SCIENTIFIC ARTICLE

A Prospective Randomized Controlled Trial of

Injection of Dexamethasone Versus Triamcinolone for
Idiopathic Trigger Finger
David Ring, MD, PhD, Santiago Lozano-Calderón, MD, Robert Shin, MD, Peter Bastian, BA,
Chaitanya Mudgal, MD, Jesse Jupiter, MD

Purpose This study was designed to test the null hypothesis that there is no difference in resolution of triggering 3
months after injection with either a soluble (dexamethasone) or insoluble (triamcinolone) corticosteroid for
idiopathic trigger finger.
Methods Eighty-four patients were enrolled in a prospective randomized controlled trial comparing dexamethasone and
triamcinolone injection for idiopathic trigger finger. Sixty-seven patients completed the 6-week follow-up (35 triamcinolone
arm, 32 dexamethasone arm), and 72 patients completed the 3-month follow-up (41 triamcinolone arm, 31
dexamethasone arm). Outcome measures included the Disabilities of the Arm, Shoulder, and Hand (DASH) questionnaire,
trigger finger grading according to Quinnell, and satisfaction on a visual analog scale. To preserve autonomy, patients were
permitted additional injections and operative treatment at any time. Twenty-five patients requested a second injection (10
triamcinolone arm, 15 dexamethasone arm), and 21 elected operative treatment (10 triamcinolone arm, 11 dexamethasone
arm) during the study period. The analysis was according to intention to treat principles.
Results Six weeks after injection, absence of triggering was documented in 22 of 35 patients in the
triamcinolone cohort and in 12 of 32 patients in the dexamethasone cohort. The rates 3 months after
injection were 27 of 41 in the triamcinolone cohort and 22 of 31 in the dexamethasone cohort. The
triamcinolone cohort had significantly better satisfaction and Quinnell grades than did the dexamethasone
cohort at the 6-week follow-up but not at the 3-month follow-up. There were no significant differences
between Disabilities of the Arm, Shoulder, and Hand scores at the 6-week follow-up and the 3-month
follow-up. After the close of the study, there were 8 recurrences among patients with documented absence
of triggering in the triamcinolone cohort and 1 in the dexamethasone cohort.
Conclusions Although there were no differences 3 months after injection, our data suggest that triamcinolone
may have a more rapid but ultimately less durable effect on idiopathic trigger finger than does dexamethasone. (J
Hand Surg 2008;33A:516–522. Copyright © 2008 by the American Society for Surgery of the Hand.)
Type of study/level of evidence Therapeutic I.
Key words Clinical trial, corticosteroid injection, trigger finger, dexamethasone, triamcinolone.

From the Harvard Medical School, Boston, MA; and the Hand
and Upper Extremity Service, Department of Orthopaedic
Surgery, Massachusetts General Hospital, Boston, MA.
Received for publication August 20, 2007; accepted in
revised form January 3, 2008.
Support was received from AO Foundation, Wright
Medical, Joint Active Systems, Smith and Nephew, Small
Bone Innovations, and Biomet.
Corresponding author: David Ring, MD, PhD, Orthopaedic
Hand and Upper Extremity Service, Massachusetts
General Hospital, Yawkey Center Suite 2100, 55 Fruit
Street, Boston, MA 02114; e-mail: dring@partners.org.
0363-5023/08/33A04-0007$34.00/0
doi:10.1016/j.jhsa.2008.01.001
C
ORTICOSTEROID INJECTION IS a widely used
treatment for idiopathic trigger finger.1– 4 Prior
research has established that 1 or more
corticosteroid injections alone can resolve an idiopathic
trigger finger in approximately 50% of patients.3–7 Many
hand surgeons express a preference for soluble
corticosteroids (such as dexamethasone), arguing that
insoluble corticosteroids (such as triamcinolone) may leave a
deposit in the flexor tendon sheath—an area dependent
upon smooth gliding for optimal function.8,9 Other hand
surgeons prefer the insoluble steroids. To our knowledge,
there are no scientific data available to guide the
decision between corticosteroids. This study was
designed to test the null hypothesis that there is no
difference in resolution of triggering 3 months after
injection with either dexamethasone or triamcinolone
for idiopathic trigger finger. Secondary study questions

516 䉬 ©  ASSH 䉬 Published by Elsevier, Inc.

