Clinical Outcomes after ORIF vs Arthrodesis for Subtle Lisfranc Injuries: A Minimum 2-Year Comparative Study

Abstract

Background:

Subtle Lisfranc injuries, defined by 2-5 mm of first webspace diastasis, pose unique treatment challenges distinct from more severe injuries. This study aimed to evaluate whether a primary open reduction internal fixation (ORIF) or a primary arthrodesis (PA) optimizes clinical outcomes and minimizes complications in treating subtle Lisfranc injuries.

Methods:

This study included patients who had a nondislocation Lisfranc injury with a proximal first webspace (between the medial cuneiform and second metatarsal base) diastasis of 2-5 mm, and underwent either a primary ORIF or primary arthrodesis. Preoperative weightbearing radiographs were reviewed to confirm subtle Lisfranc injuries. Of the 73 patients who met the inclusion criteria, 41 received a PA and 32 received an ORIF. Treatment selection was based on surgeon preference. Patient-reported outcomes via PROMIS scores were collected preoperatively and at least 2 years postoperatively. Subsequent procedures were also recorded.

Results:

We received PROMIS surveys from 57 patients (78%). The average preoperative diastasis of the ORIF group significantly differed from that of the PA group (P < .05). Both ORIF and PA cohorts demonstrated significant improvement in all physical PROMIS criteria on minimum 2-year follow-up (P < .05). Our results did not demonstrate a significant difference in patient-reported outcomes between the ORIF and PA groups. There was no significant difference in the incidence of complications between groups, but the ORIF group underwent significantly more hardware removal procedures than the PA group (P < .01).

Conclusion:

This study compared outcomes of subtle Lisfranc injuries treated with ORIF and PA. Our results demonstrated no significant differences between ORIF and PA outcomes. This study suggests that both ORIF and PA may be viable options for subtle Lisfranc injuries; however, further research is needed to determine which may be optimal for different patient populations.

Level of Evidence:

Level III, retrospective comparative study.

Introduction

Lisfranc injuries refer to a group of injuries to the tarsometatarsal joints (TMTJs), a broad selection of pathologies to the bones and ligaments of the midfoot. ² These relatively uncommon injuries are often misidentified. Subtle Lisfranc injuries refer to TMTJ injuries resulting in a small amount of displacement, often resulting from low-energy injuries during athletics. ²¹ Lower-energy injuries result in Lisfranc deformities with less significant displacement. These underresearched injuries may benefit from a different surgical approach than the higher-energy Lisfranc injuries that have already been widely investigated, as mismanagement can lead to chronic pain, instability, and osteoarthritis, significantly impacting daily functioning and quality of life.

Originally, as described by Quenu and Kuss in 1909, Lisfranc injuries were classified as either isolated, homolateral, or divergent. ¹⁸ Studies have shown, however, that this injury classification does not affect surgeons’ treatment decisions. ¹³ Only certain characteristics have been shown to significantly affect patient outcomes in Lisfranc injury cases: the amount of time in between the initial injury and surgical intervention, the type of movement that propagated the injury, and ligamentous vs bony involvement in the injury. ³ In stable cases, conservative management is usually successful. ⁸ In unstable cases surgical intervention may be necessary for proper healing.

Open reduction and internal fixation (ORIF) restores anatomical alignment and joint stability whereas primary arthrodesis (PA) sacrifices joint motion, potentially leading to altered biomechanics and increased stress on the neighboring joints. ¹ ORIF preserves flexibility and allows for easier augmentation in future surgeries because of minimal hardware implantation; however, micromotion at the fixation site can lead to painful or broken hardware and complications such as neurovascular injury, especially when soft tissue damage is extensive.^10,16,19 This results in high rates of hardware removal procedures after surgery. PA, despite sacrificing joint mobility, may be favored by some surgeons because of a lower likelihood of subsequent surgery. In addition, PA provides the ability to avoid posttraumatic arthritis by removing the joint, and has been shown to reduce time of recovery, but it may result in nonunion, and it reduces the ability to execute certain movements. ¹⁶ PA is generally recommended for the treatment of Lisfranc injuries with significant ligamentous involvement, chronic presentation, and arthritis.^11,14 Additionally, some studies have shown that PA may minimize return-to-activity duration in young populations.^4,15 There is currently a gap in the literature concerning whether or not subtle Lisfranc injuries would produce the same outcomes as higher-energy Lisfranc injuries when undergoing ORIF or PA procedures.

