Article Text

Download PDFPDF
Chronic meningitis, seizures and myoclonus
  1. James Joseph McDonald1,2,
  2. Matthew Jones3,
  3. Colin Smith4,
  4. Peter L Foley1,2,
  5. Liam Lee2,
  6. Richard Davenport1
  1. 1 Department of Clinical Neurosciences, NHS Lothian, Edinburgh, UK
  2. 2 The University of Edinburgh, Edinburgh, UK
  3. 3 Manchester Centre for Clinical Neurosciences, Salford Royal NHS Foundation Trust, Salford, UK
  4. 4 Centre for Clinical Brain Sciences, The University of Edinburgh, Edinburgh, UK
  1. Correspondence to Dr James Joseph McDonald, Department of Clinical Neurosciences, NHS Lothian, Edinburgh EH16 4SA, UK; james.mcdonald{at}


A 70-year-old retired male plumber was admitted with reduced consciousness following a recent onset of confusion and headache. He developed encephalopathy, seizures and myoclonus. He died despite treatment 10 weeks later. This article reports the clinicopathological conference presented at the Edinburgh Clinical Neurology Course 2021.

  • CSF
  • MRI
  • neurooncology
  • myoclonus
  • epilepsy

Data availability statement

No data are available.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Case presentation

A 70-year-old retired male plumber was admitted with reduced consciousness. His family reported that he had been intermittently confused for several weeks, reporting of shooting pains in his head and difficulty with navigation. His medical history included migraine, osteoarthritis, hypertension and insulin-dependent type 2 diabetes. He enjoyed unpasteurised cheese and had recently returned from a camping trip near Brisbane, Australia. He had never smoked and occasionally drank alcohol.

On examination, his Glasgow Coma Scale (GCS) score was E4 V2 M4 with responsive pupils, globally increased tone with upgoing plantar responses. There was non-purposeful movement of all four limbs. He required intubation and ventilation due to reduced conscious level. Table 1 shows his initial investigations.

Table 1

Admission investigations

He remained drowsy despite correction of hypoglycaemia and developed a fever of 38.7°C. Chest X-ray showed patchy change at the right hilum. He was started on antibiotics to cover aspiration. His consciousness improved 24 hours after admission and he was extubated. He remained confused and combative. Antimicrobial cover was broadened to cover central nervous system (CNS) infection (ceftriaxone, amoxicillin and acyclovir) and a lumbar puncture showed a cerebrospinal fluid (CSF) lymphocytosis (table 2).

Table 2

Lumbar puncture results

Antimicrobial therapy was continued. He improved to full alertness, but developed severe nausea, vomiting and dizziness requiring multiple antiemetic medications. He reported intermittent right temporal headache. Examination identified mild bilateral ptosis, globally increased tone and hyperreflexia in the right arm; ankle jerks were absent.

MR scans of the brain and spine (Day 10) showed patchy meningeal enhancement and cerebritis with prominent facial nerve enhancement bilaterally (figure 1). CT scan of the chest identified a 12 mm pulmonary nodule, not amenable to biopsy. Specialist reviews of the imaging by respiratory and radiology concluded that the lesion was unlikely to be a malignancy. Box 1 shows subsequent tests and investigations (including a further lumbar puncture). Positron emission tomography (PET) scanning was planned but he became too unwell to proceed.

Figure 1

MR scan of the brain and spine with contrast (Day 10). Selected post-contrast T1 MR images of the brain and spinal cord. Arrows indicate areas of enhancement. (A) MR T1 post-gadolinium axial section through the brain. Serpiginous enhancement of the leptomeninges over the medial frontal lobes bilaterally. (B) MR T1 post-gadolinium axial section at the level of the internal acoustic meati. The trigeminal nerves enhance bilaterally. (C) MR T1 post-gadolinium axial section showing enhancement of the facial and vestibulocochlear nerves bilaterally. (D) MR T1 post-gadolinium sagittal section through the posterior fossa and cervical spine. Solid arrows—enhancement of the cerebellar nodulus along the inferior margin of the fourth ventricle. Open arrows—leptomeningeal enhancement along the cervical and upper thoracic spinal cord.

