Assessing the Biomechanical, Kinematic, and Force Distribution Properties of the Foot Following Tarsometatarsal Joint Arthrodesis: A Systematic Review

Abstract

The Lisfranc joint connects the forefoot to the midfoot. Tarsometatarsal (TMT) arthrodesis is commonly employed for Lisfranc joint injuries; however, there is active discussion regarding the optimal method of fixation for TMT arthrodesis. The purpose of this systematic review is twofold: to assess the stability of various constructs used in TMT arthrodesis and to evaluate joint motion and force distribution in the foot following arthrodesis. The PubMed, CINAHL, MEDLINE, and Web of Science databases were searched for articles evaluating biomechanical and kinetic properties of TMT arthrodesis constructs in accordance with PRISMA guidelines. The preliminary search yielded 367 articles and the final review included 14 articles with 195 cadaveric and 70 synthetic bone constructs. Plantar plates and intramedullary screw fixation at the first TMT joint were consistently found to bear significantly greater loads and resist diastasis more effectively than crossed screws. Furthermore, whole foot and lateral column arthrodesis significantly elevate calcaneocuboid and lateral column pressures. This increase was not observed with isolated fourth or fifth TMT arthrodesis. TMT arthrodesis should aim to avoid the lateral column and fuse as few joints as possible. Overall, plantar plates are an effective construct for first TMT arthrodesis due to their ability to withstand both compressive and tensile forces while maintaining stable alignment of the foot via reinforcement of the transverse arch. Intramedullary fixation devices are an alternative option that provide a high degree of active compression across the joint space while avoiding irritation of surrounding soft tissue structures.

1. Introduction

The tarsometatarsal (TMT) joint complex, sometimes referred to as the Lisfranc joint, is a vital and anatomically complex joint, connecting the forefoot to the midfoot via articulations in the metatarsal and various cuneiform or cuboid bones [1,2,3]. The TMT joint complex can be further subdivided into the medial column (first metatarsal and medial cuneiform), middle column (second and third metatarsal and middle and lateral cuneiforms), and lateral column (fourth and fifth metatarsals and cuboid) [1,4]. The consideration of specific midfoot columns has functional importance, as each column differs in its motion, thereby lending unique implications for injury pattern and recovery [1,2,4,5]. The medial and middle columns permit only a small amount of motion, 3.5 mm in the coronal plane and 0.6 mm in the sagittal plane, respectively, and have thereby been termed “nonessential joints” [5,6]. The lateral column acts as a shock absorber on uneven surfaces and permits the most motion, allowing for 13 mm of motion in the sagittal plane [2,5].

In addition to connecting the forefoot and midfoot, the Lisfranc joint is essential for supporting the transverse arch of the foot [1,2]. A wide variety of ligaments provide additional stability, with the Lisfranc ligament, an interosseous ligament running obliquely between the medial cuneiform and second metatarsal, being the strongest [1,4]. As such, injury to the Lisfranc joint complex or supporting ligaments results in the destabilization of the midfoot complex and increases risk for arch collapse, gait abnormalities, and osteoarthritis [7,8]. TMT joint complex injury occurs in 1 in 55,000 people annually, with the most common cause of symptomatic TMT complex pathology being post-traumatic arthritis followed by primary osteoarthritis [9]. Acute trauma to the TMT complex can occur via direct and indirect mechanisms, with direct injuries being high-energy traumas such as crush injuries and indirect injuries commonly having axial forces applied to a plantarflexed foot, resulting in dorsal tension failure [2,10]. The first, second, and fourth TMT joints are the most commonly affected [9,10,11]. However, up to 20% of TMT joint complex injuries are underdiagnosed and subsequently untreated, implicating a need for timely diagnosis and treatment to prevent or reduce onset of post-traumatic arthritis [12].

