The Utility of Plain Radiography in Flexor Tendon and Digital Nerve Lacerations of the Hand

Abstract

Purpose

Physical examination during hand lacerations may sufficiently identify flexor tendon and digital nerve injuries, but radiographs are often acquired to rule out fracture, dislocation, or foreign body. This study examines the impact of plain radiography in the management of patients with isolated sharp injuries of the hand.

Methods

We retrospectively reviewed consecutive adult patients with flexor tendon and/or digital nerve lacerations who presented during 2019-2021 to a tertiary referral urgent care hand clinic. Charts were reviewed to determine the presence of bony injury or foreign body and whether radiographic evaluation affected clinical management. Comparisons were made among patients with varying mechanisms.

Results

One hundred eighty-five patients met inclusion criteria, with the mechanisms of sharp laceration (83.2%), saw injury (11.9%), and crush injury (4.9%). All patients with saw and crush injuries received radiographs compared with 45.5% in the sharp laceration group. Bony injuries and/or foreign bodies were found in 63.6% and 66.7% of cases in the saw and crush groups, respectively. In comparison, no patients with sharp lacerations had a bony injury, and one patient had a foreign body from glass shattering that was removed in the emergency department.

Conclusions

Our results suggest that the utility of plain radiography in sharp lacerations may be limited to clinical contexts in which a retained foreign body is suspected. We recommend only selective use of radiography in isolated sharp lacerations of the hand; however, radiographs regularly altered management in injuries because of saw and crush mechanisms and would be warranted in these settings.

Clinical relevance

This study supports the judicious use of radiography in simple sharp lacerations of the volar hand.

Hand injuries are common, accounting for approximately 20% of the workload seen by physicians in emergency departments.1, 2, 3, 4 Costs associated with hand injuries are considerable, affecting patients, their work productivity, and health care systems.^1,5 Lacerations are the most common presentation of hand injury,¹ with these injuries most frequently occurring by knife or glass lacerations.^1,6,7 Volar lacerations frequently involve the flexor tendons and/or digital nerves, with even small lacerations showing concomitant tendon injury in 54.8% of patients.⁶

A flexor tendon injury exhibits distinct physical findings, such loss of active flexion or extended resting posture of the digit, and its diagnosis can be made by clinical examination alone.^8,9 Prompt surgical repair of tendon lacerations is advised, as delayed treatment results in proximal tendon retraction, which may prevent primary repair and necessitate tendon grafting, tendon transfer, or multistage reconstruction.^10,11 Digital nerve injuries can occur in the presence or absence of flexor tendon injuries and require surgical repair, and these injuries can similarly be accurately identified through clinical examination without the need for additional investigations.^12,13 However, plain radiographs are commonly obtained in flexor tendon and digital nerve injuries to rule out the presence of foreign bodies, dislocations, or fractures.^14,15 Concomitant injuries need to be identified to optimize management plans and rehabilitation strategies.¹⁶

As part of the clinical work-up, the mechanism of injury should guide further investigations. We hypothesized that skeletal involvement is rare in the context of simple sharp lacerations without associated high energy mechanisms. Evidence supporting the liberal use of radiographs in these injuries largely originates from early studies, but there is no current literature available to guide the omission of imaging in patients with low-risk hand injuries.17, 18, 19 As clinical practices move toward more evidence-based, sustainable, and cost-efficient health care, evidence to limit extraneous testing is crucial.²⁰ This study aimed to examine the use of plain radiography in injuries to flexor tendons and digital nerves, as well as the frequency in which it altered the management of the injury following a laceration to the volar aspect of the hand.