Box 1

Summary of subsequent investigations

Normal/negative tests:


  • HIV, VDRL, hepatitis B, hepatitis C.

  • Blood cultures (Day 1, Day 65 and Day 66).

  • Cryptococcal antigen (serum).

  • TB QuantiFERON GOLD.

  • Cerebrospinal fluid (CSF) PCR for HSV 1 and 2/VZV/enterovirus/parechovirus.

  • CSF bacterial PCR profiles for Haemophilus influenza, Neisseria meningitidis and Streptococcus pneumoniae.

  • CSF cultures for mycology and TB.

  • CSF 16sRNA.

  • N. meningitides ctrA gene (CSF).


  • Antinuclear antibody screen.

  • MPO/PR3.

  • Serum ACE.

  • Anti-TPO.

  • NMDA-R Ab, VGKCh Ab (LGI1/CASPR2), voltage-gated calcium channel antibody, anti-neuronal screen (Oxford Panel: Yo, Hu, Ri).


CT of the chest, abdomen and pelvis (Day 12)

Right middle lobe mass with mildly enlarged right hilar and mediastinal nodes. No CT findings typical of sarcoidosis. The right middle lobe nodule well defined and not typical of a primary malignancy.


  • HbA1c—80.

  • LDH 292 (125–220 U/L).

  • Immunoglobulins—low IgM 0.33 g/L (0.35–2.9); C3 1.68 (0.81–1.57).

Three weeks into admission, fluctuating confusion returned and he reported onset of double vision. Examination showed a left third nerve palsy and his uvula deviated to the left. A full skin survey, testicular examination and systemic examination were unremarkable.

On Day 28, he remained confused. He was started on prednisolone 60 mg daily and broad spectrum anti-tuberculous treatments (rifampicin, isoniazid, ethambutol and pyridoxine). The following day he had a tonic-clonic seizure and was subsequently difficult to rouse. He was monitored in the intensive care unit and loaded with antiseizure medications.

An electroencephalogram (EEG) identified bilateral slowing but no epileptic activity. He appeared photophobic and was responding to visual hallucinations. Corticosteroids and anti-tuberculous treatments were continued.

On Day 40, he had episodes of looking up, speech arrest and unresponsiveness. EEG during such an event showed a focal seizure arising from the right hemisphere (figure 2). Antiseizure medications were increased, his conscious level and interaction improved, but he remained intermittently agitated. Isoniazid was withheld due to concerns regarding encephalopathy and prednisolone dose was doubled (120 mg) due to interaction with rifampicin. Prednisolone was then weaned (20 mg per week). We noted a possible clinical response to corticosteroids during this period, with improved conscious level. On Day 60, there was a further decline with intermittent myoclonic jerking of all limbs persisting into sleep. EEG showed no associated cortical discharges. Agitation and confusion persisted although he became more interactive.

Figure 2

Electroencephalogram recording capturing a focal seizure. There is an evolving epileptiform discharge in the right frontotemporal region (F8–T4, T4–T6), spreading to centroparietal (C4–P4) region. It meets Salzburg criterion A (epileptiform discharges at >2.5 Hz for >10 s) for diagnosis of electrographic seizure.

After 10 weeks in hospital, he deteriorated abruptly with a reduced coma score and respiratory failure. CT scan of the head showed ventricular enlargement and evolving communicating hydrocephalus. Chest imaging identified a small segment pulmonary embolism and he was started on anticoagulation. A lumbar puncture was repeated later in the week during a temporary pause in anticoagulation therapy (Day 77).

In the days following, he deteriorated with worsening myoclonus, agitation and tachypnoea. After discussion with family, palliative care was pursued; he died 79 days after admission. The family consented to an autopsy.

