Diagnosis and management of leptomeningeal disease secondary to grade IV astrocytic glioma

  1. Chon Meng Lam ,
  2. Anthony Lisacek-Kiosoglous ,
  3. Elena Paleacu and
  4. Elin Jones
  1. Hywel Dda University Health Board, Carmarthen, UK
  1. Correspondence to Dr Elin Jones; elin.m.jones@wales.nhs.uk; Dr Chon Meng Lam; chonmenglam@doctors.org.uk

Publication history

Accepted:08 Sep 2022
First published:16 Sep 2022
Online issue publication:16 Sep 2022

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

A man in his mid-40s presented to hospital with confusion, headache and feeling generally unwell. He had had a total resection of a grade IV astrocytic glioma 1 year prior. Initial observations, blood tests and CT head scan were unremarkable for acute features to explain the patient’s presentation. However, an MRI head scan on this admission demonstrated a clear communicating hydrocephalus with new abnormal leptomeningeal enhancement, consistent with leptomeningeal metastatic infiltration by glioma. Lumbar puncture cytology and biochemistry supported this interpretation. As a small district general hospital in rural Wales, we discuss the experience of diagnosis and coordination of specialist input from a multidisciplinary team. We share the challenges of managing leptomeningeal disease in the COVID-19 pandemic, in the context of the additional risks this presents with chemotherapy-induced immunosuppression.

Background

Leptomeningeal disease (LMD) secondary to astrocytic glioma is a rare complication of an uncommon condition. LMD is most commonly secondary to melanoma, breast and lung cancers in adults and has a poor prognosis; median survival after diagnosis of LMD is 3–6 months.1 Treatment is often limited to palliative options, with the goal of preserving neurological function and prolonging survival.2 Correctly diagnosing LMD is important to allow for effective symptom control. However, symptoms and signs can be non-specific and can overlap with more common pathologies such as infective meningitis. The build-up of cerebrospinal fluid (CSF) and the increase in subarachnoid pressure can lead to irreversible visual loss and be detrimental to a patient’s quality of life. Initial blood tests, observations and CT head scan findings can be unremarkable, leading to difficulty and delay in reaching the diagnosis.

Here, we report a case of LMD secondary to grade IV astrocytic glioma presenting 1-year postresection of the primary tumour, and discuss challenges faced by clinicians at a small rural district general hospital (DGH) in diagnosing and managing this condition during the COVID-19 pandemic.

Case presentation

A man in his mid40s presented to a DGH with acute confusion, photophobia, vomiting, headaches and visual changes. The patient spoke English and Welsh fluently. One year prior, the patient was diagnosed with an isocitrate dehydrogenase mutant grade IV left temporal astrocytic glioma (figure 1) and underwent total surgical resection of the lesion at a tertiary centre, followed by radiotherapy at a different tertiary centre. The primary tumour was MGMT methylated (O6-methylguanine-DNA-methyltransferase gene) and deemed favourable for temozolomide chemotherapy. However, the patient chose not to undergo chemotherapy due to concerns about immunosuppression and the risk of COVID-19 associated with attending hospital for treatment. The patient experienced receptive dysphasia postoperatively.

Figure 1

MRI head scan showing the initial primary tumour prior to surgical resection.

Observations and blood tests on this admission were unremarkable. On general neurological examination, expressive dysphasia and gait ataxia were noted. Cranial nerve examination was unremarkable. Upper limb examination revealed generalised hyper-reflexia bilaterally. Lower limb examination revealed positive right sided Babinski reflex, joint proprioception abnormality of the right first interphalangeal joint and absent vibration sense of the right first interphalangeal joint (though not the metatarsophalangeal joint). There was no light touch sensory loss or muscle power loss in the upper or lower extremities.

The patient was referred to an ophthalmologist and found to have severe papilloedema with significant visual field defects. Visual acuity was 6/6–2 on the right and 6/12+2 on the left.

Investigations

Following surgical resection of the primary tumour, the patient had had several MRI head scans for disease surveillance. The MRI head scan 1 month prior to admission was stable and the radiological report indicated no interval change (figure 2). The MRI head scan on the current admission showed new abnormal leptomeningeal enhancement involving the base of the brain and brainstem (figures 3 and 4). The quick development of these changes was deemed highly suspicious of acute meningitis by the consultant radiologist.

Figure 2

MRI head scan reported as negative for drop metastasis.

Figure 3

MRI head scan showing diffuse prominent leptomeningeal enhancement.

Figure 4

MRI head scan showing new widespread enhancement involving the posterior fossa and cerebellum.