There is active discussion regarding the optimal surgical treatment for repairing ligamentous or bony damage to the Lisfranc joint complex, with post-operative arthritis rates following open reduction and internal fixation ranging from 27% to 94% [2,3,6,13]. Likewise, closed reduction and internal fixation has been associated with a high rate of repeat dislocations [14]. Primary arthrodesis has emerged as a viable option, with multiple systematic reviews and randomized control trials (RCTs) demonstrating decreased reoperation rates, fewer hardware complications, and improved patient reported outcomes (PROs) [3,6,13,15]. However, there is little clarity regarding the efficacy of the various constructs utilized for arthrodesis [16,17,18,19]. As such, this systematic review aimed to evaluate the biomechanical integrity of different constructs employed for arthrodesis in TMT joints through simulated use cases in human cadavers and synthetic bone models. This was accomplished through two main modalities with the primary outcome being a biomechanical analysis of the stability of various constructs as measured by failure load and joint displacement following the arthrodesis procedure. A secondary outcome was a kinematic analysis of joint motion and force distribution of the foot following arthrodesis at various joint levels. To date, no study has evaluated both construct stability and kinematic and force distribution changes following arthrodesis to provide insights regarding both the optimal fixation method and level [20]. In doing so, we hope to address some of the controversy surrounding the biomechanics of this procedure, although clinical information is ultimately necessary prior to adoption.

2. Materials and Methods

2.1. Search Parameters

This systematic review adhered to the guidelines outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). A comprehensive search encompassing PubMed, CINAHL, MEDLINE, and Web of Science databases was executed from their inception until 18 June 2023. The search utilized terms such as (TMT OR TMTJ OR tarsometatarsal OR tarsal-metatarsal OR Lisfranc OR midfoot) AND biomechanic* AND (fusion OR arthrodesis). After removing duplicates, the titles and abstracts were screened for relevance. Full texts meeting the inclusion/exclusion criteria were thoroughly reviewed, and data extraction was then conducted on the included full-text articles.

2.2. Inclusion/Exclusion Criteria

The inclusion criteria encompassed articles categorized as evidence level I-III, focusing on primary outcomes associated with the biomechanical properties (failure load, construct stiffness, plantar diastasis) of the fixation construct/TMT joint or kinetic properties of the foot subsequent to arthrodesis of any TMT joint. Exclusion criteria comprised abstract-only publications, unavailability of full-text, articles in languages other than English, case reports, case series, technical notes, retrospective clinical studies, systematic reviews, meta-analyses, and book chapters.

2.3. Study Definitions

The term stability, as it is used in this review, is meant to describe a construct/joint’s resistance to separation/diastasis following load. A related term, strength, is used to refer to the maximal loads a construct/joint can handle prior to experiencing permanent deformation or absolute construct failure. By comparison, stiffness refers to the inherent elastic characteristics of a structure and represents the amount of force per unit length necessary to produce a permanent deformation.

2.4. Data Extraction

Information extracted from the chosen articles for this systematic review encompassed the first author of the article, complete article title, year of publication, construct type (cadaver, synthetic bone analogue), and the quantity of constructs employed, along with the fixation method (plates, intramedullary fixation devices, screws, shape-memory staples), and biomechanical/kinetic outcomes of TMT arthrodesis (construct stability, construct strength, plantar diastasis, adjacent joint pressures, joint mobility, etc.).

2.5. Statistical Analysis

Statistical analysis was conducted using Microsoft Excel version 16.77.1. A narrative approach was employed in this systematic review due to the heterogeneity among studies, precluding a meta-analysis. Consequently, only descriptive statistics and basic frequencies were computed.

3. Results

3.1. Preliminary Search Findings

The preliminary search generated a total of 367 articles, with 201 identified as duplicates, resulting in 166 unique articles. Among these, 150 were excluded due to a lack of relevance, leaving 16 articles for thorough full-text review. Following the review, two articles did not meet the eligibility criteria and were consequently excluded. Figure 1 below provides a visual representation of the search process and article identification for the ongoing systematic review.

