Early postoperative cervical spinal cord signal changes following percutaneous cordotomy: a clinically correlated imaging pattern

TO THE EDITOR

Percutaneous cervical cordotomy remains a highly effective palliative procedure for patients with unilateral, opioid-refractory cancer pain [1]. By selectively targeting the anterolateral spinothalamic tract, rapid and substantial analgesia can be achieved [2]. Despite its efficacy, early postoperative neurological symptoms such as dysesthesia and burning sensation are frequently encountered and are commonly regarded as procedural complications [3,4]. However, the structural correlates and clinical significance of these early findings remain insufficiently understood, and transient MRI signal changes may be misinterpreted as spinal cord injury.

The authors report three consecutive patients who underwent CT-guided percutaneous cervical cordotomy at the C1–2 level for severe cancer-related pain. Patient 1 had breast carcinoma with left hip bone metastasis (numeric rating scale [NRS] score 8), Patient 2 had malignant mesothelioma with left-sided thoracic and back pain (NRS score 9), and Patient 3 had sacral chordoma with severe left hip pain (NRS score 10), rendering supine positioning intolerable. Written informed consent for publication of clinical information and imaging data was obtained from all patients included in this case series. All images and clinical information were anonymized prior to publication. The study was conducted in accordance with the principles of the Declaration of Helsinki.

All procedures were performed under continuous axial CT guidance. Patients 1 and 2 received mild sedation with midazolam (1 mg), while Patient 3 required intraoperative fentanyl (100 mcg). After local anesthesia, the subarachnoid space was accessed percutaneously at the C1–2 level. Intrathecal positioning was confirmed, and diluted contrast agent was administered with barbotage to optimize intrathecal distribution and visualization. CT imaging verified contrast spread within the spinal canal.

The needle was advanced into the anterolateral spinal cord, targeting the spinothalamic tract. Three-dimensional CT reconstructions of the needle trajectory were generated from acquired imaging data (Fig. 1), and axial CT images confirmed accurate placement at the C1–2 level (Fig. 2). Functional confirmation was achieved using radiofrequency stimulation, reproducing concordant pain in the affected region. Motor stimulation was negative in all patients. Radiofrequency ablation was then performed at 70°C for 30 seconds using a 2-mm active tip electrode.

Fig. 1.

Procedure localization. Three-dimensional CT reconstruction demonstrating the needle trajectory during percutaneous cervical cordotomy.

Fig. 2.

Needle position (red arrows) at C1–2 level. (A–C) Axial CT images confirming accurate needle placement within the anterolateral spinal cord at the C1–2 level. (A) Patient 1, (B) Patient 2, (C) Patient 3.

Immediately after the procedure, all patients experienced complete contralateral analgesia (NRS 8–10 to 0), consistent with effective interruption of the spinothalamic tract. On postoperative day 1, cervical MRI demonstrated a consistent finding in all patients: focal T2 hyperintensity localized to the right anterolateral hemicord at the C1–2 level (Fig. 3).

Fig. 3.

Early postoperative MRI (Day 1). (A–C) Axial and sagittal T2-weighted MRI images demonstrating focal hyperintensity in the right anterolateral hemicord at the C1–2 level (red arrows). (A) Patient 1, (B) Patient 2, (C) Patient 3.

Clinically, all patients developed transient sensory symptoms on the side of the cordotomy lesion (ipsilateral to the lesion and contralateral to the original pain side) within the first postoperative hours. Patient 1 developed paresthesia and altered body perception approximately 2–4 hours after the procedure, which improved progressively, decreasing by approximately 50% on postoperative day 3 and 70% on Day 7. Patient 2 developed shock-like sensations involving the ipsilateral upper and lower extremities approximately 6 hours after the procedure, severe enough to disturb sleep, with approximately 70% improvement by postoperative day 3 and complete resolution within the first week. Patient 3 developed transient sensory disturbances approximately 3 hours after the procedure, which improved progressively and had completely resolved by the 1-month follow-up visit. The severity of postoperative sensory symptoms was not systematically quantified using a numerical rating scale or other validated assessment tool; therefore, symptom characterization was based on routine clinical documentation and follow-up records. No patient required specific treatment for postoperative sensory symptoms, and all sensory disturbances improved spontaneously during follow-up. Contralateral analgesia was maintained throughout follow-up.

One patient developed contralateral lower extremity weakness during follow-up. This patient had advanced sacral chordoma with documented sacral involvement. The weakness was anatomically inconsistent with the expected effects of a unilateral C1–2 spinothalamic tract lesion, and postoperative MRI demonstrated no evidence of lesion extension toward motor pathways. Therefore, the finding was considered more likely related to the underlying sacral disease than to the cordotomy procedure; however, a procedural contribution cannot be completely excluded.

Follow-up MRI at 1 month demonstrated marked regression of the previously observed signal changes in all patients (Fig. 4). At 3-month clinical follow-up, all patients continued to report complete pain relief (NRS 0), with no recurrence of primary pain symptoms. The demographic characteristics, procedural details, postoperative neurological findings, MRI evolution, and clinical outcomes of the three patients are summarized in Table 1.

Fig. 4.

Follow-up MRI (1 Month). (A–C) Follow-up T2-weighted MRI images demonstrating regression of the previously observed focal T2 hyperintensity (red arrows). (A) Patient 1, (B) Patient 2, (C) Patient 3.

Table 1.

Clinical, imaging, and outcome characteristics of the three patients.

