manual work demonstrated that only the diagnosis of PsA was associated with the number of thickened pulleys (OR 4.8 (95% CI 3.3 to 6.3); P=0.001).

Supplementary file 3

In order to understand if previous episodes of dactylitis could modify pulley anatomy and increase their thickness, we performed a subanalysis by comparing patients PsA with or without a dactylitis history. Only the second-digit A1 pulley was found thicker in patients who had previous dactylitis. The mean thickness of PsA pulleys still remained significantly higher compared with RA when patients with PsA with previous dactylitis were excluded, except for A1 pulley of the second finger (online supplementary table 3). Patients with trigger finger did not have more frequent dactylitis in the past and the thicknesses in any of the pulleys were not higher than those of patients without trigger finger (data not given).

Supplementary file 4


Given the emerging link between the flexor tendon pulleys in dactylitis in PsA, we hypothesised that these structures may be abnormally thickened in PsA cases compared with RA, PsO and HCs and thus may exhibit deep Koebner response. The findings herein confirmed that the pulleys were indeed thickened in subjects with PsA compared with RA. The thicker pulleys in RA compared with HCs might point towards a non-specific effect of a chronic autoimmune tenosynovitis on the adjacent pulleys. However, the greater magnitude of pulley thickening in PsA, especially in the setting of dactylitis, suggests an intrinsic pathology in the pulley contributing to PsA-related tenosynovitis.

Skin injury in PsO is associated with epidermal hyperplasia and thickening. Likewise, the skeletal phenotype in PsA and SpA is often associated with excessive repair responses that manifest not as epidermal hyperplasia but as postinflammation skeletal hyperplasia. Also, subjects with PsO, but without clinical arthritis, often exhibit thickening of non-diseased large entheses. Given that the accessory pulleys are a type of mini-entheses network that anchor the tendons to the bone and are subjected to very high physical stress, we tested the hypothesis that these likewise exhibit ‘deep Koebner’ phenomena and are thickened. Indeed, these findings suggest that such a mechanism is possible given the frequency of thickening of these structures in PsA. It will be important to show that such changes are present in early PsA where their presence would support the idea of early biomechanical alterations in tendon function that could contribute to dactylitic tenosynovitis.

This study had a number of limitations. First, we scanned three digits of the dominant hand. Including patients with PsA with hand involvement may not be representative of all PsA population, although hands are frequently involved in peripheral disease. Studies in early PsA to ascertain whether these pulley changes trigger PsA tenosynovitis or whether they represent a reactive process are needed. We defined the cut-off for the thickness of a pulley as 2 SD above the measurements in HCs; this is not a validated measure as there are no data published on it and need replication. In addition, our sample size was not very large, yet we were still able to demonstrate differences among groups. Despite these limitations, our study suggests a high frequency of micro-entheseal changes in PsA pulleys.

In conclusion, pulleys are involved in PsA, even in patients who are in minimal disease activity suggesting a finding of tissue damage rather than inflammation. The difference between PsA and RA suggests that pulley involvement is not purely dependent on tenosynovitis. The difference between PsO and HCs may point to pulley involvement being an early finding, as previously shown with enthesitis.


View Abstract


  • Handling editor Josef S Smolen

  • Contributors IT, DMG, SZA and PM gave substantial contributions to study conception and design. IT, DC, AM and PM gave substantial contributions to acquisition of data. IT, DMG, SZA and PM drafted the article and revised it critically for important intellectual content. IT, DMG, SZA, DC, AM and PM gave final approval for the version of the article to be published.

  • Funding This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests DMG received grants and/or honoraria from Abbvie, Celgene and Janssen. SZA received grants and/or honoraria from Abbvie, Novartis, UCB, Sanofi and Pfizer.

  • Patient consent Obtained.

  • Ethics approval The Ethics Committees of ‘Sacro Cuore’ Hospital Negrar (VR), Italy, and Ospedale S. Maria Nuova, Reggio Emilia, Italy, approved the study.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.