Spontaneous Rupture of the Extensor Pollicis Longus Tendon: A Systematic Review

Abstract

Background:

Extensor pollicis longus (EPL) rupture and tenosynovitis of the third dorsal compartment is often described in association with a history of rheumatoid arthritis or in the setting of a distal radius fracture. However, the literature suggests multiple other potential factors that may lead to a seemingly spontaneous rupture.

Methods:

We performed a systematic review following guidelines set forth by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement. The search consisted of headings and keywords related to tendon injuries, tendinopathy, hand surgery, tendon transfer, and injections, as published in reports and studies. Citations were screened by title and abstract against predetermined inclusion and exclusion criteria by 2 independent reviewers, with a third reviewer resolving discrepancies. To be eligible, articles had to meet the following inclusion criterion: describe cases of spontaneous EPL rupture or tenosynovitis of the third dorsal compartment. The exclusion criterion was any history of distal radius fracture or rheumatoid arthritis.

Results:

We identified 29 articles that met the inclusion criterion.

Conclusions:

A myriad of prodromal events or predisposing factors ultimately led to EPL rupture or tenosynovitis of the third compartment. Methods of reconstruction described included primary repair, tendon grafting, and tendon transfer techniques; all with generally good outcomes. These results highlight the inherent fragility of this tendon and support the historical recommendation for early release of the EPL tendon in the setting of tenosynovitis of the third dorsal compartment.

Introduction

The extensor pollicis longus (EPL) muscle is responsible for extension of the thumb at the interphalangeal joint and assists in thumb adduction at the carpometacarpal joint. It is the sole tendon of the third dorsal extensor compartment, passing over the distal radius at Lister tubercle before turning toward the thumb and inserting on the distal phalanx. ¹ This region is recognized as a poorly vascularized region.^2-5 It is well documented in the literature that rupture of the EPL is associated with distal radius fractures or a history of rheumatoid arthritis.^3,4,6-8 In addition, there is a number of seemingly spontaneous EPL ruptures reported in the literature, suggesting it is frequently at risk of rupture. Traditionally, authors have recommended compartment release in cases of tenosynovitis of the third extensor compartment to prevent EPL rupture.^9,10 Currently, little evidence has been published to support this recommendation. We have conducted a systematic review of the literature to better understand and characterize the nature of spontaneous EPL rupture, the factors which contribute to its occurrence, and the manners in which it is treated.

Methods

We performed a systematic review following guidelines set forth by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement. A comprehensive literature search was conducted by a medical librarian on December 7, 2021, using the following bibliographic databases from inception: PubMed/MEDLINE and Scopus/Elsevier. The search consisted of headings and keywords related to tendon injuries, tendinopathy, hand surgery, tendon transfer, and injections, as published in reports and studies. The search strategy is provided in Supplemental Content (Supplemental Material 1).

Results produced from the database searches were imported into Covidence, a systematic review screening tool, and deduplicated. Remaining citations were screened by title and abstract against predetermined inclusion and exclusion criteria by 2 independent reviewers, with a third reviewer resolving discrepancies. To be eligible, articles had to meet the following inclusion criterion: describe cases of spontaneous EPL rupture or tenosynovitis of the third dorsal compartment. The exclusion criterion was any history of distal radius fracture or rheumatoid arthritis. To ensure consistent data quality, an initial extraction from 3 sample articles was performed by 4 team members. The results of the extractions were compared, conflicts were clarified, and the extraction template was adjusted accordingly. Data were then extracted from the resulting articles by 2 reviewers working independently. A single reviewer compared the 2 extractions and resolved any remaining conflicts.

The following information was extracted and documented: patient demographics; rupture characteristics; workup; interventions; and outcomes. Patient demographics included the sex, age, dominant hand, hobby or occupation, and comorbidities of every patient reported in each article. Rupture characteristics included which hand was affected and the location of the rupture based on surgical or radiologic findings, inciting factors, and clinical presentation. The data on workup included any labs or imaging the patient may have received. Patient intervention data included the presence of conservative therapy and the method of surgical treatment (primary, grafting, or tendon transfer). Outcomes were categorized into one of three categories: good; fair; and poor. “Good” was defined as a return of full or near-full range of movement (ROM) and no residual pain. “Fair” outcomes had restricted ROM and/or some residual pain. “Poor” had no recovery of ROM and/or high levels of residual pain. The time to reach the outcome or follow-up time was also recorded.