 DEXAMETHASONE VERSUS TRIAMCINOLONE FOR TRIGGER FINGER 517

included comparison of Disabilities of the Arm, Patients Enrolled:

84
Shoulder, and Hand (DASH) scores 6 weeks after
injection and comparisons of the rate of absence of
triggering, Quinnell grade, and patient satisfaction at
both time points.
Dexamethasone: Triamcinolone:
MATERIALS AND METHODS 40 44
The study protocol was approved by our human research
committee. Patients with idiopathic trigger finger that
presented to the office of 1 of 3 hand surgeons between
6-week visit: 6 –week visit:
March 1, 2005, and November 30, 2005, were invited to Missing: 8 Missing: 9
participate in the study. Inclusion criteria included any adult Complete FU: 32 Complete FU: 35
Cured: 12 Cured: 22
patient (age 18 years or greater) with an isolated new Surgery: 6 Surgery: 3
diagnosis of 1 or more idiopathic trigger fingers of any Total: 40 Total: 44
Quinnell grade 2 or greater. According to the Quinnell
system, trigger fingers are rated as follows: 0, normal Injection:
15 patients
Injection:
10 patients
movement of the digit; 1, uneven movement; 2, actively 13 Dexamethasone 10 Triamcinolone
2 Triamcinolone
correctable locking of the digit; 3, passively correctible
locking; and grade 4 represents fixed deformity.3 The
3-month visit: 3-month visit:
exclusion criteria were inflammatory or autoimmune Missing: 9 Missing: 3
arthritis, prior tendon injury of the involved digit, a current Complete FU: 31 Complete FU: 41
Cured: 22 Cured: 27
or suspected pregnancy, and breast-feeding. When multiple Surgery: 5 Surgery: 7
trigger fingers were present, the most radial digit was Total: 40 Total: 44

studied. As a result, the number of fingers and the number Injection: Injection:
of patients studied is identical. During the study period, only 2 patients 7 patients

2 qualified patients declined enrollment. Informed consent

was obtained from all patients.
Eighty-four patients enrolled and were randomized to
injection with either triamcinolone (44 patients) or
dexamethasone (40 patients) according to a sequence
determined by a computer random-number generator
(Windows Excel; Microsoft, Redmond, WA) (Fig. 1). The
injections were performed according to prevailing practice
styles at our institution, where both medications are
routinely used. The injections were placed into and around
the flexor sheath using a 25-gauge or 27-gauge needle
(surgeon preference) at the level of the A1 pulley using a 1:1
mixture of 1% lidocaine (Hospira, Inc., Lake Forest, IL) and
either triamcinolone (10 mg/mL; Bristol-Myers, New York,
NY) or dexamethasone (4 mg/mL; American
Pharmaceutical Partners, Inc, Schaumburg, IL) with a total
injection volume of 1.0 to 1.5 mL. Dexamethasone is 5
times as potent and its duration of action is twice that of
triamcinolone.10 Neither the treating physicians nor the
participants were blinded to the medication. Because one
medication is white and opaque (triamcinolone) and the
other is clear (dexamethasone), blinding of the surgeon and
patient would not be straightforward. We believed that
blinding was not important because neither the patient nor
the surgeon had any preconceived notions about which type
of steroid would be superior.
At the time of enrollment, each patient completed the
DASH questionnaire11,12 and a demographic questionnaire
including patient’s age, gender, occupation, and the involved
fingers. After completing the forms, each patient was
examined by the treating hand surgeon, and the trigger
finger was classified according to the Quinnell system.3
FIGURE 1: Flowchart of patient disposition. FU, follow-up.
Subjects were reevaluated 6 weeks (⫾2 weeks) and 3
months (10 –20 weeks) after their initial injection. At each
evaluation, the trigger finger was reclassified according to
the Quinnell grade, and the patient completed the DASH
questionnaire and a visual analog scale measuring satisfaction
with treatment (SVAS). For the visual analogue scale,
patients were asked to make a mark on a line between the 2
extremes of complete dissatisfaction and complete
satisfaction, and the results was measured as the fraction of
the distance between complete dissatisfaction and the length
of the entire line. Any trigger finger with a Quinnell grade
greater than 0 was considered unresolved—Quinnell grade 0
was rated as complete absence of triggering. To ensure
patient autonomy, additional treatments including a second
injection of steroid was allowed. For subsequent injections,
patients were offered the same medication that they
originally received but were allowed to request the other
medication, again to preserve autonomy.
Statistical Methods
A power analysis indicated that a total sample size of 52
patients randomized equally (1:1 randomization) to each
treatment arm without any blocking or stratification would
provide 80% statistical power (␣ ⫽ .05, ␤ ⫽ .20) to detect a
20% difference in mean DASH scores between cohorts
assuming a common standard deviation of 25% (effect size
⫽ 20/25 ⫽ 0.8). To account for an estimated 38% loss to
follow-up, we enrolled a total of 84 patients. Power

