Current Concepts in the Management of Trigger Finger in Adults

Joseph A. Gil, MD; Andrew M. Hresko, MD; Arnold-Peter C. Weiss, MD


J Am Acad Orthop Surg. 2020;28(15):e642-e650. 

In This Article

Conservative Management

Multiple conservative modalities have been reported for TF including corticosteroid injection, oral or injectable nonsteroidal anti-inflammatory medications, and immobilization using a variety of orthoses.[3] Lunsford et al[3] conducted a systematic review of evidence supporting or refuting the utility of NSAIDs, orthoses, and steroid injection. They recommended that a single joint be immobilized for 6 weeks initially and then up to 12 weeks if symptoms fail to resolve at the 6-week follow-up.

Little has been known regarding the natural history of TF until recently. McKee et al[6] conducted a retrospective case series analysis of 343 patients with TF who were managed with observation alone (ie, no splinting or steroid injections). Loss to follow-up occurred in less than 1% of this cohort. They reported that 6% of patients had spontaneous resolution of symptoms occurring during the 6 to 8 week interval between symptom onset and presentation for evaluation. Overall, 52% of the patients had complete spontaneous resolution of symptoms. Of those with spontaneous resolution, 50% had resolution by 8 months and 90% had complete resolution within one year of initial consultation.


Immobilization of TF with single joint orthoses can effectively relieve pain and improve function.[3] Teo et al[11] conduced a randomized comparison of Green grade 2 or 3 TF (Table 1) managed with either a PIP joint blocking orthosis or MCP joint blocking orthosis. At the 2 month follow-up, a greater degree of pain reduction was observed in the PIP blocking orthosis group. The authors reported improvement by at least a single grade in 48% of patients in the PIP blocking orthosis group and 40% in the MCP blocking orthosis group. There was notable improvement in the Quick Disability of the Arm, Shoulder and Hand Score (DASH) and markedly longer duration of orthosis wear for the PIP blocking orthosis group. The authors concluded that while both effectively reduce the severity of TF, the PIP blocking orthosis is associated with superior cosmesis and functional outcomes.

Drijkoningen et al[12] assessed the effectiveness of nighttime splinting in patients with acute onset (<3 months) of TF. Thirty-four patients wore a MCP blocking splint at night for 6 weeks. Patients completed the short version of the DASH questionnaire and a numerical rating scale for pain at the initial visit, at 6 to 8 weeks, and after 3 months. They found that at the final follow-up, 18 patients (55%) had complete resolution of symptoms.


Steroid injections are an effective method for resolving signs and symptoms of TF, with the reported response rates between 45% and 80%.[13] NSAID injections have also been studied as an alternative for patients who cannot tolerate steroids but have been found to be less effective. Shakeel and Ahmad[14] designed a double-blinded RCT comparing diclofenac injection with triamcinolone and found that at the final follow-up 70% of patients in the steroid group had complete resolution of symptoms compared with 53% in the NSAIDs group. In addition, they found that patients in the steroid group had better Quinnell scores (Table 1) at the short-term (3 weeks) follow-up. More recently, Leow et al[13] repeated this study comparing triamcinolone with ketorolac injections and found similar results. At 3, 6, and 12 weeks, a higher percentage of patients in the steroid group had a complete resolution of symptoms. However, at the 24-week follow-up, no difference was noted in pain between the groups. The authors acknowledged that the resolution of symptoms in the ketorolac group at 24 weeks could be related to spontaneous resolution of TF.

Steroid injections improve TF by reducing flexor tendon and A1 pulley size. Takahashi et al[10] conducted a prospective study to determine if high resolution ultrasonography could detect differences in volumes of the tendon and pulley after steroid injection. They performed axial scans prior to and an average 30 days after an intrasynovial steroid injection in 23 digits. Participants in the study group improved by at least a single grade, and the transverse diameter and cross-sectional area of the tendon and the thickness of the pulley markedly decreased.

