Tendoscopic versus open release for de Quervain’s disease: earlier recovery with 7.21 year follow-up

De Quervain’s disease, or stenosing tenosynovitis of the first dorsal compartment of the radial styloid process, is a common affliction with a prevalence of 0.5% among men and 1.3% among women [1]. The cause of this disease is thought to be the thickening of the tendon sheath containing the abductor pollicis longus (APL) and the extensor pollicis brevis (EPB) tendons [2]. Nonsurgical management, including rest, splint immobilization, physiotherapy, and steroid injections into the tendon sheath, could provide effective symptoms relief in most of the patients [3‐5]. In resistant cases, surgical release of the first dorsal compartment is recommended. Open release has been widely studied since Fritz de Quervain firstly published the surgical technique in an article in 1895 [6]. Endoscopic release for de Quervain’s disease was firstly described by Slade and Merrell in 2007 [7] and was then studied by several researches for its good results and fewer complications [8‐10]. We prefer to call it tendoscopic rather than endoscopic as the procedure is done around the APL and EPB tendons, and to our knowledge none of them reported the long-term results of this minimum-invasive technique. We presented our study here to compare the time return to work and long-term results of tendoscopic technique with open release for de Quervain’s disease.

We retrospectively reviewed 56 consecutive patients (56 wrists) with a diagnosis of de Quervain’s disease who were treated with tendoscopic or open release from March 2005 to June 2013. All patients showed pain and tenderness over the first dorsal compartment and a positive sign of Finkelstein test despite a minimum of 3 months non-operative treatment, including rest, splinting, and steroid injection (0.5 ml betamethasone, DIPROSPAN®, MSD and 1 ml of 1% lidocaine) for up to three courses. Patients were excluded if: (1) patients had secondary arthritis to the wrist (related to rheumatoid arthritis, gouty arthritis, or post-infectious arthritis), (2) patients had previous traumatic injury to the wrist, (3) patients had bone hyperplasia at the radial styloid, and (4) patients were involved with other symptomatic upper limb disorders, such as adhesive capsulitis, rotator cuff tear, lateral epicondylitis, and carpal tunnel syndrome etc. Of a total of 56 patients, 50 patients met the study inclusion criteria; 41 (82%) were available for follow-up assessment. These 41 patients were divided into two groups: 20 patients who were treated with tendoscopic release (group A); 21 patients who were handled with open release (group B).

No significant difference of basic characteristics was found between groups regarding gender, age, profession, pre-operative duration of symptoms, pre-operative pain and disability scores or follow-up period (Table 1). The average operative time was 19.62 ± 4.17 min in group A and 17.87 ± 3.42 min in group B (P = 0.324). The average time return to work was 5.1 ± 1.37 days in group A and 15.33 ± 4.89 days in group B (P = 0.006). The average hospital cost was $362.00 ± 20.30 in group A and $429.38 ± 25.85 in group B (P = 0.137). Clinical outcomes were shown in Table 2 and Fig. 4. The mean VAS and DASH scores decreased significantly in both groups at 1 month and last follow-up visit (P < 0.001). At one month, the mean VAS score and DASH score in group A were significantly better than in group B (P < 0.05 and P < 0.001, respectively) (Table 2). There was no difference between groups at last follow-up. However, changes of the mean VAS pain score were not significantly different between groups at neither 1 month postoperative (p = 0.062) nor last follow-up visit (p = 0.862), compared with baseline. In addition, the improvement of the mean DASH score in group A was significantly greater at 1 month than in group B (34.74 ± 10.99 in group A and 23.58 ± 12.01 in group B, P < 0.01). However, the changes from baseline to last follow-up visit showed no significant difference between two groups (52.36 ± 11.87 in group A and 49.25 ± 14.55 in group B, P = 0.460). The surgical scar was evaluated using the POSAS scale, as shown in Table 2. Both the OSAS and PSAS scores were significantly better at last follow-up point in the tendoscopic release group.

One patient in group A had cephalic vein injury and finally got relieved by manual compression. Three patients in group B were involved with radial sensory nerve injury but had no signs of neuroma. No patients had tendon subluxation or symptom relapse. In addition, all patients showed negative on Finkelstein’s test at last follow-up visit.

To our knowledge, conservative treatment could effectively relieve pain and improve joint motion in acute phase of de Quervain’s disease. However, its long-term efficiency is dissatisfied with high recurrence rate. Surgical treatment is indicated following the failure of non-operative treatment for symptomatic de Quervain’s disease. Complications associated with surgical release include injury to cephalic vein or the radial sensory nerve, tendon subluxation postoperatively, hypertrophic and/or painful scarring, wound infection or delayed wound healing, and shoulder hand syndrome [13‐19]. Iatrogenic injury to the radial sensory nerve is a common complication following the decompression [17]. Therefore, searching for a minimally invasive surgery procedure is of great value for those intractable de Quervain’s disease.

