Thrombolysis for ischaemic stroke and glioblastoma multiforme: a case report

Abstract

Objective:

To report the uncomplicated use of systemic thrombolysis for stroke in a patient with a misdiagnosed glioblastoma multiforme mimicking brain ischaemia and to suggest that new clinical situations question the stated exclusion criteria for intravenous thrombolysis.

Patient:

A 57-year-old male presented at the emergency room with a sudden aphasia.

Measurement and main results:

After Safe Implementation of Thrombolysis in Stroke-Monitoring Study (SITS-MOST) exclusion criteria were ruled out, intravenous alteplase was administered. The patient presented with tonic–clonic seizures 17 min after perfusion completion, requiring phenytoine administration. Additional computed tomography scan did not show haemorrhagic transformation or brain oedema. A left temporal lobe glioblastoma multiforme was diagnosed after magnetic resonance imaging and neurosurgery. The patient became asymptomatic on the seventh day.

Conclusion:

Any history of central nervous system neoplasm is considered a contraindication to thrombolysis, but the true risk of systemic thrombolysis-precipitated intracranial bleeding is unknown. Further data are needed to establish real haemorrhage risk in this clinical condition.

BACKGROUND

Intravenous recombinant tissue plasminogen activator (rt-PA) is a safe and effective therapy when given by trained physicians in hospitals with appropriate infrastructure. Nevertheless, alteplase is severely underused; symptomatic intracranial haemorrhage is the main safety concern.¹ A review article has been published² on off-label use of intravenous thrombolysis or intra-arterial thrombolysis for stroke or other indications, despite presumed high risk of bleeding. After a systematic search, the authors did not find patients treated with thrombolysis for stroke despite the presence of an intracranial tumour. In August 2007 the first clinical case on this matter was reported,³ with a subsequent haemorrhage. We present the only intravenous thrombolysis (out of 183) performed in our centre to a patient with a misdiagnosed glioblastoma multiforme, which was not followed by intracranial bleeding.

CASE PRESENTATION

A 57-year-old hypertensive male presented at the emergency room with a sudden aphasia. His National Institutes of Health Stroke Scale (NIHSS) score was 7. Urgent blood analyses and non-contrast cranial computed tomography were performed, and Safe Implementation of Thrombolysis in Stroke-Monitoring Study (SITS-MOST) inclusion and exclusion criteria were verified. Mild left temporal lobe hypodensity was detected on brain imaging (fig 1A). Alteplase treatment was started 1.5 h after the onset of symptoms. The patient began with tonic–clonic seizures 17 min after perfusion completion, requiring phenytoine administration. Additional CT scanning did not show haemorrhagic transformation or brain oedema. A left temporal lobe glioblastoma multiforme was diagnosed after MRI and neurosurgery (fig 1B). Diffusion-weighted images did not show any hyperintensities.

Figure 1.

A. Cranial computed tomography without contrast, previous to alteplase infusion. B. Coronal gadolinium enhanced T₁-weighted MRI. A very small glioblastoma multiforme could be diagnosed after stereotaxic biopsy.

INVESTIGATIONS

• Basal cranial computed tomography: mild left temporal lobe hypodensity, which was considered as a possible acute sign of brain ischaemia despite being located in an area with frequent artefacts caused by middle cranial fossa.

• Control cranial computed tomography (after seizure): mild left temporal lobe hypodensity, without any changes to previous images.

• MRI (performed 24 h after symptoms onset): left temporal lobe glioblastoma multiforme. Diffusion and apparent diffusion coefficient (ADC) sequences did not show acute ischaemia. T2-weighted images excluded vasogenic oedema.

OUTCOME AND FOLLOW-UP

The 24-h and 7-day NIHSS scores were 2 and 0, respectively, without corticotherapy and previous to surgery. Modified Rankin score at 3 month evaluation was 1, due to minimal post surgical right-sided paresis.

DISCUSSION

To our knowledge, this is the first report on the use of alteplase in a patient with a glioblastoma multiforme mimicking a stroke without haemorrhage. Previous reports on the safe use of thrombolytic therapy in patients with brain tumours have been published, but endovenous reperfusion agents were not used for stroke treatment. The patient treated with alteplase for an acute myocardial infarction, as documented in Rubinshtein et al,⁴ had planned a pituitary adenoma surgery. Han⁵ reported a safe use of alteplase for a pulmonary embolism in a patient with a known glioblastoma multiforme. At just 2 months after partial resection of tumour, 100 mg of drug were administered. No signs of intracranial bleeding were found on further CT scan.

Previous reports have mentioned the occurrence of misdiagnosed central nervous system masses as cerebrovascular disease. In the CT era, 4.9% of the patients discharged with a diagnosis of brain tumour may initially be thought to have a stroke;⁶ if only glioblastoma multiforme cases are considered, this percentage raises to 11.1%. Diagnosis of sudden onset glioma can be challenging even when CT perfusion⁷ or MRI⁸ are available at the emergency room. Cerebral blood flow (CBF) and cerebral blood volume (CBV) perfusion imaging needs additional investigation to better differentiate between low grade intracranial neoplasms and stroke.

Making the work of the clinician more complicated, gliomas may infiltrate the Sylvian fissure and involve middle cerebral artery, leading to stroke,⁹ even when tumours are not yet visible in radiological tests.¹⁰ Moreover, tumours may show increased signal on diffusion-weighted images (DWI) and low signal intensity on ADC maps on MRI, as acute stroke.⁷ In our case DWIs did not show any hyperintensities. As the patient was symptomatic at 24 h, we considered that the symptoms were not due to brain ischaemia but to glioblastoma. Nevertheless, it is not possible to exclude an ischaemic event related to brain tumour resolved by thrombolysis.

At the moment, three prospective observational cohort studies have been published, the largest in 2007. No patients with brain tumours were reported among the protocol violations in the Standard Treatment with Alteplase to Reverse Stroke (STARS)¹¹ and Canadian Alteplase for Stroke Effectiveness Study (CASES)¹² investigations. As isolated cases are insufficient to withdraw conclusions, in order to establish frequency and safety of alteplase use on misdiagnosed brain tumours it would be essential to publish data recorded on the SITS-MOST registry¹³ referring to thrombolysis in patients that were non-stroke, as centres were able to notify on the seventh day a final diagnosis of non-ischaemic stroke in the international database.

LEARNING POINTS

• Diagnosis of sudden onset brain tumours can be challenging in the emergency room.

• The true risk of systemic thrombolysis-precipitated intracranial bleeding is unknown in patients with central nervous system neoplasms.

• It would be essential to publish data recorded on international registries referring to thrombolysis made in patients that were non-stroke in order to establish frequency and safety of alteplase use on misdiagnosed brain tumours.