Abstract
Background:
Corticosteroid injections are a first-line treatment of trigger finger and de Quervain’s tenosynovitis. Little research has evaluated preinjection patient-reported outcomes as a predictive factor for treatment success following corticosteroid injection. We hypothesized that patients with less pretreatment impairment would demonstrate greater post-treatment improvement than patients whose function was more severely impaired.
Methods:
We retrospectively reviewed prospectively collected Patient-Reported Outcomes Measurement Information System (PROMIS) upper extremity (UE) scores in patients undergoing corticosteroid injection for trigger finger or de Quervain’s tenosynovitis from 2017 to 2023. Independent variables were patient baseline characteristics, comorbidities, and baseline PROMIS UE. The primary outcome was treatment success between 30 days and 12 weeks, defined as achieving the minimal clinically important difference for PROMIS UE without undergoing surgery.
Results:
In total, 240 trigger finger and 74 de Quervain’s tenosynovitis patients (N = 314) were analyzed. Following injection, 63 (20.1%) patients achieved treatment success, 86 (27.4%) underwent surgical release, and 165 (52.5%) did not significantly improve function or undergo surgery. Each 1-point increase in baseline PROMIS UE was associated with 10% lower odds of treatment success (P < .001). Among nonoperative patients, each 1-point increase in baseline PROMIS UE was associated with a 0.51-point decrease in PROMIS UE score (P < .001) and diabetes was associated with a 2.74-point decrease in PROMIS UE after injection (P = .44).
Conclusion:
Corticosteroid injection provides meaningful improvement for a subset of trigger finger and de Quervain’s tenosynovitis patients. Corticosteroid injection remains a first-line treatment for trigger finger and de Quervain’s tenosynovitis patients, especially for those with more severe functional impairment.
Keywords: patient-reported outcomes, trigger finger, de Quervain’s tenosynovitis, corticosteroid injection, stenosing tenosynovitis
Introduction
Trigger finger and de Quervain’s tenosynovitis are common, painful hand disorders characterized by limited function and mobility that significantly affect the quality of life for individuals affected. 1 These two pathologies are commonly treated with corticosteroid injections, giving substantial relief to many patients.2,3 This approach not only provides an alternative to a surgical procedure but also significantly reduces the economic burden associated with operative interventions. 4 However, there is variability in individual responses to corticosteroid injection, which remains a challenge for providers, necessitating a thorough investigation into the predictive factors that determine treatment effectiveness. Understanding the underlying reasons of variability is crucial not only for optimizing patient care but also for managing patient expectations and mitigating potential health care costs of these prevalent hand disorders.
While many studies have explored various anatomical and clinical factors influencing the response to injection in these conditions, there is a paucity of literature using patient-reported outcomes (PROs) as a predictive measure.5,6 Previous research has shown that patients with multiple affected fingers and increased symptom severity are less likely to experience resolution of symptoms post corticosteroid injection. In addition, higher body mass index (BMI), type 2 diabetes mellitus, and male sex have also been associated with treatment failure.2,7-9 Few studies have examined the predictive value of preoperative Patient-Reported Outcomes Measurement Information System (PROMIS) scores for A1 pulley release.10,11 However, the existing literature has limited use of patient demographics in conjunction with PROMIS data to determine risk factors for corticosteroid injection, emphasizing the need for further investigation in this area.
In response to this critical gap, our purpose was to characterize the factors associated with favorable response to corticosteroid injections for trigger finger and de Quervain’s tenosynovitis as it pertains to patient-reported functional improvement. Using least absolute shrinkage and selection operator (LASSO) regression to identify factors predictive of PRO measures, this study aimed to analyze an extensive range of patient factors including demographic characteristics such as sex, age, and BMI, among others. Furthermore, we considered the influence of comorbidities, recognizing their potential impact on treatment response.