Variable	Patient 1	Patient 2	Patient 3
Age/Sex	60/F	42/M	64/F
Primary malignancy	Breast cancer with bone metastasis	Malignant pleural mesothelioma	Sacral chordoma
Pain location	Left hip and leg	Left thoracic and back pain	Left hip and leg
Baseline NRS	8	9	10
Pain side	Left	Left	Left
Cordotomy side	Right	Right	Right
Immediate analgesia	Complete (NRS score 0)	Complete (NRS score 0)	Complete (NRS score 0)
Postoperative sensory symptom side	Right (ipsilateral to lesion)	Right (ipsilateral to lesion)	Right (ipsilateral to lesion)
Postoperative sensory symptoms	Paresthesia, altered body perception	Shock-like sensations, dissociative symptoms	Dissociative symptoms
Motor findings	None attributable to cordotomy	None	Contralateral dorsiflexion weakness, considered more likely related to sacral disease; procedural contribution could not be excluded
Day-1 MRI	Right anterolateral T2 hyperintensity	Right anterolateral T2 hyperintensity	Right anterolateral T2 hyperintensity
1-month MRI	Marked regression	Marked regression	Marked regression
3-month pain outcome	NRS score 0	NRS score 0	NRS score 0

NRS: numeric rating scale.

This case series demonstrates a consistent and reproducible pattern of early postoperative spinal cord signal change following CT-guided cervical cordotomy, with strong temporal and anatomical correlation with clinical findings. Immediate and sustained pain relief further supports the functional effectiveness of targeted spinothalamic tract ablation.

Early postoperative MRI in all patients revealed focal T2 hyperintensity confined to the anterolateral hemicord at the lesion site. This finding is consistent with previous reports describing the imaging appearance of cordotomy lesions. Vedantam et al. [5] demonstrated similar T2 hyperintense lesions in the anterolateral quadrant following cordotomy; however, their study was limited to early postoperative imaging without longitudinal follow-up. In contrast, the present series demonstrates regression of these signal changes on follow-up MRI, providing direct insight into their temporal evolution.

The temporal relationship between imaging findings and clinical symptoms is particularly noteworthy. All patients developed transient ipsilateral sensory symptoms within the first postoperative hours. These symptoms improved progressively during clinical follow-up and had either resolved completely or substantially improved by the time radiological regression was documented on 1-month MRI. This temporal association suggests that the observed MRI findings represent a reversible treatment-related process. Based on the imaging appearance and clinical course, localized spinal cord edema is considered the most likely explanation; however, transient inflammatory change or the expected evolution of focal tissue injury following radiofrequency lesioning cannot be excluded [6].

The anatomical correlation further strengthens this interpretation. Ipsilateral sensory symptoms reflect local tissue response at the intervention site, whereas contralateral analgesia corresponds to the intended interruption of the spinothalamic tract [7]. This predictable and anatomically coherent pattern supports the possibility of accurate spinothalamic tract targeting and physiological specificity.

From a mechanistic perspective, these findings are consistent with the known effects of radiofrequency lesioning, which produces focal thermal injury sufficient to disrupt nociceptive transmission while inducing transient perilesional edema [8]. It should be emphasized that the proposed mechanism of transient perilesional edema is inferred from the observed imaging pattern and clinical evolution, as no histopathological confirmation or advanced imaging biomarkers were available in the present study. The absence of persistent neurological deficits clearly attributable to the cordotomy procedure and the stability of analgesic outcomes further support the selectivity and safety of this approach when performed under precise imaging and neurophysiological guidance [2].

Importantly, these observations suggest that some early postoperative neurological symptoms and MRI abnormalities may represent expected post-lesion changes rather than progressive injury, although additional studies are required to further define their clinical significance [9]. In the absence of progressive deficits or radiological worsening, focal T2 hyperintensity at the lesion site appears to represent a predictable and transient imaging correlate of effective lesioning [10]. Misinterpretation of these findings may lead to unnecessary concern, inappropriate interventions, and overestimation of complication rates [11].

From a clinical perspective, the findings observed in this series should be interpreted within the context of stable or improving neurological status and subsequent radiological regression. The presence of progressive neurological deficits, enlarging MRI abnormalities, unexpected lesion extension beyond the targeted anterolateral cord, hemorrhagic changes, or radiological progression should not be regarded as postoperative findings observed in the present cases and should prompt further clinical and radiological evaluation. Therefore, the imaging pattern described in this report may represent a postoperative correlate of lesion localization in the appropriate clinical context only when accompanied by an appropriate clinical course and radiological stability or improvement.

Furthermore, early postoperative MRI may provide supportive evidence of lesion localization following cervical cordotomy. The consistent localization of signal change within the anterolateral cord together with sustained clinical analgesia suggests a possible association between imaging findings and successful spinothalamic tract targeting. However, this observation should be considered hypothesis-generating, and larger prospective studies are required to determine whether MRI may have a role in post-procedural validation and outcome assessment following cordotomy [3,5].

Although limited by the small number of cases, the consistency of imaging and clinical findings across all three patients strengthens the validity of these observations. To the authors’ knowledge, this is the first report demonstrating both early postoperative spinal cord signal change and its radiological regression in parallel with clinical symptom improvement or resolution following cervical cordotomy.

In conclusion, early postoperative spinal cord signal change after cervical cordotomy appears to represent a clinically correlated postoperative imaging pattern characterized by the regression of signal changes over time. Recognition of this pattern may improve postoperative assessment and help avoid misinterpretation of treatment-related post-lesion changes.