The purpose of this study was to evaluate whether a primary ORIF optimizes outcomes and minimizes complications when compared to a primary arthrodesis of the TMTJ for subtle Lisfranc injuries. We hypothesized that primary ORIF would yield better patient-reported outcomes than primary arthrodesis for the surgical treatment of subtle Lisfranc injuries.

Methods

This was a single-center retrospective study conducted from the institutional review board–approved Foot and Ankle Registry data, and the protocol was approved by the steering committee at the investigators’ institution. Inclusion criteria were patients who underwent either a primary ORIF or primary arthrodesis for a ligamentous or minimally displaced avulsion fracture subtle Lisfranc injury between 2016 and 2023 with no additional injuries. The criteria used for a subtle Lisfranc was a stage II Lisfranc injury (diastasis 2-5 mm), as is commonly used in other studies.^2,4,9,17 Diastasis was measured as described by Sugino et al ²⁴ in 2021, in which the distance between the base of the second metatarsal and the medial cuneiform was recorded. Preoperative weightbearing radiographs were reviewed to measure diastasis and confirm subtle Lisfranc injuries (Figures 1 and 2). Patients with dislocation, previous ipsilateral surgeries, or concomitant procedures were excluded. Seventy-three patients met the inclusion criteria and were analyzed; 41 patients received a PA, and 32 patients received an ORIF. Procedure selection was determined by surgeon preference. Patient-reported outcomes via PROMIS scores were collected preoperatively and at least 2 years postoperatively for all patients. Subsequent procedures were also recorded.

Figure 1.

Lateral radiographic view of subtle Lisfranc injury between the first and second metatarsals.

Figure 2.

Axial radiographic view of subtle Lisfranc injury between the first and second metatarsals.

Survey Outcomes

Patient-reported outcomes were assessed using Patient-Reported Outcomes Measurement Information System (PROMIS) scores, which has been validated for various foot and ankle conditions. ¹² PROMIS is a computerized adaptive test (CAT) used to assess functional outcomes in the following domains: Physical Function, Pain Interference, Pain Intensity, Global Physical Health, Global Mental Health, and Depression. Scores have a standardized mean of 50, the reference population average, with an SD (T score) of 10. Higher scores indicate greater physical function, pain interference, pain intensity, global health, and depression. In our patient cohort, clinical outcomes were collected preoperatively and at a minimum of 1-year postoperative follow-up. All patients received PROMIS surveys at 1, 2, and 5 years (if applicable) postoperatively through the foot and ankle registry at our institution, and for the purposes of this study, an attempt was made to collect the most recent PROMIS scores for all patients.

Surgical Techniques

Open Reduction Internal Fixation

Patients were placed supine with regional or spinal anesthesia and a thigh or ankle tourniquet for hemostasis. An ~3-cm dorsal incision was made in the first web space, with careful dissection to protect the neurovascular bundle. Scar tissue was debrided as needed.

A secondary ~1-cm medial incision was made over the medial cuneiform to access the joint. The Lisfranc joint was reduced first to remove the diastasis, followed by assessment of the first tarsometatarsal (TMT) and intercuneiform (IC) joints for residual instability. After identifying and protecting the tibialis anterior tendon, fixation was achieved using either Synthes cortical-headed screws or Arthrex headless variable-pitch screws, per surgeon preference. If instability remained on stress testing, an additional screw was placed across the medial and middle cuneiforms.

Final fixation was confirmed with intraoperative fluoroscopy. A splint and soft dressing were applied postoperatively. Sutures were removed at 2-3 weeks, and range-of-motion exercises were initiated. Patients remained nonweightbearing for 6 weeks, progressed to partial weightbearing in a CAM boot, and returned to full activity by 4-5 months postoperatively (Figures 3 and 4).

Figure 3.

Lateral view of ORIF postoperative radiograph displaying a singular screw providing stability to the foot. ORIF, open reduction internal fixation.

Figure 4.

Axial view of ORIF postoperative radiograph displaying a singular screw providing stability to the foot. ORIF, open reduction internal fixation.

Arthrodesis

An ~4-cm dorsal incision was made along the first web space, and dissection was carried to the first TMT joint with retraction of the extensor hallucis longus tendon. Cartilage was removed from the first metatarsal and medial cuneiform using an oscillating saw, followed by subchondral preparation with 2.0 mm drill perforations. The second TMT joint was exposed and debrided similarly.