Materials and Methods
Study Population
From a prospectively collected registry of PRO measures, we retrospectively reviewed all patients diagnosed with trigger finger or de Quervain’s tenosynovitis from January 2017 to January 2023. NYU Langone Health institutional review board approval (IRB i23-00184) was received before conducting study activities, and patients provided informed consent when providing PROMIS data per our standardized institutional protocol. Patients were screened for diagnosis using a combination of International Classification of Diseases, 10th Revision and current procedural terminology (CPT) codes and verified by manual chart review. Inclusion criteria were patients aged 18 years or older who were clinically diagnosed with trigger finger or de Quervain’s tenosynovitis who were treated at our institution with at least 1-month follow-up. We deemed 1 month to be enough time to include all patients who would respond to a steroid injection understanding that the study was not designed or intended to answer questions related to the duration of treatment response from a steroid injection. The number of corticosteroid injections received before final study follow-up was recorded. All patients meeting the inclusion and exclusion criteria were included in the study, and a post hoc power analysis was conducted to determine study power. Initial query of the institutional database identified 9380 patients who were either diagnosed with trigger finger or de Quervain’s tenosynovitis or had an upper extremity corticosteroid injection. Of these, 751 patients had baseline PROMIS upper extremity (UE) score and 387 patients additionally completed the follow-up PROMIS UE questionnaire between 30 days and 12 weeks. These patients were manually chart reviewed, which identified 314 patients who met the inclusion criteria with a confirmed diagnosis of trigger finger or de Quervain’s tenosynovitis which was specifically treated with corticosteroid injection (Figure 1). The discrepancy between the initial number of patients identified and those with PROMIS data can be explained as the query identified all patients within a large hospital network covering the New York City metropolitan area and Long Island; however, PROMIS data were only routinely collected at the main academic campus in Manhattan.
Figure 1.
Flow diagram of participant eligibility and reason for exclusion at each stage of study participant identification. PROMIS = Patient-Reported Outcomes Measurement Information System; UE = upper extremity.
Independent Variables
Baseline patient characteristics, including age, sex, BMI, race, smoking status, and medical comorbidities, were extricated from the medical record. The PROMIS Upper Extremity (PROMIS UE) survey, a validated measure of patient upper extremity function developed by the National Institutes of Health, was used for this study. Patient-Reported Outcomes Measurement Information System UE baseline scores were used as an independent measure of patient function before treatment with corticosteroid injection. Patients were stratified into tertiles based on baseline PROMIS UE score to analyze response to corticosteroid injection.
Dependent Variables
Eligible patients had baseline PROMIS UE scores before injection and scores between 30 days and 12 weeks following their first injection. These values were used to calculate change from baseline (∆PROMIS UE), our first dependent variable. The second, binary outcome variable was “treatment success,” which we defined as achievement of PROMIS UE minimal clinically important difference (MCID) without the need for surgery. Surgical release data was collected using CPT codes and manual chart review. Patients with a ∆PROMIS UE less than the MCID value of 4.2 or those who required surgery regardless of ∆PROMIS UE score were considered treatment failures.
Statistical Analysis
Patient baseline characteristics and PROs were measured in descriptive manner. Continuous and categorical variables were compared by analysis of variance and χ2 or Fisher exact tests, respectively. For univariate analysis, patients were stratified into tertiles based on the PROMIS UE score. Regression analysis was conducted in a 2-stage manner. Least absolute shrinkage and selection operator regression was first used to select only those independent variables that improved model performance. Least absolute shrinkage and selection operator regression imposes a constraint on the complexity of linear regression by eliminating regression coefficients and thus reducing the risk of overfitting the regression model. 12 These variables selected by LASSO regression were then used for multivariate binary logistic regression to assess the dependent variable of treatment response or multivariate linear regression to analyze the ∆PROMIS UE. For models using the combined cohort, the variable for trigger finger or de Quervain’s diagnosis was included regardless of LASSO results. Post hoc power analysis of treatment success demonstrated a power of 0.94 to detect the observed effect size of 0.23 as measured by Cohen’s w statistic for the trigger finger cohort and a power of 0.93 to detect the observed effect size of 0.40 observed in the de Quervain’s tenosynovitis cohort. Given only patients with complete PROMIS UE data were included, and all patients were chart reviewed, we did not encounter missing data for participants. We used the Strengthening the Reporting of Observational Studies in Epidemiology cohort checklist when writing our report. 13
Results
In total, 314 participants met the inclusion criteria, of which 240 (76%) were diagnosed with trigger finger and 74 (24%) were diagnosed with de Quervain’s tenosynovitis (Tables 1 and 2). Average follow-up was 64.4 days for all patients. Among all patients, 63 (20.1%) achieved treatment success, while 86 (27.4%) required surgical release, and 165 (52.5%) did not achieve treatment success and did not undergo surgery.
Table 1.