In this retrospective study, we reported our findings comparing open versus tendoscopic release for 41 patients (41 wrists) with a mean follow-up period of 7.21 years (range, 4–12 years). The average time return to work was significantly less in group A than group B, benefiting from less invasive procedure and earlier symptom relief. Our data demonstrated that tendoscopic release provided better clinical outcomes in terms of VAS and DASH scale at 1 month follow-up. In Kang’s randomized trial [10], short-term advantages of tendoscopic release was also confirmed that the clinical outcomes were significantly better in tendoscopic release group at 1 month follow-up. In Kang’s randomized trial [10], short-term advantages of tendoscopic release was also confirmed that the clinical outcomes were significantly better in tendoscopic release group at 12 weeks, but there was no significant difference between tendoscopic and open release at 24 weeks. Our follow-up period was much longer, and for the long-term efficacy at last follow-up visit, both tendoscopic and open techniques provided satisfactory results, and there was no significantly difference between two groups.

Slade [7] firstly described the tendoscopic technique for de Quervain’s disease and reported satisfactory results with no significant complications. However, Kang et al. [10] reported that 3 (11%) patients in tendoscopic release group were involved with superficial radial nerve injury and that value in open release group was 9 (36%). Scheller et al. presented successful results with no nerve injury of operative treatment for de Quervain’s disease and Finkelstein’s test was negative in all 94 patients after a minimum of 15 years follow-up [19]. In our study, patients treated with tendoscopic release had lower rate of neurovascular injuries than open release: 1 (5%) in group A and 3 (14%) in group B (Table 2). No nerve injury was found in the tendoscopic release group. The only one case of cephalic vein injury occurred at the time of synovitis debridement, damaged by the shaver with excessive suction. The bleeding got relieved by manual compression finally. By contrast, 3 cases of radial sensory nerve injury occurred in the open release group but had no signs of neuroma. We started our tendoscopic technique since 2008. Of those 20 patients available for follow-up, the mean length of follow-up was 6.75 years (range, 4–10 years). Considering the anatomy of the fibro-osseous tunnel on the radial aspect, two transverse portals were made: one 1.5 cm distal and the other 2–2.5 cm proximal to the radial styloid. These two portals were alternatively used for visualization or manipulation, depending on the variations in the anatomy of the first dorsal compartment. Unlike two portals endoscopic technique, Karakaplan et al. [8] practiced one portal endoscopic release technique among 10 patients. In terms of complications, one patient had significant scar tenderness and two patients had paresthesia of the superficial radial nerve which was resolved spontaneously in 12 weeks postoperatively. In hand surgery, there is an increasing application of percutaneous needle technique for de Quervain’s disease; however, the outcomes were mixed. In a cadaver study by Güleç et al. [20], percutaneous needle release with an 18-G needle on 48 cadaver wrists was associated with high rates of incomplete release (47.9%) and tendon damage (39.6%) which was thought to be due to the presence of a septum. Lapègue et al. [21] applied ultrasound-guided percutaneous release of the first extensor tendon compartment with a 21-G needle; in this study, the clinical outcomes were feasible; however, the release was confirmed as “partial” in its cadaver study. By contrast, complete release under direct vision in our study was confirmed, which is essential to reduce postoperative recurrence of this disease.

Different with the outside-in technique of open procedure, tendoscopic technique provided inside-out release of the first extensor retinaculum, minimizing the rate of radial sensory nerve injury, as well as tendon injury and adhesion of subcutaneous tissue around the incisions. In our experience, synovial hyperplasia is with high prevalence in de Quervain’s disease, and synovial debridement could be well done under direct vision in tendoscopic technique. Meanwhile, by abducing the thumb, complete release of the retinaculum was confirmed with a 360° field view of the tendons under tendoscopy. In addition, our tendoscopic knife could be almost non-invasive for its retractable blade. The blade remained retracting inside of the blunt tip in the process of inserting and shuttling alongside the tendons.

Pregnant and perimenopausal women are more vulnerable to de Quervain’s disease, and esthetic appearance of the scar is really a considerable problem for this group that wrists are always uncovered by clothing. In our study, the POSAS score was applied for esthetic evaluation of the scar, both the patient’s subjective opinion (PSAS) and observer’s assessment (OSAS) showed that small transverse incisions in tendoscopic release group (Fig. 5) are with higher satisfaction than those longer longitudinal incisions in open release group, which exactly meets the medical esthetics. As a minimally invasive technique, tendoscopic release offers good visualization of the involved structures and equivalent decompression without the need for an extensive and more aggressive open approach. However, the manipulable space of wrist arthroscopy is much narrower than the arthroscopy of the knee and shoulder. In addition, the tendoscope and the other devices are nearly on the same axial line; this face-to-face approach has a steep learning curve and demands some arthroscopic skills.

In our experience,the tendoscopic release for De Quervain's Disease must pay attention to following several points. Understanding the surgical anatomy of the first extensor compartment and the adjacent structures is essential. Marking of the cephalic vein and the radial sensory nerve before surgery could be helpful. Sharp dissection through the subcutaneous tissue is avoided. The operator should use the shaver carefully with the blade back to the neurovascular structures and the vacuum suction should be opened after partial synovial debridement to avoid suction-induced neurovascular injury. Finally, professional arthroscopic instruments (e.g., the telescopic knife and the 3.0-mm shaver) would be favorable to the enhancement of operative efficiency and effect.

The long-term results of our study are encouraging and demonstrate that tendoscopic release technique for de Quervain’s disease could provide earlier symptom relief and earlier recovery with fewer complications and a more desirable scar, as well as equivalent successful long-term outcome when compared with traditional open release.