Discussant: Dr Matt Jones

Initial thoughts: red herrings and cognitive bias

When I read the case I was not struck with a thunderbolt of enlightenment; I had some thoughts about the diagnosis but I knew I did not have the answer. When I re-read the case, I could not inhibit the desire to spot ‘red herrings’ or pan for ‘nuggets of gold’: We heard the patient was a plumber (could this be mesothelioma?), had type 2 diabetes (think invasive fungal infections), he liked unpasteurised cheese (listeria? brucellosis?) and he had gone a long way from Edinburgh on a camping trip (melioidosis?). Near the outset he developed a possible area postrema syndrome—with intractable vomiting—increasingly associated with neuromyelitis optica spectrum disorder.

However, on reflection it occurred to me that these clues were neither red herrings nor nuggets of gold—they were merely potential cognitive biases. The literature around cognitive bias and diagnostic error has increasingly informed my thinking over recent years, ever since I was introduced to the paper by Vickrey and colleagues, “How neurologists think: A cognitive psychology perspective on missed diagnoses”.1 Box 2 shows some thoughts on applying this to the case.

Box 2

Potential points of cognitive bias—adapted from Vickrey et al1


This aspect triggers overlearned associations and is present when we make prominent upfront statements, eg, “unpasteurised cheese eater who went camping near Brisbane”. We may be closing our mind to a more reasoned path. This is common in clinical medicine.


This describes becoming latched onto a particular diagnosis and not re-evaluating it when more information comes to light.


The tendency to make diagnosis based on the ease of recall to a particular case, perhaps because it was recent or memorable.


Failing to use information regarding base rate probabilities—remember if you hear hooves, think horses, not zebras.

Blind Oobedience

Being over-reliant on test results and investigations, not using clinical reasoning to inform decision-making.

Applying the basics

Neurology can be fascinatingly complex but also gloriously simple if a common-sense approach is taken. We should fall back on the basics of clinical medicine. We need to use a theoretical framework to understand the nervous system; knowing which bits do what, and how they behave when damaged by disease. Patients tell us stories in the form of histories. We can ask additional questions in order to work out what disease they may have and which bit of the nervous system it may be affecting. Then we can use physical examination to test hypotheses and ‘localise the lesion’. This should lead to a syndromic diagnosis, and perhaps a likely cause. We can then target our investigations appropriately to confirm the diagnosis.

The syndrome

This is a 70-year-old man who presents over a few weeks with confusion, headache and falling GCS. Examination finds upper motor neurone signs and he needs intensive care support. He has a low blood sugar but correcting this does not make any immediate difference. There is a fever but only after admission, there is no mention of this before. The reason for this is thought to be aspiration, presumably due to failure to protect his airway. He is extubated within 24 hours but remains confused. The blood tests are unremarkable. A CT scan of the head shows nothing acute or dramatic. So, this sounds like a subacute encephalopathy or a subacute encephalomyelopathy.

Unfortunately, this syndrome has a vast differential diagnosis: infection, inflammation, toxic/metabolic, neoplastic, paraneoplastic or degenerative; all remain possible at this point. On the positive side however, this is a path well-worn, we see patients presenting with this commonly and we know what to do: we cover for infection, get some spinal fluid for analysis and obtain better imaging (MR imaging with gadolinium). That is exactly what happens in this case.

Refining the diagnosis—investigations and clinical course

CSF was obtained on Day 1. There’s no opening pressure recorded but there is raised protein and a reduced glucose ratio. There are too many white cells in the CSF, mostly lymphocytes. Nothing grows on culture and cytology is negative.