Demographics and Treatment Outcome of Trigger Finger Patients Stratified by Baseline PROMIS UE in Tertiles.
| Variable | Lower tertile (N = 82) | Middle tertile (N = 79) | Upper tertile (N = 79) | Overall (N = 240) | P value |
|---|---|---|---|---|---|
| Age | 62.9 (10.9) | 64.4 (11.7) | 64.0 (12.8) | 63.8 (11.8) | .695 |
| Female | 52 (63.4%) | 54 (68.4%) | 38 (48.1%) | 144 (60.0%) | .025 |
| Male | 30 (36.6%) | 25 (31.6%) | 41 (51.9%) | 96 (40.0%) | |
| BMI | 28.9 (7.3) | 27.4 (5.2) | 26.7 (5.1) | 27.6 (6.0) | .059 |
| Never smoker | 46 (56.1%) | 49 (62.0%) | 47 (59.5%) | 142 (59.2%) | .063 |
| Current smoker | 8 (9.8%) | 0 (0%) | 4 (5.1%) | 12 (5.0%) | |
| Former smoker | 28 (34.1%) | 30 (38.0%) | 28 (35.4%) | 86 (35.8%) | |
| Race | .317 | ||||
| Asian | 5 (6.1%) | 8 (10.1%) | 5 (6.3%) | 18 (7.5%) | |
| Black | 9 (11.0%) | 10 (12.7%) | 4 (5.1%) | 23 (9.6%) | |
| Hispanic or Latino | 3 (3.7%) | 1 (1.3%) | 1 (1.3%) | 5 (2.1%) | |
| White | 62 (75.6%) | 59 (74.7%) | 69 (87.3%) | 190 (79.2%) | |
| Other race | 3 (3.7%) | 1 (1.3%) | 0 (0%) | 4 (1.7%) | |
| Number of comorbidities | .384 | ||||
| 0 | 42 (51.2%) | 44 (55.7%) | 49 (62.0%) | 135 (56.3%) | |
| 1 | 29 (35.4%) | 27 (34.2%) | 25 (31.6%) | 81 (33.8%) | |
| 2 | 9 (11.0%) | 8 (10.1%) | 3 (3.8%) | 20 (8.3%) | |
| 3 | 2 (2.4%) | 0 (0%) | 2 (2.5%) | 4 (1.7%) | |
| Follow-up (days) | 62.0 (17.3) | 64.3 (15.7) | 64.5 (15.9) | 63.6 (16.2) | .597 |
| Baseline PROMIS UE | 28.8 (3.15) | 36.3 (1.56) | 48.1 (6.41) | 37.6 (9.00) | <.001 |
| Post-treatment PROMIS UE | 32.8 (6.30) | 36.8 (8.99) | 42.1 (9.28) | 37.2 (9.08) | <.001 |
| ∆ PROMIS UE | 3.9 (6.5) | 0.6 (9.1) | −6.0 (9.5) | −0.5 (9.4) | <.001 |
| Achieved MCID | 32 (39.0%) | 16 (20.3%) | 9 (11.4%) | 57 (23.8%) | <.001 |
| Treatment outcome | .015 | ||||
| Treatment success | 25 (30.5%) | 12 (15.2%) | 8 (10.1%) | 45 (18.8%) | |
| Underwent surgery | 17 (20.7%) | 20 (25.3%) | 24 (30.4%) | 61 (25.4%) | |
| Treatment failure without surgery | 40 (48.8%) | 47 (59.5%) | 47 (59.5%) | 134 (55.8%) |
Note. Categorical data are presented as count (%) while continuous data are presented as mean (standard deviation). Continuous variables include Age, BMI, Follow-up, Baseline PROMIS UE, Post-treatment PROMIS UE, and ∆PROMIS UE. Statistical significance is denoted by bold text. PROMIS = Patient-Reported Outcomes Measurement Information System; BMI = body mass index; UE = upper extremity; MCID = minimal clinically important difference.
Table 2.
Demographics and Treatment Outcome of de Quervain’s Patients Stratified by Baseline PROMIS UE.
