Radiation-Induced Third Cranial Nerve Palsy With the Longest Interval Between Radiation Therapy and Onset of Cranial Neuropathy

A 72-year-old woman noticed progressive ptosis and binocular diplopia for the past 3 months. She had a known history of stage-4 breast cancer diagnosed 3 years ago with metastases to the spine, pleura, and mediastinal lymph nodes for which she has been undergoing chemotherapy with nanoparticle albumin-bound paclitaxel. She also had a remote history of pituitary macroadenoma diagnosed 23 years ago and treated with transsphenoidal excision, followed by radiation therapy (RT) (3-field technique with 50 Gy in 25 fractions). On examination, she had complete right upper lid ptosis, 2 mm of anisocoria (right pupil larger than left), and complete limitation of adduction, supraduction, and inferoduction of the right eye in keeping with the diagnosis of complete right pupil-involving cranial nerve 3 (CN3) palsy (CN3P). Intorsion of the right eye was intact, and there were no signs of aberrant regeneration. MRI and time-of-flight MRA of the brain performed 2 weeks ago elsewhere was reported as normal with exception of unchanged pituitary gland enlargement, stable compared with imaging performed 3 years ago. When imaging was re-examined, striking bilateral enhancement of the cisternal portion of both CN3 was seen that was not present on the last available postcontrast MRI images from 3 years ago. Repeat MRI of brain and orbits with gadolinium and steady-state imaging 2 weeks later again showed marked smooth enhancement and mild thickening of cisternal segments of both oculomotor nerves but showed no other areas of leptomeningeal or intraparenchymal enhancement (Fig. 1A–C). Two large volume lumbar punctures (LPs) 1 month apart showed normal cerebrospinal fluid (CSF) composition (both with no white blood cells and normal protein and glucose), and cytological analysis was negative for malignancy on both occasions. Patient continued to do well systemically, although CN3P remained unchanged. MRI repeated 6 months later demonstrated unchanged thickening and enhancement of cisternal portion of both oculomotor nerves and unchanged pituitary gland enlargement with no evidence of leptomeningeal enhancement elsewhere (Fig. 1D, E).FIG. 1.: T2 axial SPACE sequence (A) shows the location of the oculomotor nerves (arrows). T1 axial spin echo sequences before (B) and after (C) intravenous contrast injection show diffuse smooth enhancement of the cisternal segments of the oculomotor nerves (arrows). T1 coronal sequence after intravenous contrast injection (D) shows persisting enhancement of cisternal segments of both oculomotor nerves (arrows) and T2 axial FLAIR sequence (E) shows hyperintensity of cisternal segments of both oculomotor nerves (arrows).Initially, carcinomatous meningitis was believed to be the cause of clinical 3NP and bilateral enhancement of cisternal portions of CN3s on neuro-imaging; thus high-volume LPs were performed looking for presence of malignant cells in the CSF. Although CSF cytology has poor sensitivity for detecting central nervous system (CNS) spread of solid malignancies, it does increase with subsequent LPs. Third LP was planned but patient declined further investigations. The working diagnosis was CNS spread of cancer causing seeding of CN3s. However, the patient continued to be stable clinically and had no new neurological symptoms. When MRI remained unchanged 6 months later, we concluded that the diagnosis must be postradiation cranial neuropathy (PRCN) because prognosis in patients with carcinomatous meningitis is uniformly poor, whereas our patient continued to be well with no changes on neuro-imaging. Diplopia and cranial neuropathy in patients with a history of cancer should always raise suspicion of metastases or tumor recurrence. Neuromyotonia is another rare entity that may present with episodic diplopia in patients who have undergone cranial radiotherapy. Although aneurysmal compression should be ruled out first in cases of 3NP, balanced steady-state gradient echo MRI sequences should be performed in all patients with a history of cancer presenting with cranial nerve palsy (1). Enhancement of the oculomotor nerve always indicates underlying pathology but does not always correlate with a clinically apparent CN3P (2). The differential diagnosis for enhancing CN3 consists of inflammatory entities (demyelinating if involving