Lymphomatous Meningitis in Primary Central Nervous System Lymphoma

Marc C. Chamberlain, M.D.


Neurosurg Focus. 2006;21(5) 

In This Article

Diagnosis of LM

The most useful laboratory test for diagnosing LM is investigation of the CSF. Abnormalities include increased opening pressure (> 26 mm H2O), increased leukocytes (> 4/mm3), elevated protein (> 50 mg/dl), or decreased glucose (< 60 mg/dl), parameters that, although suggestive of LM, are not diagnostic. The presence of malignant cells in the CSF is diagnostic of LM.

In up to 45% of patients with positive CSF cytology (see subsequent discussion), the initial examination will be cytologically negative.[40] The yield is increased to 80% with a second CSF examination, but little benefit is obtained from repeated lumbar punctures after two such procedures.[71] Of note, in a series including lymphomatous and leukemic meningitis reported by Kaplan, et al.,[27] the authors observed the frequent dissociation between CSF cell count and malignant cytology (29% of cytologically positive CSF had concurrent CSF counts of < 4/mm3).[36] Murray, et al.,[51] showed that CSF levels of protein, glucose, and malignant cells vary at different levels of the neuraxis, even if there is no obstruction of the CSF flow. This finding reflects the multifocal nature of LM and explains why CSF obtained from a site distant to that of the pathologically involved meninges may yield a negative cytology.

Of the 90 patients reported on by Wasserstrom, et al.,[71] a positive CSF cytology in samples obtained from either the ventricles or cisterna magna was noted in 5% of the cases. In a series of 60 patients with LM, positive lumbar CSF cytology at diagnosis, and no evidence of CSF flow obstruction, ventricular and lumbar cytological samples obtained simultaneously were discordant in 30% of cases.[31] The authors observed that in the presence of spinal signs or symptoms the lumbar CSF was more likely to be positive and, conversely, in the presence of cranial signs or symptoms the ventricular CSF was more likely to be positive. Obtaining no CSF sample from a site of symptomatic or radiographically demonstrated disease was found to correlate with false-negative cytological results in a prospective evaluation of 39 patients, as did withdrawing small CSF volumes (< 10.5 ml), delayed processing of specimens, and obtaining fewer than two samples.[35] Even after correcting for these factors, a substantial group of patients remains who have LM and persistently negative CSF cytological findings. Glass, et al.,[40] reported on a postmortem evaluation in which they assessed the value of premortem CSF cytology. They demonstrated that up to 40% of patients with clinically suspected LM proven at time of autopsy were cytologically negative. This figure increased to greater than 50% in patients with focal LM.

The low sensitivity of CSF cytological examination makes it difficult not only to diagnose LM but also to assess the response to treatment. Biochemical markers, immunohistochemical, and molecular biology techniques applied to CSF have been explored in an attempt to determine a reliable biological marker of disease.

Numerous biochemical markers have been evaluated, but in general their use has been limited by poor sensitivity and specificity.[13] Nonspecific tumor markers such as creatinine-kinase BB isoenzyme, tissue polypeptide antigen, β2-microglobulin, β-glucoronidase, lactate dehydrogenase isoenzyme-5, and, more recently, vascular endothelial growth factor can be strong indirect indicators of LM, but none is sensitive enough to improve the cytological diagnosis.[65] The use of these biochemical markers can be helpful as adjunctive diagnostic tests and, when followed serially, to assess response to treatment. Occasionally, in patients with clinically suspected LM and negative CSF cytological results, they may support the diagnosis of LM.

Use of monoclonal antibodies for immunohistochemical analysis in LM does not significantly increase the sensitivity of cytological examination alone.[10,33,45,68] However, in LM, antibodies against surface markers can be used to distinguish between reactive and neoplastic lymphocytes in the CSF.

Cytogenetic studies have also been evaluated in an attempt to improve the detection of LM. Flow cytometry and DNA single-cell cytometry, techniques that measure the chromosomal content of cells, and fluorescence in situ hybridization, which detects numerical and structural genetic aberrations as a sign of malignancy, can provide additional diagnostic information, but these modalities still have a low sensitivity.[8,25,56,67] Polymerase chain reaction can establish a correct diagnosis when cytological findings are inconclusive, but the genetic alteration of the neoplasia must be known for it to be amplified with this technique.[42]

In cases in which there is no manifestation of systemic cancer and CSF examinations remain inconclusive, evaluation of a meningeal biopsy specimen can be diagnostic. The yield of this test increases if the biopsy sample is taken from an enhancing region demonstrated on MR imaging.[24]

Magnetic resonance imaging with Gd enhancement is the modality of choice to evaluate patients with suspected leptomeningeal metastasis.[23,59,66] Because LM involves the entire neuraxis, whole-CNS imaging is required in patients considered for further treatment. Both contrast agent-enhanced and unenhanced T1-weighted sequences, combined with fat suppression T2-weighted sequences, constitute the standard MR imaging examination. Magnetic resonance imaging has been shown to have a higher sensitivity than cranial contrast medium-enhanced CT scanning in several series and is similar to CT myelography for the evaluation of the spine, but is significantly better tolerated.[12,31,58]

Any irritation of the leptomeninges will result in their enhancement on MR imaging, which is seen as a fine signal intense layer that follows the gyri and superficial sulci. Subependymal involvement of the ventricles often results in enhancement of the ventricles. Some changes such as cranial nerve enhancement and intradural extramedullary enhancing nodules (most frequently seen in the cauda equina) can be considered diagnostic of LM in patients with cancer. Because lumbar puncture itself rarely causes a meningeal reaction leading to dural-arachnoidal enhancement, imaging should be conducted preferably prior to the procedure.[50] The incidence of false-negative results on Gd-enhanced MR imaging remains 30% so that a normal study does not exclude the diagnosis of an LM. In cases involving a typical clinical presentation, however, abnormal results on Gd-enhanced MR imaging alone are adequate to establish the diagnosis of LM.

