Multifactorial Non-union After First Metatarsophalangeal Arthrodesis Revised With Structural Iliac Crest Autograft: A Case Report and Comprehensive Literature Review

Patient information and initial presentation

A 62-year-old woman presented with progressive right MTP-I pain, stiffness and walking limitation after failed conservative treatment.

Examination showed advanced hallux rigidus with hallux varus and second-toe varus/claw deformity. Weight-bearing radiographs confirmed end-stage MTP-I arthritis (Fig. 1).

She had a 45 pack-year smoking history. Distal pulses were palpable and arterial duplex ultrasonography was normal. MTP-I arthrodesis was recommended.

Index arthrodesis and early post-operative course

The index procedure used cup-and-cone preparation and low-profile dorsal plate fixation. Intraoperative alignment was considered satisfactory, with dorsiflexion relative to the floor and neutral transverse alignment relative to the second ray.

Postoperatively, she remained non-weight-bearing for 4 weeks before progressive loading. At 3 months, alignment and fixation were maintained, but osseous consolidation was incomplete; retrospectively, dorsiflexion appeared slightly greater than ideal.

Diagnosis of failure

At 4 months, weight-bearing pain and swelling increased. Radiographs were initially inconclusive, but at 6 months showed dorsal plate fracture and persistent fusion line. Computed tomography (CT) confirmed pseudarthrosis, absent bridging trabecular bone, and proximal phalangeal osteolysis (Fig 2). There were no clinical signs of infection; white blood cell, erythrocyte sedimentation rate, and C-reactive protein were normal. Symptomatic aseptic non-union with implant fatigue failure was diagnosed.

Failure analysis

Failure was considered multifactorial. Persistent smoking was the main modifiable biological risk. CT showed local biological failure with pseudarthrosis, sclerosis, and osteolysis. Plate fracture was interpreted as the endpoint of micromotion and cyclic loading, not the primary cause. Subtle sagittal malposition and untreated second-ray deformity may have increased forefoot loading.

Revision surgical technique

Revision was performed through the previous dorsal approach. Failed hardware was removed, fibrous pseudarthrosis and sclerotic margins were excised, intraoperative cultures were obtained, and all non-viable bone was debrided to bleeding cancellous surfaces. Cultures were negative.

Because compression alone would have shortened the first ray, a tricortical iliac crest graft of approximately 1.8 × 1.2 cm was used to restore length and bone stock. Fixation used an interfragmentary compression screw and a stronger dorsal revision plate.

The second-ray deformity was corrected with subcapital second metatarsal osteotomy and proximal interphalangeal arthrodesis stabilized by K-wire to improve forefoot balance.

A conservative post-operative protocol was used: Eight weeks non-weight-bearing in a boot followed by gradual loading. Smoking cessation was achieved and maintained during healing.

Outcome and follow-up

Recovery was uncomplicated. At 3 months, she walked fully weight-bearing in normal footwear without pain or swelling, and radiographs showed maintained alignment with progressive graft incorporation (Fig. 3).

At 6 months, CT showed bridging trabecular bone, graft incorporation, closure of the fusion gap, and stable fixation without recurrent osteolysis (Fig. 4). At 1 year, she remained asymptomatic in regular footwear. The American Orthopaedic Foot and Ankle Society hallux metatarsophalangeal-interphalangeal score improved from 29/100 to 85/100 (Table 1).

Patient perspective and functional assessment

The patient reported resolution of weight-bearing pain and swelling, return to regular footwear, and improved walking tolerance.

Why this case is more than an isolated hardware failure

Plate breakage should not be treated as a diagnosis. It is usually the visible endpoint of an unhealed fusion exposed to repetitive loading. Implant exchange alone may miss the underlying causes: Impaired biology, insufficient compression, bone loss, malalignment, premature loading, or adjacent-ray pathology.

Here, incomplete consolidation preceded symptoms, and symptoms preceded radiographic plate failure. CT then confirmed pseudarthrosis and osteolysis, supporting biological failure and micromotion before fatigue fracture.

Biological risk: Smoking and host factors

Smoking is a major modifiable risk factor for impaired osseous healing through vasoconstriction, reduced tissue oxygenation, impaired osteoblast activity, diminished collagen synthesis, and reduced angiogenesis [5,6]. Even when MTP-I-specific studies vary, broader foot-and-ankle fusion data support cessation before elective arthrodesis.

Normal pulses and duplex findings reduced concern for major arterial disease but did not exclude smoking-related microvascular or cellular impairment.