| Variable | Lower tertile (N = 27) | Middle tertile (N = 24) | Upper tertile (N = 23) | Overall (N = 74) | P value |
|---|---|---|---|---|---|
| Age | 54.4 (15.4) | 51.3 (14.7) | 51.2 (16.0) | 52.4 (15.2) | .694 |
| Female | 21 (77.8%) | 15 (62.5%) | 13 (56.5%) | 49 (66.2%) | .256 |
| Male | 6 (22.2%) | 9 (37.5%) | 10 (43.5%) | 25 (33.8%) | |
| BMI | 28.8 (7.90) | 25.3 (5.24) | 28.2 (5.78) | 27.5 (6.58) | .140 |
| Never smoker | 18 (66.7%) | 19 (79.2%) | 14 (60.9%) | 51 (68.9%) | .502 |
| Current smoker | 2 (7.4%) | 2 (8.3%) | 1 (4.3%) | 5 (6.8%) | |
| Former smoker | 7 (25.9%) | 3 (12.5%) | 8 (34.8%) | 18 (24.3%) | |
| Race | .096 | ||||
| Asian | 0 (0%) | 3 (12.5%) | 3 (13.0%) | 6 (8.1%) | |
| Black | 4 (14.8%) | 4 (16.7%) | 2 (8.7%) | 10 (13.5%) | |
| Hispanic or Latino | 4 (14.8%) | 0 (0%) | 0 (0%) | 4 (5.4%) | |
| White | 19 (70.4%) | 16 (66.7%) | 16 (69.6%) | 51 (68.9%) | |
| Other race | 0 (0%) | 1 (4.2%) | 2 (8.7%) | 3 (4.1%) | |
| Number of comorbidities | .317 | ||||
| 0 | 20 (74.1%) | 21 (87.5%) | 20 (87.0%) | 61 (82.4%) | |
| 1 | 6 (22.2%) | 2 (8.3%) | 1 (4.3%) | 9 (12.2%) | |
| 2 | 1 (3.7%) | 0 (0%) | 1 (4.3%) | 2 (2.7%) | |
| 3 | 0 (0%) | 1 (4.2%) | 1 (4.3%) | 2 (2.7%) | |
| Follow-up (days) | 66.7 (16.7) | 67.5 (12.4) | 65.7 (13.4) | 66.7 (14.4) | .928 |
| Baseline PROMIS UE | 25.8 (3.6) | 32.2 (1.9) | 42.0 (5.7) | 32.9 (7.8) | <.001 |
| Post-treatment PROMIS UE | 31.2 (5.7) | 34.4 (5.9) | 39.6 (10.5) | 34.8 (8.2) | <.001 |
| ∆ PROMIS UE | 5.4 (6.5) | 2.1 (6.8) | −2.4 (10.0) | 1.9 (8.4) | .003 |
| Achieved MCID | 16 (59.3%) | 9 (37.5%) | 4 (17.4%) | 29 (39.2%) | .010 |
| Treatment outcome | .015 | ||||
| Treatment success | 11 (40.7%) | 5 (20.8%) | 2 (8.7%) | 18 (24.3%) | |
| Underwent surgery | 11 (40.7%) | 7 (29.2%) | 7 (30.4%) | 25 (33.8%) | |
| Treatment failure without surgery | 5 (18.5%) | 12 (50.0%) | 14 (60.9%) | 31 (41.9%) |
Note. Categorical data are presented as count (%) while continuous data are presented as mean (standard deviation). Continuous variables include Age, BMI, Follow-up, Baseline PROMIS UE, Post-treatment PROMIS UE, and ∆PROMIS UE. Statistical significance is denoted by bold text. PROMIS = Patient-Reported Outcomes Measurement Information System; BMI = body mass index; UE = upper extremity; MCID = minimal clinically important difference.
Univariate analysis of trigger finger patients showed no differences among patients stratified into tertiles by baseline PROMIS UE score regarding age, sex, BMI, smoking status, race, number of comorbidities, or follow-up (Table 1). However, patients in the lowest tertile of baseline PROMIS UE score demonstrated a greater increase in PROMIS UE score of 3.9 points at follow-up compared with the middle tertile’s increase of 0.6 points and the upper tertile’s decrease of 6.0 points, though it should be noted the middle and upper tertiles demonstrated higher variance in PROMIS UE score (P < .001). Similarly, patients in the lower tertile were more likely to achieve treatment success (30.5%) compared with the middle (15.2%) and upper tertiles (10.1%, P = .015).
Crude analysis of de Quervain’s tenosynovitis patients demonstrated no differences among cohorts for age, sex, BMI, smoking status, race, or number of comorbidities when patients were stratified into tertiles by baseline PROMIS UE score (Table 2). Further investigation revealed that ∆PROMIS UE score following treatment was higher in the lowest tertile (+5.4) compared with the middle (+2.1) and upper tertiles (−2.4) though higher variance was again noted in the upper tertile (P = .003). Accordingly, patients in the lowest tertile of baseline PROMIS UE score were more likely to reach treatment success (40.7%) compared with patients in the middle (20.8%) and highest tertile of baseline PROMIS UE score (8.7%, P = .015).