Reviewing the imaging, there is no evidence of diffusion restriction, pointing us away from stroke or abscesses. The T2 MR imaging was unremarkable except from some deep periventricular white matter change congruent with his diabetes and vascular risk factors. The fluid attenuated inversion recovery (FLAIR) scanning is unremarkable; this is interesting—there is no sign of an exudative meningeal effusion (which we would expect to suppress incompletely on FLAIR). Pre-contrast T1 was also unremarkable except for a degree of atrophy, which is not immediately relevant. The real abnormality comes with contrast administration when we see widespread leptomeningeal enhancement present throughout the CNS with prominent enhancement of the cranial nerves. None of the enhancement appears to be nodular.

Thereafter we are given a clinical update: he becomes more alert and has severe nausea and vomiting. For what it is worth, the scan shows contrast enhancement over the area postrema. The patient develops bilateral ptosis and he continues to demonstrate upper motor neurone signs—hypertonia and hypereflexia—but with absent ankle jerks.

So by applying the above information, we can try to fine tune the diagnostic hypothesis. The imaging and CSF tell us that this man’s leptomeninges are prominently involved. There is a lack of striking parenchymal involvement on the imaging but the signs tell us that the brain must be involved. We are not told if the absent ankle jerks are a new sign or pre-existing. If this is new, it may suggest more extensive disease, involving lower motor neurone structures. In the context of the case, it is perhaps more likely to be nerve root involvement than peripheral nerve.

Our clinical syndrome at this point becomes a bit of a mouthful: a subacute meningoencephalomyeloradiculopathy. This is not terribly useful, so better to re-frame it. What this tells us is that it is a diffuse disease of the nervous system, which is coating it from top to tail. Despite this, there is not much going on outside of the nervous system: there is a lack of systemic symptoms and signs, both in the prodrome and since admission to hospital. This leads to the more sensible syndrome of diffuse leptomeningeal-based disease.

We can now narrow down our list of potential causes. The imaging and CSF picture are not compatible with a toxic or metabolic insult. The case looks less like paraneoplasia and it does not fit with a degenerative process. We are, however, still left with a broad list of potential causes.

Clinically, after a couple of weeks, things appear to be better. A further lumbar puncture is performed (Day 16); the protein is further elevated and there is ongoing inflammation on cytology, however the white cell count has improved. It is difficult to judge if this CSF picture is really an improvement—it remains very abnormal. Recognising that the situation evolved is important, but it is not clear if this is due to the natural history of the disease or treatment effect. If it is a treatment effect, we should consider what treatments he has had. He was initially treated for aspiration pneumonia and then CNS infection was covered, both for viral and bacterial pathogens. Applying the local guidelines, he would have received a few days of amoxicillin and metronidazole followed by ceftriaxone, amoxicillin and aciclovir. It is unclear if he received dexamethasone. While this would have been prescribed for typical bacterial meningitis, this case is far from typical.

Unfortunately, by Day 20, the encephalopathy deepens and he develops double vision with cranial nerve palsies. A detailed systemic examination does not reveal any further clues. Overall things appear worse. We can refine our syndromic diagnosis to a diffuse leptomeningeal-based disease causing subacute, then fluctuating meningoencephalitis. I felt unclear at this point as to the cause of his fluctuation. Could it be a treatment effect? Was medication stopped and then he worsened?

On Day 28, his clinical state remains unchanged. CSF examination is repeated and looks worse (table 2). At this point, anti-tuberculous treatment and prednisolone were started. The next day he deteriorates and develops seizures. He was loaded with antiseizure medications. Things carried on like this until Day 40, when he developed focal-onset seizures and his antiseizure medications are adjusted. A repeat EEG shows a right hemisphere focus—the EEG fulfils criteria for electrographic seizures. Antituberculous medications are altered due to concerns around isoniazid and steroids doubled due to a concern about interaction with rifampicin. Prednisolone is then tapered and he has a short-lived improvement. But by Day 60, he is agitated, confused and has myoclonus. By Day 63, he has a falling GCS and has communicating hydrocephalus on imaging. He has a pulmonary embolism. Overall, he sounds a lot worse. One final CSF analysis is performed on Day 77 at which point the opening pressure has come down a bit, protein is still very high. The glucose is, for the first time, a normal ratio, but the white cells are up (44) mostly lymphocytes. There is a final deterioration, palliative treatments ensue before he finally succumbs on Day 79.

