A rare case of secondary paroxysmal hemicrania caused by a T1 nerve root schwannoma

We describe a case of secondary paroxysmal hemicrania due to a T1 nerve root schwannoma, reinforcing sympathetic dysfunction as a key pathophysiological mechanism in paroxysmal hemicrania.

Abstract

Introduction:

Paroxysmal hemicrania (PH) is a rare headache disorder of severe intensity, which is classified within the group of trigeminal autonomic cephalalgias and is renowned for its prompt response to indomethacin.

Objectives:

Since its first description in 1974, only few cases of secondary PH have been published and even fewer with distinct lesions and a clear pathophysiological association.

Methods:

We here present a unique case of long-standing secondary PH, which resolved permanently after resection of a schwannoma of the T1 nerve root, which carries sympathetic projections to the periorbital region.

Results:

The current case stresses the importance of sympathetic/parasympathetic dysregulation in the pathophysiological origination of PH, which is far beyond purely central mechanisms. It is the first case ever published, which shows a quite remote peripheral pathology causing a trigeminal autonomic headache disorder.

Conclusion:

This case has important implications on a better understanding of the pathophysiology and future diagnostic workup of PH, which should ideally include imaging not only of the brain but also of the spine down to the thoracic level.

1. Introduction

Paroxysmal hemicrania (PH) is a rare headache disorder of severe intensity, which is classified within the group of trigeminal autonomic cephalalgias (TAC) and is renowned for its prompt response to indomethacin. This distinguishes PH along with the related hemicrania continua from other TAC like cluster headache or short-lasting unilateral headache. Epidemiological data indicate a first onset typically around the fourth decade of life and shows a slight female predominance. Much has been speculated about its pathophysiology, and there is currently intense discussion as to whether central, peripheral, or neurovascular mechanisms are primarily responsible.² Secondary symptomatic cases may help to improve our understanding of disease development and pathophysiology. But since its first description in 1974 by Sjaastad and Dale,²⁰ only few cases of secondary PH have been published and even fewer with distinct lesions and a clear pathophysiological association.

2. Methods

We here present a unique case of long-standing secondary PH, which resolved permanently after resection of a schwannoma of the T1 nerve root, which carries sympathetic projections to the periorbital region.

3. Results

The male patient, a neuropediatrician, first sought medical advice in 2013 at age 64 years with ongoing bouts of headache since 2011. The symptoms initially occurred in clusters of active episodes and had striking peculiar features: The pain was of excruciating intensity (8–10 on a 10-point rating scale), strictly bound to the left periorbital and temporal region and accompanied by tearing and miosis of the left eye. Duration of attacks was exactly 10 minutes, followed by an interictal phase of exactly 90 minutes until the next attack appeared. During the active episodes, this pattern continued for 2 to 10 weeks, meaning he had a severe headache attack every 90 minutes all day and night. In the beginning he noticed a circannual accumulation with more active episodes during spring and fall. An active episode began with constant painless tearing of the left eye for 2 days and then the headache attacks set in. The headache showed a dramatic response to indomethacin, stopping the pattern of bouts abruptly and preventing new headache attacks on a dosage of 50 mg daily. Initial diagnostic workup showed no focal-neurological deficits. An MRI of the brain and cervical spine depicted no lesions, which could explain the new onset of headaches. The patient was diagnosed with primary episodic PH based on the International Classification of Headache Disorders (ICHD) criteria.¹⁶ He was advised to take indomethacin daily. The patient was able to stop the medication after the active phase had ended, delaying the intake of the drug every 24 hours. If no headache appeared, the episode had stopped but reoccurred reliably after a few weeks. Over the next years, attacks appeared regularly with increasing frequency and were documented in detail in a digital headache diary along with blood pressure and heart rate by the patient (Fig. 1A). During attacks, these recordings indicated reproducibly a significant relative decrease of the heart rate and a slight significant increase of systolic blood pressure (Fig. 1B). Since 2020, the patient was treated for elevated blood pressure with lercanidipine (5–10 mg/d), candesartan (4–16 mg/d), and metoprolol (47.5 mg/d). Otherwise, there was no indication of cardiac disorders.

Figure 1.

Fluctuations of vital signs in relation to PH symptom trajectory. (A) Long-term tracking of vital signs by a health care app before and after surgical removal of the T1 nerve root schwannoma. (B) Statistical comparison of systolic and diastolic blood pressure and heart rate in the attack-free interval (pale colors) and attacks (bright colors). *P < 0.05 (paired t test). bp, blood pressure; PH, paroxysmal hemicrania.

In 2022, the frequency of active headache episodes further increased, and the patient had a new onset of bilateral back pain and some tingling of the legs. A contrast-enhanced T1-weighted MRI of the cervical and thoracic spine depicted an extramedullary tumor with a dimension of 2 × 1 cm at T1 level, which locally compressed the spinal cord and affected the nerve roots T1 (Fig. 2A). There were no clinical signs of a myelopathy, ie, no paresis, proprioceptive, gait, or bladder dysfunction.

Figure 2.

