[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.161.216.242. Please contact the publisher to request reinstatement.
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Download PDF
Figure 1.
Magnetic Resonance Imaging of Brain and Cervical Spine
Magnetic Resonance Imaging of Brain and Cervical Spine

A, Axial gadolinium-enhanced T1-weighted magnetic resonance image (MRI) showing multiple, loculated, cystic collections with irregular, nodular enhancement along the borders. B, Axial diffusion-weighted MRI showing no evidence of diffusion restriction. C, Axial T2-weighted MRI with 1 cyst showing a T2-isointense internal nodular structure (arrow). D, Sagittal T2-weighted cervical spine MRI showing marked compression of the upper cervical cord (arrows). E, Axial T2-weighted MRI before development of a parenchymal lesion. F, Axial T2-weighted MRI taken 5 days later showing interval development of a new parenchymal lesion (arrow).

Figure 2.
Neuropathological Findings on Brain Biopsy
Neuropathological Findings on Brain Biopsy

A, Glandular tissue reminiscent of intestinal-type epithelium (hematoxylin-eosin, original magnification ×4). B, Diffuse nuclear positivity for caudal-type homeobox 2 protein (original magnification ×200). C, Diffuse cytoplasmic positivity for cytokeratin 7 (original magnification ×200). D, Diffuse cytoplasmic positivity for cytokeratin 19 (original magnification ×200). E, Patchy cytoplasmic positivity for cytokeratin 20 (original magnification ×200).

Table 1.  
Cerebrospinal Fluid Analysis
Cerebrospinal Fluid Analysis
Table 2.  
Additional Cerebrospinal Fluid and Serum Results
Additional Cerebrospinal Fluid and Serum Results
1.
Pattisapu  JV.  Etiology and clinical course of hydrocephalus. Neurosurg Clin N Am. 2001;12(4):651-659, vii.
PubMed
2.
Scarrow  AM, Segal  R, Medsger  TA  Jr, Wasko  MC.  Communicating hydrocephalus secondary to diffuse meningeal spread of Wegener’s granulomatosis: case report and literature review. Neurosurgery. 1998;43(6):1470-1473.
PubMed
3.
Troiani  C, Lopes  CC, Scardovelli  CA, Nai  GA.  Cystic brain metastases radiologically simulating neurocysticercosis. Sao Paulo Med J. 2011;129(5):352-356.
PubMedArticle
4.
Hernandez  O, Zagzag  D, Kelly  P, Golfinos  J, Levine  PH.  Cytological diagnosis of cystic brain tumors: a retrospective study of 88 cases. Diagn Cytopathol. 2004;31(4):221-228.
PubMedArticle
5.
Osborn  AG, Preece  MT.  Intracranial cysts: radiologic-pathologic correlation and imaging approach. Radiology. 2006;239(3):650-664.
PubMedArticle
6.
Lo Re  V  III, Gluckman  SJ.  Eosinophilic meningitis. Am J Med. 2003;114(3):217-223.
PubMedArticle
7.
Jurado  R, Walker  HK. Cerebrospinal fluid. In: Walker  HK, Hall  WD, Hurst  JW, eds. Clinical Methods: The History, Physical, and Laboratory Examinations.3rd ed. Boston, MA: Butterworth; 1990.
8.
Kimura-Hayama  ET, Higuera  JA, Corona-Cedillo  R,  et al.  Neurocysticercosis: radiologic-pathologic correlation. Radiographics. 2010;30(6):1705-1719.
PubMedArticle
9.
Carpio  A.  Neurocysticercosis: an update. Lancet Infect Dis. 2002;2(12):751-762.
PubMedArticle
10.
Nash  TE, Garcia  HH.  Diagnosis and treatment of neurocysticercosis. Nat Rev Neurol. 2011;7(10):584-594.
PubMedArticle
11.
Pekmezci  M, Perry  A.  Neuropathology of brain metastases. Surg Neurol Int. 2013;4(suppl 4):S245-S255.
PubMed
12.
Barnholtz-Sloan  JS, Sloan  AE, Davis  FG, Vigneau  FD, Lai  P, Sawaya  RE.  Incidence proportions of brain metastases in patients diagnosed (1973 to 2001) in the Metropolitan Detroit Cancer Surveillance System. J Clin Oncol. 2004;22(14):2865-2872.
PubMedArticle
13.
Patchell  RA.  The management of brain metastases. Cancer Treat Rev. 2003;29(6):533-540.
PubMedArticle
14.
Delattre  JY, Krol  G, Thaler  HT, Posner  JB.  Distribution of brain metastases. Arch Neurol. 1988;45(7):741-744.
PubMedArticle
15.
Oshiro  S, Tsugu  H, Komatsu  F,  et al.  Metastatic adenocarcinoma in the brain: magnetic resonance imaging with pathological correlations to mucin content. Anticancer Res. 2008;28(1B):407-413.
PubMed
16.
Teplyuk  NM, Mollenhauer  B, Gabriely  G,  et al.  MicroRNAs in cerebrospinal fluid identify glioblastoma and metastatic brain cancers and reflect disease activity. Neuro Oncol. 2012;14(6):689-700.
PubMedArticle
Views 1,833
Citations 0
Clinical Pathologic Conference
November 2013