So that completes the clinical course, what about other investigations? Blood results show diabetes. His lactate dehydrogenase and immunoglobulins are slightly out of range but I am not sure the significance of these in someone who is very sick, so I ignored them. His immunology results were normal or negative (which is reassuring as they test for conditions which do not really fit the clinical syndrome). Lots of infection tests, including HIV, are normal or negative. These included tests for tuberculosis, cryptococcus, Neisseria and any bacterial genomic material (16sRNA). He has a right middle lobe nodule on CT scan. This sounds potentially worrisome, but we are told its judged innocuous.

It is traditional in the clinico-pathological conference to consider what is ‘missing’. We do not have neuroimaging from later time points. There is no serum electrophoresis or free light chains. In the CSF some of the white cell differential counts are unclassified, there are no oligoclonal band results, there is no tuberculosis PCR or advanced stains looking for fungi. The patient was too unwell for FDG PET (fluorodeoxyglucose positron emission tomography). There is no meningeal biopsy despite abnormal tissue seen on MR imaging, I suspect he was too unwell for this.

Taking a step back

It was while trawling through the extensive investigations that I realised it is easy with complicated cases to get caught up in the minutiae. I find it helpful to step back and look at the overall journey. This patient was in hospital for months and looking at his overall course, he seemed to be on a worsening trajectory, notwithstanding a few small improvements along the way. This relentless worsening must be telling us something about the disease (figure 3).

Figure 3

Taking a step back: clinical state of the patient over time.

A final refinement

I settled on a syndrome of diffuse leptomeningeal-based disease, causing subacute then progressive meningoencephalitis and eventually hydrocephalus. It was time to reflect on the causes, which can be classified broadly under the headings infection, inflammation or tumour.

Considering infection in more detail, we can exclude common causes such as Neisseria meningitidis and Streptococcus pneumoniae, as well as standard viruses (herpes, enterovirus, etc) as they do not fit the time course. Listeria spp and melioidosis remain in the differential—interestingly, they came up in the ‘red herring’ section—but I felt them to be unlikely. TB remains important and was clearly a major concern for the clinical team. Brucella might cause this syndrome (and was also a ‘red herring’), but you would be unlucky to catch Brucella in Edinburgh or Australia. Cryptococcus is possible but unlikely; there was no severe immunosuppression, no raised CSF opening pressure and no classic exudate seen on the MR imaging. Could it be Angiostrongylus cantonensis, a case of rat lung worm meningitis? I suspect not—this is a southeast Asian bug and I worry about that because I have seen a memorable case in the UK: this is ‘availability bias’. There are, however, problems with this being an infection. There is not much fever for someone with chronic infection and there are repeatedly negative cultures and PCR. There is limited, if any, response to antibiotics.

As for inflammation, possibilities include sarcoid, other granulomatous disease or rheumatoid. Granulomatous diseases other than sarcoid tend to cause more pachymeningeal involvement and in rheumatoid we would expect more in the way of systemic symptoms and signs, making these relatively unlikely. Sarcoidosis remained on my differential although I was surprised by the lack of response to corticosteroids. The lack of systemic features might be considered unusual, but a small proportion of sarcoid appears isolated to the nervous system at presentation.2

Finally, tumour: Diffuse leptomeningeal glioneuronal tumour is typically described in children, however occurs occasionally in adults. I feel it is unlikely to affect a patient of this age. Could this be leptomeningeal carcinomatosis? There is a lack of systemic features and no obvious primary tumour but that does not rule it out.

So, narrowing this down to a top three, I would consider infection (TB), inflammation (neurosarcoidosis) or tumour (leptomeningeal carcinomatosis).