Materials and Methods

A retrospective review was conducted of consecutive patients who visited the acute hand injury clinic at our academic level 1 trauma center between 2019 and 2021. Institutional ethics approval was obtained prior to study commencement. This clinic serves as a tertiary referral urgent care clinic for hand injuries suspected to require a minor or operative procedure. Adult patients (age ≥ 18) with acute (< 3 weeks old) flexor tendon lacerations, with or without an associated digital nerve injury, or isolated digital nerve injuries sustained from lacerations on the volar surface of the hand were included. Patients were excluded if they were < 18 years of age, had an injury proximal to flexor zone IV, or had a concomitant extensor tendon injury. Demographic data collected included age, sex (assigned at birth), handedness, occupation, smoking status, comorbidities, and Workers’ Compensation injury status. Clinical characteristics included physical examination findings, radiographic findings, operative findings, and whether there were alterations to the operative plan from the initial clinical assessment due to the radiographic findings. Etiologies were classified into one of the following mechanisms: (1) sharp laceration; (2) laceration by saw; or (3) crush injury.

The primary outcome was a change in management from the initial clinical assessment because of radiographic evaluation. This was defined as the need for additional management (eg, fracture fixation, joint reduction, or alteration of the rehabilitation program) as diagnosed radiographically and confirmed intraoperatively.

Descriptive statistics were reported as mean ± SD or frequency (%). Fisher exact tests were used to determine any significant differences in obtaining a radiograph, the presence of bony injury and/or foreign body, and change in management, by injury mechanism and by coverage under Workers’ Compensation. These outcomes of interest were also assessed through pairwise comparison between each mechanism with a Bonferroni correction to accommodate for type I error. All other tests were considered significant where P < .05.

Results

Of the 4,412 patients who visited the acute hand injury clinic between 2019 and 2021, 185 had a volar laceration and met the inclusion criteria (Fig. 1). One hundred twenty-four patients had flexor tendon injuries, 75 of whom also had a concomitant digital nerve injury, and the remaining 61 patients had isolated digital nerve injuries. Plain radiographs were ordered for 101 (54.6%) patients, demonstrating bony injuries and/or foreign bodies in 21 (11.4%) patients and resulting in a change in management in 12 (6.5%) patients (Table 1). Mechanisms of injury were categorized into sharp laceration (eg, glass, knife; n = 154; 83.2%), laceration by saw (n = 22; 11.9%), or crush injuries (n = 9; 4.9%). The study population had a mean age of 39.5 ± 15.6 years, were more frequently male (63.2%) and right-handed (86.5%), and were most commonly employed in labor (43.2%) or administrative occupations (30.3%). Injuries were often sustained on the index (25.9%) and small (25.9%) fingers within flexor zone II (63.2%). When comparing demographic characteristics between patients with flexor tendon and digital nerve injuries, those with isolated digital nerves were more likely to be employed in labor or office positions, more likely have injuries to the index finger and more proximally, and less likely to have received a radiograph (42.6% vs 60.5%).

Figure 1.

Flowchart of patient selection.

Table 1.