Headache and Focal Neurologic Deficits in a 37-Year-Old Woman

Author Affiliations
  • 1Department of Neurology, University of California, San Francisco
  • 2Division of Neuropathology, Department of Pathology, University of California, San Francisco
  • 3Department of Neurosurgery, University of California, San Francisco
  • 4now with the Department of Neurology, Lahey Hospital and Medical Center, Burlington, Massachusetts
JAMA Neurol. 2013;70(11):1445-1449. doi:10.1001/jamaneurol.2013.3933
Abstract

A 37-year-old woman presented with progressively worsening headache. She had no headache history, and initial evaluation revealed hydrocephalus of unclear etiology. A ventriculoperitoneal shunt was placed with improvement. However, her headache returned and she developed neurologic deficits. Imaging studies demonstrated multiple cystic lesions in the posterior fossa. Serum and cerebrospinal fluid studies were unrevealing and a biopsy of the cystic lesions was performed. The clinical approach, differential diagnosis, and neuropathological findings are discussed.

A 37-year-old woman with no headache history presented to her primary care physician with worsening daily headache ongoing for 1 month. The headache was holocephalic, exacerbated by lying down, and relieved by sitting up or standing. She had associated vertigo, nausea, vomiting, and gait impairment. Her medical history included hypertension, hyperlipidemia, type 2 diabetes mellitus, and schizophrenia. Her current medications were lisinopril, simvastatin, metformin hydrochloride, bupropion hydrochloride, and risperidone. She had no known allergies. She was born and raised in California, was briefly homeless, and had never traveled outside the United States. She was a current smoker with a 40-pack/y smoking history. She drank 6 to 12 alcoholic beverages per week. She used marijuana occasionally and denied other drug use, including intravenous drugs. There was no family history of neurologic disorders. General medical examination findings were normal.

A noncontrast head computed tomographic (CT) scan revealed panventricular enlargement consistent with hydrocephalus. Nonenhanced brain magnetic resonance imaging (MRI) showed no mass lesion. She had no history of risk factors for hydrocephalus, including no previous head trauma, intracranial hemorrhage, or central nervous system (CNS) infection.

A ventriculoperitoneal (VP) shunt was placed with clinical resolution of her symptoms and radiographic improvement. Cerebrospinal fluid (CSF) analysis at the time of the procedure revealed no cells, normal protein and glucose levels, and negative bacterial cultures.

The patient’s symptoms resolved until 6 months later, when she developed an occasional headache. She reported no visual symptoms, nausea, or vomiting. Her headache was holocephalic, without positional effects, and easily relieved with ibuprofen. A noncontrast head CT revealed no intracranial pathology, stably decompressed ventricles, and appropriate VP shunt placement. During the next 4 months her headache worsened, occurring daily and associated with photophobia, nausea, and vomiting. There was intermittent diplopia in multiple directions of gaze. The headache was worse in the morning and at times woke her from sleep. She reported no fevers, chills, rash, weight loss, or abdominal pain. There was no bowel or bladder dysfunction. She presented to the emergency department 1 week after developing neck stiffness.