The initial LASSO selection model analyzing treatment success among all patients identified sex, baseline PROMIS UE score, and additional corticosteroid injections as independent variables (Table 3). These factors, along with diagnosis of trigger finger or de Quervain’s tenosynovitis, were subsequently analyzed using multivariate binary logistic regression, which demonstrated an area under the receiver operating curve (AUC) of 0.73. After adjusting for other factors, the model failed to detect an association between diagnosis and treatment response (P = .929). Men showed 87% higher odds of achieving treatment response (P = .043) while every single-point increase in baseline PROMIS UE was associated with a 10% decrease in odds of achieving treatment response (P < .001).
Table 3.
Multivariate Logistic Regression of Predictors of Treatment Response.
| Variable | Odds ratio | 2.5% | 97.5% | P value | |
|---|---|---|---|---|---|
| Combined | de Quervain’s (reference) | — | — | — | — |
| Trigger finger | 1.03 | 0.53 | 2.05 | .929 | |
| Additional corticosteroid injection | 0.64 | 0.39 | 1.01 | .061 | |
| Female (reference) | — | — | — | — | |
| Male | 1.87 | 1.02 | 3.44 | .043 | |
| Baseline PROMIS UE | 0.90 | 0.86 | 0.94 | <.001 | |
| DQ only | Peripheral vascular disease | 3.29 | 0.26 | 80.80 | .369 |
| Never smoker (reference) | — | — | — | — | |
| Current smoker | 0.86 | 0.04 | 7.76 | .905 | |
| Former smoker | 4.15 | 1.06 | 17.14 | .041 | |
| Baseline PROMIS UE | 0.87 | 0.77 | 0.95 | .009 | |
| TF only | Female (reference) | — | — | — | — |
| Male | 2.28 | 1.13 | 4.66 | .022 | |
| Never smoker (reference) | — | — | — | — | |
| Current smoker | 1.33 | 0.31 | 5.00 | .684 | |
| Former smoker | 0.71 | 0.32 | 1.51 | .385 | |
| Additional corticosteroid injections | 0.64 | 0.36 | 1.11 | .117 | |
| Baseline PROMIS UE | 0.91 | 0.87 | 0.96 | <.001 |
Note. Statistical significance is denoted by bold text. PROMIS = Patient-Reported Outcomes Measurement Information System; UE = upper extremity; DQ = de Quervain’s tenosynovitis; TF = trigger finger.
Additional regression models analyzed factors predictive of treatment response for either de Quervain’s or trigger finger patients. Least absolute shrinkage and selection operator regression for de Quervain’s patients selected peripheral vascular disease, smoking status, and baseline PROMIS UE score as informative factors (Table 3). The resulting multivariate binary logistic regression model using these coefficients demonstrated an AUC of 0.77. Increased PROMIS UE score was consistently associated with decreased odds of treatment success (P = .009), while former smokers were more likely than never smokers to respond to treatment (P = .041). Analysis of trigger finger patients identified sex, smoking status, additional corticosteroid injections, and baseline PROMIS UE as factors predictive of response with a model AUC of 0.73 (Table 3). Increased baseline PROMIS UE (P < .001) and male sex (P = .022) were again associated with lower and higher odds of treatment response, respectively.
In a separate analysis of patients who did not undergo surgery (N = 240), including those who achieved and did not achieve treatment success, the primary LASSO regression model selected age, sex, race, diabetes, osteoporosis, peripheral vascular disease, additional corticosteroid injections, and baseline PROMIS UE as predictor variables. The resulting linear regression model using these coefficients with the dependent variable of ∆PROMIS UE demonstrated an R2 value of 0.28 (Table 4). Diagnosis of de Quervain’s or trigger finger was not associated with ∆PROMIS UE. In contrast, type II diabetes was associated with a 2.74-point decrease in ∆PROMIS UE (P = .044) and a 1-point increase in baseline PROMIS UE was associated with a 0.51-point reduction in ∆PROMIS UE (P < .001). No difference in ∆PROMIS UE was detected between patients who received additional corticosteroid injections.
Table 4.