Fixation began with reduction and stabilization of the first TMT joint using screws in compression from both distal-to-proximal and proximal-to-distal. The second TMT joint was stabilized using either screws or a plate with locking screws. If intercuneiform instability persisted, an additional IC screw was placed. As with ORIF, screw selection (Synthes or Arthrex) depended on surgeon preference. Locking plates were also often used. Calcaneal bone autografts were frequently used to support fusion.

Final fluoroscopic images were obtained. Postoperative care included a splint and soft dressing, with suture removal and ROM exercises at 2-3 weeks. Weightbearing protocols matched those of the ORIF cohort (Figures 5 and 6).

Figure 5.

Lateral view of PA postoperative radiograph showing significant hardware requirement for fixation. PA, primary arthrodesis.

Figure 6.

Axial view of PA postoperative radiograph showing significant hardware requirement for fixation. PA, primary arthrodesis.

Statistical Analysis

Descriptive statistics were reported as means and SDs for continuous variables and counts and percentages for categorical variables. A priori power analysis required at least 24 patients per cohort for 80% power. ¹¹ Paired t tests were used to compare preoperative and postoperative PROMIS scores after normality was assessed and confirmed using the Shapiro-Wilk test. Listwise deletion was conducted for patients with missing data. Postoperative scores were also compared between study groups. Statistical significance was determined with an alpha of 0.05. Analysis was conducted on R: A Language and Environment for Statistical Computing (R Core Team 2021, Vienna, Austria).

Results

Demographic Data

Chart review was performed to collect demographic information and to record any postoperative complications. The average age for all patients was 41.0 (range, 14-75) years, with an average of 35.2 (range, 14-70) years for the ORIF group and an average of 45.3 (range, 20-75) years for the PA group (P = .004). The average body mass index (BMI) for all patients was 25.6 (range, 17.5-46.1), with an average of 25.1 (range, 17.5-30.2) for the ORIF group and an average of 25.9 (range, 17.7-46.1) for the PA group (P = .49). The average time to clinical follow-up was 4.2 years (range, 2.0-7.4), with an average of 4.0 (range, 2.0-7.4) for the ORIF group and an average of 4.4 (range, 2.0-7.3) for the PA group (P = .48) (Table 1).

Table 1.

Demographic Cohort Comparison.

	Open Reduction and Internal Fixation (n = 36)	Primary Arthrodesis (n = 42)	P Value a
Age, y, mean (SD)	35.2 (15.5)	45.3 (16.2)	.009
BMI, mean (SD)	25.3 (3.5)	25.9 (5.6)	.57
Sex, number of female patients (%)	19 (53%)	25 (60%)	.63
Clinical FU, years, mean (range)	3.9 (2.0-7.4)	4.4 (2.0-7.3)	.36
Radiographic FU, y, mean (range)	1.0 (0.3-6.7)	1.2 (0.3-3.9)	.76
Initial diastasis (SD)	3.26 (0.99)	3.63 (0.80)	.08

Abbreviations: BMI, body mass index, FU, follow-up.

^aBoldface type indicates statistical significance.

Clinical Outcomes

There was a significant difference in preoperative diastasis between the ORIF group (3.32 mm) and the PA group (3.68 mm) (P = .04). Of the 73 subtle Lisfranc patients included in the study who were at least 2 years postsurgery, we were able to collect 57 follow-up surveys (78%), including 31 from the PA group (73%) and 28 from the ORIF group (84%) (P = .25). Both groups demonstrated significant improvement in Physical Function, Pain Interference, Pain Intensity, Global Physical Health, and Depression from their most recent postoperative PROMIS scores. There was no significant difference between the PA and ORIF groups at preoperative or postoperative surveys in any PROMIS criteria (Table 2).

Table 2.

Comparison of Most Recent PROMIS Scores Between ORIF and PA Patients. ^a

	Preoperative Score, Mean ± SD	Postoperative Score, Mean ± SD	P Value	Score Change
Physical function
ORIF	34.8 ± 14.1	54.9 ± 10.8	<.01	20.1
PA	32.5 ± 7.4	53.6 ± 10.3	<.01	21.1
P value	.45	.64
Pain interference
ORIF	61.9 ± 8.3	47.5 ± 7.6	<.01	−14.4
PA	64.4 ± 6.4	50.7 ± 9.4	<.01	−13.7
P value	.24	.18
Pain intensity
ORIF	48.4 ± 8.9	38.8 ± 6.9	<.01	−9.6
PA	51.4 ± 7.0	41.9 ± 9.2	<.01	−9.5
P value	.22	.17
Global physical health
ORIF	45.5 ± 8.8	55.1 ± 9.3	<.01	9.6
PA	43.2 ± 6.7	52.8 ± 9.5	<.01	9.6
P value	.36	.38
Global mental health
ORIF	55.4 ± 9.6	53.2 ± 9.3	.51	−2.2
PA	53.3 ± 6.5	53.9 ± 9.9	.78	0.6
P value	.41	.79
Depression
ORIF	53.1 ± 11.0	46.7 ± 6.7	.027	−6.4
PA	51.0 ± 8.6	46.6 ± 7.6	.036	−4.4
P value	.48	.97