JHS 䉬 Vol A, April 

518 DEXAMETHASONE VERSUS TRIAMCINOLONE FOR TRIGGER FINGER
calculations were determined using nQuery Advisor
software program (ver. 5.0; Statistical Solutions, Boston,
MA). The adequacy of randomization was tested by
comparing demographic and disease characteristics at
enrollment using Student’s t-test for continuous variables
and chi-square and Fisher’s exact test for dichotomous and
categorical variables.
The analysis was performed according to intention to
treat principles, meaning that patients were counted as
belonging to the treatment group assigned even if other
treatments (injections, surgery, etc., or no treatment) were
administered. Intention to treat analysis tests the effectiveness
of prescribing a certain treatment rather than the efficacy of
a specific and ideally administered treatment protocol and
therefore is better reflection of actual practice. It also insists
on the prestudy analysis and prohibits poststudy alteration of
the data, which might introduce bias.
Two-way repeated-measures mixed model analysis of
variance (ANOVA)13 was used to assess differences in
DASH scores in the same group over time. A compound
symmetry covariance structure was fitted to account for the
within-subjects variation, and the Akaike information
criterion revealed good model fit to the longitudinal data.
The mixed procedure in SPSS, version 15.1 (SPSS Inc.,
Chicago, IL) was used for analysis.
The SVAS scores were compared between groups at the
6-week and-3 month evaluations using Student’s t-test and
within groups using paired Student’s t-test. Resolution rates
at the 6-week follow-up and the 3-month follow-up times
were compared between groups and within groups using
Fisher’s exact test.
Determinants of absence of triggering, satisfaction, and
DASH scores at each time point were sought with
univariate analysis (t-tests, Fisher’s exact tests, and Pearson
correlations). All variables with significant or near significant
(p ⬍ .08) relationships except for marital status and years of
education were then evaluated using multivariate binary
logistic regression using the backwards conditional method
for determinants of absence of triggering (nominal discrete
variable) and multivariate linear regression using the stepwise
method for determinants of satisfaction and DASH scores 6
weeks and 3 months after injection (numeric continuous
variables). Both types of multivariate analyses aim to
determine the smallest combination of variables that
accounts for the majority of the variation in the dependent
variable. The adjusted R-squared that results from a
multivariate model is an estimate of the percentage of the
variation in the dependent variables accounted for or
“explained” by the combination of explanatory or
independent variables. All the statistical analysis was done in
SPSS, version 15.1.
RESULTS
Baseline Patient Characteristics
There were no statistical differences in demographic and
baseline study data between cohorts. (Table 1) Among the
JHS 䉬 Vol A, April 
40 patients in the dexamethasone cohort, the thumb was
involved in 8, the index finger in 4, the long finger in 16,
the ring finger in 7, and the small finger in 5. Among the
44 patients in the triamcinolone cohort, the thumb was
involved in 13, the long finger in 14, the ring finger in 16,
and the small finger in 1.
The average age of patients in the dexamethasone group
was 63 years (range, 30 –93 years) compared with 65 years
(range, 36 – 86 years) in the triamcinolone group (p ⫽ .47).
There were 24 women and 20 men in the triamcinolone
group compared with 16 women and 24 men in the
dexamethasone group (p ⫽ .58). The average duration of
symptoms was 15 weeks (range, 2–52 weeks) in the
triamcinolone group and 10 weeks (1– 42 weeks) in the
dexamethasone cohort. (p ⫽ .67) Seven patients in the
triamcinolone cohort and 3 in the dexamethasone cohort
had diabetes mellitus (p ⫽ .32; Fisher’s exact test).
The baseline DASH score averaged 24 points (range,
0 –90 points; SD, 19.9; 95% CI, 18.1–29.8 points) in the
dexamethasone cohort and 22 points (range, 0 – 86 points;
SD, 22.4; 95% CI, 15.5–28.1 points) in the triamcinolone
cohort (t ⫽ 0.36, p ⫽ .71) The initial Quinnell grades were
as follows: 11 patients were rated grade 4 (7 in the
dexamethasone cohort and 4 in the triamcinolone cohort);
52 patients (26 in each cohort) were rated grade 3; and 21
were rated grade 2 (7 in the dexamethasone cohort and 14
in the triamcinolone cohort) (p ⫽ .22).
Additional Treatment
Fifteen patients in the dexamethasone cohort and 10 patients
in the triamcinolone cohort requested a second injection for
the trigger finger under study at the 6-week evaluation (p ⫽
.21). Two patients originally injected with dexamethasone
requested that the second injection be performed with
triamcinolone. Two patients in the dexamethasone cohort
and 7 patients in the triamcinolone cohort requested a
second injection at the 3-month evaluation (p ⫽ .12).
Eleven patients in the dexamethasone cohort and 10 in
the triamcinolone cohort elected operative trigger finger
release during the study period. Nine patients (6
dexamethasone, 3 triamcinolone) elected surgery within 6
weeks, and 12 patients (5 dexamethasone, 7 triamcinolone)
elected surgery between 6 weeks and 3 months (p ⫽ .45).
The time between injection and surgery averaged 9 weeks
in the dexamethasone cohort and 11 weeks in the
triamcinolone cohort (p ⫽ .21). Univariate statistical analysis
found no patient or disease characteristics predictive of a
choice for surgery.
Data Analyzed
Sixty-seven patients (35 in the triamcinolone cohort and 32
in the dexamethasone cohort) completed the 6-week
follow-up. Seventy-two patients (41 in the triamcinolone
cohort and 31 in the dexamethasone cohort) completed the
3-month follow-up (Fig. 1).
 DEXAMETHASONE VERSUS TRIAMCINOLONE FOR TRIGGER FINGER 519