3.2. Study Demographics

In the 14 articles analyzed, a total of 195 human cadaver and 70 synthetic bone models were used [16,17,18,19,21,22,23,24,25,26,27,28,29]. A combination of dorsal plates, plantar plates, crossed screws, intramedullary fixation devices, shape-memory staples, and locking plates with lag screws were utilized to achieve arthrodesis. Arthrodesis was performed at each TMT joint as well as the medial cuneiform–second metatarsal and intercuneiform joints. Construct stability, strength, and resulting changes in foot kinematic motion and pressure distribution were the primary outcomes assessed.

Table 1. Biomechanical data regarding the type of arthrodesis construct, location of primary arthrodesis, and biomechanical testing data for each article included in this systematic review.

First Author (Year)	Full Title	Model Used (Cadaver, Synthetic Bone Model, etc.)	Number of Construct	HPrimary Fusion Joint	Fixation Method(s)	Diastasis Mean (mm)	Plantar Gapping (mm)	Displacement (mm)	Average Diastasis Decrease (mm)	Ultimate Failure Load (N)	Axial Stiffness (N/mm)
Amiethab Aiyer (2015) [29]	The Impact of Nitinol Staples on the Compressive Forces, Contact Area, and Mechanical Properties in Comparison to a Claw Plate and Crossed Screws for the First Tarsometatarsal Arthrodesis	Synthetic Bone	20	1 TMT	Single Shape Memory Staple, Double Shape Memory Staple, Claw Plate, Crossed Screws		Single Staple: 0.95 Double Staple: 0.69 Claw Plate: 2.61 Crossed Screws: 0.6				Single Staple: 206.78 Double Staple: 226.24 Claw Plate: 184.47 Crossed Screws: 347.85
Ashleen R. Knutsen (2017) [17]	Biomechanical Comparison of Fixation Devices for First Metatarsocuneiform Joint Arthrodesis	Synthetic Bone	32	1 TMT	IO Fixation, Crossed Screws, Dorsal Plate	IO Fix: 2.8 Crossed Screws: 0.4 Dorsal Plate: 5.5				IO Fix: 460 Crossed Screws: 430 Dorsal Plate: 110	IO Fix: 4 Crossed Screws: 10 Dorsal Plate: 5.9
Cynthia Prieto-Diaz (2019) [22]	Biomechanical Comparison of First Tarsometatarsal Arthrodesis Constructs Over Prolonged Cyclic Testing	Cadaver	40	1 TMT	Intramedullary fixation, medial plate with crossing screw, crossed screws	Intramedullary 4 hole plate: 0.14 Intramedullary 3 hole plate: 0.17 Medial Plate: 0.23 Crossing Screws: 0.54
Ernest Pope (2013) [26]	Midfoot Fusion: A Biomechanical Comparison of Plantar Planting vs. Intramedullary Screws	Cadaver	7	1/2/3/4/5 TMT	Plantar Plating, Intramedullary Screws					Plantar Plate: 344 Intramedullary Screws: 313	Plantar Plate: 9.7 Intramedullary Screws: 11.2
Genbin Wu (2019) [21]	Effect of Different Fusion Types on Kinematics of Midfoot Lateral Column: A Comparative Biomechanical Study	Cadaver	10	4/5 TMT	3.5-mm Fully Threaded Screws
Hugo Perez (2007) [24]	Effects on the Metatarsophalangeal Joint After Simulated First Tarsometatarsal Joint Arthrodesis	Cadaver	5	1 TMT	K-wires
Jaeyoung Kim (2022) [25]	Kinematic Analysis of Sequential Partial-Midfoot Arthrodesis in Simulated Gait Cadaver Model	Cadaver	10	1/2/3 TMT	Dorsal Lisfranc Plate
Joshua Nadaud (2011) [28]	Plantar and Calcaneocuboid Joint Pressure After Isolated Medial Column Fusion Versus Medial and Lateral Column Fusion: A Biomechanical Study	Cadaver	12	1/2/3/4/5 TMT	4.5-mm Fully Threaded Screws
Nathan C. Ho (2019) [23]	Biomechanical Comparison of Fixation Stability Using a Lisfranc Plate Versus Transarticular Screws	Cadaver	13 pairs	1/2 TMT	Dorsal Plate, Crossed Screws
Rene Burchard (2018) [16]	Biomechanics of Common Fixation Devices For First Tarsometatarsal Joint Fusion—A Comparative Study With Synthetic Bones	Synthetic Bone	9	1 TMT	Dorsal Locking Plate, Plantar Locking Plate, IO Fixation			Dorsal Plate: 0.06 Plantar Plate: 0.03 IO Fix: 0.08		Dorsal Plate: 324 Plantar Plate: 377 IO Fix: 173
Richard Marks (1998) [27]	Midfoot Fusion Technique for Neuroarthropathic Feet: Biomechanical Analysis and Rationale	Cadaver	8 pairs	1 TMT	Plantar Plate, Crossed Screws			Plantar Plate: 0.2 Crossed Screws: 0.89		Plantar Plate: 2917 Crossed Screws: 1480
Sarah Ettinger (2022) [19]	Biomechanical Evaluation of Tarsometatarsal Fusion Comparing Crossing Lag Screws and Lag Screw With Locking Plate	Cadaver	30	1/2/3 TMT	Locking Plate with Lag Screw, Crossed Screws
VC Shen (2022) [30]	In Situ Deformation of First Tarsometatarsal Arthrodesis Implants with Digital Image Correlation: A Cadaveric Study	Cadaver	6 pairs	1 TMT	Locking Plate with Lag Screw, Shape-Memory Staples			Locking Plate: 3.1 Nitinol Staples: 3.9
Zachary A. Koroneos (2022) [18]	Biomechanical Comparison of Fiber Tape Device Versus Transarticular Screws for Ligamentous Lisfranc Injury in a Cadaveric Model	Cadaver	8 pairs	Medial Cuneiform—2nd Metatarasal	Crossed Screws, Interosseous Fiber Tape	Crossed Screws: 0.68 Fiber Tape: 0.9			Crossed Screws: −2.61 Fiber Tape: −2.81