On general examination, she was afebrile and her blood pressure was 145/83 mm Hg. Her heart rate, respirations, and room air oxygen saturation were normal. She was diaphoretic. She endorsed pain with passive neck movements but had full range of motion. Cardiovascular, pulmonary, abdominal, and skin examination findings were normal.

On neurologic examination, she appeared sleepy but was arousable to voice with a normal orientation. Pupillary light reflexes and funduscopic examination findings were normal. There was saccadic breakdown of smooth pursuit. Extraocular movements were intact. She had bifacial weakness and spastic dysarthria. There was mild neck flexor weakness. Tone was spastic in the bilateral upper extremities, with full strength throughout. Reflexes were diffusely brisk with bilateral Hoffman signs, no jaw jerk, and flexor plantar responses. Sensory examination findings and appendicular coordination were normal. Her gait was wide based and ataxic.

Laboratory and Radiological Data

The patient’s serum metabolic profile and complete blood cell count with differential were normal. A noncontrast head CT revealed multiple cystic posterior fossa lesions. Gadolinium-enhanced brain MRI demonstrated loculated, extra-axial, and intraparenchymal fluid collections with irregular, nodular enhancement along their borders within the cerebellar hemispheres and along posterior fossa structures (Figure 1A). There was mass effect with marked compression of the inferior cerebellum and vermis, fourth ventricle, medulla, and upper cervical cord. The cystic fluid was of the same signal intensity as CSF. There was no evidence of diffusion restriction (Figure 1B). One cyst showed a T2-isointense internal nodular structure (Figure 1C). Gadolinium-enhanced spine MRI revealed marked compression of the upper cervical cord from C1 to C4 by extramedullary cystic collections (Figure 1D). There was extensive, diffuse, nodular leptomeningeal enhancement along the entire spinal cord. There was no intrinsic cord signal abnormality.

A lumbar puncture was performed, with initial CSF results presented in Table 1. Serum and CSF enzyme-linked immunosorbent assay for Taenia solium, among other studies, were sent for analysis. The results of additional CSF and serum studies are presented in Table 2.

The patient was premedicated with dexamethasone for a day and then began treatment with albendazole for presumed neurocysticercosis (NCC). Six hours after receiving her first dose of albendazole, she had an acute decline in mental status and was transferred to the intensive care unit. Repeated head CT revealed interval development of hydrocephalus compared with imaging 5 days earlier, and an external ventricular drain was placed. Albendazole was discontinued and dexamethasone was continued. Gadolinium-enhanced MRI demonstrated development of a new rim-enhancing cystic parenchymal lesion within the left cerebellar hemisphere (Figure 1E and F).

Despite external ventricular drain placement, the patient clinically declined during the next 4 days, with signs of worsening brainstem compression. Serum and CSF enzyme-linked immunosorbent assay for T solium returned negative. Given her worsening brainstem compression and nondiagnostic serum and CSF studies (Table 2), she underwent a suboccipital craniotomy with leptomeningeal/cyst biopsy.

Clinical Discussion

This woman’s presentation with new-onset daily headache that worsened with recumbency and was associated with vertigo, nausea, vomiting, and gait impairment suggests increased intracranial pressure. The temporal course of the illness implies a subacute process. Her initial neuroimaging revealed hydrocephalus with panventricular enlargement without any evidence of an obstructing mass lesion, suggesting a process affecting the subarachnoid spaces or CSF absorptive surfaces. This may be seen in conditions causing acute inflammation or delayed scarring or fibrosis (ie, trauma, infection, or hemorrhage) as well as with leptomeningeal infiltration from granulomatous disease or leptomeningeal metastases from a systemic cancer.1,2 Gadolinium-enhanced imaging may reveal abnormal dural enhancement in some of these conditions.2