The final analysis

This is a disease that diffusely covers the nervous system. While initially there was little parenchymal involvement, it emerged eventually. In terms of clinical course, it was a relentlessly progressive process over months, with no clear treatment response despite the best efforts of the team. I think this sounds like a cancer.

Final diagnosis

Leptomeningeal carcinomatosis of unclear primary.

For the primary tumour I considered lung cancer, but given the benign looking nodule seen on CT scan I could not be sure. Haematological malignancy can present in this way but there was limited paraclinical evidence for this.

Pathology: Professor Colin Smith

At postmortem examination the brain was retained and detailed neuropathological examination undertaken. There were no obvious macroscopic abnormalities, but histological examination revealed a cellular infiltrate in the leptomeninges and in a perivascular distribution within the brain parenchyma (figure 4A,B). The cells showed significant pleomorphism, consistent with a neoplastic infiltrate, and in some cells a dark pigment consistent with melanin was seen (figure 4C).

Figure 4

Histological appearances of the infiltrating tumour identified postmortem. A dense cellular infiltrate was seen in the leptomeninges, filling the subarachnoid space (A: H&E ×5) and extending along Virchow-Robin spaces into the brain parenchyma with a perivascular distribution (B: H&E ×10; arrow highlighting a vessel surrounded by pleomorphic cells). The cells showed marked nuclear pleomorphism and melanin pigment was seen as brown dots within the cytoplasm of some cells (arrows) (C: H&E ×20). Immunohistochemical staining for HMB-45 highlighted a subset of cells (arrows) within the malignant infiltrate confirming the diagnosis of malignant melanoma (D: HMB-45 immunohistochemistry ×20).

A range of immunohistochemistry was undertaken which identified expression of HMB-45 within a subset of neoplastic cells (figure 4D), HMB-45 being an antibody which predominantly reacts with melanomas, although it can also been seen in other rare tumours showing evidence of melanogenesis, such as clear cell sarcoma. In this case the cytology and immunoprofile were in keeping with malignant melanoma.

While no primary site was identified, the lung nodule was assessed histologically and was found to be a melanoma metastasis with localised infiltration. The primary site can be difficult to find with metastatic melanoma, in this case the brain pathology was considered to be metastatic. A primary dural origin of malignant melanoma was considered but felt to be unlikely in the absence of a primary dural mass.

Final pathological diagnosis

Malignant meningitis, malignant epithelioid melanoma (no primary site identified but additional lung metastasis with localised infiltration).

Malignant meningitis: Dr James McDonald

Malignant meningitis is caused by infiltration of neoplastic cells into the subarachnoid space. These cells can wreak havoc by depletion of nutrients (glucose) within the CSF, by direct irritation and infiltration of the brain itself or by disruption to the normal fluid dynamics within the CSF.

Malignant meningitis is reported to be present in 4%–15% of the patients with solid tumours, the so called carcinomatous meningitis3 and between 5%–15% of the patients with haematological malignancies—leukaemic or lymphomatous meningitis.4 Meningitis may also be seen with primary CNS tumours although this is relatively rare.

Of the different adenocarcinomas, melanoma accounts for most malignant meningitis cases (22%–46%) followed by small cell lung cancer (10%–25%) and breast.3 Due to the prevalence of breast cancer in the background population, meningitis caused by breast adenocarcinoma is most commonly seen. Within this group, young age, triple negative status and the presence of brain metastasis are noted risk factors.4

Of the protean presentations of malignant meningitis, features related to raised intracranial pressure are common. Headache, nausea and vomiting and changes in mental state are common followed by gait abnormalities and cranial neuropathies.5 Other presenting features such as radiculopathy and endocrine effects (syndrome of inappropriate antidiuretic hormone release) are also reported.5

Most patients present with malignant meningitis in the context of a known disseminated malignancy or after a period of disease-free survival. This may, as demonstrated, be the presenting feature of an occult malignancy in around 5%–10% of cases.4

