A case report of a transient splenial lesion related to HaNDL syndrome

Abstract

Background

Transient lesions in the splenium of the corpus callosum have been identified in many clinical cases, and often correspond to a metabolic insult to the brain. The syndrome of transient headache and neurological deficits with cerebrospinal fluid lymphocytosis (HaNDL syndrome) is a rare but under-recognised headache syndrome.

Case

A 47-year-old man presented to our hospital with a 2-week history of intermittent headache, and acute right sided hemisensory deficit. A CSF lymphocytosis was found and a diagnosis of HaNDL was made. A lesion in the splenium of the corpus callosum was identified on MRI. CSF lymphocytosis and the splenial lesion resolved on follow up 4 weeks later.

Conclusion

These two entities are uncommon but increasingly recognised. The co-incidence in this patient raises the possibility of similar underlying pathological mechanisms, including vasomotor changes in blood vessels, cortical spreading depression and glutamate excitotoxicity leading to intra-myelinic oedema. Awareness of these entities will allow prompt diagnosis, preventing unnecessary tests and treatment, and allow appropriate patient management.

Introduction

The syndrome of headache and neurological deficits with cerebrospinal fluid lymphocytosis (HaNDL) is a rare yet increasingly described entity since it was first reported in 1981 (1). It occurs in otherwise healthy individuals without a history of cerebrovascular disease or migraine.

With widespread use of MRI, lesions in the splenium of the corpus callosum are increasingly recognised. Importantly, the lesion resolves within weeks, implying only oedema rather than axonal damage. These lesions have been described in numerous clinical scenarios. The underlying pathophysiology is poorly understood, and may vary between its myriad inciting factors.

We present a case that involved both these two entities and allows us to speculate on the nature of the pathology that links the two. This is only the second case in the literature of HaNDL described with a transient splenial lesion (2).

Case history

An otherwise well 47-year-old Caucasian male flight attendant was brought to our hospital with headache, speech disturbance and right sided sensory symptoms. He had a history of flu-like illness three weeks prior to presentation on holiday in Vanuatu. Two weeks prior to presentation, he had experienced the onset of holocephalic bifrontal/occipital pressure-type headache with gradual onset over minutes. Then, 10 days prior to this presentation, he was assessed at another hospital with similar symptoms of confusion, bilateral hand paraesthesia and facial numbness with the headache. He was discharged following a normal CT brain and resolution of headache. In the intervening period, he had experienced ongoing intermittent headache with right facial numbness and paraesthesia, and a general sense of cognitive slowing and vagueness.

This particular episode was characterised by the sudden onset of right-sided paraesthesia of the face and arm more than leg, associated with some speech difficulty which was described as effortful and hesitant. He had minor right-sided motoric symptoms, including minimal facial weakness and a tendency to lean to the right whilst walking. A bifrontal headache had preceded the paraesthesia and evolved over minutes reaching a severity of 8/10. The weakness and speech difficulty lasted approximately 20 minutes and were witnessed by paramedics. These had resolved by the time of hospital presentation, but the headache persisted.

The patient was not on regular or anti-inflammatory medication, and had only a history of mild asthma. He had not been exposed to any unwell contacts. Prior to this, the patient had not experienced any prior similar episodes, nor suffered from migraine, though his father suffered from migraines without aura.

On arrival at our hospital, his only neurological deficits were paraesthesia to the right cheek (preserved objective sensation) and mild bifrontal headache. He had no meningism. He underwent acute stroke protocol imaging with CT brain, CT cerebral angiogram and CT perfusion. This demonstrated delayed perfusion (time to peak) of the left occipito-parietal region, not conforming to a vascular territory.

The following day, his headache persisted, but there was no recurrence of other features. Electroencephalography (EEG) was normal. Cerebrospinal fluid (CSF) analysis demonstrated a lymphocyte pleocytosis with 89 mononuclear cells per mm³ with mildly elevated protein of 0.87 g/L. No viral cause was identified (PCR testing of HSV 1/2, CMV, VZV, enterovirus, adenovirus). Matched serum and CSF oligoclonal bands were seen. Blood tests showed normal electrolytes and were negative for inflammatory markers, autoantibodies and HIV serology. An MRI was performed that showed a solitary, non-enhancing ovoid T2 hyperintense midline splenial lesion (Figure 1) with associated diffusion restriction. No vascular abnormality or established left hemisphere infarction was identified. The patient was discharged and received a follow-up contrast MRI brain and spine 4 weeks later that demonstrated minimal residual T2 FLAIR hyperintensity in the splenium with resolution of diffusion restriction; the spine was normal. He underwent further CSF testing, which revealed five mononuclear cells, normal protein (0.4 g/L) and no oligoclonal bands.

Figure 1.

MRI demonstrating splenial lesion. (a) Coronal diffusion weighted image (DWI); (b) axial DWI; (c) sagittal T2 fluid attenuated inversion recovery (FLAIR); (d) axial T2 FLAIR.

