Flexor Pollicis Longus Palsy Following Both-Bone Forearm Fracture Fixation: A Case-Based and Cadaveric Evaluation

Abstract

Background:

Flexor pollicis longus (FPL) palsy following both-bone forearm fracture (BBFF) is a rare complication.

Methods:

A retrospective review of acute BBFF treated with open reduction internal fixation by a single surgeon from 2005 to 2023 was performed. Injury and surgical characteristics of patients with documented FPL palsy were reviewed. In addition, 10 cadaveric dissections were performed to evaluate the anatomy of the anterior interosseous nerve (AIN) and its branches. The distance of these branches from palpable elbow landmarks and variability in branching pattern were evaluated.

Results:

Twenty-nine patients underwent surgery for acute BBFF. Of these, 5 (17%) had evidence of an FPL palsy either at the time of injury presentation (n = 2) or immediately following surgery (n = 3). All patients with FPL palsy sustained fractures in the middle one-third of the radius. All palsies resolved after an average of 33 days of observation. In cadaveric dissections, the average distance from the lateral epicondyle to the AIN takeoff and branch to the FPL was 5.5 and 7.6 cm, respectively. The AIN takeoff and branch to the FPL were never less than 4 and 7 cm from the lateral epicondyle, respectively.

Conclusion:

Flexor pollicis longus palsy following BBFF can occur at the time of injury or following surgery. All FPL palsies involved midshaft radial fractures and were likely neurapraxia. The etiology of FPL palsy remains unclear, but cadaveric dissection suggests the FPL motor branch may be at risk from mid-to-proximal radius fracture fragments or excessive traction during surgery.

Introduction

Both-bone forearm fractures (BBFF) are common injuries, typically occurring from high-energy trauma across the forearm in both the pediatric and adult population. Management of BBFF in skeletally mature patients has long been open reduction internal fixation (ORIF) with plate fixation, with a goal of achieving anatomic reduction and primary bone healing. ¹ Common complications following BBFF include hardware removal given the subcutaneous location of ulnar hardware, nonunion, malunion, refracture after hardware removal, heterotopic ossification, and synostosis.^1,2 Little has been reported on nerve injuries in the adult population following BBFF beyond case reports.^3-11 A study of complications following surgical and nonsurgical management of pediatric BBFF found a 5.4% incidence of nerve injuries, 4.1% in nonoperatively treated fractures, and 7% in operatively treated fractures. ¹² The most commonly affected nerve was the ulnar nerve, and all nerve complications except one patient had complete recovery without surgical intervention. ¹²

Palsy of the anterior interosseous nerve (AIN) results in weakness of the flexor pollicis longus (FPL), flexor digitorum profundus (FDP) to the index, pronator quadratus, and varying contributions of strength to the long-finger FDP without sensory changes in the hand. Despite the rare reports of nerve injury complications following BBFF in the adult population, recent observations of patients reporting absent FPL function following BBFF led us to question the incidence of AIN palsy, particularly affecting the FPL, following BBFF and the clinical outcomes of these palsies. Therefore, the purpose of this study was to evaluate surgically managed BBFF for evidence of FPL palsy and the clinical outcome of this palsy. In addition, a cadaveric study evaluating the position of the AIN and its motor branches, particularly to the FPL, in relation to the forearm was performed to better understand anatomic risk factors for FPL palsy following BBFF.

Methods

Patient Data

Following institutional review board approval, a retrospective review of patients surgically treated for a BBFF by a single surgeon from 2005 through 2023 was performed. Patient demographics, injury mechanism, operative reports, radiographs, and clinical notes were reviewed. For the purpose of this study, FPL palsy was characterized as complete inability to actively flex the thumb interphalangeal joint without sensory changes in the median nerve distribution. Clinical and operative notes were reviewed to conform there was no structural cause of the weakness (ie, no flexor tendon or muscle injury). Timing of the first documented exam of a deficit was recorded, and any additional studies such as advance imaging or electrodiagnostic studies were reviewed. Additional surgeries, concomitant injuries, and any complications were documented. Injury and surgical characteristics were compared between patients with documented FPL palsy and those without.