Characteristics of Study Cohort

Characteristics	Total (n = 185)	Flexor Tendon Laceration (n = 124)	Digital Nerve Injury (n = 61)	P Value
Age (y, mean ± SD)	39.5 ± 15.6	39.6 ± 16.6	39.3 ± 13.5	.915
Sex				.872
Female	68 (36.8%)	45 (36.3%)	9 (45.0%)
Male	117 (63.2%)	79 (63.7%)	11 (55.0%)
Handedness				.414
Left	23 (12.4%)	13 (10.5%)	10 (16.4%)
Right	160 (86.5%)	109 (87.9%)	51 (83.6%)
Unknown	2 (1.1%)	2 (1.6%)	0
WCB status	26 (14.1%)	15 (12.1%)	6 (30.0%)	.272
Occupation				.002
Unemployed	23 (12.4%)	22 (17.7%)	1 (1.6%)
Laborer	80 (43.2%)	49 (39.5%)	31 (50.8%)
Office Worker	56 (30.3%)	34 (27.4%)	22 (36.1%)
Student	9 (4.9%)	5 (4.0%)	4 (6.6%)
Retired	9 (4.9%)	9 (7.3%)	0
Unknown	8 (4.3%)	5 (4.0%)	3 (4.9%)
Comorbidities	50 (27.0%)	34 (27.4%)	16 (26.2%)	>.999
Smoking	46 (24.9%)	34 (27.4%)	12 (19.7%)	.282
Mechanism of injury				.237
Sharp laceration	154 (83.2%)	101 (81.5%)	53 (86.9%)
Laceration by saw	22 (11.9%)	18 (14.5%)	4 (6.6%)
Crush	9 (4.9%)	5 (4.0%)	4 (6.6%)
Side of injury				.634
Left	109 (58.9%)	75 (60.5%)	34 (55.7%)
Right	76 (41.1%)	49 (39.5%)	27 (44.3%)
Digits affected				.002
1st	34 (18.4%)	24 (19.4%)	10 (16.4%)
2nd	48 (25.9%)	25 (20.2%)	23 (37.7%)
3rd	17 (9.2%)	10 (8.1%)	7 (11.5%)
4th	16 (8.6%)	7 (5.6%)	9 (14.8%)
5th	48 (25.9%)	41 (33.1%)	7 (11.5%)
Multiple	22 (11.9%)	17 (13.7%)	5 (8.2%)
Zones				.002
Zone I	44 (23.8%)	33 (26.6%)	11 (18%)
Zone II	117 (63.2%)	83 (66.9%)	34 (55.7%)
Zone III	20 (10.8%)	7 (5.6%)	13 (21.3%)
Zone IV	4 (2.2%)	1 (0.8%)	3 (4.9%)
X-ray	101 (54.6%)	75 (60.5%)	26 (42.6%)	.028
Bony injury	21 (11.4%)	17 (13.7%)	4 (6.6%)	.217
Change in management	12 (6.5%)	10 (8.1%)	2 (3.3%)	.342

Values are shown as n (%) unless specified otherwise.

WCB, Workers' Compensation Board.

Rates of obtaining a radiograph (P < .001), the presence of a concomitant bony injury and/or foreign body (P < .001), and change in management (P < .001) varied significantly among patients with different mechanisms of injuries (Table 2). During the pairwise comparisons of outcomes by mechanism of injury (Fig. 2), patients with sharp lacerations were less likely to have had a plain radiograph (n = 70; 45.5%) when compared with patients with laceration by saw (n = 22; 100%; P < .001) or with crush injury (n = 9; 100%; P < .001) (Fig. 2A).

Table 2.

Comparison of Plain Radiography, Bony Injury/Foreign Body, and Change in Management by Mechanism of Injury

Factor	Sharp Laceration (n = 154)	Laceration by Saw (n = 22)	Crush Injury (n = 9)	P Value
Plain radiograph performed	70 (45.5%)	22 (100%)	9 (100%)	<.001
Bony injury present	1 (0.6%)	14 (63.6%)	6 (66.7%)	<.001
Change in management	1 (0.6%)	8 (36.4%)	3 (33.3%)	<.001

Figure 2.

Rates of A obtaining a plain radiograph, B presence of bony injury and/or foreign body, and C change in management following plain radiographic evaluation by mechanism of injury. Pairwise comparisons with Fisher exact test were performed with Bonferroni correction, and analyses were considered significant where P < .017.

Patients with sharp lacerations also experienced less frequent bony injuries/foreign bodies (0.6%) and fewer changes in management (0.6%) compared to saw and crush injuries. A single patient with an isolated sharp laceration had broken a glass bottle in his hand and was unsure of any retained fragments. A radiograph was obtained that identified a 3 mm glass shard that was subsequently removed in the emergency department. In comparison, most patients who had a laceration by saw or a crush injury had a bony injury and/or foreign body (n = 14; 63.6%; P < .001 and n = 6; 66.7%; P < .001, respectively) (Fig. 2B), which resulted in a change in management (n = 8; 36.4%; P < .001 and n = 3; 33.3%; P < .001, respectively) (Fig. 2C).

Of the patients with a laceration by saw, six patients with radiographic findings had unicortical fractures that were managed non-operatively, and no deviations were required from the standard management of flexor tendon and/or digital nerve injury. Of the three crush injury patients who had radiographic findings but no change in management, one patient had a nonsurgical 1 mm extra-articular fracture fragment, a second patient had a tuft fracture that did not impact treatment plan, and the final patient had small submillimeter metallic foreign bodies which were irrigated per usual fashion during a digital nerve repair. No notable differences in the rates of radiographs, bony injuries and/or foreign body, or change in management were noted between the saw laceration and crush injury groups.

The mechanisms of injury varied significantly by Workers’ Compensation status (P = .012; Fig. 3). Patients who had Workers’ Compensation-associated injuries were less likely to have a sharp laceration (65.4% vs 86.2%) but more likely to have a laceration by saw (19.2% vs 10.7%) or a crush injury (15.4% vs 3.1%). When comparing outcomes by Workers’ Compensation status (Table 3), patients with Workers’ Compensation claims were more likely to have a bony injury/foreign body (26.9%; P = .014) than those without this coverage (8.8%); however, rates of obtaining a radiograph and change in management were similar between the two groups.

Figure 3.

Comparison of mechanism of injury by Workers’ Compensation insurance status. Fisher exact test demonstrated a significant difference (P = .012) in the distribution of mechanism of injury between patients with and without Workers’ Compensation coverage.

Table 3.

Comparison of Plain Radiography, Bony Injury/Foreign Body, and Change in Management by Workers’ Compensation Insurance Status

Factor	Workers’ Compensation Injury (n = 26)	No Workers’ Compensation Injury (n = 159)	P Value
Plain radiograph performed	17 (65.4%)	84 (52.8%)	.290
Bony injury/foreign body present	7 (26.9%)	14 (8.8%)	.014
Change in management	2 (7.7%)	10 (6.3%)	.678

Discussion

This study identified variability in the acquisition of plain radiographs in patients with volar lacerations of the digits and hand. Only 1 patient of 154 with simple sharp lacerations had a foreign body found on a plain radiograph, and no patients sustained a concurrent bony injury on radiography. Nevertheless, nearly half of the patients with sharp laceration had plain radiographs performed. In comparison, all patients with laceration by saw or crush injuries received plain radiographs and were more likely to have a bony injury and/or foreign body and consequential change in management. Aside from a single patient where a glass foreign body was identified, plain radiography did not impact the management of patients with sharp lacerations; thus, these radiographs were unnecessary.

The clinical diagnosis of flexor tendon and digital nerve injuries through physical examination is generally sufficient to indicate injury to these structures.^8,12 Volar lacerations can occur in many different types of injuries, but the most common mechanism in our study was an accidental laceration to the digit with a sharp object (83.2%). The mechanisms and demographics are comparable to the descriptions in other studies.^1,6,7,12 Cases of laceration either from sharp glass or knife are intuitively at a lower risk of fracture compared with lacerations from higher energy mechanisms such as a laceration by saw blade. Clinical judgment should be used in assessing the need for imaging, which was evident in the variable rates of radiographs ordered within our study cohort. Patients with sharp lacerations were much less likely to receive a plain radiograph compared with those with crush injuries or saw injuries (45.5% vs 100%). In keeping with our hypothesis, patients with lacerations from saws or crush injuries benefited from the radiographs, demonstrating higher rates of bony injury/foreign body and ultimately change in management. This supports the continued use of plain radiography in these patients.