Magnetic resonance imaging of T1 schwannoma and schematic overview of sympathetic pathways potentially involved in PH. (A) Axial and transversal T1-weighted MRI sections (turbo-spin-echo sequence with application of contrast agent) depicting a contrast-enhancing tumor with a dimension of 2 × 1 cm at T1 level. (B) Schematic drawing of the central and peripheral sympathetic networks projecting to the periorbital region, potentially involved in the pathophysiology of secondary paroxysmal hemicrania. IML, intermediolateral nucleus; L, left; PVN, paraventricular nucleus; R, right; RVLM, rostral ventrolateral medulla; SG, stellate ganglion; SCG, superior cervical ganglion; T1, nerve root T1.

Surgical removal of the tumor was performed in March 2023 to treat the enduring back pain and prevent further spinal cord compression in case of tumor progression. Intraoperative inspection showed an attachment of the tumor to the nerve roots T1. The tumor could be removed with a partial preservation of the nerve root. Pathological analysis confirmed the diagnosis of a schwannoma grade 1. Interestingly, the patient perceived a sudden complete cessation of his headache from the day of the surgery. He was able to stop the indomethacin treatment completely. On a 24-month follow-up, no further headache attacks appeared. A reevaluation of the patient case led to the diagnosis of a secondary PH provoked by the schwannoma of the nerve root T1.

4. Discussion

How could a tumor at this level cause the picture of PH, and what could we possibly learn from this case for the general pathophysiology of PH?

It is well known that the nerve root T1 carries sympathetic fibers, which travel with the sympathetic trunk to the stellate ganglion and superior cervical ganglion. From there, the tertiary fibers reach the forehead and periorbital region, where they branch with fibers from the trigeminal nerve. Further fibers travel to the heart to contribute to its sympathetic innervation and rise along the external carotid artery to reach meningeal vessels.²⁵ An affection of the T1 nerve root by tumor growth or compression, therefore, could well impair the sympathetic system. The clinical signs in our case with tearing and miosis of the left eye speak for a transient imbalance of the parasympathetic and sympathetic input during symptomatic phases. This view is reinforced by a well-documented decrease of heart rate within the attacks. This decrease in sympathetic tone matched the initial descriptions by Sjaastad et al.²¹ who discussed a sympathetic deficit as a decisive pathophysiological feature of PH, as the autonomic symptoms during the attack and interictal findings pointed to a transient functional sympathetic deficit or sympathetic/parasympathetic imbalance. In line, Drummond⁸ suggested a sympathetic deficit and parasympathetic overactivation during active episodes in chronic PH. This hypothesis was also confirmed in other TAC. A reduction of sympathetic tone would cause a relative parasympathetic predominance and therefore an ipsilateral vasodilation of the meningeal vessels, which could contribute to the manifestation of hemicrania through an activation of trigeminal nociceptive fibers, as proposed also for the pathophysiology of TAC.¹¹ This theory is facilitated by one case report, where the onset of secondary PH was associated with the intake of vasodilative drugs.²³ The proposed action of indomethacin in our case could be through its comparably potent vasoconstrictive potential in combination with its anti-inflammatory action.¹ The gradual increase of the attack frequency over a decade fits with the suspected slow growth of a benign schwannoma. The instantaneous and long-term enduring cessation of headaches after its removal strongly suggests a causal relation. In the literature, very few cases of secondary PH exist, of which only 3 cases completely fulfilled the ICHD-3 criteria and provided adequate MR imaging.^12,19,24 There are 12 more lesional cases, which do not fully cover the ICHD-3 criteria mostly because indomethacin was not applied (“probable PH”)^{3–5,7,9,10,13–15,17,18,22,26} (Supplementary Table 1, http://links.lww.com/PR9/A382).

Until now, it was presumed that all these lesions with a PH phenotype cannot be aligned to a certain cerebral region or network, making it difficult to establish a clear pathophysiological relation to the headache.² However, most of the reported lesions affect central or peripheral sympathetic networks, including the pituitary-hypothalamic region (ie, the paraventricular nucleus), the brainstem (ie, the nucleus of the solitary tract, caudal, and rostral ventrolateral medulla), and the sympathetic trunk. The current case—being the first with a distinct focal lesion at the crossroad of central and peripheral sympathetic networks at T1 level—shows clearly that a rather remote sympathetic lesion can lead to a dysfunction of autonomic networks through the mechanisms described above.

The slight circannual rhythmicity described by the patient could be explained by the well-known circannual oscillation of the autonomic system in higher mammals and humans,⁶ which may contribute to a vulnerability because of physiological seasonal changes, making it more susceptible to damage from tumor associated pressure or other affections.

Some specifics of the current case may impact its generalizability to other cases of PH: Lercanidipine, candesartan, and metoprolol may have affected heart rate and blood pressure data because they inhibit the sympathetic effects on cardiovascular regulation. However, the relative decrease in heart rate during attacks cannot be explained sufficiently by drug effects only. It cannot be excluded that the tumor removal caused some indirect effects on pain dynamics including placebo effects. The patient did not expect an effect of the surgery on PH based on pretreatment consultation though.

Overall, the current case reinforces previous hypotheses on the contribution of the autonomic nerve system to the pathophysiology of PH (Fig. 2B). To our knowledge, it is the first to show that a pathology at T1 level can cause a trigeminal autonomic headache disorder. Consequently, diagnosis of exclusion in PH should ideally include imaging of the brain, the spinal cord to the thoracic level, and the mediastinal and neck region, especially if the demographic features of the patients do not align with the “typical” epidemiology of primary PH such as female preponderance and mid age.

Disclosures

The authors have no conflict of interest to declare.