Primary leptomeningeal lymphoma masquerading as infectious tubercular meningitis

  1. Salini Sumangala 1,
  2. Thidar Htwe 2,
  3. Yousuf Ansari 3 , 4 and
  4. Lidia Martinez- Alvarez 3 , 5
  1. 1 Department of Neurology, Queen Elizabeth Hospital Birmingham, Birmingham, UK
  2. 2 Department of Respiratory Medicine, Sandwell and West Birmingham Hospitals NHS Trust, Birmingham, UK
  3. 3 Department of Ophthalmology, Queen Elizabeth Hospital Birmingham, Birmingham, UK
  4. 4 Department of Ophthalmology, Birmingham and Midland Eye Centre, Birmingham, UK
  5. 5 Department of Ophthalmology, Sandwell and West Birmingham Hospitals NHS Trust, Birmingham, UK
  1. Correspondence to Dr Thidar Htwe; Thidar.htwe@nhs.net

Publication history

Accepted:02 Sep 2021
First published:13 Sep 2021
Online issue publication:13 Sep 2021

Case reports

Case reports are not necessarily evidence-based in the same way that the other content on BMJ Best Practice is. They should not be relied on to guide clinical practice. Please check the date of publication.

Abstract

Primary central nervous system lymphoma (PCNSL) is infrequent and often poses diagnostic conundrums due to its protean manifestations. We present the case of a South Asian young man presenting with raised intracranial pressure and a lymphocytic cerebrospinal fluid (CSF) with pronounced hypoglycorrhachia. Progression of the neuro-ophthalmic signs while on early stages of antitubercular treatment led to additional investigations that produced a final diagnosis of primary leptomeningeal lymphoma. Treatment with chemoimmunotherapy (methotrexate, cytarabine, thiotepa and rituximab (MATRix)) achieved full radiological remission followed by successful autologous transplant. This case highlights the difficulties and diagnostic dilemmas when PCNSL presents as a chronic meningeal infiltrative process. While contextually this CSF is most often indicative of central nervous system tuberculosis and justifies empirical treatment initiation alone, it is essential to include differential diagnoses in the investigation work-up, which also carry poor prognosis without timely treatment. High suspicion, multidisciplinary collaboration and appropriate CSF analysis were the key for a correct diagnosis.

Background

Primary central nervous system lymphoma (PCNSL) is a rare extra nodal non-Hodgkin’s lymphoma accounting for 3%–4% of newly diagnosed central nervous system (CNS) tumours. As the name implies, it is confined to the brain, spinal cord and eyes.1 PCNSL poses frequent diagnostic challenges due to its location, varied and non-specific clinical manifestations and lack of diagnostic tests with a high sensitivity. Clinical presentations reported in the literature include impairment of higher functions, features of raised intracranial pressure, focal neurological deficit, seizures and headache.2 Visual loss is a predominant feature in intraocular (retinal) lymphoma; however, this can also occur indirectly as a consequence of raised intracranial pressure or direct cranial nerve involvement (optic nerves) in other forms of PCNSL with meningeal infiltration.3 4 While leptomeningeal seeding of lymphoma can occur as a late manifestation of systemic lymphoma or synchronously with PCNSL, lymphoma isolated to the leptomeninges is exceedingly rare.5

We report the case of a young immunocompetent man with the diagnostic dilemma associated with a chronic meningitis presentation with prominent hypoglycorrhachia. The first diagnostic suspicion was meningeal tuberculosis (TB), considering the South Asian origin with high endemic prevalence and cerebrospinal fluid (CSF) analysis. However, further investigations disclosed a primary leptomeningeal lymphoma (PLML). Persevering with investigations when appropriate, along with early adequate CSF analysis, is pivotal in establishing timely diagnosis. This is particularly important at a time where new therapeutic regimes of highly effective chemotherapy combinations with autologous stem transplantation improve prognosis in this aggressive lymphoma subset with a potential to achieve durable remission or cure.6

Case presentation

A previously healthy 37-year-old man of South Asian origin presented with an insidious increase in fronto-occipital headaches over several weeks, which displayed high-pressure features at presentation. There was no associated fever or weight loss. He was a non-smoker and did not drink alcohol. He was born in Pakistan, lived in the UK for 17 years, and his last visit to Pakistan was 3 years ago. He had never received BCG vaccine.