below highlights the key characteristics of the included studies.

3.3. Single TMT Joint Arthrodesis

Of the fourteen articles included, seven studies evaluated the efficacy of various fixation constructs at a single TMT joint, with six studies assessing the first TMT joint, and one study assessing the Lisfranc joint. Koroneos et al. evaluated fixation constructs at the Lisfranc joint, comparing an interosseous fiber tape device to traditional dual crossing screw fixation [18]. They found no difference in diastasis of the Lisfranc joint with either fixation method (p = 0.55).

Knutsen et al. sought to evaluate a novel intramedullary fixation device (IO fixation) consisting of a post through the first metatarsal with a screw passing at 60 degrees through the medial cuneiform relative to both traditional crossed-screw fixation and a dorsal locking plate [17]. Both the IO fixation device and crossed-screw fixation required a significantly higher failure load and demonstrated significantly less plantar gapping than the dorsal plate (p = 0.01). Prieto-Diaz and colleagues continued to study intramedullary fixation, comparing two intramedullary fixation devices (three-hole and four-hole) to a medial plate with crossing screw and crossed screw fixation [22]. Both intramedullary devices and the medial plate with crossing screw exhibited significantly less plantar gapping than the crossed screw constructs following a 20,000-cycle fatigue test (p < 0.05).

Marks et al. was the first to report on the biomechanics of plantar plating in TMT arthrodesis, comparing it to traditional crossed screw fixation and finding that plantar plating required a significantly higher failure load (p = 0.03) and experienced significantly less displacement prior to and following joint loading (p = 0.05) [27]. Burchard et al. synthesized the various aforementioned constructs into one comparative study, assessing the relative strengths of a plantar locking plate, medial locking plate, and the same IO fixation device [16]. IO fixation provided the greatest compressive force, which was significant when compared to the plantar plate only (p = 0.033). However, the plantar plate provided significantly greater resistance to displacement than both the medial locking plate and IO fixation (p < 0.000) while also requiring a significantly greater failure load than IO fixation (p < 0.002).