Despite VP shunt placement, her headaches progressively worsened. The onset of associated nuchal rigidity suggests meningeal irritation. She also developed diplopia, dysarthria, bifacial weakness, and a depressed level of consciousness, suggesting dysfunction at the level of the pons including the facial nuclei and possibly the reticular activating system. Neck flexor weakness with hyperreflexia and increased tone in the arms implicate corticospinal tract dysfunction in the brainstem or upper cervical spine. There was also probable midline cerebellar dysfunction given her wide-based, ataxic gait that was not associated with abnormal sensory findings. Her MRI findings support this clinical picture and revealed cystic lesions affecting the cerebellum, brainstem, and upper cervical cord.

The differential diagnosis for cystic brain lesions is broad and includes benign, inflammatory, neoplastic, and infectious processes. Here, MRI revealing both extra-axial and intraparenchymal lesions focuses the differential on neoplastic and infectious etiologies as these can more commonly have wide leptomeningeal dissemination concurrent with intraparenchymal involvement.

Primary CNS tumors typically present with solitary lesions. Those that commonly have a cystic appearance include pilocytic astrocytoma, ganglioglioma, ependymoma, and hemangioblastoma.3,4 Cystic brain metastases have been described from primary cancers involving the lung, breast, prostate, and pancreas.4

Infectious etiologies include multiple abscesses, which may be from severe bacterial, mycobacterial, or fungal infections. Additionally, various parasitic infections (ie, NCC, echinococcosis, amebiasis) may also appear as cystic lesions within the CNS.5

Analysis of CSF later demonstrated a mononuclear- and lymphocyte-predominant pleocytosis consistent with chronic meningeal irritation. While the proportion of eosinophils fell short of the 10% required for eosinophilic meningitis, finding any eosinophils in the CSF is abnormal. Causes of CSF eosinophilia include parasitic and fungal infections, meningeal involvement of certain malignant neoplasms (ie, Hodgkin disease, non-Hodgkin lymphoma, eosinophilic leukemia), or lastly a foreign-body reaction such as to a VP shunt.6 A markedly elevated CSF protein level (>1 g/dL; to convert to grams per liter, multiply by 10.0) may be seen with any purulent meningitis or intracerebral hemorrhage or when there is a spinal block to CSF circulation.7 A low CSF glucose level (<60% serum) suggests acute bacterial, fungal, or tuberculous meningitis, lymphomatous or carcinomatous meningitis, neurosarcoidosis, or NCC.7 The MRI finding of a cyst with a T2-isointense internal nodular structure combined with the CSF profile raises suspicion for NCC.

Neurocysticercosis occurs with infection by the larval cystic form of the pork tapeworm T solium. The MRI findings here implicate the vesicular and racemose forms of NCC, which may coexist in the same patient. A hallmark of the vesicular form of NCC is MRI revealing intraparenchymal, thin-membranous cystic structures with fluid of CSF-like intensity along with a mural nodule, which is aT2-isointense to T2-hyperintense nodular structure along the cyst wall representing the viable larval scolex.8 In racemose NCC, subarachnoid infiltration produces cystic lesions within the basal cisterns seen deforming neighboring structures on MRI. The cyst wall in vesicular and racemose NCC generally does not enhance.8

Neurocysticercosis may only be definitively diagnosed through pathologic analysis of excised cysts. Less invasive tests to detect the presence of serum or CSF antibodies to T solium are the enzyme-linked immunosorbent assay and enzyme-linked immunoelectrotransfer blot. The appropriate treatment of NCC depends on the presentation, location, and burden of disease. Antiepileptic drugs, antihelminthic drugs (most commonly albendazole and praziquantel), corticosteroids, and surgical interventions all play a role.9,10 Antihelminthic therapy in those with a large burden of subarachnoid/cisternal or intraventricular disease is controversial, as acute destruction of the parasite may trigger an inflammatory response leading to clinical worsening from edema and increased intracranial pressure. Premedication with corticosteroids or stabilization with surgical decompression and/or shunt placement prior to antihelminthic treatment is debated.10 In this case, the rapid clinical deterioration at the time of treatment for NCC was concerning for reactive edema around the cyst vs progression of her disease, and a decompressive suboccipital craniotomy with diagnostic cyst biopsy was performed.