An assessment by the ophthalmology emergency services confirmed the presence of papilloedema. There were no localising neurology signs. The remainder of the systemic examination and biochemistry results were normal. His visual acuity on formal ophthalmology assessment was 6/6 in both eyes. His visual fields were full bilaterally and he had no visual symptoms. A slit lamp examination did not show any evidence of anterior segment inflammation or corneal keratic precipitates. Dilated fundus examination disclosed only Frisén grade 2 papilloedema with a sphincter haemorrhage along the superior part of the disc in the right eye and Frisén grade 1 in the left (figure 1A).

Figure 1

(A) Red-free fundal photos of the optic nerves on first presentation showing papilloedema Frisén grade 2 in the right eye and grade 1 in the left eye. Vision was accordingly normal. (B) Six weeks later, the papilloedema has worsened, with Frisén grade 3 in both eyes. (C) The right optic nerve has developed atrophy and loss of nerve fibre layer with established poor central vision. There is persisting papilloedema in the left eye.

Investigations

Preliminary work-up revealed a normal CT of the head and CT venogram. MRI of the head and spine with contrast failed to demonstrate any significant findings at that stage. A lumbar puncture and CSF analysis revealed raised opening pressure of 32 cmH2O, raised CSF protein of 1.72 g/L, low glucose of 1.0 mmol/L and a white cell count of 126 cells/μL (10% polymorphs and 90% lymphocytes). CSF smears, TB PCR and cryptococcal antigen were negative (details shown in table 1). A whole body CT scan was normal. HIV, hepatitis B and C screen, and syphilis antibodies were negative. Likewise, autoimmune screen including antinuclear antibodies, extractable nuclear antigen and rheumatoid factor were within normal titres. He had normal full blood counts, ACE levels, and thyroid, liver and renal function tests. B12 and folate levels were also normal, and C reactive protein was 1 mg/L.

Table 1

Serial lumbar punctures with CSF analysis

Date
19 May 2020 10 July 2020 03 August 2020 Normal range
Opening pressure (cmH2O) 32 Not done 17 8–15
CSF protein (g/L) 1.72 2.94 8.27 0.15–0.45 
CSF glucose (mmol/L) 1.0  1.0 0.3 2.8–4.2 
CSF while cell count (cells/μL) 126  190 275 0–3 
Lymphocyte predominance (%) 90 100 100
Polymorph (%) 10 0 0
CSF bacterial culture, Acid fast bacilli, fungal stain, cryptococcal antigen, Mycobacterium tuberculosis DNA PCR All negative All negative All negative
  • CSF, cerebrospinal fluid.

Based on the CSF characteristics (lymphocytic pleocytosis with marked hypoglycorrhachia) and absence of other evident alternative causes, he was diagnosed with probable TB meningitis, and empirical treatment immediately started with rifampicin, isoniazid, ethambutol, pyrazinamide with pyridoxine and a tapering dose of oral dexamethasone.

He re-presented with worsening headache 6 weeks after initiation of anti-tubercular treatment (ATT) and steroids and was assessed in a specialist neuro-ophthalmology clinic. An increase in the papilloedema by one grade was identified during this presentation with ongoing headaches (figure 1B). His visual acuity was still normal in keeping with mild to moderate papilloedema with no relative afferent pupillary defect (RAPD), full colour vision and normal intraocular pressure. A QuantiFERON TB assay had returned negative, and there were no other manifestations of pulmonary or milliary TB. Another lumbar puncture was instigated to include cytology and flow cytometry. CSF analysis (details shown in table 1) showed a similar biochemical profile with raised protein of 2.94 g/L, very low glucose of 1 mmol/L, white cell count of 190 cells/μL and red cell count of 386 cells/μL (0% polymorphs and 100% lymphocytes). Both cytology and cytospin were requested but not adequately processed due to a laboratory sampling error, with inconclusive results in the analysed lymphocytes. Beta-glucan and histoplasma antibodies were negative in CSF with no antibodies in serum against dimorphic fungi or histoplasma. Cultures for mycobacteria and fungi did not grow any organisms.

