Is Pulley Reconstruction Better Than Pulley Release for De Quervain's Tenosynovitis? A Double-Blind Randomized Controlled Trial

Abstract

Purpose  The aim of this study was to compare the results of surgeries of the De Quervain's disease (DQD) through a randomized control trial.

Materials and Methods  We treated 40 cases of De Quervain's tenosynovitis in 2 groups: 20 patients by pulley release method and another 20 patients by pulley reconstruction method. The patients were selected as per the random table number. The clinical data, numeric Visual Analogue Scale (VAS) score, Quick DASH—Disabilities of the Arm, Shoulder, and Hand—score, Mayo Wrist Score, and subluxation of the tendons in dynamic ultrasonography (USG) during wrist hyperflexion and abduction of the thumb test were noted preoperatively and 6 months after the operation. All patient charts were reviewed and data analysis was done after completion of the study.

Results  All patients improved after surgery. There was no difference in clinical outcome data, numeric VAS score, Quick DASH Score, and Mayo Wrist Score ( p -value > 0.05). There were four patients with tendon subluxation under USG in pulley release group, out of which only one patient was clinically symptomatic. There was no tendon subluxation in pulley reconstruction group. The tendon subluxation between the two groups was not found to be statistically significant ( p -value: 0.661).

Conclusion  Although, higher numbers of subluxation were found in release group, there was no statistically significant difference in the outcome of the two surgical procedures for DQD. This study is a pilot study, and it may act as the groundwork over which further studies may be performed.

Level of Evidence  This is a Level I study.

De Quervain's disease (DQD) is a stenosing tenosynovitis of the abductor pollicis longus (APL) tendon and the extensor pollicis brevis (EPB) tendon within the first extensor compartment of the wrist. ^{1 2 3} It is named after a Swiss surgeon, Fritz de Quervain, who noted five such patients in 1895. The patients had tender and thickened first dorsal compartment (DC1) of the wrist with pain during activities over this area. Initially thought to be an inflammatory condition, pathological specimens showed collagen disorientation and mucoid change. ^{4 5} It may be more appropriate to regard DQD as tendinosis rather than tendinitis. DQD has been managed by different treatment modalities, which include splints, physiotherapy, local steroids, and surgery. Other modalities include ultrasound-guided percutaneous tenotomy, acupuncture, platelet-rich plasma injections, hyaluronic acid injection, and prolotherapy. ⁶ The operative treatment has been considered as an option by surgeons when noninvasive or less invasive modalities have failed to give results.

The operative treatment includes the release of constricted fibrous pulley of DC1, and in some systemic conditions, excision of hypertrophied tenosynovium. This procedure has the potential complication of subluxation of the APL and EPB tendons, following a release of the first extensor compartment. Several authors have supported reconstruction of the compartment through different methods to prevent the tendon subluxation. ^{7 8 9} The Z-plasty of the pulley is one of the methods for DC1 pulley reconstruction. ¹⁰ The DC1 pulley stabilizes the tendons during flexion of the wrist and abduction of the thumb. These movements place the maximum stress on the inflection point of the tendons within the first extensor compartment. At this specific movement, the pulley prevents bowstringing of the tendons, and therefore its anatomic features should be respected. ¹¹ There are remote chances of recurrence with pulley reconstruction as the tenosynovitis may recur if potential large space is not created by reconstruction of the pulley.

DC1 pulley reconstruction is supposed to be better surgical technique as it prevents the subluxation. A carefully performed release of the pulley has been reported to give good outcomes. To the best of our knowledge, there are no published studies that have compared the results of these two different surgical techniques. The purpose of this study was to compare the results of the two surgical techniques, pulley release and pulley reconstruction, through a double-blind randomized control trial.

Materials and Methods

Approval for the study was taken from ethics committee of the institute. The informed, written consent was obtained from all the participants of the study. The sample size was decided based on the article by Arons. ¹² The prevention of tendon subluxation was taken as primary outcome measure, so according to the incidence of tendon subluxation of 1 in 16 cases, the absolute risk reduction was calculated to be 0.625. Hence, the sample size came to be 16 per group, with p -value = 0.05 and power of study = 80%. Assuming a 20% attrition risk in view of the prospective nature of the study, a sample size of 20 per group was taken.