TABLE 1: Demographics

Characteristic Dexamethasone Triamcinolone Significance

Age (y) 63 (range, 30–93) 65 (range, 36–87) t ⫽ ⫺0.71, p ⫽ .47
Gender ␹2 ⫽ 1.77, p ⫽ .13
Female 16 24
Male 24 20
Marital status ␹2 ⫽ 2.85, p ⫽ .58
Single 5 8
Married 28 26
Widowed 1 4
Living with partner 2 1
Divorced 3 43
Unknown 1 1
Years of education (y) 14 (range, 6–20) 14 (range, 6–19) t ⫽ ⫺0.01, p ⫽ .98
Expected recovery time (d) 54 (range, 2–365) 38 (range, 1–180) t ⫽ ⫺0.82, p ⫽ .41
Duration of symptoms (wk) 10 (range, 1–42) 15 (range, 2–52) t ⫽ 0.42, p ⫽ .67
Finger Fischer’s exact test ⫽ 0.14,
p ⫽ .46
Small 5 1
Ring 7 16
Long 16 14
Index 4 0
Thumb 8 13
Initial DASH score (points) 24 (range, 0–90) 22.4 (range, 0–86) t ⫽ 0.36, p ⫽ .71
Quinnell grade (no. patients) Fischer’s exact test ⫽ 2.98,
p ⫽ .22
0 0 0
1 0 0
2 7 14
3 26 26
4 7 4
Total patients 40 44

Results 6 Weeks After Injection triamcinolone cohort requested operative treatment.