Diagnostic lumbar puncture remains the standard for diagnosis: series from the literature report non-specific abnormality in the CSF in around 90% of the cases, with positive findings of malignant cells on cytology in around two-thirds of the cases.5 Large volume CSF sampling is important to achieve optimal diagnostic sensitivity with best practice being reported as taking >10 mL.6 Should initial CSF sampling be negative, there is value in repeating the lumbar puncture; however the optimal number of lumbar punctures in the face of negative samples remains a matter of discussion, European Association of Neur0-Oncology (EANO) guidelines suggest limiting this to two in the face of cytology negative CSF.5

Imaging also plays a limited role in the investigation of malignant meningitis. MR scan of the neuroaxis may show characteristic patterns of leptomeningeal and nerve root enhancement as well as identifying potential targets for biopsy. Complications, such as hydrocephalus may also be identified. Clearly, body imaging protocols have a role in the search for the primary tumour.

The goal of treatment in malignant meningitis is to prolong life, improve quality of life and arrest neurological deterioration.7 Therapeutic options are dictated by the underlying biology of the neoplasm. Targeted systemic approaches based on the tumour genetics (such as tyrosine kinase inhibition) are preferred where available however untargeted chemotherapies are available both for intrathecal and systemic administration.7 8

The use of checkpoint inhibitor immunotherapy is now part of the standard care for many malignancies and has shown favourable results for some forms of CNS metastasis, including melanoma.8 Focal radiotherapy has a role for the targeting of bulky, symptomatic deposits but there is a lack of randomised controlled trial evidence to show efficacy.5 Should there be extensive nodular disease then whole brain radiotherapy can be considered, this approach is favoured in those with concomitant solid brain metastasis5 and has been shown to be useful for cranial neuropathies and hydrocephalus.7 In cases where hydrocephalus develops, shunt placement may be employed.5

The prognosis remains guarded even with optimal therapy. Without targeted therapies directed at the underlying tumour, median survival is quoted as between 6 and 8 weeks. Survival improves somewhat with the addition of targeted treatments but depends ultimately on the underlying tumour biology (breast 1.75–4.5 months, lung 3–6 months and melanoma 1.7–2.5 months).5

Key points

  • Breaking down presentations into syndromic diagnosis can help to narrow the differential diagnosis.

  • Malignant meningitis may have protean manifestations and may be the first manifestation of cancer.

  • Imaging and cerebrospinal fluid analysis in malignant meningitis are helpful in making the diagnosis but may not themselves be diagnostic.

  • Outcomes for malignant meningitis remain poor even with treatment.

Further reading

  1. Taillibert S, Chamberlain MC. Leptomeningeal metastasis. 2018. 169–204. doi:10.1016/B978-0-12-811161-1.00013-X

  2. Le Rhun E, Weller M, Brandsma D, et al. EANO–ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up of patients with leptomeningeal metastasis from solid tumours. Ann Oncol 2017;28:iv84–99. doi:10.1093/annonc/mdx221

  3. Milburn-McNulty P, Michael B, Moxham N, et al. How to do it: How to get the most out of cerebrospinal fluid cytology. Pract Neurol 2012;12:241–3. doi:10.1136/practneurol-2012–000264

Data availability statement

No data are available.

Ethics statements

Patient consent for publication

Ethics approval

Not applicable.


We would like to extend our gratitude to the family of the late patient who have consented that the details of the case be used for educational purposes. Thanks also to Professor Andrew Farrall, Neuroradiologist (University of Edinburgh/NHS Lothian) for help with the selection and reporting of MRI images for publication and to Dr Fahad Shaikh (Consultant Neurophysiologist) for input with neurophysiology.



  • Contributors JJM: initial draft, manuscript revisions and literature review. MJ: case discussant and manuscript revision. CS: provision of pathology images and manuscript revision. PLF: manuscript revision and clinical care of patient. LL: provision of clinical data tables and figures and manuscript revisions. RD: overall concept and manuscript revisions.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed by Michael Halmagyi, Sydney, Australia.