Findings from this study support the selective use of radiography in patients with hand lacerations by mechanism. To the best of our knowledge, there is little research in current literature to guide the use of radiography in hand injuries, and evidence in the management of flexor tendon injuries remains poor.²¹ Physicians may have concerns that their clinical examinations have not captured the full extent of the injury, and the risk of misdiagnosis from a plain radiograph is low.22, 23, 24 Given that a more severe hand injury may confer greater impact on functional recovery and quality of life, a physician may understandably desire further assurance of their diagnosis.²⁵ Concerns of litigation from a missed diagnosis may also be a consideration in driving the excess use of plain radiography.²⁶

There are substantial costs associated with flexor tendon and digital nerve injuries. Direct costs of a single case of flexor tendon laceration were estimated at $13,725 USD, but increased to upwards of $112,888 when including indirect costs such as missed days of work.²⁷ Although a plain radiograph does not individually bear a substantial expense, considerations of costs on a system level are more relevant when factoring in the large volume of hand injuries, which accounted for $48.6 billion in health care spending in the United States alone in 2012.⁵ Within our local publicly funded health care system, a single hand radiograph series confers costs of $107.48, inclusive of facility fees and physician reimbursements.²⁸ Of the 155 patients with sharp lacerations who were seen at our center, 70 received radiographs, which cumulated to a total of $7523.60 with no notable change in outcome. This number is likely higher in jurisdictions with multiple private payors (eg, United States); thus, the potential for cost savings may be even greater in those practice settings. Although this study examined hand lacerations associated with tendon and nerve injuries, it is likely that our findings also apply to more benign presentations without clinical findings of deeper injuries, which would capture a much broader patient population and resultant costs of radiography than appreciated in our study. The benefits of any suggested treatment must outweigh the risks, and any increased burden on the health care system must be justified. Routine use of plain radiography without clear benefit accumulates to a costly drain of health care dollars.²⁹ As we found minimal impact of plain radiography in the treatment of patients with sharp lacerations, the use of imaging in this cohort of patients largely provided no additional value.

There is conflicting literature describing Workers’ Compensation as a predictor for poorer health outcomes.30, 31, 32, 33 Within this study, injuries relating to a Workers’ Compensation claim seemed to be more severe and involving machinery, and these injuries were more likely to have associated bony injuries and/or foreign bodies. An association between Workers’ Compensation status and change to management was also expected. However, the relationship failed to reach statistical significance, possibly owing to small numbers within this subgroup. Patients with Workers’ Compensation-associated injuries are more likely to have high-risk features that contribute to their extent of injury, and we found that patients covered under Workers’ Compensation more frequently sustained lacerations by saw or crush mechanisms.³⁴ This is suspected as a mediator of the effect of Workers’ Compensation status on bony injuries and/or foreign bodies, although further analyses were not available.

This study has some limitations. Hand radiographs were deferred for half of the patients with sharp lacerations, presumably because of low suspicion of positive findings by the clinician, which may contribute toward confirmation bias within this study. However, the less than 1% frequency of radiographic findings within this patient group supports more judicious acquisition of hand radiography. Secondly, patients within this study were selected from our acute hand injury clinic shortly after an initial assessment in the emergency department. Our findings do not encompass patients with multisystem trauma or more extensive hand injuries, as these patients would likely be admitted and instead managed by inpatient services, nor do they capture more benign lacerations which were believed not to require specialist referral. Lastly, in addition to unnecessary costs, it is possible that the use of radiography in patients with sharp lacerations could affect other patient metrics, such as increased length of stay secondary to interruptions of physician workflow.³⁵ However, additional data on the increased duration of hospital stay and resultant costs were not available for this study. Economic analyses aimed at determining the number needed to screen to detect an abnormality on plain radiography in volar hand injuries, as well as the costs of missing a substantial injury, would be useful future additions to the literature.

In conclusion, the use of plain radiography did not impact the management of flexor tendon and digital nerve injuries secondary to a low energy sharp laceration of the hand. Where sharp lacerations occur from glass or knife, clinical judgment should be exercised to identify contexts in which plain radiographs are likely to confirm the presence of a bony injury and/or foreign body and to result in subsequent changes to management. Our findings support the more judicious acquisition of plain radiographs for patients with low energy sharp lacerations to the volar hand, such as for those suspected to have retained glass fragments. However, we recommend routine use of radiography in lacerations by saw or crush injuries, as these patients have high rates of concomitant bony injury and/or foreign body.

Conflicts of Interest

No benefits in any form have been received or will be received related directly to this article.