At the 6-week follow-up, there was no significant difference
in the average DASH score (triamcinolone, 15 points
[range, 0 – 66.7 points; SD, 18.5 points; 95% CI, 9.2–23.8];
dexamethasone, 19 points [range, 0 –78 points; SD, 18.2
points; 95% CI, 11.3–24.7]; t ⫽ 0.78, p ⫽ .43). The
absence of triggering rate 6 weeks after injection was 22 of
35 patients in the triamcinolone cohort and 12 of 32
patients in the dexamethasone cohort (p ⬍ .05). According
to an alternative analysis that considers surgery a failure of
the injection, the absence of triggering rates were 19 of 35
patients for triamcinolone and 6 of 32 patients for
dexamethasone (p ⬍ .05). Six of 40 patients in the
dexamethasone group and 3 of 41 patients in the
Satisfaction was significantly greater in patients in the
triamcinolone cohort as measured on the SVAS than in
patients in the dexamethasone cohort 6 weeks after injection
(average 8.4 points [range, 2–10 points; SD, 2.3; 95% CI,
8.2–9.4 points] vs 6 points [range, 0 –10 points; SD, 3.5;
95% CI, 4.8 –7.1 points]; t ⫽ ⫺3.28, p ⬍ .01).
Results 3 Months After Injection
There was no significant difference in DASH scores at the
3-month evaluation: average 11 points in the
dexamethasone cohort (range, 0 – 42 points; SD, 14.6; 95%
CI, 5.5–20.5 points) and 13 points in the triamcinolone
cohort (range, 0 –77 points; SD, 19.7; 95% CI, 5.2–20.6