Differential diagnosis

Profound hypoglychorrachia is often considered synonymous with the presence of glucophagic micro-organisms7; however, other conditions can also cause this type of CSF. Infectious meningitis from mycobacteria or fungi followed by a haematological malignancy or neurosarcoidosis8 was the differential diagnosis when considering the insidious presentation with raised intracranial pressure, lymphocytic pleocytosis and high protein. CSF pleocytosis with low glucose may also harbour other different carcinomatous meningitis and infections (parasitic, viral and atypical bacteria and spirochetes).9

Eight weeks after first presentation, the patient reattended emergency services with visual drop in the right eye and a worsening of his headache as steroid was being tapered. Neuro-ophthalmology examination found a new retrobulbar right optic neuropathy with an evident afferent pupillary defect. A reduction of the nerve fibre layer in this eye in the following days from ensuing atrophy was seen, while there was persistent swelling in the fellow eye (figure 1C). Automated perimetry (Humphrey 24/2 visual fields) of both eyes showed a right dense central scotoma with marked reduction of sensitivity to light stimulus. The left visual field is normal (figure 2). Other cranial nerves remained intact. This was thought to represent a likely direct retrobulbar optic nerve infiltration from progressive meningeal disease. Ethambutol toxicity is a differential for visual loss in this context; however, this is typically bilateral and not compatible with a purely unilateral dense optic neuropathy.10 Another MRI of the brain with contrast showed mild cavernous sinus and subtle trigeminal nerve enhancement of the right side (figure 3A,B) with no significant hydrocephalus. A repeat MRI spine with contrast found subtle leptomeningeal enhancement and thickened cauda equina nerve roots (figure 4).

Figure 2

Automated perimetry (Humphrey 24/2 visual fields) of both eyes shows a right dense central scotoma with marked reduction of sensitivity to light stimulus. The left visual field is normal. AGIS, Advanced Glaucoma Intervention Study; PSD, pattern standard deviation; RX, prescription; SITA, Swedish interactive threshold algorithm; VFI, visual field index;NEG, negative; POS, positive.

Figure 3

(A) Magnetic Resonance Imaging (MRI) demonstrating meningeal enhancement (blue arrow) extending from the cavernous sinus, right more than left. (B) MRI demonstrating right trigeminal nerve enhancement (blue arrow).

Figure 4

Magnetic resonance imaging (MRI) of the spine showing subtle leptomeningeal enhancement.

A third lumbar puncture with CSF analysis (details shown in table 1) showed an increase in the lymphocytosis to 275 cells/μL with a further drop in glucose to 0.3 mmol/L and very high protein of 8.27 g/L. Cytology found highly suspicious malignant cells, and CSF flow cytometry confirmed a diffuse large B-cell lymphoma (table 2). Bone marrow biopsy did not show any peripheral involvement, and staging with CT of the thorax, abdomen and pelvis was negative.

Table 2

CSF flow cytometry from last CSF sample confirming idffuse large B cell lymphoma (DLBCL)

B lymphoproliferative panel analysis
B cells 27%
B-cell phenotype 17% CD19+20+79b+CD23wkCD43+CD10+CD38+CD11c-CD5-CD200-lambda weak expression
10% CD19+CD20+CD5+CD23wkCD43+CD10+CD38+CD79b-CD200-CD11c-lambda weak expression but brighter than the first population
Surface Ig’s Lambda positive, weak expression
Clinician interpretation Consistent with central nervous system lymphoma clonal B-cell CD19+ population identified which are lambda restricted; appearance is of a dual population which has two subpopulations, one of which is lambda weak CD79b+CD5 and second population of CD79b-CD5+overall CD20+CD10+CD38+CD200−
Morphologically larger cells plus cleated cells which are smaller raises possibility of clonal progression of DLBCL
  • CSF, cerebrospinal fluid.

A positron emission tomography (PET) scan confirmed that the lymphoma was confined to the meninges, showing multiple deposits within the spinal canal, intervertebral foramina and nerve roots (figure 5A,B).

Figure 5

(A) PET scan showing spinal cord increased uptake. (B) PET scan showing increased dorsal ganglia uptake at multiple levels.

