Trigger finger or flexor stenosing tenosynovitis of the finger flexor apparatus is a common reason for consultation in primary care, rheumatology and hand surgery. Its prevalence in the general population has been estimated at 2% and up to 7% in diabetics, with a slight predominance in the female sex and in the dominant hand1.

Trigger finger is a limitation for the extension of the finger once it has been flexed2. It is due to a biomechanical incongruence between the space left by the pulleys (flexor canal)3, mainly A1, and the volume occupied by a specific portion of the finger flexor apparatus composed of the flexor digitorus superficialis and flexor digilatil profundus in the long fingers and by the flexor pollicis longus in the thumb. This incongruence may be due to a thickening of the pulley, a thickening of the tendinosous apparatus or a combination of both these factors4.

Diagnosis is eminently clinical, based on the history of episodes of an impossibility to extend the finger which requires passive assistance2,4,5. Occasionally the thickening is palpable and infrequently the patient has pain or loss of strength2,6. Performing imaging tests is not mandatory for diagnosis. but due to its low cost, innocuousness and relative accessibility, ultrasound is able to distinguish the aetiology of the process on a level of pulley-tendon complex and can be used to guide local treatment3,4,7. Stratification of symptom severity is made using the Quinnell8 clinical scale. This scale has 4 ordinal categories: 1) Crepitus without blockage of mobilisation, 2) actively resolving extension blockage, 3) extension blockage resolved by assistance and 4) irreducible blockage. Traditionally, when conservative management fails, and before release surgery, treatment with corticosteroid infiltrations at the level of the flexor sheath at pulley height isused9. This procedure does not, however, ensure a response in all cases, as approximately 30% will require surgery10.

The aim of this study was to determine the factors associated with the therapeutic success of corticosteroid infiltrations in patients diagnosed with trigger finger at A1 pulley level.

A prospective longitudinal study was conducted between May 2019 and June 2021. Consecutive patients were recruited from a musculoskeletal practice, a rheumatology practice and a general orthopaedic and trauma practice. Inclusion criterion was clinical diagnosis of the trigger finger at A1 height, confirming in the corresponding consultation any finger of the hands and the therapeutic failure of general measures. Patients with trigger fingers on specular fingers of both hands were excluded as they made comparative ultrasound study impossible. Patients who had been subjected to infiltrations with corticosteroids or fibrinolytic agents, radiotherapy or previous trigger finger operations, or previous palmar fascia disease in either hand were also excluded.

Once the patient had signed a consent form to participate in the study an ultrasound study was performed and an injection was made in accordance with standard clinical practice of a preparation of 20 mg of triamcinolone acetate and 0.5 cc of 2% mepivacaine. The total volume of the mixture was 1 ml, although only 0.3-0.5 ml was administered during infiltration. The needle was placed at the proximal palmar crease as a starting point for position adjustment using ultrasound guidance, coinciding with the anatomical position of the A1 pulley11. The ultrasound was performed using a Logiq 9S (General Electric®) ultrasound scanner with an 18 MHz “Hockey stick” probe. During the ultrasound scan, the thickness of the A1 pulley and the flexor tendon of the affected and contralateral finger were measured. Measurements were made with the finger in full extension. All infiltrations were performed under real-time ultrasound guidance with the probe positioned transverse to the longitudinal axis of the corresponding flexor tendon. The infiltrations was made inside the flexor tendon sheath in its adjacent portion and at the level of the A1 pulley, using a subcutaneous needle and standard aseptic measures.

Patients were grouped according to the detected ultrasound finding: thickening of the pulley (> 15% of the thickness of the pulley of the affected finger compared with the contralateral one) or thickening of the flexor tendon (> 15% of the thickness of the affected finger compared with the contralateral one). To determine the severity of the clinical symptoms, the original Quinnell scale was used before and 8 weeks after infiltration.

Statistical analysis was fixed at 2 primary endpoints: achieving Quinnell grade I and lowering to at least one ordinal category on the Quinnell scale. A bivariate association analysis was performed between the independent demographic, clinical and baseline ultrasound variables. A predictive (forward Wald) modelling approach was planned in which the variables that reached statistical significance P < .10 were included.

This study was approved by the Research Ethics Committee of our centre on 5th June 2019 (EXP 126/19).

The study included 74 patients. Three patients rejected the therapeutic approach with infiltrations, 2 did not return for a check-up and one was referred for surgery with a Quinnell IV. Forty-eight patients (64.8%) came from a musculoskeletal practice, 15 (20.2%) from a general orthopaedic and trauma practice and 11 (14.8%) from a rheumatology practice. Mean ± standard deviation of the 68 patients who participated in the study was 59.08 ± 6.91 years. Forty (58.8%) were men. Twenty-six patients (38.2%) were classified as Quinnell II and 42 (61.8%) as Quinnell III. Table 1 summarises the participant group characteristics according to the Quinnell scale medical symptom grading.

Eight weeks after infiltration, 22 patients (32.4%) achieved a Quinnell grade I and 50 (73.5%) recorded an ordinal reduction of their clinical scale of at least one grade.