Multivariate Linear Regression of Predictors of ∆PROMIS UE.
| Variable | Estimate | 2.5% | 97.5% | P value |
|---|---|---|---|---|
| de Quervain’s (reference) | — | — | — | — |
| Trigger finger | −0.41 | −3.16 | 2.34 | .769 |
| Age | 0.05 | −0.04 | 0.13 | .288 |
| Female (reference) | — | — | — | — |
| Male | 1.57 | −0.72 | 3.85 | .178 |
| White (reference) | — | — | — | — |
| Asian | −0.25 | −4.43 | 3.92 | .905 |
| Black | 0.41 | −2.95 | 3.77 | .811 |
| Hispanic | −5.27 | −11.23 | 0.69 | .083 |
| Other race | −5.52 | −12.63 | 1.59 | .128 |
| Type II diabetes | −2.74 | −5.40 | −0.08 | .044 |
| Osteoporosis | 1.87 | −1.46 | 5.19 | .269 |
| Peripheral vascular disease | 3.04 | −3.54 | 9.61 | .364 |
| Additional corticosteroid injection | −0.69 | −2.29 | 0.92 | .400 |
| Baseline PROMIS UE | −0.51 | −0.63 | −0.39 | <.001 |
Note. Statistical significance is denoted by bold text. PROMIS = Patient-Reported Outcomes Measurement Information System; UE = upper extremity.
Discussion
Trigger finger and de Quervain’s tenosynovitis are some of the most treated in-office conditions in a hand surgeon’s practice. Initial treatment with corticosteroid injections remains standard of care. However, there is a paucity of literature using PROs as a predictive measure. It is well known that PRO measures are critical regarding patient perception of treatment effectiveness.5,14,15 In this study, we sought to identify patient and disease-specific factors associated with corticosteroid injection treatment success. We aimed to enhance the clinical decision-making process and refine treatment strategies for patients diagnosed with trigger finger or de Quervain’s tenosynovitis. Through this analysis, we strove to contribute to the growing body of knowledge, ultimately improving care for patients with these debilitating conditions.
Our study found that approximately 20% of patients who received corticosteroid injections for trigger finger achieved treatment success, defined as achievement of PROMIS UE MCID without the need for surgery. Ultimately, 27% of patients required surgical intervention for this pathology. The variability in outcomes we report have not yet been shown in the literature and may suggest that patients with more severe functional limitations at baseline may demonstrate greater functional increase following corticosteroid injections as measured by PROs. In a subgroup analysis of patients who did not require surgery, we found that higher baseline PROMIS UE and diabetes were associated with decreased PROMIS UE scores post injection, suggesting that patients with these factors are less likely to achieve a large improvement in upper extremity function.
Notably less research has focused on the prognostic value of PROs following corticosteroid injection for treatment of de Quervain’s tenosynovitis. Smolyak et al evaluated PROMIS Physical Function (PF) and Pain Interference (PI) scores to predict treatment response. Their findings associated low PROMIS PF or high PROMIS PI with increased odds of treatment failure and subsequent surgical release, particularly in women with ages ranging from 40 to 60 years. 5 Interestingly, our results suggest that patients with the lowest baseline function were more likely to achieve clinically significant symptom improvement in function compared with patients with less functional impairment. However, this group was also more likely to ultimately require surgical release than patients in the middle and upper tertiles of baseline function. Taken together, these findings suggest that a substantial portion of patients with pronounced functional impairment from de Quervain’s tenosynovitis may benefit from corticosteroid injection, though non-responders who experience refractory symptoms may be more likely to require surgical release.
Our findings suggest that patients with the greatest limitation in upper extremity function at baseline are most likely to see improvements in their symptoms and function after corticosteroid injection. It should be acknowledged that patients with higher baseline functional status may demonstrate a smaller increase in patient-reported function following corticosteroid injection that does not reach the MCID despite improvement in symptoms. These findings contrast with Phan et al, who showed that the 64% of patients who had improvement in triggering also had higher PROMIS UE scores after treatment. In that cohort, patients who exhibited no improvement were associated with lower baseline scores. Some studies have shown that patients with increased severity are associated with worse outcomes, in contrast to our results. This may be due to differing patient population’s perception of pain and the demands of their work and life generally. Ultimately, this can inform how clinicians counsel patients before injection, setting appropriate expectations for possible outcomes.
Diabetes is frequently noted a risk factor for treatment failure in various pathologies, including trigger finger. Baumgarten et al 16 assessed the efficacy of corticosteroid injections for trigger finger in both diabetic and nondiabetic patients and found them significantly more effective in nondiabetic patients. In contrast, Grandizio et al 17 found diabetes was not a risk factor for recurrent trigger finger corticosteroid injection, suggesting that patients with diabetes can be effectively managed with corticosteroid injections and have similar outcomes compared with patients without diabetes. We found an association between baseline PROMIS scores and diabetes, suggesting these patients have poorer function compared with those nondiabetic patients. Furthermore, these patients were less likely to demonstrate improvement in PROMIS UE scores after corticosteroid injection. It is possible that these patients showed a less significant improvement post injection due to their lower baseline PROMIS. While diabetes may be a risk factor for the treatment success, further study into this area is warranted to elucidate this relationship more clearly.