Cadaveric Evaluation

Cadaveric data collection was approved by the Institutional Biospecimens Committee. Cadaveric dissections of 10 fresh-frozen forearms with no documented history of forearm injury or pathology were performed to evaluate the anatomy of the AIN. The distance from the lateral epicondyle to the AIN origin, branch to the index FDP, and branch to FPL was measured using calipers. The lateral epicondyle was chosen as a palpable bony landmark to measure the AIN branch distances for its reproducably identifiable bony prominence, which would allow for clinical application of estimating the nerve branch locations without making an incision.

Statistical Analysis

Given the relatively small sample sizes of the impacted patients and the cadaveric cohort, we report descriptive statistics only.

Results

Patient Data

Twenty-nine patients at an average age of 42 years (range 16-70 years) were included in the study, of which 20 (69%) were male. Nine (31%) BBFF were open injuries, and 6 (21%) were BBFF in the setting of polytrauma. Two patients (6.9%) underwent forearm fasciotomies for concerns of compartment syndrome. Five patients (17%) had evidence of an FPL palsy either at the time of injury presentation (n = 2) or immediately following surgery (n = 3). Of the 5 patients with FPL palsy, 1 was an open injury, 2 were polytrauma patients, and 1 patient had forearm fasciotomies for concerns of developing compartment syndrome. All palsies resolved after an average of 33 days (range 6-90 days) of observation. Details regarding injury characteristics, surgery, and FPL palsy characteristics are included in Table 1. Most patients (n = 22, 76%) had fractures of the middle one-third of the radius, and all patients with an FPL palsy sustained fractures in the middle one-third of the radius. No patient required advanced imaging or electrodiagnostic studies for their palsy. Similar surgical approaches (volar Henry approach to the radius and subcutaneous split between the extensor and flexor carpi ulnaris to the ulna) and fixation methods (6-9 hole limited dynamic contact compression plates) were used in patients with and without FPL palsy (Figure 1).

Table 1.

Demographic and Clinical Data for Patients With Symptoms of Flexor Pollicis Longus Deficits.

Case	Age at injury (years)	Sex (M/F)	Injury characteristics	Surgical approach (R/U)	Surgical fixation (R/U)	FPL symptoms	Time to resolution (days)
Case 1	25	M	Midshaft radius and ulna	Volar Henry/ECU-FCU split	6 H LC-DCP/8 H LC-DCP	POD1: FPL palsy, large hematoma concerning for compartment syndrome s/p hematoma evacuation	20
Case 2	20	M	Midshaft radius and ulna	Volar Henry/ECU-FCU split	8 H LC-DCP/8 H LC-DCP	Preoperative FPL palsy	6
Case 3	16	M	Midshaft radius and ulna	Volar Henry/ECU-FCU split	7 H LC-DCP/7 H LC-DCP	POD1 FPL palsy	35
Case 4	43	F	Midshaft radius and ulna with concomitant ipsilateral elbow dislocation and open ipsilateral distal radius and distal ulna fractures.	Volar Henry/ECU-FCU split	Metadiaphyseal volar locking plate/Lag screws plus 7 H LC-DCP	Preoperative FPL palsy	90
Case 5	68	F	Midshaft radius. Polytrauma with subdural hematoma, operative pelvic and thoracic spine fracture	Volar Henry	8 H LC-DCP	FPL palsy noted upon extubation POD1	14

Note. M = male, F = female, LC-DCP = limited contact dynamic compression plate, ECU = extensor carpi ulnaris, FCU = flexor carpi ulnaris, FPL = flexor pollicis longus, POD1 = postoperative day 1, R = radius, U = ulna.

Figure 1.

Preoperative and postoperative radiographs of a both-bone forearm fracture treated with open reduction internal fixation in a patient who was diagnosed with a flexor pollicis longus (FPL) palsy on postoperative day 1.

Note. This patient’s FPL palsy resolved at 3 weeks postoperatively without intervention.