On the initial lumbar puncture (LP), a CSF protein of 20.11 g/L (normal range: 0.15–0.45), CSF glucose 5.9 mmol/L (normal range: 2.2–3.9), CSF lactate 8.2 mmol/L (normal: <3.0) with an opening pressure of >35 cmH2O were recorded. CSF culture was negative. The cytology was limited by extensive cell degeneration, which is not uncommon in CSF preparations, and was analysed both locally and sent to a tertiary centre for a second opinion. Atypical cells with a few multinucleate forms were identified on both the Grocott methenamine silver and periodic acid-Schiff stain preparations, some with larger nuclei and significant pleomorphism. MRI spine, which was performed of the complete neuraxis to rule out further sinister pathology, did not reveal spinal metastases (figure 5A–C).3 The neuroradiology multidisciplinary team (MDT) reviewed all the prior imaging, and reported that the most recent imaging showed significantly increased T2 hyperintensities indicating more extensive caking of the posterior fossa and cerebral hemispheres when compared with previous MRI scans (figure 4).

Figure 5

MRI cervical, thoracic and lumbar spine without significant findings.

Differential diagnosis

The initial neurological examination findings were suspicious of upper motor neuron pathology, which could not be fully explained by the existing knowledge of the patient’s condition. Following consultation with neurologists at a tertiary centre, it was deemed that the right side upgoing plantar reflex and expressive dysphasia could be explained by the surgical resection of the primary tumour. However, the other findings meant that further investigation was warranted. The new acute changes on the MRI head compared with the images 1 month prior meant that an acute meningitis secondary to infection was the working differential diagnosis. One of the most alarming CSF findings was the extremely high protein level, and therefore, a diagnosis of tuberculous meningitis had to be considered. However, considering the low prevalence of tuberculosis in the area, lack of foreign travel in the clinical history and the high glucose level on the CSF (normally low in tuberculous meningitis), this diagnosis was felt unlikely.

To elucidate the full extent of the disease and rule out a neoplastic cause, an MRI whole spine was requested. An urgent ophthalmology referral was made to definitively characterise the visual deficits, which identified severe papilloedema. This finding reinforced the urgency to find a diagnosis and provided a baseline from which we could monitor the patient’s intracranial pressure and consequent visual symptoms. Monitoring was conducted by the DGH, and updates were provided to the neurosurgical and neuro-oncology MDT at the tertiary centre. From CSF cytology alone, it was difficult to confidently diagnose a malignant infiltrate. When correlated with clinical and radiological features, and after discussion at the neuro-oncology MDT, a diagnosis of leptomeningeal metastatic infiltration by glioma was made.

Treatment

Treatment was initiated with high-dose intravenous dexamethasone with intravenous antibiotics for meningitis, and symptoms initially improved but intermittently recurred. After discussion with the neurosurgical team and the neuro-oncology MDT, it was felt that the focus of treatment should be palliative as no curative options were available. Of immediate concern was the threat to the patient’s vision due to the high subarachnoid pressure. It was agreed that a permanent CSF diversion with a ventriculoperitoneal (VP) shunt was the ideal way to relieve this pressure, though noting that there would be a significant risk of blockage given the elevated CSF protein levels. Daily therapeutic LPs to remove 30–50 mL of CSF were advised, with the vast majority of these being performed in a sterile operating theatre environment to reduce the risk of infection.

The patient’s symptoms initially improved and the LPs were therefore reduced to alternate days for the patient’s comfort. A temozolomide regimen with palliative intent was started on the ward at the DGH, with the aim to bring the protein level down to allow for the VP shunt to be performed and to improve prognosis.

The therapeutic LPs became increasingly ineffective and within 2 weeks, visual acuity deteriorated to 6/18 on the right and worse than 6/60 on the left. A COVID-19 outbreak on the neurosurgical ward at this time led to a detailed virtual discussion between the neurosurgeons in the tertiary centre, the local oncology team, the patient and their next of kin. The discussion pertained to the treatment options available at the time, which included: (1) a return to daily LPs until the COVID-19 outbreak on the ward had settled; (2) transfer to a tertiary centre for a lumbar or external ventricular drain to alleviate the pain of having daily LPs or (3) insertion of a VP shunt, accepting the risk of acute blockage. All involved understood that options 2 and 3 had an increased risk of COVID-19 infection with the added risk that the completed cycle of temozolomide could dampen the patient’s immunity. The patient informed us that he was not concerned about the discomfort of the LPs and decided to continue with option 1.