JHS 䉬 Vol A, April 

520 DEXAMETHASONE VERSUS TRIAMCINOLONE FOR TRIGGER FINGER
points) (t ⫽ ⫺0.51, p ⫽ .61). There was no significant
difference in absence of triggering rates 3 months after
injection when including patients treated surgically
(intention to treat): 22 of 31 patients in the dexamethasone
cohort and 27 of 41 patients in the triamcinolone cohort (p
⫽ .87). According to an alternative analysis that considers
patients receiving surgery as failures of the injection, the
rates of absence of triggering were also comparable between
cohorts: 17 of 31 patients in the dexamethasone cohort and
20 of 41 patients in the triamcinolone cohort (p ⫽ .42)—
numbers that reflect the success of injections alone. Five of
40 patients in the dexamethasone cohort and 7 of 44
patients in the triamcinolone cohort requested operative
treatment. Finally, 11 patients in the dexamethasone cohort
and 10 in the triamcinolone cohort had no triggering after a
single injection (p ⫽ .79). There was no significant
difference in satisfaction at the 3-month follow-up: average
SVAS of 7.7 points in the dexamethasone group (range,
4 –10 points; SD, 3.1; 95% CI, 6.2– 8.8 points) and 7.4 in
the triamcinolone group (range, 0 –10 points; SD, 3.3; 95%
CI, 6.3– 8.4 points) (t ⫽ 0.40, p ⫽ .69).
Improvement Within Each Group
Two-way repeated-measures ANOVA indicated a highly
significant time effect within each treatment group,
indicating significant improvement in DASH scores at 6
weeks (p ⬍ .01) and at 3 months (p ⬍ .01) compared with
initial DASH scores. Improvement in SVAS scores within
groups showed a trend of increased values that was nearly
significant in the dexamethasone group (p ⫽ .08), and
statistically significant in the triamcinolone group (p ⬍ .05).
Cure rates within groups showed also a trend of increment
for both arms of the study. The improvement was nearly
significant in the dexamethasone group (p ⫽ .09) and
significant in the triamcinolone group (p ⬍ .01).
Later Recurrence
Participation in the study officially ended with the 3-month
evaluation, and we did not have institutional review board
approval for additional patient contact. Nine patients (1
injected with dexamethasone and 8 with triamcinolone; p ⬍
.05) that had a documented absence of triggering during the
study period returned after study completion with a
recurrent trigger finger. Average time for presentation with
recurrence in this set of patients was 7 months (range, 2–13
months).
Predictors of Absence of Triggering
Univariate analysis identified triamcinolone (␹2 ⫽ 4.30, p ⬍
.01), older age (t ⫽ ⫺2.9, p ⬍ .01), and thumb
involvement (␹2 ⫽ 7.13, p ⫽ .07) as significant predictors
of absence of triggering 6 weeks after injection (variables
with p values lower than .08). The best multivariate model
included triamcinolone, older age, and thumb involvement
as predictors of absence of triggering at 6 weeks (adjusted
R2 ⫽ 0.37; p ⬍ .01), meaning that these 3 factors account
JHS 䉬 Vol A, April 
for 37% of the variability in absence of triggering among the
patients.
Univariate analysis identified thumb involvement (p ⬍
.01) and a second injection (p ⬍ .05) as significant
predictors of absence of triggering at 3 months. The best
multivariate model included both variables and accounted
for 21% of the observed variation in absence of triggering.
(adjusted R2 ⫽ 0.21, p ⬍ .01).
Predictors of DASH Scores
Initial DASH score was the only variable that correlated
with the DASH score 6 weeks after injection in the
univariate analysis (r ⫽ 0.63, p ⬍ .01). Because we found
only 1 variable presenting statistically significant correlation
in the univariate analysis, we did not perform a multivariate
analysis to establish a model of prediction for the DASH
score 6 weeks after the injection.
Univariate analysis identified the initial DASH score (r ⫽
0.65, p ⬍ .01) and the election of surgery (t ⫽ ⫺2.3, p ⬍
.05) as predictors of the 3-month DASH score. The best
multivariate model included the initial DASH score alone
and accounted for 63% of the variation in DASH scores.
(adjusted R2 ⫽ 0.63; p ⬍ .01).
Predictors of Satisfaction
Univariate analysis identified triamcinolone (t ⫽ ⫺3.3, p ⬍
.01), initial DASH scores (r ⫽ ⫺0.33, p ⬍ .01), and
election of surgery (t ⫽ 1.91, p ⫽ .06) as predictors of
satisfaction at 6 weeks. The best multivariate linear
regressional model included triamcinolone and initial DASH
score and explained 52% of the variation in satisfaction
(adjusted R2 ⫽ 0.52, p ⬍ .01).
Univariate analysis identified election of surgery (t ⫽ 3.7,
p ⬍ .01), a second injection, and thumb involvement (F ⫽
3.33, p ⬍ .05) as predictors of increased satisfaction 3
months after injection. The best multivariate linear