Local biology and joint preparation

Fusion requires cartilage and sclerotic bone removal, viable cancellous surfaces, and stable contact. Cup-and-cone preparation preserves bone stock and assists alignment, whereas planar cuts may improve cancellous apposition and compression [7,8,9,10]. In revision, the specific preparation method is less important than complete debridement to bleeding bone. In this case, excision of fibrous tissue and sclerosis converted an inactive pseudarthrosis into a viable graft bed.

Fixation and compression in revision arthrodesis

Stable fixation and compression are central to MTP-I fusion. Evidence generally supports constructs combining compression with dorsal plating, particularly in revisions or biologically compromised patients [11,12]. The revised construct used an interfragmentary compression screw and stronger dorsal plate to compress the graft interfaces and resist bending during rollover.

Structural grafting: Restoring biology, bone stock and length

Revision often involves bone loss, osteolysis, shortening and sclerosis. When compression would shorten the first ray, structural grafting is required. Tricortical iliac crest autograft provides support and osteogenic, osteoconductive and osteoinductive properties [13]. CT-confirmed incorporation and bridging trabeculation showed that the graft served both structural and biological roles.

Forefoot biomechanics and second-ray deformity

MTP-I arthrodesis should be assessed within the whole forefoot. Adjacent lesser-ray deformity may alter rollover mechanics and plantar pressure, although direct causality cannot be proven from a single case. Correcting the second-ray deformity was intended to improve balance and reduce abnormal loading across the revised first ray.

Post-operative protection

Although early protected weight-bearing may be safe after selected primary fusions [14,15], this evidence should not be applied uncritically to high-risk revisions. Smoking history, pseudarthrosis, bone loss, and structural grafting justified 8 weeks of non-weight-bearing before gradual loading.

Proposed revision framework

On the basis of the present case and the available literature, failed MTP-I arthrodesis should be assessed using a structured framework. The surgeon should determine whether failure is symptomatic, confirm non-union with CT when radiographs are equivocal, exclude infection, evaluate host biology, assess bone loss and first-ray length, analyze fixation and alignment, identify adjacent-ray deformity, and individualize the post-operative protocol. Revision should be designed to restore both biology and mechanics (Table 2).

Comparison with published revision literature

Literature on failed MTP-I arthrodesis is limited. Takács and Swierstra used inlay grafting for pseudoarthrosis, supporting biological augmentation and stable graft-host contact [16]. Larger revision cohorts by Waizy et al. and Gaudin et al. show that revision is clinically important and carries higher complication risk than primary fusion [4,17].

Saccomanno and Bitterman treated infected non-union with staged revision, antibiotic spacer placement, and tricortical iliac crest autograft [18]. Their report supports structural grafting but differs because infection required staging. In the present aseptic case, normal inflammatory markers, absence of purulence, and negative cultures supported single-stage revision.

Malunion is another failure mode. Hirao et al. used CT-based planning and a custom guide for malunited pronation deformity after MTP-I arthrodesis, emphasizing three-dimensional alignment assessment [19]. In this case, suspected excessive dorsiflexion prompted careful sagittal restoration during grafting and fixation.

For recalcitrant non-union with hostile biology, Kaiser and Levin reported medial femoral condyle free-flap reconstruction [20]. By contrast, this case united with non-vascularized tricortical iliac crest autograft because infection was excluded, macrovascular supply was preserved, bone loss was reconstructable and stable compression fixation was achievable.

Salvage arthrodesis after failed MTP-I arthroplasty also informs bone-loss reconstruction. Mao et al., Yurteri et al., Johnson et al. and Adamson et al. highlight strategies including structural grafting, custom implants and length maintenance when bone stock is compromised [21,22,23,24] (Table 3).

This case was distinct because aseptic failed primary arthrodesis occurred in a heavy smoker with CT-confirmed pseudarthrosis, osteolysis, and plate fatigue and second-ray deformity. Union was achieved without staged infection treatment, vascularized bone transfer, or arthroplasty conversion by combining debridement to viable bone, structural autograft, compression, enhanced fixation, adjacent deformity correction, smoking cessation, and prolonged protection.

Revision series support a whole-forefoot, stepwise escalation approach (Table 4). Gaudin et al. identified revision indications beyond hardware problems, including non-union, malunion, metatarsalgia, claw-toe deformity, and interphalangeal disorders [17]. Takács, Swierstra, and Malhotra et al. support graft-based repair and length-preserving bone-block arthrodesis [16,25], whereas Kaiser and Levin represent escalation to vascularized bone for rare hostile non-unions [20] (Table 4).

Conflict of Interest:

Source of Support:

Nil

Consent:

The authors confirm that informed consent was obtained from the patient for publication of this article