Bilateral Medial Medullary Syndrome Following Anterior Odontoid Screw Fixation for a Type II Fracture: A Case Report

Presentation and initial imaging

A 28-year-old man presented with neck pain after a road traffic accident. Neurological examination at admission was entirely normal. Computed tomography (CT) of the cervical spine demonstrated a posteriorly displaced type II odontoid fracture together with a fracture of the C2 spinous process, and surgery was planned for later the same day.

His vascular risk factors and co-morbidity profile are summarized in Table 1.

Severe atherosclerosis (grade IV plaques in the bilateral vertebral and basilar arteries) was identified on CT angiography performed as part of the etiological work-up.

Perioperative course

During induction of anesthesia, the patient developed anisocoria with left-sided mydriasis. Anesthesia was stopped and urgent CT angiography was performed, which showed no vascular abnormality – in particular, the vertebral artery was normal – and no change in fracture position. Once he had woken, the neurological examination was again normal and the pupillary asymmetry had resolved. Surgery was therefore postponed, and halo-ring traction (5 kg) was applied to allow gradual, controlled reduction of the fracture over the following 3 days.

Definitive fixation was carried out on day 3, at the completion of the halo-ring traction period. The operative technique is detailed below.

The patient was positioned supine on a radiolucent table with the neck in neutral alignment. General anesthesia was induced with orotracheal intubation under fiber-optic guidance, and manual in-line stabilization was maintained throughout. Halo-ring traction (5 kg) was applied, and gentle longitudinal traction with slight neck extension achieved anatomical reduction, confirmed on lateral and open-mouth fluoroscopy. A right-sided Smith–Robinson anterior cervical approach was used at the C1–C2 level; the prevertebral fascia was opened and the longus colli retracted laterally to expose the C2 body. Two 3.5 mm partially threaded cannulated cancellous screws (DePuy Synthes) were inserted anterior-to-posteriorly along the axis of the odontoid under biplanar fluoroscopic guidance, with entry points in the inferior half of the C2 body as a paramedian pair on either side of the midline, directed superiorly and posteriorly to the tip of the odontoid. Screw positions were confirmed on anteroposterior and lateral fluoroscopy, with no canal breach. Intraoperative neuromonitoring (somatosensory evoked potential and motor evoked potential) remained stable throughout. The operative time was 110 min, with an estimated blood loss of <50 mL and no intraoperative complications. Closure was performed in standard layered fashion with a suction drain.

Postoperative course and neurological deterioration

Weaning from ventilation was prolonged and complicated. CT of the head and cervical spine confirmed satisfactory fracture reduction and correct screw positions, with unremarkable brain parenchyma. Because the clinical course remained unexplained, cerebral and cervical magnetic resonance imaging (MRI) was obtained and revealed acute bilateral ischemia of the medial medulla. Tracheostomy and percutaneous endoscopic gastrostomy were subsequently performed, after which weaning was achieved successfully.

Neurological findings

Examination showed a bilateral incomplete tetraparesis, predominantly left-sided, with pyramidal signs. Cranial nerve examination was otherwise normal apart from deviation of the tongue to the right, in keeping with a right-sided hypoglossal nerve palsy. Formal sensory testing (light touch and pinprick) and anorectal examination were performed once the patient was extubated; voluntary anal contraction and deep anal pressure were both preserved. Detailed International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) motor and sensory scores, neurological level, and ASIA grade at postoperative day 2 (the time of worst neurological deficit) and at 6-month follow-up are summarized in Table 2.

Clinical timeline

The overall sequence of injury, surgery, deterioration, and recovery is summarized in Table 3.

Imaging findings

Preoperative CT confirmed the displaced type II odontoid fracture, and postoperative CT confirmed satisfactory reduction with correctly positioned screws (Fig. 1a and b). MRI obtained after the failed weaning attempts showed restricted diffusion with a high diffusion-weighted imaging (DWI) signal and corresponding high T2 signal in the anterior medulla, consistent with acute bilateral medial medullary ischemia (Fig. 2). The infarcted area lay close to both the surgical field and the vertebral artery.

Treatment and etiological work-up

Enoxaparin (Clexane) 1 mg/kg subcutaneously twice daily was started on postoperative day 2, and aspirin 150 mg once daily was added on postoperative day 3 once a thrombotic vascular mechanism was considered likely; no unfractionated heparin was used at any stage. The decision to institute this antithrombotic regimen was made after weighing the risk of extending the medullary infarct against the risk of epidural or wound hemorrhage so soon after cervical instrumentation; a prophylactic-dose regimen was chosen for this reason, and no hemorrhagic transformation was seen on subsequent imaging. Neurorehabilitation began promptly, and the tracheostomy was successfully decannulated three months after the injury. Screening for an underlying cause was otherwise unremarkable except for severe atherosclerotic disease, with grade IV atheromatous plaques identified in the bilateral vertebral and basilar arteries on CT angiography (Table 1). Given the patient’s young age, dedicated screening for causes of premature vasculopathy – including familial hypercholesterolemia and other inherited dyslipidemias – was undertaken as part of this etiological work-up; no such disorder was identified, and the atherosclerotic risk-factor profile summarized in Table 1 remained the only identifiable contributor.