Radionuclide studies involving either 111Indium-diethylenetriaminepentaacetic acid or 99Tc macroaggregated albumin constitute the technique of choice for evaluating CSF flow dynamics.[14,15,18,22,37,49] Abnormal CSF circulation has been demonstrated in 30 to 70% of patients with LM, with blocks commonly occurring at the skull base, the spinal canal, and over the cerebral convexities. In several clinical series of patients in whom interruption of CSF flow was demonstrated by radionuclide ventriculography, the authors observed decreased survival compared with those in whom CSF flow was normal.[18,37] Involved-field radiotherapy targeting the site of CSF flow obstruction restores flow in 30% of patients with spinal disease and 50% of those with intracranial disease. Reestablishment of CSF flow with involved-field radiotherapy and subsequent intra-CSF chemotherapy led to longer survival, lower rates of treatment-related morbidity, and a lower mortality rate from progressive LM compared with the group in which there were persistent CSF blocks.[18,37] These findings may reflect the possibility that CSF flow abnormalities prevent homogenous distribution of intra-CSF chemotherapy, resulting in 1) protected sites where the tumor can progress and 2) the accumulation of drug at other sites resulting in neurotoxicity and systemic toxicity. Based on this, many authors recommend that intra-CSF chemotherapy be preceded by a radionuclide flow study and, if a block is found, that radiotherapy be administered in an attempt to reestablish normal flow.[14]

In summary, patients with suspected LM, newly diagnosed primary CNS lymphomas, or recurrent primary CNS lymphomas should undergo evaluation of the CSF compartment to include one or two lumbar punctures, cranial Gd-enhanced MR imaging, spinal Gd-enhanced MR imaging, and a radioisotope CSF flow study to rule out sites of CSF blockage. If cytological findings remain negative and neuroimaging results are not definitive, consideration may be given to ventricular or lateral cervical CSF analysis based on the suspected site of predominant disease. If the clinical scenario or imaging studies are highly suggestive of LM, treatment is warranted despite persistently negative CSF cytological findings.

The treatment of LMs is complicated by the lack of standard therapy, the difficulty of determining response to treatment because of the suboptimal sensitivity of the diagnostic procedures, the fact that most patients die of progressive parenchymal disease, and the fact that most studies of LM are small, nonrandomized, and retrospective. However, it is clear that treatment of LM can provide effective palliation and in some cases result in prolonged survival. Treatment in most cases requires the combination of surgery, radiotherapy, and chemotherapy.

Surgery is used in the treatment of LM for the placement of 1) intraventricular catheters and subgaleal reservoirs for administration of cytotoxic drugs and 2) ventriculoperitoneal shunts in patients with symptomatic hydrocephalus.

Drugs can be instilled into the subarachnoid space by lumbar puncture or via an intraventricular reservoir system. The latter is the preferred approach because it is simpler, more comfortable for the patient, and safer than repeated lumbar punctures. It also results in a more uniform distribution of the drug in the CSF space and produces the most consistent CSF levels. In up to 10% of lumbar punctures the drug is delivered to the epidural space, even if there is CSF return after placement of the needle, and the distribution of the drug has been shown to be better after reservoir-based drug delivery.

The two basic types of reservoirs are 1) the Rickham-style reservoir, a flat rigid reservoir placed over a bur hole, and 2) the Ommaya reservoir, a dome-shaped reservoir that can be easily palpated.[7,57] These reservoirs are generally placed over the right (nondominant) frontal region after making a small C-shaped incision. The catheter is placed in the frontal horn of the lateral ventricle or close to the Monro foramen through a standard ventricular puncture. In most cases, anatomical landmarks suffice, but ultra sonography or CT guidance can be helpful in some situations. It is important to be sure that the tip and the side perforations of the catheter are inserted completely into the ventricle to avoid drug instillation into the brain parenchyma. Correct placement of the catheter should be checked by CT scanning without contrast medium prior to its use for drug administration. This frequently will show a small amount of air in both frontal horns.

Lymphomatous meningitis often causes communicating hydrocephalus leading to symptoms of raised intracranial pressure. Relief of sites of CSF flow obstruction with involved-field radiation should be attempted to avoid the need for placing a CSF shunt. If hydrocephalus persists, a ventriculoperitoneal shunt should be placed to relieve the pressure because relief of pressure often results in clinical improvement. If possible, an in-line on/off valve and reservoir should be used to permit the administration of intra-CSF chemotherapy, although some patients cannot tolerate having the shunt turned off to allow the circulation of the drug.

Finally, in patients with a persistent blockage of ventricular CSF flow, a lumbar catheter and reservoir can be used in addition to a ventricular catheter, to allow treatment of the spine with intra-CSF chemotherapy (although as discussed earlier, cases involving postirradiation persistent CSF flow blocks are probably best managed using supportive care alone).


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: