Comparison of the effect of ketorolac versus triamcinolone acetonide injections for the treatment of de Quervain’s tenosynovitis: a double-blind randomized controlled trial

De Quervian’s disease is tenosynovitis of the first dorsal compartment described firstly Swiss surgeon Fritz de Quervain in 1895. A large UK-based study of this disease found that it commonly affected woman more than men, particularly in the 4th and 5th decades of life, with rate of 1.3% in women and 0.5% in men [1]. This condition causes severely painful radial-side wrist pain and impaired function. Treatment is directed at reducing inflammation through immobilization with the use of physiotherapeutic modalities, systemic nonsteroidal anti-inflammatory agents (NSAIDs), and/or local corticosteroid injections. Splinting with a thumb spica brace is rarely successful for long term relief, and failure and recurrence are often high [2‐4]. Corticosteroid injections have been reported to provide near complete relief with one or two injections, with symptomatic relief reported by 83% of patients with a single injection in one study [5]. However, although the corticosteroid injections have a very high success rate, there are some adverse side effects in some cases including skin hypopigmentation, tendon rupture and superficial radial nerve injury, and elevated blood glucose in diabetic patients.

In recent years, ketorolac tromethamine, an injectable non-steroidal anti-inflammatory drug (NSAID) has been gaining some popularity as either an adjunct or replacement for corticosteroid injections. Previous studies have found no detrimental side effects of intra-articular ketorolac injections on articular cartilage, ligaments or kinematic function of native knees in animal models [6, 7]. For de Quervain’s disease specifically, there has been only one study, a double-blinded randomized controlled trial which demonstrated the superiority of bethamethasone injection over ketorolac injection in the treatment of De Quervain’s tenosynovitis. However, the study population was too small to provide significance [8].

There is a lack to date of high quality, randomized controlled trials to provide data regarding the effectiveness of ketorolac compared with local corticosteroid injections for de Quervain's tenosynovitis. The present study was designed to compare the effect of these drugs on de Quervain’s disease, by comparing pain scores, functional responses between and grip and pinch strength between local injections of a corticosteroid and ketorolac. Our hypothesis was that ketorolac would be superior to the corticosteroid due to the direct anti-inflammation mechanism of action and its recent history of showing better outcomes in many aspects of other orthopedics diseases.

We screened 72 patients, eight of whom declined to participate (Fig. 1). The groups were not significantly different at baseline with regards to age, duration of symptoms, pre-operative VNRS scores for pain, DASH scores, and grip/pinch strength. At the follow-up, four participants were lost from the study because their pain had subsided (Fig. 1). Baseline data were not different between the participants who completed the study.

A total of 64 participants were enrolled in the study and randomized according to an allocation ratio of 1:1 to the ketorolac injection treatment group (31 patients) and the triamcinolone injection treatment group (33 patients). At the follow-up visits, accounting for the 4 patients who had withdrawn, there were 31 participants in the ketorolac group and 29 participants in the triamcinolone group who completed the study and were included in the analysis (Table 1). The analysis was performed on the original assigned groups, and there was no crossover between groups. There were no significant difference in age (p value = 0.67) or sex (p value = 1), hand dominance (p value = 0.73), occupation (p value = 0.42), underlying disease(s) (p value = 0.42), alcohol drinker (p value = 1), smoker (p value = 0.73), or symptomatic side (p value = 0.95) between the groups. The average duration of symptoms for the ketorolac group and triamcinolone group were 30 ± 37 and 30 ± 47.1 days, respectively (p value = 0.88). There were also no significant differences in oral medications (p value = 0.93). All of the study outcomes including pain, DASH score, grip strength, and pinch strength are shown in Table 2. There were no significant differences in all parameters at baseline. At 6 weeks the patients in the triamcinolone group had a lower average pain score (0.7) than the ketorolac group (5.3) (Fig. 2). An additional injection at six weeks with triamcinolone was given to most patients in the ketorolac group, while no-one in the triamcinolone group (25/31 versus 0/34) needed a second injection. With the DASH test, a lower score reflects better function. Patients injected with triamcinolone had a significantly lower average score than the ketorolac group (4.4 vs 34.1). For grip strength, there were significant improvements in both groups (p value 0.02 and 0.04 in right grip and left grip strength, respectively). There was no difference in pinch strength between the groups at the 6-week examinations.

