Acute transverse myelitis as a manifestation of lyme neuroborreliosis: two cases

Abstract

Lyme neuroborreliosis (LNB) is a recognized manifestation of Borrelia burgdorferi infection, although spinal cord involvement is rare and may pose a diagnostic challenge. We report two female patients presenting with subacute myelitis due to LNB. The first patient developed a focal cervical spinal cord lesion, while the second presented with a longitudinally extensive thoracic lesion.

In both cases, cerebrospinal fluid (CSF) analysis showed lymphocytic pleocytosis and intrathecal synthesis of Borrelia burgdorferi-specific antibodies. Corticosteroid therapy was initially administered prior to diagnosis due to suspected inflammatory myelitis. Following microbiological confirmation of LNB, both patients were treated with intravenous ceftriaxone, resulting in marked clinical improvement and regression of radiological findings.

One patient later developed recurrent neurological symptoms with MRI signal changes at the same spinal level, without evidence of new lesions.

These cases highlight the variability of spinal cord involvement in LNB and the importance of considering infectious etiologies in patients presenting with myelitis.

Introduction

Lyme borreliosis is the most common tick-borne disease in Europe and a significant cause of neurological morbidity [1], [2]. Nervous system involvement occurs in approximately 10–15% of untreated patients and includes meningitis, cranial neuropathy, and radiculoneuritis [3], [4].

Endemic areas, including parts of Central and Eastern Europe such as Croatia, represent regions of increased exposure risk [1], [2].

Spinal cord involvement in the form of transverse myelitis is rare, with fewer than 50–70 cases reported to date [5], [6], [7], [8], [9].

More recent cohort and epidemiological studies suggest that spinal cord involvement remains an uncommon manifestation of Lyme neuroborreliosis, whereas cranial neuropathies and radicular syndromes account for the majority of neurological presentations in adults [10], [11], [12].

Due to its nonspecific clinical and radiological presentation, it may be misdiagnosed as autoimmune or demyelinating disease [7], [13].

Case 1

A 76-year-old woman presented with subacute progressive lower limb weakness, gait disturbance, and urinary retention. She did not recall a recent tick bite.

Neurological examination revealed predominantly left-sided lower limb weakness and impaired coordination.

Table 1.

Comparison of clinical and radiological features with previously reported cases.

Feature	Literature	Case 1	Case 2
Number of reported cases	Approximately 50–70 reported cases	–	–
Age / sex	Variable	76 / F	55 / F
Clinical presentation	Subacute myelitis	Paraparesis, urinary retention	Sensory deficit, monoplegia, bilateral facial palsy
Spinal involvement	Cervical or thoracic	Cervical (C3/C4)	Thoracic (Th5–Th8)
Lesion type	Focal or longitudinally extensive	Focal	Longitudinally extensive
Cranial nerve involvement	Occasionally present	No	Yes
MRI findings	T2 hyperintensity	Small focal lesion	Extensive intramedullary lesion
Contrast enhancement	Variable	Absent	Present (subtle)
CSF findings	Pleocytosis, intrathecal antibodies	Present	Present
CSF protein	Elevated	1.35 g/L	2.1 g/L
Initial steroid therapy	Occasionally	Yes	Yes
Antibiotic response	Favorable	Favorable	Favorable
Radiological evolution	Regression	Partial regression (1 month)	Marked regression (9 months)
Long-term course	Usually monophasic	Monophasic	Later worsening without new lesions

MRI of the cervical spine demonstrated a focal T2-hyperintense intramedullary lesion at the C3–C4 level (Fig. 1A), without post-contrast enhancement on contrast-enhanced imaging (Fig. 1B).

Fig. 1.

Case 1 – Cervical spinal cord involvement. (A) Sagittal T2-weighted MRI of the cervical spine demonstrating a focal intramedullary T2-hyperintense lesion at the C3/C4 level (arrow). (B) Sagittal post-contrast T1-weighted image showing absence of post-contrast enhancement at the corresponding level. (C) Follow-up sagittal T2-weighted MRI performed after one month showing partial regression of the lesion at the same level (arrow).

Cerebrospinal fluid (CSF) analysis revealed lymphocytic pleocytosis (202 cells/µL, 86% lymphocytes) and elevated protein levels (1.35 g/L). Intrathecal synthesis of Borrelia burgdorferi-specific antibodies was confirmed (IgG antibody index ≈41), with positive serum IgG and negative IgM serology.

Additional infectious and neurological differential diagnostic evaluation included testing for tick-borne encephalitis virus, CMV, EBV, HSV−1/2, VZV, HIV, hepatitis viruses, Toxoplasma gondii, syphilis (VDRL/TPHA), routine CSF bacteriology, and neuronal/paraneoplastic antibodies; all results were negative.

Due to suspected inflammatory myelitis, corticosteroid therapy was initially administered prior to establishing the diagnosis. Following microbiological confirmation of LNB, treatment was switched to intravenous ceftriaxone for 28 days.

The clinical course was complicated by urinary retention and urinary tract infection, as well as transient antibiotic-associated neutropenia.

Follow-up MRI performed after one month demonstrated partial regression of the lesion (Fig. 1C).

The patient showed marked clinical improvement, with residual mild gait disturbance but no subsequent relapse.

Case 2

A 55-year-old woman presented with a two-month history of progressive neurological symptoms. Initial symptoms included pain in the hips and lower limbs, followed by thoracic dermatomal sensory disturbances.

She subsequently developed bilateral peripheral facial nerve palsy and weakness of the left lower limb.

Brain MRI with contrast demonstrated nonspecific supratentorial white matter lesions without a characteristic dissemination pattern together with mild enhancement of the left vestibulocochlear nerve, interpreted as compatible with neuroborreliosis in the clinical context.