There continued to be a serious risk of the patient losing their vision before the CSF protein level was low enough for the VP shunt, with no guarantee that it would ever reach the required level. With the COVID-19 situation improved on the neurosurgery ward and the CSF protein levels reaching 4.01 g/L, the lowest recorded level for the patient since the LMD diagnosis, the decision was made in conjunction with the patient’s wishes to proceed with a VP shunt.

Outcome and follow-up

The patient’s postoperative recovery was uneventful. He was repatriated to his local hospital and discharged home. Due to a low platelet count, the dose for the palliative temozolomide regimen was reduced to 85% in cycle 2. The patient’s symptoms improved, and LPs were no longer required. For the next 3 months, the patient enjoyed a good quality of life without requiring any further hospital admissions. The patient continued to have a degree of expressive dysphasia but was able to maintain a conversation. He was able to mobilise safely with a stick without any balance issues and reported returning to reading as a hobby. He was able to read large letters clearly, but found smaller letters blurry. On follow-up with ophthalmology, unaided visual acuities were 6/9–1 on the right and 6/60 on the left, improving to 6/36 with pinhole view. Fundus examination showed well defined disc margins in the right eye with definite pallor but no oedema. On the left there was old swelling of the optic nerve nasally with temporal pallor. There were optic nerve changes present in both eyes secondary to hydrocephalus, but the changes were felt by the ophthalmologist to be old.

Symptoms gradually worsened 3 months post-VP shunt insertion with the patient developing increased confusion, agitation, vomiting, blurred vision and seizures. Symptom control with medications including antiepileptics was initiated. A CT head scan in the patient’s final hospital admission showed the ventricles were dilated more than previously. An MRI head scan showed that extensive leptomeningeal dural disease had developed. The patient passed away peacefully on the ward, 6 months after the VP shunt was inserted.

Discussion

Collectively, the presence of neoplastic cells in the subarachnoid space secondary to a primary cancer is termed LMD, but is also known as neoplastic meningitis and carcinomatous meningitis. LMD is relatively rare with the the most common origins of neoplastic cells in the subarachnoid space being from melanoma, breast and lung cancer primaries.4

LMD is primarily a clinical diagnosis, with symptoms resulting from raised subarachnoid pressure caused by the presence of neoplastic cells. This can cause non-specific symptoms such as headaches, visual changes, gait disturbance, nausea and confusion. In a small case series from one centre, 12% of patients with grade III or IV glioma had LMD at the time of initial tumour diagnosis. In the other patients, the time to progression to LMD ranged from 2.8 to 221 months with a median of 16.4 months.5 In the case presented, the patient had a grade IV glioma with symptoms of LMD starting 11 months after initial diagnosis.

CSF cytology can be helpful for diagnosis but has limitations, with one case series reporting that five out of seven cases with leptomeningeal enhancement on MRI had negative CSF results.6 Glantz et al suggest that false-negative CSF cytology results can be minimised by: (1) withdrawing a minimum of 10.5 mL of CSF; (2) minimising the time between CSF sample collection and processing; (3) obtaining the CSF sample from a known site of LMD and (4) repeating CSF collection if initial cytology is negative.7 However, even if malignant cells are not seen on cytology, the CSF result can be informative and aid diagnosis. High protein levels and opening pressures can be a predictor of a malignant pathology, however, as discussed above, infectious causes such as tuberculous meningitis must also be included in the differentials.1

The imaging modality of choice for diagnosis of LMD is a T1-weighted MRI scan with gadolinium contrast of the brain and whole spine, as LMD can impact the whole neuraxis (figures 2–5C).1 3 It has been reported that there is an association between an initial location in the temporal lobe for the primary tumour and a shorter time to development of LMD. This has been attributed to its close proximity to conduits through which malignant cells can enter CSF, such as the superior temporal sulcus.5 In the case reported here, the initial tumour was in the temporal lobe.

Visual disturbance is regarded as one of the signs of LMD, however, reported evidence does not quantify the proportion of cases of LMD that have visual disturbance at diagnosis. Furthermore, there is no reported consensus on whether visual disturbances precede changes on MRI, potentially identifying an area for further research. This is significant as oncology teams could include monitoring of visual changes by an optician as part of routine monitoring in addition to MRI. For DGH settings, optician services are more readily available than MRI, oncology, neurology or neurosurgery services, and this could be used to improve monitoring and early detection of LMD for patients. This case highlights the importance of utilising face to face appointments with an optician.