the fascicle of the nerve and radiation-induced and ophthalmoplegic migraine if involving cisternal portion), infiltrative or neoplastic lesions (primary CN3 schwannoma, lymphoma, leukemia, or infiltration by perineural spread or carcinomatous meningitis presenting with irregular thickening along with enhancement) as well as infectious etiologies (1–3). One case series reviewed 13 patients with enhancement of cisternal portion of 3CN on MRI; in 6, enhancement was bilateral and of these 4 had unilateral CN3P clinically, like the patient in our case, with remaining 2 not showing any abnormalities of ocular motility. The main causes of enhancement were hematological malignancies (5/13), infections (2/13), and nonspecific inflammatory entities (6/13) (2). None of these patients had PRCN. As radiation is commonly used for treatment of skull base and nasopharyngeal tumors, there is a lot of literature on PRCN involving lower (IX–XII) cranial nerves, reporting its incidence of ∼5% with 5–7 years being an average time from RT to onset of cranial neuropathy (4). Literature on PRCN involving oculomotor (third, fourth, and sixth) cranial nerves is scanty, however, because of the anatomical proximity of sella to cisternal and cavernous portions of oculomotor nerves; RT for treatment of sellar tumors can produce PRCN involving these nerves. In one recent retrospective case series of 6 patients who have undergone transsphenoidal resection followed by RT for pituitary adenoma, one patient had third, 4 had sixth, and one had both sixth and fourth cranial nerve palsies (5). The time between RT and onset of cranial neuropathy was anywhere between 4 and 96 months, the longest interval between RT and onset of cranial neuropathy to our knowledge to date (5). Although it is well established that the risk for PRCN is related to a radiation dose (with 20–68 Gy range being the total dose previously described), peculiarly, interval between RT and onset of PRCN is inversely related to a dosage with patients receiving a lower dosage having later onset of cranial neuropathy thus emphasizing the need for a long-term follow-up of these patients (5,6). It is also possible that patients receiving chemotherapy might be at an increased risk of developing PRCN. PRCN less commonly affects oculomotor nerves compared with the more sensitive anterior visual pathway (optic nerves and chiasm) (3,7). This is likely because oculomotor cranial nerves are myelin free and their function depends on the integrity of more radioresistant Schwann cells compared with the radiosensitive oligodendrocytes that myelinate the optic nerves and chiasm (5). Although exact mechanism for PRCN is unknown, histologically fibrosis and Wallerian degeneration are seen in cranial nerves after RT (4,5). Progressive soft tissue fibrosis of the areas adjacent to the nerve causing delayed injury to the vasa nervosum has also been implicated (4). It is not clear why in PRCN there is often bilateral enhancement of cranial nerves yet clinically only one nerve is involved, this could be related to the threshold of accumulated damage to the axons before the function of the nerve is impaired where the impairment is enough to produce a nerve palsy on one side and spare the function on the other. In summary, we present a unique case with the longest reported interval between administration of RT and onset of PRCN that reminds us that PRCN should always be on the differential diagnosis of all patients with new onset of cranial neuropathy and history of RT. It also emphasizes importance of obtaining high-resolution balanced steady-state gradient echo MRI sequences in all patients who do not fit a typical profile for microvascular cranial nerve palsy and do not demonstrate improvement in ocular motility 2–3 months after onset of symptoms. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: E. Margolin, D. Mandell, T. Jeeva-Patel, T. Le; b. Acquisition of data: E. Margolin, D. Mandell, and T. Jeeva-Patel; c. Analysis and interpretation of data: E. Margolin, D. Mandell, T. Jeeva-Patel, and T. Le. Category 2: a. Drafting the manuscript: E. Margolin, D. Mandell, T. Jeeva-Patel, and T. Le; b. Revising it for intellectual content: E. Margolin, D. Mandell, T. Jeeva-Patel, and T. Le. Category 3: a. Final approval of the completed manuscript: E. Margolin, D. Mandell, T. Jeeva-Patel, and T. Le.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.