This study was undertaken to compare ketorolac and triamcinolone injections in the treatment of radial styloid tenosynovitis. Both of these drugs decrease the inflammatory process, ketorolac by inhibiting cyclooxygenase production and triamcinolone by interfering with the arachidonic acid pathway. Therefore, we hypothesized that ketorolac could have positive results similar to triamcinolone in treating radial styloid tenosynovitis. The primary outcome of this study was to compare functional scores and pain scores between the two injection techniques. Both groups reported a decrease in both the DASH scores and pain scores, meaning improvement in the overall functional scores and decreases in the pain scores.

The study found that patients who received triamcinolone had lower pain scores and better functional scores, which was consistent with a previous published study [8] that a triamcinolone injection provided better short-term outcomes in terms of reduction in pain and improving functional scores at 6 weeks. Our study found that at 6 weeks nearly 90% of the triamcinolone patients had complete resolution of their symptoms while only 40% of those who received the ketorolac reported complete resolution. A study by Chadderdon C et al. [8] also found that 52.6% and 33.3% of patients injected with ketorolac and betamethasone, respectively, required an additional injection at 6 weeks. For the DASH scores, our study found that patients injected with triamcinolone had a significantly lower average score than the ketorolac group (4.4 vs 34.1) which was consistent with another study [8] which reported that patients injected with triamcinolone had a significantly lower average score than the ketorolac group (11.1 vs 32.5). Apart from the anti-inflammatory effect, corticosteroids may affect the connective tissue and adhesions between the tendon and the surrounding peritendinous tissues by inhibiting the production of collagen, other extracellular matrix molecules, and granulation tissue in these sites [9]. These reasons support the more potent and faster action of corticosteroids compared with NSAIDs. Other studies have also compared ketorolac and a corticosteroid injection in trigger finger [10‐12] and radial styloid tenosynovitis [8], and also reported the steroid injection gave better results, similar to our study, thus it may be concluded that NSAID injections may not be the best choice for use in tendon stenosis conditions. This could be because the cause of trigger finger and radial styloid tenosynovitis both result from stenosing tenosynovitis which may not respond well to anti-inflammatories and require more intensive interventions to deal with the symptoms.

However, there have been many studies which have compared the effectiveness between ketorolac and corticosteroids in various settings and conditions, including osteoarthritis of the hip [13], osteoarthritis of the knee [14], and adhesive capsulitis [15], all of which found that ketorolac provided superior results, in contrast with our results which found that triamcinolone was superior to the ketorolac. Even though our study favored triamcinolone over ketorolac in treating radial styloid tenosynovitis, there may still be a role for this drug in the treatment of radial styloid tenosynovitis patients contraindicated for steroids, for example, uncontrolled diabetes mellitus patients [10]. Another hypothesis is that ketorolac injections might reduce recurrence if given in combination with a steroid or if given the second time the patient develops recurrence symptoms, which should be investigated with further studies.

Following a ketorolac injection the patient should be monitored for adverse events such as local skin reactions, but none of our patients developed any such conditions. For the triamcinolone injection, there have been a few reports of documented adverse reactions experienced after steroid injections [8, 16‐22], including flare reaction, skin hypopigmentation and sensory radial nerve impairment. In our study almost half the patients in the triamcinolone group developed some degree of hypopigmentation.

The present study is believed to be the first full published study comparing ketorolac and triamcinolone injections in treating patients diagnosed with radial styloid tenosynovitis, in terms of pain scores, functional scores, grip and pinch strength scores, and also side effects. However, several limitations must be acknowledged, including the small sample size and lack of long-term follow-up. For example, a 6-month follow-up would have been better to assess treatment response with the ketorolac group. Further studies should include longer follow-ups in order to assess the long-term effect of these drugs. The other potential limitation concerns the injection site used, at the point of maximum tenderness as indicated by the patient as this may have interfered with outcomes. In order to select a more precise injection site, ultrasonography could help identify more precisely the site of the most stenosis tenosynovitis. Further studies could be designed to assess the value of using ultrasonography-guided injections in these patients.

Ketorolac injections resulted in inferior pain reduction, functional score and grip improvement than triamcinolone in patients with radial styloid tenosynovitis. Future studies are required to examine the effects of ketorolac in larger group and with longer follow-up periods to further elucidate the findings of this study.