Cadaveric Data

Ten fresh-frozen cadavers were carefully dissected to identify the origin and branching patterns of the AIN (Figures 2-4). The lateral epicondyle was used as a reference landmark for all measurements (Table 2). The average length of the forearms, as measured from the lateral epicondyle to the radial styloid, was 26.2 cm (range 24-28.5 cm). The AIN takeoff was on average 5.5 cm from the lateral epicondyle (range 4-7 cm). The branch to the FPL and index FDP distances were, on average, 7.6 cm (range 7-8 cm) and 7.1 cm (range 5-9 cm), respectively. The AIN takeoff, branch to the FPL, and branch to the index FDP distances were never less than 4, 7, or 5 cm from the lateral epicondyle, respectively (Table 2). The motor branch to the index FDP was the first branch in 50% (n = 5) specimens, and 40% (n = 4) specimens had branches to the FPL and index FDP which begin at the same level in the forearm. The motor branch to the FPL originated proximal to the motor branch of the index FDP in one specimen (10%). The motor branch to the FPL takes off laterally to the AIN and enters the muscle belly on the medial or ulnar side in a slightly oblique fashion from proximal ulnar to distal radial (Figures 3 and 4). In contrast, the motor branch to the index FDP takes off medially to the AIN and runs slightly oblique from proximal-radial to distal-ulnar. The AIN continues along the interosseous membrane parallel to the axis of the forearm.

Figure 2.

Cadaveric dissection of a right anterior forearm. The hand is to the left. The pronator teres is retracted ulnarly to expose the takeoff of the anterior interosseous nerve (*).

Note. The motor branch to the flexor pollicis longus is marked with a blue vessel loop.

Figure 3.

Cadaveric image of the motor branch to the flexor pollicis longus (held in closed forceps) and the continuation of the anterior interosseous nerve (held in open forceps). Flexor pollicis longus muscle belly (*).

Figure 4.

Cadaveric dissection of the distal forearm with the flexor pollicis longus muscle (held in forceps) lifted off the radius (star). A motor branch of the flexor pollicis longus is marked with a solid arrow.

Table 2.

Cadaveric Dissections of the AIN and Its Branches to the FPL and Index FDP.

Specimen	Lateral epicondyle to radial styloid (cm)	Lateral epicondyle to AIN takeoff (cm)	Lateral epicondyle to FPL (cm)	Lateral epicondyle to index FDP (cm)
1	27	6.5	9	9
2	26	6	7.5	7.5
3	24.5	4.5	8	7.5
4	26.5	4.5	7.5	5
5	27.5	5	7	7
6	24.5	4	7	6.5
7	24	5	7	5
8	26.5	6	7.5	7.5
9	26.5	7	7.5	8
10	28.5	6.5	8	7.5
Average (cm)	26.2	5.5	7.6	7.1

Note. AIN = anterior interosseous nerve; FPL = flexor pollicis longus; FDP = flexor digitorum profundus; cm = centimeters.

Discussion

Flexor pollicis longus palsy following BBFF is an uncommon complication that frequently resolves with observation. While case reports have described isolated incidents of AIN palsy following BBFF,^3,4,6,9-11 there is limited literature reporting overall prevalence. While nerve injuries following BBFF in adults are not commonly discussed,^1,2 a study of pediatric patients reported a 0.7% (33 of 4868) risk of nerve injury following forearm fractures. ¹³ Open fractures and BBFF were associated with higher risk of nerve injuries, and the ulnar nerve was the most commonly affected nerve (19 of 33 nerve injuries). ¹³ In our series, we found that 17% of patients had evidence of an FPL palsy, suggesting this complication may occur more frequently than previously described. Diligent preoperative and postoperative physical examination of the motor and sensory function in patients is warranted.