Shen et al. compared shape-memory staple fixation intended to sustain a high level of compression under direct loading due to high elasticity with a locking plate with lag screw. Shape-memory staples demonstrated a larger joint gap before (p = 0.001) and after (p = 0.044) loading compared to locking plate with lag screw [30]. Aiyer and colleagues compared both a single and double shape-memory staple configuration to crossed screws and a claw plate [29]. Both staple configurations had significantly greater contact surface area (p < 0.01) and contact force (p < 0.01) then the crossed screw and claw plate constructs. Crossed screws did require the greatest failure load (p < 0.001) and experienced the least plantar gapping under load (p < 0.05). Following load, however, both staple groups recovered the experienced gap entirely which was significant when compared to the claw plate group (p < 0.001).

3.4. Multilevel TMT Arthrodesis

There were three included articles that assessed TMT arthrodesis at multiple joint levels. Ho et al. evaluated transarticular screw fixation and a rhomboidal dorsal plate for arthrodesis of the first and second TMT joints [23]. There was no significant difference in the load tolerated between the two groups nor any difference in motion under abduction or axial loading at either TMT or Lisfranc joint for either group. Ettinger et al. compared a locking plate with lag screw with two crossing screws across for arthrodesis of the medial column. Following arthrodesis of the first TMT joint with a locking plate and lag screw, subsequent arthrodesis of the second and third TMT joints with a locking plate and a lag screw significantly increased joint stability [19]. On the other hand, following first TMT arthrodesis with crossed lag screws, the addition of a lag screw only resulted in significantly increased joint stability at the second TMT joint (p = 0.0067), but not when added to the third TMT joint. Furthermore, while joint stability was not significantly different across the first two TMT joints between the two methods, the locking plate and lag screw demonstrated significantly greater stability at the third TMT joint (p < 0.001), indicating that utilizing a locking plate with a lag screw affords a greater initial compressive foundation that contributes added stability to arthrodesis of adjacent joints. Finally, Pope et al. studied arthrodesis of all TMT joints by comparing fixation with intramedullary screws to plantar plating [26]. Screws spanned from the distal metatarsal to the proximal cuneiform/cuboid. The plantar plate was placed below the first TMT joint while each subsequent TMT joint was held reduced and fixed via a 3.5 mm cortical screw. There was no difference in stiffness or failure load between either construct.

3.5. Kinetic Behavior and Force Distribution of Foot Following Arthrodesis

While the majority of included studies assessed the biomechanical properties of various constructs following implantation, four studies evaluated the kinetic behavior and force distribution of the foot following arthrodesis. Following arthrodesis of the first TMT joint, Perez et al. found that first MTP joint dorsiflexion increased by 25% (p = 0.01) [24]. Additionally, scatter, or the deviation between the center of rotation of the first MTP joint during various arcs of its overall range of motion, significantly decreased (p = 0.05). Kim and colleagues studied joint motion more extensively across the foot at the ankle, subtalar, and talonavicular, as well as at the first MTP joint, following sequential arthrodesis of the medial column [25]. While there was no significant change in joint motion following sequential arthrodesis of the first two joints, there was a significant, albeit minor, increase in subtalar inversion (p = 0.032), ankle motion by 2.1 degrees (p = 0.009), and subtalar motion by 2.8 degrees (p = 0.014).

Regarding force distribution in the foot, Nadaud et al. compared fusion of the medial columns to whole-foot TMT arthrodesis in order to assess changes in pressure in various aspects of the foot [28]. Whole foot fusion resulted in significantly elevated pressures along the lateral aspect of the foot in a neutral and everted position (p < 0.001). Calcaneocuboid pressure also increased significantly following whole-foot fusion in the neutral and inverted positions (p = 0.001). Wu and colleagues took a more targeted approach to evaluating force distribution in the foot, focusing on changes in pressure patterns following isolated fourth TMT arthrodesis, isolated fifth TMT arthrodesis, and joint lateral column fusion [21]. The main finding was that lateral column resulted in significant increases in lateral foot (p < 0.05) and calcaneocuboid pressures (p < 0.05) compared to the intact, isolated fourth, and isolated fifth TMT fusions.

