Review of Lyme Neuroborreliosis

Introduction

A series of 3 talks on the second day of the 13th Lyme Disease Conference examined the latest findings in the diagnosis and treatment of neurologic Lyme disease.

Brain Imaging in Lyme Disease

Structural brain imaging. MRI scans among patients with neurologic Lyme disease may demonstrate punctate white matter lesions on T₂-weighted images, similar to those seen in demyelinating or inflammatory disorders such as multiple sclerosis, systemic lupus erythematosus, or cerebrovascular disease. In Lyme disease, this is most often the case among patients with evidence of meningitis or encephalitis. In a report of patients with CSF-confirmed European Lyme disease with CNS symptoms, 50% of 14 patients had abnormal CT findings, most commonly hypodense areas corresponding with ischemic lesions caused by putative vasculitis. In each of the 3 studied cases, MRI revealed smaller, multiple sclerosis-like lesions and larger lesions also suspected to indicate vascular involvement. The authors concluded that, comparable to meningovascular and cerebrovascular syphilis, CNS micro- and macrovasculitis may cause both clinical symptoms and MRI changes in patients with CNS borreliosis.

The usefulness of MRI scans in American chronic Lyme encephalopathy is less clear. The general impression is that in chronic neurologic Lyme disease, brain MRI scans may be abnormal less often. In a report of a series of 24 patients with chronic neurologic abnormalities of at least 3 months' duration after Lyme disease, 4 (16.6%) had MRI abnormalities characterized as small round periventricular white matter lesions and 3 of these 4 patients had abnormal CSF studies. Among patients with Lyme encephalopathy in particular, the rate of MRI scans with white matter hyperintensities varied from 15% of 13 patients to 41% of 17 patients. In the latter study, when scans were repeated after treatment, half of the patients showed resolution of the signal hyperintensity. Among 8 children with neurologic Lyme disease, 25% had MRI abnormalities consisting of multiple focal areas of increased signal intensity in the white matter on long TR (both proton-density and T2-weighted) images.

In late-stage encephalomyelitis, MRI scanning often demonstrates focal areas of inflammation, most commonly in the white matter. In a series of 34 patients with acute or indolent encephalomyelitis, 26 (76.5%) had small multifocal hyperintense T2 signals, most commonly in the white matter and occasionally in the cortical and subcortical gray matter of the brain. Because encephalomyelitis clinically may result in prominent pyramidal, sensory, or cerebellar syndromes and because the MRI lesions resemble demyelination, patients with Lyme disease may be mistakenly diagnosed as having multiple sclerosis.

Despite the preponderance of evidence indicating a substantial rate of MRI hyperintensities in some patients with CNS Lyme disease, as of yet there have been no studies to examine the pathophysiology of these hyperintense areas (perfusion, reactivity to hypercapnia, metabolism) and whether they have prognostic significance. Do these hyperintensities represent demyelination or perivascular inflammation? Is the disease process underlying the hyperintensities primarily neuronal metabolic or vascular? These questions can be examined by a study that couples structural imaging with functional imaging, comparing cerebral blood flow and cerebral metabolic rate deficits. The more sensitive FLAIR sequence and magnetization transfer techniques can be used to maximize the yield on identifying white matter hyperintensities.

In other disease states, such hyperintensities have usually been attributed to ischemic cerebrovascular disease secondary to increased water content in perivascular space, axon and myelin loss, astrocyte proliferation (gliosis), and/or frank infarction. If the hyperintensities occur primarily in the arteriole-supplied watershed areas, then the most likely cause is vascular insufficiency, as these areas receive limited collateral supply. Risk factors that increase the likelihood of having hyperintensities include older age, hypertension, diabetes, coronary heart disease, and other vascular risk factors. The presence of hyperintensities has been shown to predict subsequent stroke, new-onset dementia, myocardial infarction, and vascular death.

Histopathologic studies of hyperintensities in neurologic and normal samples commonly show arteriolar hyalinization, ectasia, enlarged perivascular space, gliosis, spongiosis, and/or lacunar infarcts. Van Swieten found that white matter hyperintensities in the elderly were invariably accompanied by demyelination and gliosis, and less consistently with increased perivascular space. The demyelination was strongly associated with increased wall thickness of small arterioles. They concluded that arteriosclerosis in small arterioles (<150 microns) is the primary cause, leading to demyelination, and then cell loss with progression.

The hypothesis that MRI hyperintensities in Lyme encephalopathy are attributable to impaired blood flow (vascular insufficiency) in subcortical areas can be tested with in vivo perfusion imaging. In stroke samples, PET and SPECT studies consistently have shown correspondence between the identified areas of hypoperfusion and the structural abnormalities identified by CT, MRI, or pathology. The spatial extent of the perfusion deficit is typically larger than the areas of tissue necrosis defined by pathology or structural imaging, and remote functional changes may be observed (diaschisis).