Discussion

Our patient presented with recurrent episodes of headache with associated focal sensorimotor symptoms. Though his motor symptoms did not persist for more than 4 hours, the character of the presentation with exclusion of other causes led us to a diagnosis of HaNDL syndrome (3). This syndrome is characterised by unilateral sensory (75% of cases) and/or motor (50% of cases) deficits arising concurrently or shortly after headache. It may rarely mimic a migrainous aura by preceding the headache onset. Speech disturbance including aphasia is also common (65%), when the dominant hemisphere is affected. Visual symptoms are rare. Some 25% will have a viral prodrome and a post-viral autoimmune aetiology is postulated. The diagnosis of HaNDL relies on the presence of CSF lymphocytosis. Additionally, CSF protein is elevated in 90%, and an elevated CSF pressure is seen in more than 50% (3). The condition is generally benign and there is no specific treatment required other than symptomatic headache relief. Advising the patient regarding likely recurrent episodes is important, as well as reassurance that the condition likely resolves within weeks. They should also be counselled regarding a change in character of the headache, since symptomatic intracranial hypertension has rarely been described during the recovery phase (4).

EEG, CT or MRI perfusion, transcranial doppler and SPECT can all provide information in HaNDL cases. EEG shows focal slowing in 86% of cases. Multimodal MRI and SPECT can show decreased blood flow in brain areas in 93% (5).

Hypoperfusion has been commonly demonstrated in HaNDL, and often crosses vascular boundaries in a similar fashion to the hypoperfusion of migrainous cortical spreading depression (6). As in our case, if the clinician is not considering a diagnosis of HaNDL, some subtle perfusion changes can be dismissed as artefact.

Transient lesions of the corpus callosum, otherwise known as Reversible Splenial Lesion Syndrome (RESLES), have been described in myriad clinical scenarios but particularly related to seizures, antiepileptic medication toxicity or withdrawal, metabolic conditions such as hypoglycaemia or hyponatraemia, and as a sequela of infection (either of the CSF or systemically) (7). They are characterised by a midline/off-midline ovoid T2/T2 FLAIR hyperintensity with associated diffusion restriction on DWI/ADC and a lack of enhancement with gadolinium. Radiological appearances are consistent with cytotoxic oedema. Recently, the term “cytotoxic lesions of the corpus callosum” (CLOCCs) has been proposed to reflect this pathology. Importantly, the lesion often resolves within weeks and does not reflect permanent demyelination, axonal loss or infarction. It is thought to be caused by intra-myelinic oedema and inflammatory infiltrate (and IL-1, IL-6 cytokines) related to oxidative stress or glutamate excitotoxicity (8). It is thought that the splenium is susceptible due to the high density of oligodendrocytes expressing glutamate receptors (9).

RESLES often manifests with encephalopathy symptoms, and sometimes hemisensory or motor deficits (10,11). In fact, some authors prefer the name “Mild encephalopathy/encephalitis with reversible splenial lesion” (MERS) (7). This raises an intriguing link with some manifestations of HaNDL syndrome, since confusion/encephalopathy can occur with HaNDL, and could indicate that there may be overlap in the pathophysiology of these conditions. Case series have identified patients with CSF pleocytosis, confusion and headache; however, none met HaNDL criteria (7,12). To our knowledge, there has been only one other case described of a transient splenial lesion being related to HaNDL syndrome. The clinical features reported are similar to our patient (2).

In consideration of our case and the one of Raets, HaNDL syndrome can be added to the varied conditions that can lead to transient splenial lesions. Given such varied aetiologies, a transient splenial lesion is not symptomatic per se, but rather a radiological signal of metabolic disturbance. The varied aetiologies imply a common final process causing the splenial lesion, and yet there are clear differences between the causes, principally a recognised pathogen in the case of infective encephalitis whereas HaNDL (by definition) lacks a known pathogen and is thought perhaps to be a post-viral autoimmune phenomenon. How this relates to the pathophysiology of HaNDL remains unclear. There is nothing distinctive regarding the characteristics of our case (or the case of Raets) that might pre-dispose to a splenial lesion. Indeed, it is unclear why they occur in other conditions, in some patients but not others. We would propose that symptoms of HaNDL relate to a monophasic autoimmune (possibly post-viral) vasoactive antibody that induces vasomotor changes followed by cortical spreading depression leading to deficits across vascular territories due to oxidative stress and glutamate excitotoxicity. It is likely that the splenial lesion relates to some metabolic disturbance leading to intramyelinic oedema, perhaps related to the metabolic stress induced by a wave of cortical spreading depression and hypoperfusion, yet without any ischaemia leading to infarction.

Conclusion

We present the second ever reported case that has linked a case of HaNDL syndrome with that of a transient splenial lesion. Both entities are rare, but becoming increasingly recognised, the latter particularly with the frequent use of MRI. By making this connection, it adds an extra piece to the puzzle of both their underlying pathophysiology, and emphasises the diffuse metabolic disturbance that bouts of HaNDL induce in the brain, despite only focal perfusion changes and clinical manifestations. Perfusion changes crossing vascular boundaries may prompt the clinician to consider a diagnosis of HaNDL, and recognition may prevent unnecessary investigation or treatment.

Clinical implications

Recognition of HaNDL syndrome will minimise unnecessary investigations and provide appropriate patient counselling.

Transient lesions of the splenium of the corpus callosum can occur in myriad clinical scenarios.

Their co-occurrence indicates shared pathophysiology, which may include vasomotor changes, cortical spreading depression and glutamate excitotoxicity.