regression analysis included election of surgery and second
injection and accounted for 61% of the variation (adjusted
R2 ⫽ 0.61; p ⬍ .01).
DISCUSSION
The injection of corticosteroids into the flexor sheath was
initially advocated as a method of treatment for trigger
finger in 1950.14,15 The reported effectiveness has ranged
between 25% and 73% (average, 49%) among 12 studies we
identified.8,14 –24 In the largest retrospective study we are
aware of (338 patients), the overall improvement or absence
of triggering rate was 77%: 49% after a single injection, 23%
after 2 injections, and 5% after 3 injections.8 A prospective
study of 58 patients reported a 61% absence of triggering
after a single injection with a 27% recurrence rate
(successfully treated with repeat injection) and a 12% rate of
surgery.16 Two clinical trials have compared corticosteroid
injection with placebo: the success rates were 60% and 64%
versus 16% and 20% in the placebo controls,
respectively.19,24 Second or repeat injections are believed to
be half as effective as initial injections.8,15,17,25
 DEXAMETHASONE VERSUS TRIAMCINOLONE FOR TRIGGER FINGER
The current investigation should be interpreted with the
understanding that patients were free to choose other
treatments (patient autonomy was respected), and the
analysis was according to intention to treat principles. This
was necessary for approval by the human research
committee but also reflects practice-based treatment
effectiveness, which, it can be argued, is more relevant than
the pure efficacy of a single steroid injection in particular or
strict adherence to a treatment program in general. Second
injections and election of operative treatment were not
significantly different between cohorts, and a statistical
analysis excluding the 2 patients that changed steroids for
their second injection did not affect the results. Our
investigation answers the question, “What happens when a
single injection of either dexamethasone or triamcinolone is
given for trigger finger and then patients are allowed
autonomy in subsequent care?” It therefore does not
definitively address the difference between the 2 types of
corticosteroid under ideal circumstances.
It should be kept in mind that we included patients with
multiple digit involvement and patients with diabetes in
order to increase the generalizability (external validity) of the
findings. To be certain of the effects on patients with single-
digit involvement and those without diabetes, the study
would have to be repeated with different inclusion and
exclusion criteria.
The primary null hypothesis (that the average DASH
score 3 months after injection was not significantly different
between cohorts) was not rejected. Secondary analyses found
that triamcinolone had significantly better absence of
triggering rates, Quinnell grades, and patient satisfaction at
the 6-week evaluation but not at the 3-month evaluation.
Our data document that the effect of triamcinolone
injection may be more rapid but is ultimately no better than
that of dexamethasone injection. In additional secondary
analyses, thumb involvement was associated with a greater
likelihood of absence of triggering 3 months after injection.
Because these findings represent secondary hypotheses (the
DASH score was the primary focus of the study), they
would be strengthened by investigations designed to address
absence of triggering as a primary study question.
Our study cannot comment on the effectiveness of
subsequent injections because of the 3-month final
evaluation point. The short duration of the study was
selected because, at least in our practices, most patients with
persistent triggering 3 months after surgery elect surgery.
The much higher recurrence rate of triamcinolone injection
greater than 3 months after initial injection should be
interpreted in light of the fact that the study had ended and
these were elective patient visits. Considering the nature of
our practice and the successful randomization, we are
confident in our conclusion that triamcinolone has a higher
recurrence rate greater than 3 months after injection, but we
are not confident of the overall rate of such recurrence.
The observation that DASH scores at 6 weeks and 3
months were predicted by the initial DASH score is
JHS 䉬 Vol A, April 
521
consistent with prior research demonstrating that DASH
scores are determined to a large extent by psychosocial
factors such as depression and effective coping skills that may
not change as objective disease status changes.26 –28
On the basis of these data, patients can be advised that
the effect of corticosteroid injection may not be apparent for
greater than 6 weeks and patience is therefore merited,