MRI of the thoracic spine demonstrated a longitudinally extensive intramedullary lesion extending from Th5 to Th8 with subtle post-contrast enhancement (Fig. 2A–C).

Fig. 2.

Case 2 – Thoracic spinal cord involvement. (A) Sagittal STIR MRI of the thoracic spine demonstrating a longitudinally extensive intramedullary STIR-hyperintense lesion involving the Th5–Th8 segments (arrow). (B) Axial T2-weighted image at the thoracic level demonstrating involvement of a large portion of the spinal cord cross-section (arrow). (C) Axial post-contrast T1-weighted image demonstrating subtle focal enhancement of the thoracic intramedullary lesion (arrow). (D) Follow-up sagittal T2-weighted MRI performed nine months later showing marked regression of the lesion (arrow).

CSF analysis revealed lymphocytic pleocytosis (62 cells/µL, 78% lymphocytes) and elevated protein levels (2.1 g/L). Intrathecal synthesis of Borrelia burgdorferi-specific antibodies was confirmed (IgG index 18.13), with positive serum IgM and IgG serology confirmed by Western blot.

Additional infectious, immunological, and neurological differential diagnostic evaluation included testing for CMV, EBV, Chlamydia spp., Mycoplasma spp., Toxoplasma gondii, HIV, hepatitis B and C viruses, HSV−1/2, tick-borne encephalitis virus, syphilis (VDRL/TPHA), anti-ganglioside antibodies, autoimmune serology, neuromuscular antibodies, and onconeuronal/paraneoplastic antibodies; all results were negative.

Prior to diagnosis, corticosteroid therapy was administered with only partial and transient clinical improvement.

Following treatment with intravenous ceftriaxone, the patient experienced significant clinical recovery, including regression of facial nerve palsy and improvement in motor function. However, mild residual sensory deficits persisted.

Follow-up MRI performed nine months later showed marked regression of the lesion (Fig. 2D).

Approximately two years after the initial episode, the patient presented with worsening gait disturbance, paresthesias in the lower limbs, and urinary retention. MRI demonstrated signal changes at the same spinal level, without evidence of new lesions suggestive of dissemination in space. The patient remained under the care of a neurologist and physiatrist for further evaluation, management, and rehabilitation.

Discussion

Spinal cord involvement in LNB is uncommon. Recent cohort data suggest that myelitis remains an uncommon manifestation of neuroborreliosis, whereas cranial neuropathies and radicular syndromes predominate [10], [11], [12]. Clinical presentation is typically subacute, with motor and sensory deficits [6], [7], [8]. Cranial nerve involvement, particularly bilateral facial nerve palsy, is less frequently observed in patients with myelitis and may further complicate the clinical picture [6], [7], [8].

CSF findings in our patients were consistent with previously reported cases of neuroborreliosis-associated myelitis, typically characterized by lymphocytic pleocytosis and elevated protein levels, although the degree of inflammatory response may vary [6], [7], [8].

MRI findings are often subtle and non-specific, particularly in focal cervical involvement, while more extensive lesions may be observed in thoracic disease [6], [9]. This variability may complicate differentiation from demyelinating disorders such as multiple sclerosis and neuromyelitis optica spectrum disorders, as well as other infectious or post-infectious myelitides [6], [9], [14]. In both patients, imaging findings together with the absence of dissemination in space and the response to antibiotic therapy supported an infectious rather than a primary demyelinating etiology.

Importantly, diagnosis was not based on isolated serological positivity but on concordant clinical presentation, inflammatory CSF findings, confirmed intrathecal synthesis of Borrelia-specific antibodies, and structured neurological and infectious differential diagnostic evaluation [15].

A key clinical issue in both cases was the initial administration of corticosteroid therapy prior to diagnosis, reflecting a common diagnostic pitfall. Infectious causes such as LNB should be excluded before initiating immunosuppressive treatment [7], [12].

Both patients demonstrated a favorable response to antibiotic therapy with subsequent radiological regression on follow-up imaging.

Notably, one patient developed later clinical deterioration with MRI signal changes confined to the previously affected spinal segment. Although a subsequent tick bite was reported, early doxycycline treatment makes reinfection-related neuroborreliosis less likely [16]. The absence of new lesions and lack of radiological progression make a primary demyelinating disorder less likely. This pattern may reflect residual or recurrent inflammatory activity within previously affected tissue rather than representing a new disease process [8], [9].

Residual neurological deficits and fluctuating symptoms following neuroborreliosis have been described and may complicate long-term interpretation of disease course [7], [8]. These features highlight the importance of integrating clinical, laboratory, and imaging findings, as well as a multidisciplinary approach in order to establish an accurate diagnosis.

Conclusion

Lyme neuroborreliosis should be considered in patients presenting with myelitis, particularly in endemic regions. Clinical improvement together with radiological regression following antibiotic therapy may support differentiation of LNB from alternative etiologies.

Ethical approval

Not required.

Consent

Informed consent could not be obtained despite reasonable efforts. The manuscript is fully anonymized.

Conflicts of interest

None.

Author contributions

All authors contributed to the manuscript preparation and approved the final version.

CRediT authorship contribution statement

Mari Rončević Filipović: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Data curation, Conceptualization. Đurđica Cekinović Grbeša: Writing – review & editing, Supervision. Domagoj Ćužić: Resources, Investigation.

Funding

None.

Declaration of Generative AI and AI-assisted technologies in the writing process

During the preparation of this manuscript, the authors used ChatGPT (OpenAI) exclusively to assist with language editing, text organization, and manuscript formatting. All scientific content, interpretation of findings, clinical decisions, and final manuscript revisions were performed and verified by the authors, who take full responsibility for the content of the submitted work.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.