Multi-parametric MRI for the evaluation of cranial nerves dysfunction in hematological malignancies

Abstract Background In hematological malignancies which include a heterogeneous group of neoplastic lymphoid or myeloid disease, cranial nerve dysfunction (CND) is not uncommon and can occur secondary to a wide range of potential etiologies. CND may be due to infiltration, infection, and therapeutic complications with broad spectrum of clinical manifestation. Aim of work Magnetic resonance imaging (MRI) is essential for evaluating cranial nerves. Our aim was to investigate multi-parametric MRI approach for the evaluation of cranial nerve dysfunction, to differentiate between possible etiologies and illustrate the potential diagnostic challenges in the context of hematological malignancies. We aimed to differentiate between cranial nerve infiltration and non-infiltrative causes, thereby avoiding unnecessary radiation or chemotherapy. Methods Our retrospective study included 30 patients (13 females and 17 males) presented with CND into hematology–oncology clinic from February 2019 to March 2022. The patients have been categorized into three groups: infiltration, infection, and treatment-induced cranial neuropathy, according to the pathology results, clinic–laboratory findings, response to treatment or by exclusion. The MR imaging criteria included cranial nerve involvement, solitary versus multiple, unilateral versus bilateral, size of the nerve, signal intensity on T1, T2, FLAIR, and diffusion-weighted imaging, and pattern of enhancement. The imaging features correlated to each category. Results The cohort included 15 patients with leukemia, 14 patients with lymphoma, and only one patient with juvenile xanthogranuloma. Bilateral cranial nerves involvement was observed in 46.7% ( n = 14) with the optic nerve being the most frequently affected. The infiltration group was the most prevalent and represented 66.7% ( n = 20). Eleven patients (55%) presented with isolated cranial nerve involvement while the remaining 45% had perineural spread along the course of the cranial nerve from a remote focal mass. Optic nerve was the most common involved cranial nerve 63% ( n = 19), followed by trigeminal nerve 40% ( n = 12). Diffusion restriction along with heterogeneous pattern of enhancement was evident in all cases of neoplastic infiltration. Conclusion Imaging criteria from multiple MR sequences are helpful in the evaluation of CND in patients with hematological malignancies which would promptly guide patient management and minimize unnecessary and risky biopsies. Diffusion-weighted imaging (DWI) appeared to be a good imaging marker, as diffusion restriction was exclusively appreciated in the malignant infiltrative group of CND.