The patients with clinically diagnosed DQD, with pain more than 6 months and age above 18 years, were recruited. Pregnant, postpartum females and DQD due to secondary causes were excluded ( Table 1 ). The CONSORT—Consolidated Standards of Reporting Trials—diagram explains the reasons for exclusion ( Fig. 1 ). Randomization was performed using a computer-generated random table. Allocation of the patients was concealed using opaque sealed envelopes according to the random number table generated by biostatics department of our institute. It was done so that the surgeon is kept blind before the surgery. This is the highest possible way of keeping the surgeon blind in a clinical trial. The surgeon and the patient were kept blind for this study (double-blind randomized control trial). The written consent of the patient was taken before enrolment in this study. The envelope was handed over to the operating surgeon just before the surgery inside the operation theater.

Table 1. Inclusion and exclusion criteria used for recruitment of the patients with DQD.

Inclusion criteria	Exclusion criteria
1. Patients with De Quervaiǹs tenosynovitis having the pain for more than 6-month duration and not responding to conservative management 2. Age of the patients should be more than 18 years	1. Pregnant females with DQD 2. Inflammatory arthritis with DQD 3. Patients with DQD in postpartum period 4. Patients not willing to participate in the study

Abbreviation: DQD, De Quervain's disease.

Fig. 1.

The CONSORT—Consolidated Standards of Reporting Trials—diagram showing the recruitment of the patients.

The patients enrolled for study were assessed preoperatively, and postoperatively at 1 week, 3 months, and 6 months. These patients were assessed clinically and radiologically. In the clinical evaluation, patient-related outcome measures such as Visual Analogue Scale (VAS) score, Mayo Wrist Score, and Quick DASH—Disabilities of the Arm, Shoulder, and Hand—score were recorded pre- and postoperatively. The clinical evaluation also included the following tests: Finkelstein test, Eichhoff's test, wrist hyperflexion and abduction of the thumb (WHAT) test, and Hitch hiker's sign in pre- and postoperative periods. Pre- and postoperative ultrasonography (USG) was done to look for subluxation of tendons. The postoperative USG was done at 6 months after surgery.

Surgical Techniques

The patients were operated under wide awake local anesthesia and no tourniquet (WALANT) technique. A single injection of WALANT cocktail (total of 11 mL: 5 mL of 2% lignocaine + 5 mL of normal saline + 1 mL of soda bicarbonate) was given subcutaneously at the incision site after sensitivity testing for lignocaine. The patients were operated after 30 minutes of the injection. The surgical procedures were done under 4.2X loop magnification. All the surgeries were performed by fellowship trained hand surgeon, MTA.

Release (Group 1)

A 2 cm long, longitudinal skin incision was made over the first extensor compartment, directly over the radial styloid process. Great care was taken in identification and retraction of sensory nerve branches of the radial nerve. The pulley was released from its dorsal margin. The EPB and APL tendons were exposed and identified just distal to the first extensor compartment. If there was a septum between the EPB and APL tendons, the septum was incised longitudinally and excised ( Fig. 2 ). Synovectomy was performed. Any indentation or multiple slips were noted ( Fig. 3 ).

Fig. 2.

The septum is visible, which was excised during release of the pulley.

Fig. 3.

Indentation over the tendons is prominent and the tendons have wider diameter proximal to this constriction.

Reconstruction (Group 2)

We used the technique as described by van der Wijk et al. ¹⁰ Skin and subcutaneous dissection was performed as mentioned for the release group. Consecutively the tendon sheath was dissected and incised proximal and distal to the pulley. A Z-shaped incision of the pulley was made from proximal dorsal to distal palmar. If there was a septum between the EPB and APL tendons, the septum was excised. Synovectomy was performed. To enlarge the pulley, the bases of the flaps were incised up to half their size. To avoid translation at the inflection point, the palmar flap was mobilized over half of its length in its proximal portion and the dorsal flap in the same fashion over its distal portion. The transversely mobilized sides of the pulley were consequently sutured together, thereby enlarging the pulley. The volar flap was sutured in a layered fashion, taking care to put it under the dorsal flap, using PDS 4–0 suture (Ethicon, Calle Durango, Chihuahua, Mexico) ( Fig. 4 ). The enlarged pulley was checked for patency by passing the tip of the curved mosquito artery forceps between the new pulley and the tendons. Patients were then asked to flex their wrist while abducting their thumb actively to verify the stability of the tendons of the first extensor compartment. After both the procedures, a palmar, below elbow slab was applied, and the wrist was immobilized for 10 days.