Results

A total of 1987 records were identified in our database search. Of the total, 395 duplicates were identified and removed. In total, 1592 title and abstracts were screened for spontaneous EPL rupture, and 44 full-texts were reviewed, with 29 studies included for data extraction. The 29 studies accepted for data extraction were case reports and case series, within which 47 patients and 51 hands that met the inclusion and exclusion criteria were analyzed (Figure 1).

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram demonstrating article selection.

Demographics

There were 18 female and 29 male patients. Age of the patients ranged from 17 to 82 years (mean 45 years). Dominant hand was identified in 25 cases: right-handed (n = 23) and left-handed (n = 2) (Table 1). Occupations and hobbies were identified in most cases but absent in 11 cases (Table 2). Comorbidities were only identified in 9 cases, and these include hypertension, valvular heart disease, substance use disorder, Crohn’s disease, diabetes mellitus, chronic renal failure, testicular germ cell tumor, and gout.

Table 1.

Demographics With Number and Percentage of Patients Reported.

Female	Male	Ages	Right-hand dominant	Left-hand dominant
18 (38%)	29 (62%)	17-82 (µ = 45)	23	2

Table 2.

Occupations and Hobbies With Number of Patients Reported.

Occupation or hobby	Number
Home maker	8
Police officer	3
Farmer	2
Student	2
Businessman	2
Tennis player	2
Sales person	1
Fisherman	1
Office worker	1
Information technology	1
Goalkeeper	1
Lab technician	1
Professional skier	1
Manual labor	1
Physician	1
Interscholastic diver	1
Nursing supervisor	1
Carpenter	1
Ice skater	1
Kick-boxer	1
Retail worker	1
Machine operator	1
Video gamer	1
Body building	1

Note. Some patients are represented in more than 1 category.

Rupture Characteristics

Of the 47 patients, symptoms presented: bilaterally n = 4; right hand only n = 18; left hand only n = 16; side unidentified n = 9.

Thriteen ruptures occurred in the dominant hand, 27 occurred in the nondominant hand, and in 2 of the cases that presented with bilateral symptoms, no rupture had occurred in the nondominant hand (Table 3).

Table 3.

Rupture Occurrence With Number and Percentage of Patients Reported.

Right hand	Left hand	Bilateral	Dominant	Nondominant
18 (50%)	16 (44%)	2 (6%)	13 (33%)	27 (67%)

The most prominent presenting symptoms were loss of extension (n = 49) and pain (n = 33). Swelling (n = 6), popping/cracking (n = 2), and erythema (n = 1) were also reported (Figure 2).

Figure 2.

Presenting symptoms in general groups with number reported.

Intraoperative identification found that in 21 cases, the EPL tendon had ruptured at the site of Lister tubercle, and in 6 cases, it was described rupturing at a different location: at carpal level; at the base of the metacarpophalangeal joint; at the base of the first metacarpal; at the musculotendinous junction; beneath the retinaculum in the third extensor; and 3 cm proximal to the wrist joint.

Inciting factors were identified in all but 3 of the cases, and these were separated into 6 broad categories: “drugs,” referring to steroid injections, fluoroquinolone treatment, and intravenous substance use (n = 8); “trauma,” with distal radius facture excluded by x-ray (n = 10); “mechanical,” such as repetitive motions (n = 8); “inflammatory,” with rheumatoid arthritis excluded with appropriate labs (n = 3); “primary” (n = 9); and “other,” such as anatomical defects or recent surgery in the region (n = 6) (Figure 3).

Figure 3.

Inciting factors in general groups with number reported.

Workup

In 23 of the cases, labs were completed, and these labs generally looked for unidentified rheumatoid arthritis or other underlying comorbidities such as infectious diseases, diabetes mellitus, and gout. They were consistently within normal limits for all cases with two exceptions: One patient exhibited increased beta-2 microglobulin and parathyroid hormone in the setting of chronic renal failure, and another had increased uric acid in the setting of known gout.

Thirty-six of the cases had radius fractures ruled out with x-ray; 12 cases had magnetic resonance imaging performed; and 3 cases performed ultrasounds. These studies were largely insignificant or confirmed EPL rupture or tenosynovitis.