Abbreviations: PROMIS, Patient-Reported Outcomes Measurement Information; ORIF, open reduction and internal fixation; PA, primary arthrodesis.

^aBoldface type indicates statistical significance. Postoperative scores represent the latest available survey follow-up scores (at least 2 years postoperatively).

Radiographic Outcomes

Patients’ postoperative weightbearing radiographs were reviewed. Fifty-nine patients had radiographs of at least 3 months after surgery (81%), of which the group had an average of 13 months after surgery in their most recent radiograph. There was no significant difference in the number of hardware failures between groups (P = .11) on radiographic review. There were no cases of loss of reduction in either group on radiograph analysis.

Complications

No symptomatic nonunion or infection occurred in either group. There were 2 returns to surgery in the PA group including 1 planned removal of hardware and 1 symptomatic removal of hardware. In the ORIF group, there were 14 returns to surgery including 4 planned removal of hardware procedures and 10 symptomatic removal of hardware procedures. There was 1 case of chronic regional pain syndrome (CRPS) in the ORIF group, but no cases of PA conversion. The difference in symptomatic hardware removals was statistically significant between groups (Table 3).

Table 3.

Revision and Complication Data. ^a

	Open Reduction and Internal Fixation, n (%)	Primary Arthrodesis, n (%)	P Value
Subsequent procedures	18 (50)	2 (5)	<.01
Hardware removal	18 (50)	2 (5)	<.01
Planned hardware removal	7 (19)	1 (2)	.01
Symptomatic hardware removal	11 (30)	1 (2)	<.01
Chronic regional pain syndrome	1 (3)	0 (0)	.28
Hardware failures	2 (6)	0 (0)	.11

^aBoldface type indicates statistical significance.

Discussion

This study sought to compare the patient-reported outcomes and complication rates among subtle Lisfranc patients who received PA or ORIF surgeries. Our findings demonstrate that although both treatments achieve similar patient-reported outcomes at minimum 2-year follow-up, ORIF patients experienced significantly higher rates of hardware removal procedures (50% vs 5%, P < .01), which has important implications for patient counseling and treatment selection. There was no significant difference in hardware failures between groups on radiographic analysis (P = .11). To our knowledge, this study represents the comparison of patient-reported outcomes in PA and ORIF procedures in subtle Lisfranc patients with the longest follow-up.

The PA group was significantly older (P < .01), possibly inflating its results because of differing expectations. The lower average age of patients in the ORIF group may reflect the surgeons’ preference to preserve joint motion in younger patients, and it shows the preference toward PA in older patients. The difference in diastasis between groups likely showed surgeon preference for PA in more significant injuries. We did not account for operating time or hardware costs in each group, which is generally higher for arthrodesis procedures than for ORIF procedures. There also may have been some attrition bias because of the 22% of patients lost to follow-up. Although there were no documented symptomatic nonunions in the PA group, because there were no postoperative CT scans conducted, there may have been asymptomatic nonunions of which we were not aware. Contrary to our hypothesis, the 2 groups demonstrated similar outcomes in their most recent postoperative PROMIS scores.

Open reduction and internal fixation is often used for the treatment of bony high-energy Lisfranc injuries, and it has demonstrated good results in patient-reported outcomes as compared to alternative measures.^{7,8,13,16,19,23} In a study by Crates et al, ⁵ at an average of 3 years after ORIF surgery, a group of 20 patients had significantly better AOFAS outcomes than a group of 16 patients undergoing conservative measures during the same duration (92.3 vs 75.3). In a study of 48 patients with general Lisfranc injuries conducted by Kuo et al, ¹³ at an average of 52-month follow-up, ORIF patients had significantly better AOFAS outcomes than patients who received a secondary fusion procedure (80.2 vs 58.2). In a study of 107 Lisfranc fractures treated with ORIF conducted by Sroka et al, ²³ patients who received a hardware removal procedure (n = 77) at an average of 4.43 months experienced a significantly greater improvement in Physical Function and Pain Interference PROMIS scores as compared to patients who received no such procedure. This explains the prominence of hardware removal procedures demonstrated in the ORIF procedures in this study. Dong et al ⁷ described a group of 182 patients who received ORIF for the treatment of general Lisfranc injuries, and these patients experienced significant improvement in physical function PROMIS scores at consistent intervals from surgery until 48 weeks following surgery. This trend shows the importance of using outcomes of at least 2 years after surgery to avoid substantial discrepancies in stages of recovery between patients.