Fig. 4.

The pulley has been reconstructed by doing Z-plasty and the flaps are sutured with PDS 4–0 suture.

Statistical analysis was performed using Statistical Package for Social Sciences (Virginia, USA), Version 20. Quantitative data were analyzed using Fisher's exact test. For continuous data, Student's t -test was used if the data followed a normal distribution. However, if the continuous data did not follow a normal distribution, Wilcoxon signed-rank test was employed. The results of both the groups have been compared at 6 months in this study.

Results

Sixty-eight patients were assessed for the study. Twenty-eight patients were not included in the study, as 18 patients did not meet the inclusion criteria ( Table 1 ), 8 patients did not give the consent to participate in the study, and 2 patients dropped out due to other reasons. Forty patients were recruited in the study, with 20 patients in each group, as per the random table number generated at the beginning of the study. The total 40 patients were followed up and were analyzed at the conclusion of the study.

The mean age in group 1 (release group) was 43.2 years (range: 25–60 years) and in group 2 (reconstruction group) the mean age was 41.3 years (range: 26–63 years). There were 4 males and 16 females in group 1, while 6 males and 14 females in group 2. The mean duration of symptoms in group 1 was 10.2 months, and 6.9 months in group 2. In group 1, 3 (15%) patients had taken steroid injections twice, and 1 (5%) patient had taken 4 steroid injections while rest 16 (80%) patients had not taken steroid injection in DC1. In the reconstruction group, 1 (5%) patient had taken single steroid injection while 19 (95%) patients had not taken steroid injection. There was no statistically significant difference between the two groups in terms of age, sex, duration of the symptoms, and previous treatment received in this study ( Table 2 ).

Table 2. Patient demographics in both groups comparing background including age, sex, dominant/nondominant hand, and preoperative outcome measures.

Parameter	Release group	Reconstruction group	p -Value
Age (years)	43.2 (range: 25–60)	41.3 (range: 26–63)	0.5
Sex	Males: 4 (20%) Females: 16 (80%)	Males: 6 (30%) Females: 14 (70%)	0.7
Affected side	Left: 8 (40%) Right: 12 (60%)	Left: 14 (70%) Right: 6 (30%)	0.1
Duration of symptoms (months)	10.2 (range: 6–24)	7.0 (range: 6–12)	0.3
Dominant hand	Right: 20 (100%)	Right: 20 (100%)	1
Comorbidities	4 (20%)	3 (15%)	1
Swelling over DC1	17	16	1
Tenderness over DC1	19	20	1
Finkelstein test	20	19	1
Hitch hiker's test	19	16	0.3
Eichhoff's test	20	19	1.0
WHAT test	20	20	1
VAS Score	7.7	7.6	0.7
Mayo Wrist Score	64.5	63	0.7
Quick DASH score	52.4	52.1	0.9

Abbreviations: DASH, Disabilities of the Arm, Shoulder, and Hand; DC1, first dorsal compartment; VAS, Visual Analogue Scale; WHAT, wrist hyperflexion and abduction of the thumb.

Note: Both the groups were essentially same and there was no significant statistical difference.

The Clinical Evaluation Tests

Upon preoperative clinical evaluation ( Table 3 ), in group 1, 85% patients have swelling over DC1, 95% patients have tenderness over DC1, 100% patients have positive Finkelstein test, 95% patients have positive Hitch hiker̀s sign, 100% patients have Eichhoff̀s test positive, and 100% patients have WHAT test positive. In group 2, 80% patients have swelling over DC1, 100% patients have tenderness over DC1, 95% patients have positive Finkelstein test, 80% patients have positive Hitch hiker̀s sign, 95% patients have Eichhoff̀s test positive, and 100% patients have a positive WHAT test. There was no statistically significant difference ( p -value > 0.05) in the preoperative comparison of the clinical evaluation tests between the groups.