An LMD is difficult to treat and the focus of treatment is palliative. Primary treatment goals are to improve neurological function, survival and quality of life, however assessing the efficacy of treatment is limited by the lack of randomised control trials.1 4 Treatment options can be divided into supportive care, medical management, radiotherapy, chemotherapy and surgical management. Supportive care should be offered to all patients and be tailored to the needs of the patient and their families. Medical treatment is primarily focused on symptom control such as using antiepileptics for seizure control, antiemetics for management of nausea and analgesia for pain. Radiotherapy can be used effectively for palliative relief by using focal radiotherapy for bulky disease or to target obstructive lesions causing hydrocephalus.1

In this case, decisions on chemotherapy were taken during the COVID-19 pandemic and the risk of immunosuppression was a factor in the patient opting against postradiotherapy adjuvant temozolomide after the initial tumour resection. It is impossible to tell whether this decision would have influenced the development of LMD. Temozolomide was started with a palliative intent after the diagnosis of LMD and continued after the VP shunt was inserted. A VP shunt can be of particular benefit when hydrocephalus is present and a small case study of patients with LMD secondary to lung adenocarcinoma found that 90.3% of patients showed an improvement after shunt surgery.8 This outcome was observed with the patient no longer requiring regular LPs, and experiencing a marked improvement in performance status.

As DGH clinicians, it is important that the patient and their next of kin are provided with opportunities to speak with tertiary centre specialists. In this case, in-person and virtual consultations were arranged for the patient and his next of kin with tertiary centre specialists and a member of the local oncology team present. This was found to be beneficial to both DGH clinicians and the patient group, as it improved understanding of the DGH management options and limitations. The DGH clinicians played a vital role in coordinating the MDT and were an important point of contact for the patient and their family.

Patient’s perspective

The initial shock of having a brain tumour made him very uncertain of life and he questioned what kind of a future would await him. He had suffered with what appeared to be sinusitis over a period of many years. Making the 4 hour return journey to the tertiary centre most days for radiotherapy over a 6 week period was very tiring for him and refusing chemotherapy tablets with the radiotherapy played on his mind. He asked at the tertiary centre what would happen if he had the COVID-19 virus while having chemotherapy and was informed that there would be a significant risk of being unwell and dying. This had a lasting effect on him for the remaining period of his life. After the initial tumour surgery, he was more confident—life for him returned to some form of normality. From around 1 month before his LMD was diagnosed, he had periods of regular sickness and vision problems and was admitted to the local DGH on a few occasions for scans and LPs. Around this time he was very confused and found great difficulty expressing himself. He became very agitated and cross when we spoke to him by telephone and could not understand why we did not visit him. He was not aware that he had been to the tertiary centre for the shunt operation. The onset of the COVID-19 pandemic played constantly on his mind and worsened from after the diagnosis of LMD. Periods in hospital after the LMD diagnosis made him confused and disorientated and unable to understand what was explained to him by hospital staff. The period prior to his transfer to the tertiary centre for the shunt operation was extremely difficult for him—he did not realise nor understand what had happened nor what was going to happen. A few months after the shunt operation, he began to develop regular seizures and this was a traumatic period for him and us. He had no recollection of the seizures he experienced during the last months of his life until we told him afterwards. The change in personality in the last months of his life was very prominent—he easy going and relaxed person he used to be became more aggressive.

Learning points

  • Be aware that leptomeningeal disease (LMD) can present with intermittent and non-specific symptoms that have unremarkable findings on initial blood tests, observations and CT head scans. Accordingly, clinicians should have a low threshold for considering further investigations such as MRI head and lumbar puncture when clinical findings do not correlate with said investigations.

  • Multidisciplinary team input is essential for LMD management. District general hospital clinicians can and should facilitate effective communication between the multidisciplinary team and the patient and their family.

  • Clinicians should have a low threshold for advising patients to attend for regular eye checks following initial therapy to monitor for papilloedema, and advise patients to self-refer should they notice any deterioration.

Ethics statements

Patient consent for publication

Acknowledgments

With thanks to Dr Mark Gavartin for his contribution towards revising the language in this case report.

Footnotes

  • Contributors CML: planning and conducting case report, collecting data from the case, reviewing of literature and drafting manuscript. AL-K: had a leading role in making the revisions to manuscript and played a leading role in responding to reviewers comments. EP: collecting and reviewing radiological images from the case, senior doctor involved in case, contribution to planning of manuscript content. EJ: consultant responsible for care of the patient, overall management of patient’s condition and insight into patient’s care.

  • 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.

  • Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

References

Use of this content is subject to our disclaimer