The efficacy of steroid injections for TF varies based on the number of affected digits and the clinical severity.[5] Using a prospective cohort of 99 digits, Shultz et al[5] found that multiple affected digits and stage severity were predictive of patient response to a steroid injection. Patients with multiple affected fingers were 5.8 times more likely not to have a response to steroid compared with those with a single affected digit. For every stage increase in severity, the odds of having no symptom resolution doubled.

Dardas et al[15] conducted a retrospective case series analysis to quantify the long-term success of repeat steroid injections in patients with TF and to identify patient characteristics that are predictive of treatment outcomes. They included 292 repeat injections in their analysis. Second injection provided long-term success, defined as no need for additional injection or surgical release of the A1 pulley, in 111 patients (39%). Eighty-six (86) patients (30%) required an additional injection, and 108 (38%) underwent surgical release. Of patients receiving a third injection, 39% (24 of 62) had long-term success, 35% received a fourth injection, and 27% underwent surgical release. Sex, TF grade, presence of multiple TFs, and diabetes status were not associated with the success of second or third injections. Although most patients ultimately required a surgical release, 50% of patients who had a repeat injection had at least a year of symptomatic relief. Sobel et al[16] prospectively followed 160 patients with 186 trigger digits after first, second, and third injection and similarly found that 81 of 160 (51%), 16 of 45 (37%) and 3 of 10 (30%), respectively, did not require any additional intervention. Therefore, repeat injections should be offered to patients who prefer a nonsurgical option.

No difference in outcomes has been demonstrated between steroid injections given subcutaneously (extrasheath) versus within-the-flexor tendon sheath (intrasheath). Mardani-Kivi et al randomized 180 patients to receive either an intrasheath or extrasheath injection under ultrasonography guidance. They found no difference in the rate of TF resolution at one year, reinjection rate, or the final Quinnell grade between the techniques.[17]

Steroid injection may increase the risk of postoperative infection in subsequent open TF release. Ng et al[18] retrospectively reviewed the effect of steroid injection on the outcomes of 999 open TF releases, aiming to identify the risk factors for postoperative infections. They found that older age and decreasing days between injection and surgery correlated with infection rates. Patients without infection had a mean 260 days between injection and surgery compared with 79 days in those who had a postoperative infection. Lutsky et al[19] used retrospectively matched cohorts to investigate the risk of surgical site infection after steroid injection given intraoperatively during another soft-tissue hand procedure. They found that injections ipsilateral to the surgical procedure markedly increased the infection risk. The authors concluded that injections should not be provided intraoperatively during other procedures on the ipsilateral hand. Flexor tendon rupture after corticosteroid injection has been reported, although this complication is rare.[20]

Efficacy of Single and Repeat Injection in Diabetic Patients

Historically, the literature has suggested that steroid injections are less effective in patients with diabetes compared with those without, with success rates in diabetic patients reported to be between 32% and 66%.[4,21] Most notably, a prospective randomized trial conducted by Baumgarten et al[21] in 2007 revealed that steroid injections were markedly more effective in nondiabetic patients (86% effective) than in diabetic patients (63%). However, more recently, Castellanos et al[22] conducted a long-term follow-up of steroid injections in patients with TF and demonstrated no difference in success between injections in patients with diabetes (57%) and without diabetes (72%). Similarly, Dardas et al[15] found no difference between the efficacy of second or third repeat injection in diabetic patients and nondiabetics. Corticosteroid injections for TF have been noted to cause transient hyperglycemia in diabetic patients, but this risk is considered acceptable given their efficacy.[4]

Cost Effectiveness of Injection

Halim et al[23] recently conducted a cost analysis on a prospectively collected cohort of patients with TF to identify a cost effective treatment strategy. Costs reflected reimbursements for all care related to the TF, and cost savings were calculated by comparing actual cost incurred to theoretical cost of performing TF release instead of a second or third steroid injection. They included 88 digits that had at least a single steroid injection. Offering up to three injections resulted in a potential cost savings of $72,730. Of the patients who underwent more than one injection, a second injection resulted in a potential cost savings of $15,956 and a third injection resulted in a potential cost savings of $1,986. They concluded that from a cost stand point, up to three injections should be offered before surgical release.