The relationship between sex and treatment success of corticosteroid injection has not been well defined. We observed that men had significantly higher odds of achieving treatment response compared with women. Mixed findings have been reported in the literature regarding this topic. Oh et al 2 reported that women were approximately 3.23 times more likely than men to fail corticosteroid treatment for de Quervain’s tenosynovitis. Conversely, in studies assessing treatment for trigger finger, Marks and Gunther 18 and Wojahn et al 19 reported that women had an increased percentage of corticosteroid injection treatment success compared with men. In addition, the observations of Dardas et al 20 stand in the middle ground, reporting no difference between men and women. More highly powered future studies may help to elucidate the role of sex as a factor in treatment success.
Our study has limitations that merit consideration. Although this study is retrospective in nature and carries the risk of bias inherent to retrospective reviews, our main outcome data were collected prospectively, which improved the reliability of our analysis. The standardized administration of surveys to all patients minimized variations in data collection, ensuring a consistent approach across the study cohort. In addition, the study was constrained by a limited sample size, notably in de Quervain’s cohort, though this did not in practice limit the power of this analysis as demonstrated by the post hoc power analysis. In addition, we acknowledge potential response bias associated with using PRO measures in general, though standardized administration of the PROMIS UE at our institution aimed to limit this effect. Finally, patients were treated by multiple different health care providers, which introduced an element of potential technique variability for corticosteroid injections. However, these injections are incredibly common, thus contributing to the consistency and reliability of our study.
Although we did not analyze trigger finger patients’ outcomes following surgery, studies have shown that preoperative PROMIS scores can predict significant postoperative score improvement in hand surgeries including trigger finger release. Bernstein et al 10 observed that patients with lower baseline PROMIS PF scores and higher PROMIS PI and Depression scores improved their postoperative scores. Regarding de Quervain’s tenosynovitis, Zamri et al. found that patients with high PROMIS PI scores reported poor surgical outcomes more frequently after surgery than those with lower baseline scores. 11 Nevertheless, our findings share similarities with reports in the literature, suggesting the possible prognostic value of PROMIS UE assessments for treatment success.
Conclusions
In conclusion, our findings suggest that baseline PROMIS UE scores can serve as predictors of conservative treatment success with corticosteroid injections for trigger finger and de Quervain’s tenosynovitis. In addition, diabetes was confirmed as a risk factor for poorer outcomes, and sex, particularly men, were found to have better treatment success. As is the trend in hand and upper extremity literature, PROs have become key in informing appropriate management. Patient-Reported Outcomes Measurement Information System scores provide a standardized and patient-centered measure of outcomes and understanding the significance of PROMIS scores can guide clinical decision-making and patient expectations. In addition, our results show that corticosteroid injections remain an indicated first-line treatment of trigger finger and de Quervain’s tenosynovitis, especially in patients with higher baseline functional impairment.
Footnotes
Ethical Approval: This study was approved by our institutional review board.
Statement of Human and Animal Rights: All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008 (5). Informed consent was obtained from all patients as previously described.