Most FPL palsy cases were documented postoperatively, suggesting that injury occurred perioperatively. Previously, proposed etiologies of AIN palsy following BBFF include AIN trauma secondary to a displaced bony fragment,^4,6 compression of the nerve at the fracture site,^4,6 fracture hematoma, constrictive dressings perioperatively, ³ and excessive intraoperative nerve traction during fracture exposure. ⁹ Our case series had similar possible etiologies, including one patient with a large postoperative hematoma requiring emergent return to the operating room, patients with FPL palsy immediately following injury, and patients with FPL palsy postoperatively.

Our cadaveric study found the motor branch to the FPL to be an average of 7.6 cm distal to the lateral epicondyle, similar to prior anatomic studies. ¹⁴ The standard anatomic configuration of the AIN demonstrates that the branch to the FDP is typically proximal to the branch to the FPL. ¹⁵ This pattern was found in 5 (50%) of the cadaveric specimens in our study. However, there were variant cases wherein the branches to the FDP and FPL emerged at the same level (n = 4), or the branch to the FPL was proximal to the branch to the FDP (n = 1), suggesting variability of AIN anatomy. The motor branch to the FPL ran along the ulnar side of the muscle belly, as previously well described. ¹⁶ The likelihood of iatrogenic injury to the FPL motor branch is low if surgeons are in the correct surgical interval during the volar approach to the radius, carefully elevating the FPL off the radius for bony exposure in a subperiosteal fashion. However, excessive retraction of the FPL could cause a neurapraxia, and we believe this is likely a common cause of postoperative FPL palsy.

All cases of FPL palsy following ORIF recovered completely without further intervention within 6 weeks of surgery in this series, consistent with neurapraxia. Similar findings were reported in multiple case reports of AIN palsy following BBFF.^3,6,9-11 Prior authors suggested emergent fixation of BBFF as well as pronator teres release in patients presenting with AIN palsy. ¹⁰ These case reports found hematomas or bony fragments putting tension/pressure on the AIN at the time of surgery and no reports of disruption of the nerve. Two patients in our series presented with preoperative AIN palsy and were treated within 24 hours following their injury, but did not undergo proximal exploration and a release of the pronator teres. The approach to management of these two patients was not changed based on preoperative findings of an FPL palsy. Both patients had complete resolution of their AIN palsy postoperatively. Surgeons should consider the risks and benefits of proximal exploration of the AIN in the setting of BBFF fixation, including possible increased soft-tissue damage, iatrogenic injury to the nerve, and prolonged operative time. We do not have sufficient evidence to suggest the necessity or detriment to AIN exploration, and surgeons should decide based on patient presentation and injury characteristics.

Our study should be viewed in the context of its limitations. The patient data for this study were collected retrospectively and represent a single surgeon’s experience with BBFF-associated FPL palsy in a relatively small patient cohort. A single surgeon’s clinical population may not be representative of the overall patient population, potentially inflating the prevalence of palsies identified. Furthermore, many preoperative evaluations were performed by resident physicians and may have missed subtle palsies due to incomplete examination or documentation. Accurate strength testing of the finger flexors at the onset of unstable BBFF may be limited due to patient’s discomfort or bulky splints as well. Due to the retrospective nature of the study and the variability in treatment and recovery, the precise etiologies underlying FPL palsies remains nebulous in certain cases, particularly when deficits manifested postoperatively. Furthermore, as the study included only patients treated with ORIF to limit factors related to surgical strategies, we were unable to describe the potential for FPL palsy and its resolution following other management strategies. Lastly, the cadaveric study includes a relatively small number of specimens, limiting the generalizability of the results.

Flexor pollicis longus palsy following BBFF is an uncommon complication, and it is associated with reliable recovery without subsequent intervention. Flexor pollicis longus palsy may occur at the time of BBFF or following ORIF, with a variety of etiologies implicated. There is a variation in the anatomy of the AIN, with variant branching patterns and proximal-most origin of the AIN potentially acting as risk factors for AIN palsy, particularly the motor branch to the FPL near the radial shaft. Fortunately, all FPL palsies in our series resolved without intervention within 6 weeks of injury, consistent with neurapraxia. Surgeons can appropriately council patients on expected symptom resolution.