Neuropathology

Microscopic examination of surgical samples revealed columnar epithelium with scattered goblet cells and increased mitotic activity, consistent with well-differentiated adenocarcinoma (Figure 2A). Immunohistochemical stains showed that the tumor cells were diffusely positive for caudal-type homeobox 2 (CDX-2) (Figure 2B), cytokeratin (CK) 7 (Figure 2C), and CK19 (Figure 2D) and focally positive for CK20 (Figure 2E). Dual expression of CK7 and CK20 is consistent with a metastatic adenocarcinoma arising from the upper gastrointestinal tract or pancreaticobiliary system.11

Patient Outcome

The pathologic diagnosis was metastatic adenocarcinoma, consistent with a primary neoplasm involving upper gastrointestinal tract. Staging CT of the chest, abdomen, and pelvis as well as whole-body positron emission tomography showed no primary or additional tumor. Serum cancer markers of carcinoembryonic antigen and cancer antigens 125, 19-9, and 15-3 were not elevated. Despite suboccipital decompression, the patient’s neurologic examination findings continued to deteriorate and she developed worsening cervical myelopathy and quadriparesis. The patient and family elected to pursue palliative radiation therapy, after which she was transferred to hospice care and died 3 weeks later from progressive CNS metastases.

Conclusions

Metastasis to the CNS is much more common in those with known advanced systemic cancers and is estimated to occur in about 15% of patients with cancer.12 Gastrointestinal tract tumors are especially unlikely to have CNS metastasis. Infratentorial seeding, as seen in this case, occurs in about 20% of CNS gastrointestinal metastases.13 This disproportionately high incidence of posterior fossa metastases suggests that these primary malignant neoplasms more often gain CNS access through the posterior circulation.14

Multicystic CNS metastases from adenocarcinoma of the lung, breast, and, rarely, gastrointestinal organs (ie, stomach, pancreas) have been described.3,4,15 Characteristics of cystic CNS metastatic adenocarcinomas on MRI are similar to those in this case with isointensity to hypointensity on T1-weighted sequences, isointensity to hyperintensity on T2-weighted sequences, and nodular enhancement on postgadolinium sequences.15

This case is also unusual in that our patient presented with CNS metastases as an initial manifestation and the evolution over 11 months was much slower than expected for untreated CNS metastases, especially with diffuse leptomeningeal involvement. She also developed hydrocephalus without evidence of a mass lesion on brain imaging. Her initial hydrocephalus was possibly the result of a small cyst obstructing the central canal or leptomeningeal infiltration by tumor cells. Her CSF profile was difficult to interpret and likely reflected CSF block (elevated protein level) and chronic inflammation in the presence of a foreign body, ie, the VP shunt (eosinophilia). Leptomeningeal carcinomatosis could explain the low CSF glucose level, although her CSF cytological findings were negative.

The MRI finding here of both extra-axial and intraparenchymal cystic lesions was important in focusing the differential on neoplastic and infectious etiologies as these may seed both locations. This case further highlights the need for consideration of CNS metastasis, even in those without a known history of malignancy. A thorough investigation should include a determination of immune status, contrast-enhanced brain and/or spine imaging, and cytological, bacterial, mycobacterial, and fungal studies of the CSF in the appropriate clinical context. Ultimately, biopsy confirmation may be required for diagnosis. In the future, CSF biomarkers may be more sensitive than cytology for early detection of neoplastic meningitis.16 While this patient’s condition was inexorably progressive, early diagnosis and prompt treatment may improve the clinical outcome for others.

Back to top
Article Information

Corresponding Author: Brian J. Scott, MD, Department of Neurology, Lahey Hospital and Medical Center, 41 Mall Rd, Burlington, MA 01805 (brian.j.scott@lahey.org).

Accepted for Publication: August 6, 2013.

Published Online: September 30, 2013. doi:10.1001/jamaneurol.2013.3933.