particularly when using dexamethasone. They can also be
advised that two thirds of patients have no triggering within
3 months of the injection: 29% from a single injection
alone, and 51% with either 1 or 2 injections. Furthermore,
there appears to be an approximately 10% risk of recurrence
of the trigger finger after presumed resolution via
corticosteroid injection, particularly when using
triamcinolone.
REFERENCES
1.Maneerit J, Sriworakun C, Budhraja N, Nagavajara P.
Trigger thumb: results of a prospective randomised study of
percutaneous release with steroid injection versus steroid
injection alone. J Hand Surg 2003;28B:586 –589.
2.Murphy D, Failla JM, Koniuch MP. Steroid versus placebo
injection for trigger finger. J Hand Surg 1995;20A:628 – 631.
3.Quinnell RC. Conservative management of trigger finger.
Practitioner 1980;224:187–190.
4.Marks MR, Gunther SF. Efficacy of cortisone injection in
treatment of trigger fingers and thumbs. J Hand Surg 1989;
14A:187–190.
5.Saldana MJ. Trigger digits: diagnosis and treatment. J Am
Acad Orthop Surg 2001;9:246 –252.
6.Lambert MA, Morton RJ, Sloan JP. Controlled study of the
use of local steroid injection in the treatment of trigger
finger and thumb. J Hand Surg 1992;17B:69 –70.
7.Anderson B, Kaye S. Treatment of flexor tenosynovitis of
the hand (‘trigger finger’) with corticosteroids. A prospective
study of the response to local injection. Arch Intern Med
1991;151:153–156.
8.Newport ML, Lane LB, Stuchin SA. Treatment of trigger
finger by steroid injection. J Hand Surg 1990;15A:748 –750.
9.Pope DF, Wolfe SW. Safety and efficacy of percutaneous
trigger finger release. J Hand Surg 1995;20A:280 –283.
10.Brunton LL J, Parker K. Section IV: drug therapy of
inflammation. In: Brunton L, Lazo J, Parker K, eds.
Goodman & Gilman’s the pharmacological basis of
therapeutics. New York: McGraw-Hill Medical; 671–716.
11.Germann G, Wind G, Harth A. [The DASH (Disability of
Arm-Shoulder-Hand) Questionnaire—a new instrument for
evaluating upper extremity treatment outcome]. Handchir
Mikrochir Plast Chir 1999;31:149 –152.
12.Hudak P, Amadio P, Bombardier C. Development of an
upper extremity outcome measure: the DASH (Disabilities of
the Arm, Shoulder and Hand). The Upper Extremity
Collaborative Group (UECG). Am J Ind Med 1996;29:602–
608.
13.Diggle P, Liang K, Zeger S. Two-way repeated-measures
mixed modl analysis of variance (ANOVA). Analysis of
522 DEXAMETHASONE VERSUS TRIAMCINOLONE FOR TRIGGER FINGER
longitudinal data. Oxford: Oxford University Press;
1994:131–145.
14.Howard LD Jr, Pratt DR, Bunnell S. The use of compound
F (hydrocortone) in operative and non-operative conditions
of the hand. J Bone Joint Surg 1953;35A:994 –1002.
15.Akhtar S, Bradley MJ, Quinton DN, Burke FD.
Management and referral for trigger finger/thumb. BMJ
2005;331:30 –33.
16.Anderson B, Kaye S. Treatment of flexor tenosynovitis of
the hand (’trigger finger’) with corticosteroids. A prospective
study of the response to local injection. Arch Intern Med
1991;151:153–156.
17.Clark DD, Ricker JH, MacCollum MS. The efficacy of local
steroid injection in the treatment of stenosing tenovaginitis.
Plast Reconstr Surg 1973;51:179 –180.
18.Kazuki K, Egi T, Okada M, Takaoka K. Clinical outcome
of extrasynovial steroid injection for trigger finger. Hand
Surg. 2006;11:1– 4.
19.Lambert MA, Morton RJ, Sloan JP. Controlled study of the
use of local steroid injection in the treatment of trigger
finger and thumb. J Hand Surg 1992;17B:69 –70.
20.Buch-Jaeger N, Foucher G, Ehrler S, Sammut D. The
results of conservative management of trigger finger. A series
JHS 䉬 Vol A, April 
of 169 patients. Ann Chir Main Memb Super
1992;11:189 –193.
21.Fauno P, Andersen HJ, Simonsen O. A long-term follow-up
of the effect of repeated corticosteroid injections for
stenosing tenovaginitis. J Hand Surg 1989;14B:242–243.
22.Kraemer BA, Young VL, Arfken C. Stenosing flexor
tenosynovitis. South Med J 1990;83:806 – 811.
23.Quinnell RC. Conservative management of trigger finger.
Practitioner 1980;224:187–190.
24.Murphy D, Failla JM, Koniuch MP. Steroid versus placebo
injection for trigger finger. J Hand Surg 1995;20A:628 – 631.
25.Patel MR, Bassini L. Trigger fingers and thumb: when to
splint, inject, or operate. J Hand Surg 1992;17A:110 –113.
26.Doornberg JN, Ring D, Fabian LM, Malhotra L,
Zurakowski D, Jupiter JB. Pain dominates measurements of
elbow function and health status. J Bone Joint Surg 2005;
87A:1725–1731.
27.Ring D, Guss D, Malhotra L, Jupiter JB. Idiopathic arm
pain. J Bone Joint Surg 2004;86A:1387–1391.
28.Ring D, Kadzielski J, Fabian L, Zurakowski D, Malhotra
LR, Jupiter JB. Self-reported upper extremity health status
correlates with depression. J Bone Joint Surg
2006;88A:1983–1988.