Table 3. Patient clinical evaluation test data.

S. no.	Pre-S	Post-S	Pre-t	Post-t	Pre-F	Post-F	Pre-H	Post-H	Pre-E	Post-E	Pre-W	Post-W
Release (Group 1)
Percent positive	85	0	95	5	100	0	95	0	100	0	100	0
Reconstruction (Group 2)
Percent positive	80	0	100	10	95	0	80	0	95	5	100	0
p -Value	>0.05	>0.05	>0.05	>0.05	>0.05	>0.05	>0.05	>0.05	>0.05	>0.05	>0.05	>0.05

Abbreviations: Post-E, postoperative Eichhoff's test; Pre-E, preoperative Eichhoff's test; Post-F, postoperative Finkelstein test; Pre-F, preoperative Finkelstein test; Post-H, postoperative Hitch hiker's test; Pre-H, preoperative Hitch hiker's test; Post-s, postoperative swelling over first dorsal compartment (DC1); Pre-s, preoperative swelling over DC1; Post-t, postoperative tenderness over DC1; Pre-t, preoperative tenderness over DC1; Post-W, postoperative WHAT—wrist hyperflexion and abduction of the thumb—test; Pre-W, preoperative WHAT test.

Upon postoperative evaluation at 6-month follow-up of the clinical evaluation tests, in group 1, 0% patients have swelling over DC1, 0% patients have tenderness over DC1, 0% patients have positive Finkelstein test, 5% (one patient) patients have positive Hitch hiker̀s sign, 0% patients have Eichhoff̀s test positive, and 0% patients have WHAT test positive. In group 2, 0% patients have swelling over DC1, 10% (2/20) patients have tenderness over DC1, 0% patients have positive Finkelstein test, 0% patients have positive Hitch hiker̀s sign, 5% (1/20) patients have Eichhoff̀s test positive, and 0% patients have a positive WHAT. There was no statistically significant difference ( p -value > 0.05) in the postoperative comparison of the clinical evaluation tests between the groups. The statistical evaluation was done using Fischer exact test.

Patient-Related Outcome Measures

Upon preoperative evaluation of the patient-related outcome measures ( Table 4 ), in group 1, the average Mayo Wrist Score was 64.5, the average Quick DASH score was 52.4, and the numeric VAS score at work was 7.7. In group 2, the average Mayo Wrist Score was 63, the average Quick DASH score was 52.1, and the numeric VAS score at work was 7.6. All the 40 participants had no subluxation of tendons in the preoperative USG. There was no statistically significant difference ( p -value > 0.05) in the preoperative comparison of the patient-related outcome measures between the groups.

Table 4. Patient-related outcome measure data.