Author Contributions: Dr Scott had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Aghi, Scott.

Acquisition of data: Ayodele, Pekmezci, Aghi.

Analysis and interpretation of data: Pekmezci, Aghi.

Drafting of the manuscript: Ayodele, Scott.

Critical revision of the manuscript for important intellectual content: Pekmezci, Aghi, Scott.

Administrative, technical, or material support: Ayodele, Pekmezci, Aghi.

Study supervision: Aghi, Scott.

Conflict of Interest Disclosures: None reported.

References
1.
Pattisapu  JV.  Etiology and clinical course of hydrocephalus. Neurosurg Clin N Am. 2001;12(4):651-659, vii.
PubMed
2.
Scarrow  AM, Segal  R, Medsger  TA  Jr, Wasko  MC.  Communicating hydrocephalus secondary to diffuse meningeal spread of Wegener’s granulomatosis: case report and literature review. Neurosurgery. 1998;43(6):1470-1473.
PubMed
3.
Troiani  C, Lopes  CC, Scardovelli  CA, Nai  GA.  Cystic brain metastases radiologically simulating neurocysticercosis. Sao Paulo Med J. 2011;129(5):352-356.
PubMedArticle
4.
Hernandez  O, Zagzag  D, Kelly  P, Golfinos  J, Levine  PH.  Cytological diagnosis of cystic brain tumors: a retrospective study of 88 cases. Diagn Cytopathol. 2004;31(4):221-228.
PubMedArticle
5.
Osborn  AG, Preece  MT.  Intracranial cysts: radiologic-pathologic correlation and imaging approach. Radiology. 2006;239(3):650-664.
PubMedArticle
6.
Lo Re  V  III, Gluckman  SJ.  Eosinophilic meningitis. Am J Med. 2003;114(3):217-223.
PubMedArticle
7.
Jurado  R, Walker  HK. Cerebrospinal fluid. In: Walker  HK, Hall  WD, Hurst  JW, eds. Clinical Methods: The History, Physical, and Laboratory Examinations.3rd ed. Boston, MA: Butterworth; 1990.
8.
Kimura-Hayama  ET, Higuera  JA, Corona-Cedillo  R,  et al.  Neurocysticercosis: radiologic-pathologic correlation. Radiographics. 2010;30(6):1705-1719.
PubMedArticle
9.
Carpio  A.  Neurocysticercosis: an update. Lancet Infect Dis. 2002;2(12):751-762.
PubMedArticle
10.
Nash  TE, Garcia  HH.  Diagnosis and treatment of neurocysticercosis. Nat Rev Neurol. 2011;7(10):584-594.
PubMedArticle
11.
Pekmezci  M, Perry  A.  Neuropathology of brain metastases. Surg Neurol Int. 2013;4(suppl 4):S245-S255.
PubMed
12.
Barnholtz-Sloan  JS, Sloan  AE, Davis  FG, Vigneau  FD, Lai  P, Sawaya  RE.  Incidence proportions of brain metastases in patients diagnosed (1973 to 2001) in the Metropolitan Detroit Cancer Surveillance System. J Clin Oncol. 2004;22(14):2865-2872.
PubMedArticle
13.
Patchell  RA.  The management of brain metastases. Cancer Treat Rev. 2003;29(6):533-540.
PubMedArticle
14.
Delattre  JY, Krol  G, Thaler  HT, Posner  JB.  Distribution of brain metastases. Arch Neurol. 1988;45(7):741-744.
PubMedArticle
15.
Oshiro  S, Tsugu  H, Komatsu  F,  et al.  Metastatic adenocarcinoma in the brain: magnetic resonance imaging with pathological correlations to mucin content. Anticancer Res. 2008;28(1B):407-413.
PubMed
16.
Teplyuk  NM, Mollenhauer  B, Gabriely  G,  et al.  MicroRNAs in cerebrospinal fluid identify glioblastoma and metastatic brain cancers and reflect disease activity. Neuro Oncol. 2012;14(6):689-700.
PubMedArticle
×