Statement of Informed Consent: Informed consent was received before collection of PROMIS data as is standard at our institution and the requirement for informed consent specific to this study was waived by the institutional review board.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
ORCID iDs: Walter D. Sobba 
https://orcid.org/0000-0002-0754-2262
Gerardo Sánchez-Navarro
https://orcid.org/0000-0003-4381-4656
References
- 1. Gil JA, Hresko AM, Weiss A-PC. Current concepts in the management of trigger finger in adults. J Am Acad Orthop Surg. 2020;28(15):e642-e650. doi: 10.5435/jaaos-d-19-00614 [DOI] [PubMed] [Google Scholar]
- 2. Oh JK, Messing S, Hyrien O, et al. Effectiveness of corticosteroid injections for treatment of de Quervain’s tenosynovitis. Hand. 2017;12(4):357-361. doi: 10.1177/1558944716681976 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Usmani RH, Abrams SS, Merrell GA. Establishing an efficient care paradigm for trigger finger. J Hand Surg Asian Pac. 2018;23(3):356-359. doi: 10.1142/S2424835518500364 [DOI] [PubMed] [Google Scholar]
- 4. Zhuang T, Wong S, Aoki R, et al. A cost-effectiveness analysis of corticosteroid injections and open surgical release for trigger finger. J Hand Surg Am. 2020;45(7):597-609. doi: 10.1016/j.jhsa.2020.04.008 [DOI] [PubMed] [Google Scholar]
- 5. Smolyak G, Qiu B, Cora Jones CM, et al. Association of patient-reported outcomes measurement information system measures with injection and surgical treatment response in patients with de Quervain tenosynovitis. J Hand Surg Am. 2023;48(11):1098-1104. doi: 10.1016/j.jhsa.2023.07.005 [DOI] [PubMed] [Google Scholar]
- 6. Peters-Veluthamaningal C, van der Windt DA, Winters JC, et al. Corticosteroid injection for trigger finger in adults. Cochrane Database Syst Rev. 2009;1:CD005617. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Chang CJ, Chang SP, Kao LT, et al. A meta-analysis of corticosteroid injection for trigger digits among patients with diabetes. Orthopedics. 2018;41(1):e8-e14. doi: 10.3928/01477447-20170727-02 [DOI] [PubMed] [Google Scholar]
- 8. Hollins AW, Hein R, Atia A, et al. Symptom duration and diabetic control influence success of steroid injection in trigger finger. Plast Reconstr Surg. 2022;150(2):357e-363e. doi: 10.1097/PRS.0000000000009320 [DOI] [PubMed] [Google Scholar]
- 9. Basar B, Aybar A, Basar G, et al. The effectiveness of corticosteroid injection and splint in diabetic de Quervain’s tenosynovitis patients: a single-blind, randomized clinical consort study. Medicine. 2021;100(35):e27067. doi: 10.1097/MD.0000000000027067 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Bernstein DN, Houck JR, Gonzalez RM, et al. Preoperative PROMIS scores predict postoperative PROMIS score improvement for patients undergoing hand surgery. Hand. 2020;15(2):185-193. doi: 10.1177/1558944718791188 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Zamri M, Lans J, Eberlin KR, et al. Reintervention, PROMs, and factors influencing PROMs following surgery for de Quervain’s tenosynovitis. J Hand Microsurg. 2023;15(3):165-174. doi: 10.1055/s-0041-1731105 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Ranstam J, Cook JA. LASSO regression. Br J Surg. 2018;105(10):1348-1348. doi: 10.1002/bjs.10895 [DOI] [Google Scholar]
- 13. von Elm E, Altman D, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ. 2007;335(7624):806-808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Phan A, Calderon T, Hammert W. Responsiveness of PROMIS instruments for trigger digit after corticosteroid injection or A1 pulley release. J Hand Surg Am. 2023;48(10):1064.e1-1064.e7. doi: 10.1016/j.jhsa.2022.03.015 [DOI] [PubMed] [Google Scholar]
- 15. Shetty PN, Hawken J, Sanghavi KK, et al. Correlation of Patient-Reported Outcomes Measurement Information System Questionnaires with the Brief Michigan Hand Questionnaire in patients with 5 common hand conditions. J Hand Surg Am. 2021;46(8):709.e1-709.e11. doi: 10.1016/j.jhsa.2020.11.024 [DOI] [PubMed] [Google Scholar]
- 16. Baumgarten KM, Gerlach D, Boyer MI. Corticosteroid injection in diabetic patients with trigger finger: a prospective, randomized, controlled double-blinded study. J Bone Joint Surg Am. 2007;89(12):2604-2611. doi: 10.2106/jbjs.G.00230 [DOI] [PubMed] [Google Scholar]
- 17. Grandizio LC, Speeckaert A, Brothers J, et al. Predictors of recurrence after corticosteroid injection for trigger digits. Hand. 2017;12(4):352-356. doi: 10.1177/1558944716668862 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Marks MR, Gunther SF. Efficacy of cortisone injection in treatment of trigger fingers and thumbs. J Hand Surg Am. 1989;14(4):722-727. doi: 10.1016/0363-5023(89)90199-8 [DOI] [PubMed] [Google Scholar]
- 19. Wojahn RD, Foeger NC, Gelberman RH, et al. Long-term outcomes following a single corticosteroid injection for trigger finger. J Bone Joint Surg Am. 2014;96(22):1849-1854. doi: 10.2106/JBJS.N.00004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Dardas AZ, VandenBerg J, Shen T, et al. Long-term effectiveness of repeat corticosteroid injections for trigger finger. J Hand Surg Am. 2017;42(4):227-235. doi: 10.1016/j.jhsa.2017.02.001 [DOI] [PMC free article] [PubMed] [Google Scholar]