S. no.	A	S	D/N	Pre-MWS	Post-MWS	Pre-QDS	Post-QDS	Pre-VAS	Post-VAS	(Post-Sublux-ten)
Release (Group 1)
1.	57	F	D	60	100	68.2	0	10	0	No
2.	29	F	D	85	100	25	0	5	0	No
3.	39	F	D	75	100	56.8	0	6	0	No
4.	29	F	D	95	100	25	0	4	0	No
5.	48	F	N	40	85	79.5	18.2	10	2	No
6.	25	F	D	70	100	52.3	0	7	0	No
7.	29	F	N	40	100	88.6	4.5	8	0	No
8.	49	F	N	75	100	25	4.5	7	1	No
9.	41	M	D	75	95	34.1	6.8	7	2	Yes
10.	60	M	D	60	95	61.4	6.8	8	1	No
11.	50	F	D	65	95	47.7	9.1	8	3	Yes
12.	59	F	N	65	100	47.7	0	8	0	No
13.	60	F	D	65	95	61.4	4.5	8	1	No
14.	29	F	N	60	85	56.8	18.2	9	2	No
15.	52	F	D	60	95	56.8	4.5	9	1	No
16.	39	M	N	55	100	65.9	0	9	0	No
17.	48	M	D	70	100	36.4	0	7	0	No
18.	41	F	D	50	95	63.6	4.5	8	1	Yes
19.	40	F	N	55	100	59.1	0	9	0	Yes
20.	40	F	N	70	100	36.4	0	7	0	No
Average	43.2			64.5	97	52.4	4.1	7.7	0.7	4
Reconstruction (Group 2)
1.	30	M	D	90	100	47.7	0	6	0	No
2.	38	F	N	55	100	52.5	0	7	0	No
3.	26	F	D	50	95	70.5	0	6	1	No
4.	26	M	D	55	100	36.4	0	6	0	No
5.	44	F	N	45	100	72.7	0	7	0	No
6.	37	M	N	60	95	38.6	6.8	7	2	No
7.	63	F	N	85	100	36.4	0	8	0	No
8.	42	M	D	85	100	27.3	0	8	0	No
9.	46	F	N	40	95	79.5	9.1	10	3	No
10.	50	F	D	70	95	47.7	4.5	8	1	No
11.	39	F	N	80	100	25	0	4	0	No
12.	48	F	D	50	95	52.3	4.5	8	1	No
13.	55	F	N	75	100	50	0	8	0	No
14.	26	M	N	55	95	54.5	6.8	8	2	No
15.	39	F	N	75	100	50	0	8	0	No
16.	47	F	N	60	95	54.5	4.5	8	1	No
17.	50	F	N	65	100	59.1	0	8	0	No
18.	35	F	N	55	95	65.9	4.5	9	1	No
19.	39	M	N	55	100	56.8	0	8	0	No
20.	46	F	N	55	95	63.6	4.5	9	1	No
Average	41.3	M	D	63	97.8	52.1	2.3	7.6	0.7	No

Abbreviations: A, age; D, dominant hand; F, female; M, male; N, nondominant hand; Post-MWS, postoperative Mayo Wrist Score; Pre-MWS, preoperative Mayo Wrist Score; Post-QDS, postoperative Quick DASH—Disabilities of the Arm, Shoulder, and Hand—Score; Pre-QDS, preoperative Quick DASH Score; Post-Sublux-ten, postoperative subluxation of the tendons at 6 months seen by ultrasonography; Post-VAS, postoperative Visual Analogue Scale; Pre-VAS, preoperative Visual Analogue Scale; S, sex.

Upon postoperative evaluation at 6-month follow-up of the patient-related outcome measures, in group 1, the average Mayo Wrist Score was 97, the average Quick DASH score was 4.1, and the numeric VAS score at work was 0.7. In group 2, the average Mayo Wrist Score was 97.8, the average Quick DASH score was 2.3, and the numeric VAS score at work was 0.7. There was no statistically significant difference ( p -value > 0.05) in the postoperative comparison of the patient-related outcome measures between the groups.

Upon intragroup analysis of Mayo Wrist Scores, in group 1, 18 patients had excellent results and 2 patients had good results, while in the group 2, all the 20 patients had excellent results. In both the groups there was significant improvement in all the patient-related outcome measures in postoperative period. The statistical evaluation was done using chi-squared test for VAS and Quick DASH scores while Mann–Whitney U test was used for Mayo Wrist Score.

Tendon Subluxation ( Table 4 )

There were four patients in group 1 (pulley release) where tendon subluxation was demonstrated in USG at 6-month follow-up ( Fig. 5 ). None of the patient required an intervention for the tendon subluxation.

Fig. 5.

The tendon has subluxed 1.05 cm (0.86 cm + 0.19 cm) over the radial styloid in WHAT (wrist hyperflexion and abduction of the thumb) test.

Discussion

DQD is a common cause of radial-sided wrist pain. Initially, the mainstay of treatment is conservative, which includes splint immobilization, physiotherapy, and rehabilitation. Upon failure of these methods, a steroid injection may be given into the tendon sheath. However, symptom resolution after corticosteroid injection may take up to 18 months. ¹³ Repeated steroid injections have potential complications such as atrophy of the subcutaneous tissue and tendon rupture. In cases where conservative management has failed or the patient is willing for one-time solution, surgical intervention may be performed.