In the bivariate correlation study with complete resolution (Quinnell I), tendon thickening was detected in 18/22 patients (81.8%) who achieved this objective and in 13/46 (28.3%) who did not achieve the primary objective (OR: 1.43; CI 95%: 3.24-40.24; P < .001). Progression time of patients who achieved clinical resolution was 11.5 ± 3.56 months and 8.86 ± 3.41 of those who did not achieve the primary objective (t = 2.92; P = .005). In the bivariate correlation study with partial resolution (reduction of a Quinnell grade), tendon thickening was detected in 28/50 patients (56%) who achieved this objective and in 3/18 (16.7%) who did not (OR: 6.36; 95% CI: 1,63-24,78; P = .005). Progression time in the group who achieved partial resolution was 10.36 ± 3.54 and in the group who did not achieve it this was 7.94 ± 3.45 (t = 2.49; P = .015). The other variables were not statistically significant. Table 2 summarises the results of the bivariate correlation studies.

For binomial logistic modelling, the tendon thickening variables (HR: 10.72; 95% CI: 2.88-39.93; P < .001) and progression time (HR: 1.23; 95% CI: 1.02-1.49; P = .027) were included in the predictive model of treatment success (primary objective: Quinnell I). For the partial success model (reduction by one Quinnell grade), the variables which were statistically significant within the model were tendon thickening (HR: 5.57; 95% CI: 1.38-22.41; P = .016) and progression time (HR: 1.18; 95% CI: .99-1.41; P = .051). No multicollinearity was identified between the variable time of progression and tendon thickening.

Regarding adverse events reported by patients, these were presented by 6 individuals over follow-up: 4 cases of pain and functional limitation, which required anti-inflammatory treatment and which in all cases was limited to under 2 of the first weeks, one case of vagal decompensation without loss of consciousness during the procedure and one case of local erythema which persisted for 72 h before subsiding without the need for any treatment. No infections, tendon ruptures or hyperglycaemic decompensations were reported.

The detection of tendon thickening at pulley level is the most important factor in the outcome of trigger finger infiltrations with corticosteroids. The chronic nature of the process also appears to lead to favourable infiltration outcomes.

This study was not exempt from limitations. On the one hand, we exclusively focused on the diagnosis of A1 pulley dependent trigger finger which —although the most common— does not represent all cases. The A1 pulley is larger and easier to identify by ultrasound. This case selection homogenized the type of condition studied. Another limitation was that the infiltration was ultrasound guided to guarantee administration of the treatment in the location recommended by the literature. However, this type of approach is not the one most commonly used in standard clinical practice. For this reason, the effectiveness described in our results could be higher than that observed in settings where infiltration was performed using only anatomical references. Notwithstanding, ultrasound guidance is relevant in putting treatment at pulley height without demonstrating that intra-synovial administration really is more effective than extra-synovial9. Finally, our study did not cover recurrence as an outcome variable. Other studies report that trigger fingers resolved with infiltrations may recur in 30% of cases after 6 months and in 65% after a year10. The population included in our study had a low representation of patients with immune mediated rheumatological diseases. This was due to asymmetry in the proportion of patients who came from the 3 different practices involved in this study.

The flexor canal, where the obstruction in patients with trigger finger originated, comprises the pulley and the flexor tendon apparatus. In our study we did not identify that the thickening of the pulley was relevant in infiltration effectiveness. Two explanations may be given for this phenomenon: firstly, that the increase in thickness was produced at the expense of the collagen fibres which were most distant from the flexor canal. Secondly, that the thickening of the pulleys was in keeping with a non-inflammatory adaptive phenomenon and is therefore not affected by corticosteroid exposure. This second hypothesis is supported by the absence of inflammatory findings in the histopathology of the studied pulleys of patients with trigger finger7. Permeability of the tendinous synovium and the similar efficacy outcomes in comparative studies between the administration with or without ultrasound guidance support both hypotheses9,12. However, it has been demonstrated that in the tenosynovial tissue of specimens obtained from patients with trigger finger there is inflammatory infiltration, neovascularisation and synovial hyperplasia. Although the most prevalent histological finding is the infiltrate of CD44+ hyaluronic acid-producing chondrocytoid cells3, these findings would account for the greater efficacy of infiltration in patients with tendon rather than pulley thickening.

To a lesser extent, progression time is another predictor of infiltration treatment success. The longer the progression time the more probable it is that the patient achieves complete or partial resolution. Theoretically this observation seems difficult to explain, but if we go back to the recognized histological findings in the trigger finger, it may be interpreted that the first episode to take place is the thickening of the pulley (non inflammatory process, probably induced by the physiological adaption of repeated action) and later, the tendon response with the chondrocytoid and inflammatory infiltrate, which is susceptible to the anti-inflammatory treatment.

There are no existing studies that identify patients who would benefit from treatment with corticosteroids. Considering the greatest weakness of the use of this technique is the risk of recurrences, the identification of tendon thickening would mean that patients who would benefit from infiltration could be selected. Further longitudinal studies could back up whether this finding would also prevent long-term recurrences.

This research did not receive any specific grants from public sector, commercial sector agencies or not-for-profit entities.

The authors have no conflict of interests to declare with this study.