Outcome

He was discharged from acute care to inpatient rehabilitation 3 weeks after injury, remaining ASIA C at that point, wheelchair-bound, tube-fed, and requiring intermittent catheterization, with recurrent aspiration pneumonia during this early period. His ISNCSCI grade had improved to ASIA D by 3 months, at which stage he mobilized with a walker and tolerated a soft diet following tracheostomy decannulation. By 6-month follow-up, he was ambulating short distances with a single cane, tolerating a normal diet, voiding volitionally, and had regained independent bowel function, although mild bilateral distal weakness (hand and toe) persisted (Table 2). Detailed functional, mobility, and continence outcomes at each time point are summarized in Table 4.

Differential diagnosis

The transient anisocoria and mydriasis during induction initially raised concern for an occult vertebral or basilar artery injury; however, CT angiography at the time excluded dissection or occlusion of the basilar artery, and the pupillary asymmetry had resolved by the time the patient woke, making it unlikely to be mechanistically linked to the medullary infarction that followed. Isolated basilar artery occlusion was also considered, but this typically produces a broader pattern of cranial nerve, cerebellar, and consciousness disturbance that was not seen here. A demyelinating or inflammatory process was thought unlikely given the acute, vascular-territory-restricted appearance on DWI and the absence of any preceding systemic illness. The bilateral, paramedian distribution of the lesion, together with its close temporal relationship to surgery, was most consistent with a perioperative ischemic event.

Proposed mechanism

The etiology in this patient remained uncertain, and the only vascular abnormality identified on systematic work-up was advanced, grade IV atherosclerosis of the bilateral vertebral and basilar arteries, identified on CT angiography performed as part of the etiological work-up (Table 1). Given the anatomical proximity of the medullary infarct to both the surgical field and the vertebral artery (Fig. 3), we consider it plausible that perioperative mechanical manipulation of the upper cervical spine – during intubation, positioning, fracture reduction, or sustained traction – contributed to atherothrombotic occlusion of small perforating branches of the vertebral or anterior spinal artery in a vessel already compromised by atherosclerotic disease. Experimental work on cervical manipulation in atherosclerotic animal models has shown that rotational and manipulative forces can destabilize plaque and reduce arterial tensile properties ^[17,18], and manipulation of the neck has separately been linked to stroke in patients with underlying vascular disease ^[19]. Perioperative ischemic stroke is itself a recognized, if uncommon, complication of surgery unrelated to the cervical spine ^[10,11], which lends further plausibility to a perioperative vascular mechanism in this case.

The specific biomechanical steps used in this case merit closer consideration. Sustained halo-ring traction combined with slight neck extension during fracture reduction and positioning is mechanistically distinct from the head rotation implicated in most manipulation-related vertebral artery injury: Extension preferentially loads the V3 segment of the vertebral artery as it curves around the lateral mass of the atlas, and a systematic review of craniocervical positioning found that combined maximal extension and rotation, more than either movement alone, was associated with a measurable reduction in vertebral artery hemodynamics ^[20]. Isolated case reports have similarly documented reversible positional occlusion of the vertebral artery in extension ^[21], and intraoperative traction, even within conventional limits, has been implicated in vertebral artery dissection during cervical spine surgery ^[22]. This combination of sustained axial traction and extension differs from posterior cervical procedures, in which the neck is generally maintained in a neutral or flexed position and reported vertebral artery injuries are more commonly attributed to direct instrumentation – lateral mass or pedicle screw malposition – than to traction- or extension-related hemodynamic compromise ^[23]. In a vessel already narrowed by grade IV atherosclerotic plaque, even a modest, sub-critical reduction in flow of the kind described with extension may have been sufficient to precipitate thrombosis of small perforating branches, whereas the same maneuver in a normal artery would be unlikely to have hemodynamic consequences. This distinction has practical relevance for surgical teams managing trauma in vascularly compromised patients: where feasible, minimizing the duration and degree of neck extension during traction and positioning, and maintaining a low threshold to reassess vertebral artery patency if traction is prolonged, may help reduce the risk of this rare but serious complication.The clinical picture – bilateral, left-predominant limb weakness with pyramidal signs and a unilateral (right-sided) hypoglossal palsy – is consistent with bilateral medial medullary infarction, in which the corticospinal tracts on both sides are affected, producing quadriparesis that was more marked on the left, while hypoglossal fascicular involvement can remain asymmetric depending on the precise distribution of the paramedian lesions ^[13]. Bilateral disease, as seen here, has previously been reported almost exclusively with dissection, atherosclerosis, or vasculitis as the underlying cause ^[12,14], rather than as a perioperative event, which is what makes this presentation unusual.

Literature comparison

Table 5 compares the clinical, radiological, and treatment features of this case with previously reported cases of bilateral medial medullary infarction.

Limitations

This report has several limitations. It describes a single patient, and the proposed link between cervical manipulation and medullary infarction is inferential rather than proven; no dedicated imaging of the vertebral artery or its perforating branches was obtained at the presumed time of infarction, so a focal dissection or occlusion at that level cannot be entirely excluded. The relative contribution of each perioperative maneuver – intubation, positioning, fracture reduction, or sustained traction – cannot be separated out from this single case, and longer-term functional and swallowing outcomes beyond 6 months are not available.

Conflict of Interest:

Nil

Source of Support:

Nil

Consent:

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

How to Cite this Article

Raj S, Kumar D, Kumar V, Sen AK, Kumar SR, Goni VG. Bilateral Medial Medullary Syndrome Following Anterior Odontoid Screw Fixation for a Type II Fracture: A Case Report. Journal of Orthopaedic Case Reports 2026 October;16(10): 252-259.