It is essential to realize that many complications are associated with this surgery. The complications include radial sensory nerve related (neuroma, scarring, and adhesions of the nerve), tendon subluxation, unsatisfactory scar, minor causalgia, and shoulder hand syndrome. ¹² It is also essential to diagnose the associated or related conditions to it, which include proximal/distal intersection syndromes, trigger finger, nonarticular rheumatism, fibromyositis, and tennis elbow. To minimize the complications, different efforts have been made to modify the surgical techniques. The use of longitudinal incision is one such modification, where there are lesser chances of injury to the sensory branch of radial nerve as it is easy to explore the nerve. There are lesser chances of tendon subluxation also as the sheath is released from its dorsal margin. There are higher chances of unhealthy scar as the Langerhans's lines are crossed at 90 degrees. A randomized control trial was conducted by Abrisham et al; they compared the clinical outcomes of surgical treatment of DQD with longitudinal and transverse incision, where they found that the longitudinal incision was better than the transverse incision for the surgical treatment of DQD. ¹⁴ In another short-term study (3-month follow-up), there were fewer complications with longitudinal surgical incision than with transverse incision, although there was one case of hypertrophic scar with longitudinal incision. ¹⁵ It is difficult to compare the two surgical approaches as the surgeoǹs technique, experience of the surgeon, training of the surgeon, and frequency of following a particular surgical procedure also play an important role. Lee et al used transverse incision for the release of DC1 without any significant complication. ¹⁶ We used longitudinal incision in our surgeries and this also may be a factor behind fewer complications in both the groups.

Lee et al conducted a study in 33 patients to look for the outcome of surgical release of the DC1 using DASH score. ¹⁶ They concluded that the outcome was good with minimal morbidity. They also reported no subluxation of the tendon of the first extensor compartment. However, their study was limited by the fact that it was a retrospective, noncomparative study, and the ultrasonographic evaluation to look for subluxation of tendons was not conducted. In this study, all 20 patients improved after DC1 pulley release, as measured through the clinical parameters and outcome scores. We also found that four patients had a subluxation of tendons on WHAT test on the 6-monthly USG follow-ups. However, out of the four patients, only one patient had a clinical subluxation of the tendon. The patient had a 6-monthly VAS score of 3, Quick DASH score of 9.1, and Mayo Wrist Score of 95. She complained of an occasional feeling of snapping at the wrist. None of the other three patients had a clinical subluxation or feeling of snapping at the wrist and had good outcome scores. Lesser number of patients in group 1 had subluxation as the DC1 pulley was released from its dorsal attachment while its volar attached part prevented the subluxation.

Van der Wijk et al conducted a study on 45 DQD patients having DC1 pulley reconstruction and concluded that none of the patients required reoperation for tendon instability or incomplete decompression. ¹⁰ However, two patients had a click around radial styloid after surgery. In our study, we did a DC1 pulley reconstruction in 20 patients and found that the clinical parameters and outcome scores improved significantly at the 6-month follow-up. None of the patients had a subluxation of the tendons in USG or clinically. Although there were higher number of the patients who had subluxation of tendons in DC1 release group than the DC1 pulley reconstruction group, the difference was not found to be statistically significant ( p -value: 0.661).

In this study, the secondary objective was to compare the clinical parameters and patient-related outcome scores between the two groups. There was no significant difference in the clinical parameters or outcome scores at 6-month follow-up between both the groups.

As described by Belsole, and also by Scheller, complications following surgery for De Quervain's tenosynovitis include sensory deficit due to damage to the superficial radial nerve, painful or adherent scar, incomplete decompression of the compartment, and the troublesome volar subluxation of tendons. ^{17 18} We encountered two complications during our study; one patient had a clinical subluxation of tendon associated with occasional snapping or clicking sound. The other patient had developed a hypertrophic scar and is currently taking treatment for the same under the dermatology department in the form of local triamcinolone injections and silicone sheath.

This study gives us important information about increased chances of tendon subluxation with DC1 pulley release procedures. Although the difference was not statistically significant, it may be related to small sample size. If a multicentric study with large sample size is conducted, then we may be able to find the difference between the two procedures. The other limitation of this study is short follow-up, as our results focused on the medium-term outcomes for both the procedures. A longer follow-up may have a different outcome. This study is a pilot study, and it may act as the groundwork over which further studies may be performed.

Funding Statement

Funding None.