Primary arthrodesis is a surgical procedure that sacrifices joint mobility by fusing the affected tarsals and metatarsals, which has been clinically proven to improve function and reduce pain levels.^6,15,16,20 There are some concerns, however, that the altered biomechanics following PA can lead to increased stress on the neighboring joints. ¹ Generally, PA is recommended for patients with mostly ligamentous involvement, severe diastasis (high-energy injury onset), failed primary ORIF surgery, or severe cartilage compromise or comminution. ⁶ In a study of 187 general Lisfranc patients conducted by Saxena et al, ²⁰ 111 patients were able to return to sports, but there were 35 documented nonunions (18.6%). In a study conducted by MacMahon et al ¹⁵ of 46 patients with general Lisfranc injuries, on 5.2-year average follow-up after PA, 25% of patients experienced increased difficulty in physical activity. They reported zero instances of symptomatic nonunion, but 5 patients (13%) required hardware removal. PA can be a good option for Lisfranc injury repair if a patient is willing to sacrifice joint mobility, especially if the patient has significant ligamentous injury, but there may be a risk of nonunion to consider.

Many studies have investigated the outcomes of PA vs ORIF in general Lisfranc injuries. Ly and Coetzee presented a randomized clinical trial of 20 ORIF and 21 PA patients with ligamentous injury character. On 42.5 months’ average follow-up, the PA group had significantly higher AOFAS scores (88.0 vs 68.6) and fewer cases of persisting pain (5 in the ORIF group vs 0 in the PA group). ¹⁴ In addition, 75% of the ORIF group underwent hardware removal operations. The ligamentous injury involvement may have contributed to the midfoot degeneration witnessed in many of the ORIF patients, requiring additional procedures. In a systematic review of general Lisfranc repair by Smith et al ²² (n = 102), ORIF procedures resulted in significantly more hardware removals, but there was no difference in postoperative patient-reported outcomes. Henning et al ¹⁰ presented a prospective randomized study in which 20 ORIF and 20 PA general Lisfranc patients were evaluated 53 months after surgery. There was no difference in patient-reported outcomes, but there were significantly more hardware removals required in the ORIF group (79% vs 17%). Most of the studies comparing PA to ORIF show similar patient-reported outcomes, despite the need for significantly more hardware removal operations in the ORIF group.

Very little previous research had been conducted in determining the best procedure for subtle Lisfranc injuries. Historically, subtle Lisfranc injuries have been treated using the same principles as general Lisfranc injuries. ² Cochran et al ⁴ compared return to duty times and fitness test times in 14 PA and 18 ORIF patients in active military duty. They found that the PA group returned to duty 2.2 months faster (4.5 vs 6.7 months) and they ran the test 30 seconds per mile faster than the ORIF group. There was no significant difference in Foot and Ankle Ability Measure (FAAM) scores on a 35-month average follow-up. Like the study conducted by Ly and Coetzee, the time to return was measured from the initial surgery, so the hardware removal procedures received by the ORIF patients likely contributed to their longer times. ¹⁴

This study has several limitations that may impact generalizability and internal validity. First, there were no objective outcome measurements such as range-of-motion or strength tests, thus allowing presurgery expectations to be a potential biasing factor. Second, the retrospective study was conducted by using data from 10 different surgeons, introducing another potential confounding variable. Third, the absence of postoperative CT analysis limited the ability to evaluate osteoarthritis or adjacent joint disease. Lastly, there were no documented return to activity times in the database, so we could not compare objective results with the Cochran study. ⁴

Conclusion

In conclusion, both procedures resulted in significant improvement in clinical outcomes. ORIF and PA groups produced comparable PROMIS scores, whereas the ORIF group required more reoperations for hardware removal. Despite having significantly more removal of hardware surgeries, the ORIF group showed similar mean patient-reported outcomes as the PA group. We concluded that both treatments can be viable options, and further research is needed to determine which may be optimal for different subtle Lisfranc patient populations.