Interventions

In only 3 cases was conservative management attempted first or reported. Of the 47 patients, 44 received surgical treatment (46/51 hands). These interventions revealed three mainstays of EPL rupture treatment—primary repair (n = 7); tendon transfer (n = 29), generally extensor indicis propius (EIP) to EPL but with one exception, extensor digitorum communis (EDC) to EPL; and graft (n = 8), generally palmaris longus but with one exception, semitendinosus graft. Two prophylactic tendon decompressions were performed.

Outcomes of these procedures were generally good; of the 46 procedures performed, 40 had “good” outcomes. 4 were classified as “fair.” One of these patients did not adhere to recovery guidelines, and this patient had undergone a palmaris longus graft repair. The other three had slightly decreased ROM. Two did not have a reported outcome, of these, one was a graft patient and the other a transfer patient. There were no reported “poor” outcomes (Figure 4).

Figure 4.

Outcomes of each intervention with number of patients reported.

Recovery time was reported varyingly, and many cases reported only full follow-up time, which was generally much longer than the time expected for full recovery. Follow-up period was reported as far out as 140 months. Reported time to full recovery ranged from 6 weeks to 12 months. The mode time to full recovery, with follow-up time incorporated, was 2 months.

Discussion

Our systematic review suggests that multiple etiologies for spontaneous EPL rupture exist. Although there may not be sufficient numbers from these reports to draw reliable conclusions, they suggest that men may be affected more often than women and that the patients appear to cover a broad range of ages, occupations, and hobbies. Although it may be difficult or impossible to prevent spontaneous EPL rupture, we may be able to draw some general conclusions regarding management from the experience presented in the literature.

Lister’s tubercle is recognized as a poorly vascularized area of the EPL.^2-5 Tendon trauma, compression, and tenosynovial invasion in this vulnerable area have been suggested as causes of spontaneous rupture.^6,11 Without a secondary cause, EPL tendonitis and rupture has been associated with mechanical wear on the tendon due to osteophytes, minor trauma, anatomic variants, and specific activities.^11,12 There is no consensus on how often EPL tendonitis will lead to EPL rupture, and it appears that some ruptures occur without any heralding signs.

Most recorded patients reported wrist pain accompanied by a loss of thumb extension. In all these cases, a ruptured EPL was found on exploration. When reported, most cases discovered rupture at the site of Lister’s tubercle. Although not commented on frequently, pathological findings were often present at the tubercle. Treating surgeons should be aware of possible bone spur or protrusion around the Lister’s tubercle that could result in a point of friction with tendon gliding leading to the rupture.

Several authors suggest urgent third-compartment release for symptomatic tendonitis while others recommend a steroid injection but acknowledge concern over increased risk of tendon rupture.^9,10 These data do not support the use of steroid injection for treatment of EPL tenosynovitis. It is possible that injecting an inflamed and damaged EPL tendon to reduce symptoms makes the tendon more vulnerable to rupture.

In patients with minor damage and fraying of the EPL tendon, a direct repair is possible; however, those presenting extensive damage to the EPL tendon may fare better with an EIP to EPL tendon transfer. Although transfer was the predominant technique used in EPL repair, it does not appear superior to other techniques in terms of quality or time to result; thus, we recommend that surgeon experience or judgment guide selection of the repair technique. ¹³

Regarding immobilization and physical therapy following a surgery for complete ruptures, a short-arm thumb spica cast or splint appears to be effective, followed by initiation of passive and then active ROM protocols starting at 4 to 6 weeks postoperatively. Recovery from surgery can be expected within 2 months’ time although some cases have required up to a year.

Due to the infrequency of these cases, it is unlikely that a prospective study will be available in the near future to track outcomes of these patients in a more rigorous manner. It seems, however, that the literature has provided several examples from which some general considerations regarding spontaneous EPL rupture diagnosis and management can be derived.

Conclusions

1. Look for loss of thumb extension, pain, swelling, and potential associated popping noise when evaluating for spontaneous rupture of the EPL tendon.

2. Steroids, antibiotic use, repetitive exercises and motions, trauma, and inflammation are common inciting factors.

3. Surgical decompression for tenosynovitis of the third compartment may be superior to conservative management.

4. Primary repair, transfer, and graft all have good outcomes for surgical repair; thus, intraoperative findings or surgeon preference can guide the technique used.