Solitary extensor tendon in the third compartment has a companion: variant course of extensor pollicis brevis in third extensor compartment

Anomalous variants of the extensor tendons in the first and third compartments of the wrist have been described following cadaveric study. Being aware of such variants during either an elective or emergency hand surgery will prove beneficial in the execution of the extensor tendon repairs. A very few clinical reports are available in the literature. We report a rare finding of extensor pollicis brevis tendon accompanying the solitary extensor pollicis longus tendon in the third compartment observed during an emergency trauma surgery. Level of evidence V.


Introduction
Nine extensor tendons originate from the dorsal aspect of the forearm and course across the dorsal wrist to insert either to the metacarpals or the phalanges of the digits and thumb. These tendons are covered by the synovial sheaths which in turn pass through the fibro-osseous tunnels formed by the extensor retinaculum at the distal radio-ulnar joint (DRUJ), organised as the six extensor compartments [5].
Extensor pollicis brevis (EPB) traverses in the first compartment along with abductor pollicis longus (APL) tendon. Extensor pollicis longus (EPL) runs in the third compartment. Many variations of the course of the EPB have been reported in the literature. Suguira et al. [8] reported a variant course of EPB tendon out of the first compartment, in the third compartment along with EPL, as observed in a cadaveric study. A very few reports are found in the literature describing these variations in clinical scenarios. We report the rare finding of EPB tendon accompanying EPL tendon in the third compartment observed during an emergency trauma surgery.

Case report
A gentleman aged 60 years, right-handed presented to the emergency with an injury to the left thumb. He sustained a deep laceration over the dorsum left hand (anatomical snuff box region) while cutting fruit with a long kitchen knife. He was unable to extend the thumb and complained of pain over the injured site. Sensation over the first webspace and thumb was intact. He was able to flex the thumb at the inter-phalangeal (IP) joint as well perform abduction and adduction. There was no bony deformity. Plain radiographs of the left hand showed no fractures. Plans of the surgery, namely, exploration of the laceration, extensor tendon repair, post-operative immobilisation, and physiotherapy, were explained.
Surgery was performed under Regional Anaesthesia (axillary block). The oblique laceration was explored, and the transected distal end of the extensor tendon was identified. A passive pull of the cut end of the tendon demonstrated extension of the IP joint of the thumb, and therefore, the tendon was noted to be EPL. The latter was accompanied by another tendon located on the ulnar side, with its insertion to the base of the proximal phalanx. The passive traction or pull  1). There was no demonstrable abduction of the thumb. On proximal exploration, two tendons were retrieved from the same fascial pulley of the third compartment. The first compartment showed only one tendon with its insertion to the base of the first metacarpal, which is that of APL. EPB was absent in the first compartment. Therefore, tendon traversing ulnar to EPL in the third compartment with its distal insertion to the base of the proximal phalanx is EPB, demonstrating its variant course in the third compartment (Fig. 2). The anomalous tendon was thin and flat. The tendons were repaired using the Modified Kessler's technique using Polypropylene 3-0 for the core sutures and Polypropylene 4-0 for the epitendinous repair. Skin was closed using Polypropylene 4-0.
Post-operatively, the hand was immobilised with thumb in full extension and wrist in 20-30 degree extension using the Plaster of Paris (POP) slab. He was treated with oral antibiotics for a week. Skin sutures were removed on 12 th post-operative day. POP was continued for 4 weeks. Mobilisation of the thumb with active and passive physiotherapy was initiated from the 5 th post-operative week. At the end of 8 weeks, patient had complete recovery of extension of the IP and MCP joints of left thumb.

Discussion
Numerous descriptions regarding the anomalies of the extensor tendons in their course have been reported in the literature. Brunelli et al. [1] carried out cadaveric research to study the anatomy of EPB on 52 hands. They observed anatomical variations of the EPB with regards to its presence in the first extensor compartment, size, insertion, and function. EPB was noted to be single, thin, ran along with APL in the first compartment except in five cases where EPB traversed in a separate compartment divided from APL by a fibrous septum.
Various studies in the literature have reported the presence of accessory tendons associated with EPB [2,4]. They have also documented different insertion sites of the tendon. None of these studies have described the presence of EPB in other compartments other than in the first compartment either with APL or divided from APL by a septum.
Takayu et al. [6] reported a rare case where they found two slips of EPL tendon originating from the same muscle belly and inserting to the base of the distal phalanx. EPB was absent in the first compartment. It was considered that one of the two slips of the EPL on the radial side performed the functions of EPB.
Papaloizos et al. [3] reported a finding of supernumerary EPL during a cadaveric dissection. The additional extensor tendon traversed ulnar to the principal EPL, in the fourth compartment, separated from the latter by a bony prominence like Lister's tubercle. This tendon led to the extension  of the IP joint of the thumb. EPB was present in its usual first compartment.
EPL is an undeviating extensor tendon in the third compartment according to Yoshida et al. [9]. An absent EPB in the first compartment was noted only in two of the 52 cases studied by Brunelli et al. [1]. This suggests that an absent EPB in the first compartment during hand reconstructive surgery warrants its exploration in the other compartments just as seen in our case report.
Suguira et al. [8] described an unusual course of EPB in the third compartment in a cadaver. EPB tendon was identified radial to EPL with insertion to the base of the proximal phalanx. APL was the only tendon in the first compartment. In our patient, an additional tendon was seen traversing ulnar to the EPL in the third compartment, having a separate fascial sheath and inserting to the base of the proximal phalanx. The first extensor compartment had only one tendon with its insertion to the base of the first metacarpal being the APL. EPB was absent in the first compartment. Therefore, the additional tendon found traversing along with EPL must be EPB with anomalous variant course.
Straus [7] explained the evolution of human forearm extensors. He reported that the precursor extensor muscle mass of the forearm divides into 3 layers-(1) radial layer which forms the brachioradialis, extensor carpi radialis longus, and brevis, (2) superficial layer that forms extensor digitorum communis, extensor carpi ulnaris, and extensor digiti minimi, and (3) deep layer that forms the APL, EPB, EPL, and extensor indicis proprius. Studies that have reported variations of EPB and the present study suggest that the deep layer of the primitive extensor muscle mass is not a constant and undergoes changes in its phylogeny.
Knowledge of the anomalous variants of the EPB tendon will assist in the reconstructive hand surgery, namely, repair of the tendon during traumatic injury, spontaneous rupture of the tendon, tendon transfer surgery in spastic hands, and reconstruction of ulnar collateral ligament. This also means that a pre-operative evaluation of the tendon by ultrasound or magnetic resonance imaging becomes imperative to be certain of the anatomy and its possibility as a donor in tendon transfer surgery.
Author contributions AR: data collection, data compilation, and manuscript writing. KGB: manuscript writing.

Funding
The authors did not receive support from any organisation for the submitted work.
Availability of data and materials Available with the corresponding author.

Declarations
Conflict of interest Both authors declare they have no financial or nonfinancial interests. No conflicts of interest.

Ethical approval No ethical approval required.
Informed consent Informed consent was obtained from the patient for the use of data and photographs for publication. The consent has been formally documented in the medical record.