Treatment

The patient has had high-dose chemotherapy designed to penetrate the blood–brain barrier with a MATRix regimen and has received an autologous peripheral blood stem cell transplant.

Outcome and follow-up

To date, the patient had received a full course of MATRix chemotherapy with prophylactic antimicrobials and granulocyte colony-stimulating factor. He was admitted with sepsis and thrombocytopenia following the fourth cycle, which responded to treatment. MRI of the brain and spine has shown full remission, and treatment has progressed towards an autologous peripheral blood stem cell graft. The vision in the right eye remained, counting fingers with an established optic atrophy, while normal vision in the left eye was maintained throughout with a complete remission of the papilloedema.

Discussion

In this case, the main clinical features include an insidious onset of headache of progressive, increasing severity with papilloedema and raised intracranial pressure with CSF features corresponding to chronic meningitis. Lymphocytic pleocytosis, with high protein with marked hypoglycorrhachia in a patient originally from a high-burden area, made TB meningitis a likely preliminary diagnosis, and empirical ATT treatment was started. The identification of acid–fast bacilli in CSF and cultures has low sensitivity; thus, the diagnosis of TB meningitis is often based on clinical suspicion and compatible presentation especially in endemic areas where empirical decision making is often necessary. Following ATT initiation, TB meningitis may worsen and progressive hydrocephalus may ensue in some cases.11 However, in the absence of a definitive diagnosis, it is crucial to explore alternative differentials (fungal infection, lymphoma and malignancy) that harbour an equally poor prognosis when untreated.

Despite continuing adequate ATT treatment with good compliance, 6 weeks following presentation, our patient’s clinical features had progressed with worse papilloedema and headaches shortly followed by a new cranial nerve deficit (right optic neuropathy). This required a challenge of the previous diagnosis and repetition of CSF analysis with recommendations for high-volume specimens for cytology, flow cytometry, and addition of targeted mycobacterial and fungal cultures.

Multiple CSF samples with high volumes might be required before establishing diagnosis of CNS malignancy, especially if this has been previously suppressed with steroids. Cytology, combined with flow cytometry and immunophenotyping, has the best diagnostic yield in this setting.12 In this case, negative bone marrow biopsy and results of PET scan served to clinch the diagnosis further as a PLML.

While PCNSL has been classically associated with an immunocompromised state, it is now increasingly diagnosed in immunocompetent individuals. Variability in clinical presentation in the absence of clear structural lesions amenable to biopsy in imaging can lead to delay in diagnosis.13–15 Treatment and prognosis of these highly aggressive lymphomas have improved with highly effective, high-dose chemoimmunotherapy regimes followed by autologous transplant, which can now offer a real potential for long-term remission.6

PCNSL presenting as lymphomatous meningitis is a very rare variant with an estimated incidence of 7% of all PCNSLs, which are in themselves only 2% of all primary CNS tumours. Publications are mostly case reports and small case series, with only very few reports in the literature of PLML in association with visual loss and papilloedema and as mimickers of other conditions, including TB and idiopathic intracranial hypertension.16–18 In the largest series of PLML with n=48, a median CSF glucose of 47 mg/dL with low glucose was found in half of the cases, and this could be marked in some (ranges of 10 mg/dL–150 mg/dL).5 Headache and cranial neuropathies, often multifocal (most commonly VI and VII), are the most common presentations.

In conclusion, we present a very rare case of PLML with extreme hypoglychorrachia (0.3 mmol/L) initially misdiagnosed as TB meningitis. Mutidisciplinary cooperation was key for a correct interpretation of the neurological and neuro-ophthalmic signs and symptoms. It is important to consider the possibility of lymphoma in a patient presenting with features of lymphocytic meningitis with high protein and low glucose. In this situation, a peripheral lymphoma with meningeal spread is most common, followed by PCNSL with meningeal dissemination. Most of the radiological findings were very subtle and confirmed in retrospect in multidisciplinary meeting. Steroids were likely contributory to partial suppression of the lymphoma masking more evident radiological findings of meningeal infiltration following ATT and dexamethasone initiation. Though this may be absent in CNS TB, the lack of a definite diagnosis of pulmonary or milliary TB with absent response to treatment mandated repeat CSF sampling, including cytology and flow cytometry, which led to the diagnosis.