4. Discussion

TMT arthrodesis is an effective procedure with potential to correct instability in the midfoot and restore functional anatomy while preserving the range of motion. More than 100 surgical procedures for addressing the pathology of the first TMT joint have been described, with ongoing discussion as to which method is the most effective [16,17,18,19]. This systematic review is the first to synthesize the existing literature regarding the various constructs used to fuse TMT joints and describe their biomechanical and kinetic properties. Our review found that, while crossed screws have traditionally been the chosen method of fixation for TMT arthrodesis, various intramedullary fixation devices and plantar plates are stronger, more stable options. In addition, when performing arthrodesis, kinematic and force distribution analysis revealed that medial/middle columns or individual TMT joints should be prioritized over whole foot or lateral column fusion, as doing so increases pressure on the lateral and posterior aspects of the foot by decreasing range of motion through the critical laterally based joints of the foot.

Crossed screw configurations have traditionally been the method of fixation for primary arthrodesis of TMT joints [17,22,23]. A total of seven studies compared a variety of constructs against crossed screws across the first three TMT joints [17,18,19,22,23,27,29]. Of those seven studies, five found a significant advantage to alternative constructs compared to crossed screws, with the other two studies finding no significant difference between alternative methods and screw fixation at the first TMT joint [17,18,19,22,23,27,29]. The two constructs that demonstrated no significant difference when compared to crossed screws were a rhomboidal dorsal plate intended to provide transverse and longitudinal stability and an interosseous fiber tape device [18,23,29]. Plates placed on the plantar or medial aspect of the foot, locking plates with lag screws, shape-memory staples, and intramedullary fixation devices were the constructs that demonstrated greater biomechanical stability than crossed screws. Specifically, plantar plating was capable of handling greater loads while intramedullary fixation and locking plates with lag screw provided greater resistance to diastasis [19,22,27]. These findings are likely explained, at least in part, by the greater cortical bone apposition afforded by these alternative constructs when compared to crossed screws. For example, intrameduallary fixation with a three-hole construct or plantar fixation with a four-hole plate allows for six or eight points of bone contact as opposed to only 4four for crossed screw fixation [22]. In addition, the longitudinal stability along the length axis of the bone provided by the post in the intramedullary fixation device and the plantar plate enables greater force distribution that reduces likelihood of failure at the construct-bone interface [17].

A more in-depth, direct comparison of plantar plating and intramedullary fixation found that plantar plating was able to tolerate significantly higher loads than intramedullary fixation, and that intramedullary fixation was able to provide greater compressive force across the first TMT joint [16]. However, this comparison also found that plantar plating provided more resistance to diastasis than intramedullary fixation. This is likely because intramedullary fixation acts as a load sharing internal splint, resisting both tensile and compressive forces, while plantar plating supports the base of the foot and stabilizes longitudinal arches [27,31,32]. By and large, plantar plating is a more stable method of joint fixation than crossed screws, providing greater strength and greater resistance to displacement under load. Similarly, intramedullary fixation is a newer, more stable fixation method that provides a greater area of stabilization across the joint surface, resisting a wide variety of applied forces. Additional study is necessary to further validate these results in clinical applications prior to large-scale adoption, however.

Although they did not demonstrate broad statistical significance when compared to crossed screws, shape-memory staples were another construct that showed promise as a novel fixation method. While crossed screws showed greater rigidity and construct stiffness when compared to staples, the staples provided greater contact force and contact surface area. Staples also demonstrated full recovery of the plantar gapping following loading, indicating an ability to respond to and recover from changes that occur at the joint interface during normal physiological activity post-operatively [29]. Interestingly enough, staples were found to be significantly weaker than crossed screws or various plates in a systematic review of the biomechanical properties of contructs used for first metatarsophalangeal arthrodesis [33]. Further study in biomechanical models is necessary to validate the use of staples in TMT arthrodesis prior to clinical trial.