Patient’s perspective

I developed a headache in March 2020 after a week of self-isolation with suspected COVID-19. The headache progressively worsened until it was an unbearable pain and I made several trips to the general practitioner, who initiated me on migraine treatment. My wife suggested checking my eyes. On 18 May 2020 I visited the opticians and was sent as an emergency with optic nerve swelling to a tertiary care eye hospital. From here I was referred to the main hospital Accident & Emergency; at this point my only symptom was a headache. Following a lumbar puncture performed by the medical team, my headache ceased on 21 May 2020. I was diagnosed with TB meningitis, although no TB was isolated by culture or PCR and no TB symptoms. I was started on TB antibiotics and high-dose steroids for a year. I started developing numerous aliments which required constant visits to the hospital, and each time I was told these are TB treatment symptoms. I had several lumbar punctures which I have extremely bad experiences and these were the most difficult times of my illness, with excruciating leg and back pain. It was devastating when I became unable to walk; we were at a loss, routinely taking medications but symptoms still worsening, my wife and two small children watching as I deteriorated. One and a half months after starting TB treatment I lost vision in my right eye, later affecting my left eye also; I visited the consultant neuro-ophthalmologist for my vision, who sent me for further tests. I was eventually diagnosed with primary CNS lymphoma on 05 August 2020 and referred to haematology. I was hospitalised for over a month during my first chemotherapy treatment and recovery; this was a particularly difficult time as we were in COVID-19 pandemic, and having a language barrier I was unable to have any visitors, felt very lonely and isolated. I started initial chemotherapy on 07 August 2020, after which I started regaining my mobility, strength and general health; however, my vision has not improved. I have had a stem cell collection on 29 September 2020 and due to have my final chemotherapy and stem cell transplant in December 2020/January 2021. My chemotherapy cycles have gone remarkably well with only a few infections.

Looking back at my journey, I am grateful to the ophthalmologist for having the expertise to request the malignancy testing markers which led to my correct diagnosis and treatment. I do think due to the COVID-19 pandemic I was diagnosed too quickly and without justification, and this diagnosis was not challenged during the 3 months of TB treatment in which I had many hospital visits and lack of recovery. I am thankful that I am recovering well and my treatment is to cure the lymphoma.

Learning points

  • Lymphocytic cerebrospinal fluid (CSF) (type C) with high protein and marked hypoglychorrachia is not pathognomonic, though highly suggestive of infectious meningitis.

  • Infectious meningitis is a life-threatening disease with poor prognosis in the absence of targeted treatments. ATT should be started empirically if tuberculosis is suspected in an adequate context and presence of compatible CSF features. At the same time, efforts should be made to encompass other infectious and non-infectious differentials (fungal infection, lymphoma and malignancy) pretreatment with reconsideration of diagnosis and repeat CSF sampling if clinical response is absent.

  • The diagnosis of primary central nervous system lymphoma (PCNSL) with leptomeningeal infiltration and its subset primary leptomeningeal lymphoma relies on CSF analysis. As with other carcinomatoses, yields may be low and the diagnosis proves challenging, especially when preliminary cytology is negative, the patient has been previously treated with steroids or samples are inadequate (small volume of CSF).

  • Multiple CSF analysis for cytology with adequate volumes (minimum of 5–10 mL or more for best yield) may be required to establish the diagnosis of PCNSL.

  • PCNSL has a worse prognosis than other B-cell lymphomas, but this has improved in recent years with therapeutic combinations of intrathecal high-dose chemotherapy and autologous stem cell transplant. An early diagnosis will therefore be crucial to improve prognosis and patient survival.

Ethics statements

Patient consent for publication

Acknowledgments

We gratefully acknowledge Dr Satheesh Ramalingam, consultant neuroradiologist at Queen Elizabeth Hospital for his kind support with neurology images.

Footnotes

  • Contributors SS and TH contributed equally to this case report in terms of literature search, writing, discussion and learning points, and share first authorship. YA contributed to the initial literature search and ophthalmology findings and images. LM-A reviewed and revised the article, and contributed to the discussion part and the references.

  • 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.

References

Use of this content is subject to our disclaimer