Another key outcome addressed in this review was kinematic and force distribution changes in the foot following TMT arthrodesis. A principal finding was that, while fusion of the medial and middle columns does not significantly alter foot joint kinematics or pressures, whole foot TMT arthrodesis significantly increases calcaneocuboid and lateral pressures of the foot [25,28]. This finding holds true when observing the lateral column alone, given that isolated lateral column fusion also increases calcaneocuboid and lateral column pressures [21]. This may occur due to a loss of motion in the lateral column, which has been implicated as the most mobile Lisfranc joint [2,5]. It is for this reason that Weatherford and colleagues suggest that every effort be made to preserve the mobility of this joint [2]. The lack of significant change in adjacent joint motion following medial and middle column arthrodesis and isolated fourth or fifth TMT arthrodesis indicates that these fusions do not compromise the structural balance of the foot or offload weightbearing responsibilities on less mobile joints of the foot. Whole foot and isolated lateral column arthrodesis should be avoided, especially if clinical improvement can be obtained with media/middle column or fourth or fifth TMT arthrodesis.

Early clinical trials have validated the results of our review, confirming the effectiveness of plantar plating. Fraser et al. studied 62 patients that underwent plantar plating as the method of choice for midfoot arthrodesis and reported a 94% union rate [34]. There were also improvements in angular alignment of the talus, metatarsals, and plantar arches. Likewise, Klos et al. reported a 98% union rate and a similar improvement in joint angles following plantar plating in 58 patients [35]. Bayam et al. used the intramedullary IO fix device and reported a 94% fusion rate at the first TMT joint, with only one patient complicated by superficial skin infection and non-union [36]. While there is a significant need for further prospective clinical studies assessing fixation constructs, early clinical data validates our findings that plantar plating provides rigid fixation and high degree of resistance to deformity while maintaining better alignment of the foot. As suggested earlier, this is likely due to its ability to stabilize the soft tissue arches of the midfoot. Similarly, intramedullary fixation demonstrated comparable clinical outcomes likely owing to the greater compressive forces afforded by the device while avoiding surrounding soft tissue irritation and interruption of circulation necessary for joint healing and stabilization.

The primary limitation of this review and the included articles was the limited clinical applicability of strict, isolated biomechanical analysis in cadaveric or synthetic bone constructs. Without accounting for clinical comorbidities or kinematic and force distribution contributions from adjacent muscular and osseous structures, it is difficult to ascertain how these constructs will perform postoperatively. This was further complicated by small sample sizes and variable study design, as some studies resected the articular surfaces prior to fusion while others disrupted capsular attachments to induce midfoot instability. Second, the included studies did not assess outcome parameters uniformly. Some focused on load to failure while others chose to report on construct stiffness or plantar gapping. This variability in outcome parameters paired with different methods of measuring the same parameters created a high level of study heterogeneity. This heterogeneity, stemming from lack of shared outcome parameters, made meta-analysis impractical. Lastly, constructs such as shape-memory staples, medial plates, and fiber tape devices were underrepresented in our review. Further study in the form of RCTs and well-performed, large-sample cadaver studies is needed to better understand how these constructs will function when interacting with dynamic forces postoperatively, in turn guiding future clinical practice.

5. Conclusions

TMT arthrodesis is a complex procedure that ideally provides stability to the midfoot while resisting multidimensional applied forces. Plantar plating of the first TMT joint is an effective construct given its ability to withstand these forces and potentially maintain stable alignment of the foot by supporting the soft tissue arches. Intramedullary fixation devices are similarly effective and have demonstrated a high capacity for providing active compression across the joint space while avoiding irritation of surrounding soft tissue structures. When considering TMT arthrodesis, care should be taken to avoid fusion and maintain mobility in the lateral column, as loss of mobility here results in suboptimal force distribution that predisposes future injury. Further research applying these findings in a clinical setting is needed to validate biomechanical and kinetic analysis.