Functional brain imaging. Single photon emission computerized tomography (SPECT) and positron emission tomography (PET) provide a dynamic picture of the brain's functioning: metabolism, blood flow, and chemistry. In comparison to SPECT scans, PET scanning is able to provide better spatial resolution images (4-6 mm vs 6-9 mm) and can be used to provide an absolute quantitative assessment of regional perfusion or metabolic abnormalities. SPECT has recently been reported to be a useful tool in the evaluation of patients with Lyme disease, showing multifocal areas of decreased perfusion in both the cortex and the subcortical white matter.

Logigian contrasted the brain perfusion patterns of 13 patients with definite Lyme encephalopathy (defined by objective memory deficits on cognitive testing and/or CSF with intrathecal Ab production or positive PCR), 9 patients with possible Lyme encephalopathy (no objective deficits), and 26 normal controls. Patients with definite Lyme encephalopathy had significantly more perfusion deficits than patients with possible Lyme encephalopathy, who in turn had significantly more deficits than normal controls. After the patients with definite Lyme encephalopathy were given 4 weeks of IV ceftriaxone, a partial reversal in brain perfusion deficits was observed. These results suggest that perfusion deficits are greater with more severe disease and that perfusion deficits may be seen in the absence of objective neuropsychological deficits.

Notably, although the treated patients did show improved perfusion, significant perfusion deficits remained. Several limitations of this study should be noted. First, because this study did not have age- and sex-matched controls, the results that suggested regions of deficit specific to Lyme encephalopathy (subcortical frontotemporal white matter and basal ganglia, frontal cortex, cingulate gyrus) need to be confirmed in a better-controlled investigation. Second, the relatively older control group did not have neuropsychological testing at baseline, thereby risking the inclusion of cognitively impaired individuals. This may explain why 4 of the 26 "normal" controls had perfusion deficits in the range of those seen in the Lyme encephalopathy patients. Third, because in this study 11 of the 13 treated patients with Lyme encephelopathy had never previously received a course of IV ceftriaxone of at least 3 weeks, the results of this study cannot be generalized to the larger group of patients with Lyme encephalopathy who have already received this standard course of treatment. The authors' impression that "patients with Lyme encephalopathy who have already been treated with oneor at most two 1-month courses of IV ceftriaxone rarely improve after further courses of ceftriaxone" needs to be tested in a carefully designed way. Fourth, patients did not undergo follow-up cognitive testing, thereby precluding objective investigation of the clinical significance of the improvement in perfusion.

Hypoperfusion defects visualized on SPECT scans may result from any process that alters the radiotracer distribution, including vascular delivery to neurons, transport of the tracer into the cells, and retention of the radioactive tracer in the cells. Problems may arise secondary to direct infection of neurons, from cellular dysfunction due to the indirect effects of neurotoxic immunomodulators such as cytokines, or from decreased perfusion through arterioles secondary to vasculitis. In other words, areas of hypoperfusion may result from a cellular-metabolic and/or a vascular problem. Although SPECT reports suggest that the use of imaging pre-and post-acetazolamide may be helpful in the indirect determination of a vascular from a metabolically induced area of hypoperfusion, only PET technology using glucose and oxygen studies is capable of addressing this question directly in a fully quantitative fashion.

In what ways, then, are functional imaging scans helpful in the diagnostic assessment? First, a scan with diffuse abnormalities may confirm that an objective abnormality is present in a patient considered to have a factitious disorder. Second, a normal scan in a patient with prominent neuropsychiatric symptoms may suggest that a psychiatric disorder is the primary cause of a patient's cognitive or emotional distress, and therefore may lead the clinician to recommend a re-evaluation of the patient's psychiatric treatment. Third, an improvement in a scan after treatment provides objective evidence of physiologic change.

It should be noted that one cannot conclude from a PET or a SPECT scan that a patient has Lyme disease, as similar patterns of abnormality may be seen with other diseases as well. A diffusely abnormal scan should alert the clinician to search for the presence of an organic etiology other than that which causes primary psychiatric disorders. Other disease processes that demonstrate a heterogeneous tracer uptake include vascular dementia, chronic fatigue syndrome, CNS lupus, HIV encephalopathy, and chronic or acute stimulant abuse.

There have been no metabolic imaging studies in Lyme disease; the limited data pertain to cerebral blood flow only. It remains to be determined how quickly the hypothesized perfusion and metabolic deficits improve and whether improvement in functional imaging is correlated with neuropsychological change. For example, it is known that metabolic or flow defects may persist after a stroke or trauma despite the presence of a normal neurologic exam. The Columbia Presbyterian Medical Center's clinical experience with Lyme disease patients indicates that improvement in SPECT abnormalities may occur rapidly or lag behind clinical improvement by many months.

Questions Raised by the Structural and Functional Imaging Data Studies of Chronic Lyme Encephalopathy. Several critical public health questions are raised. First, do patients with previously treated encephalopathy show an improvement in brain imaging (both functional and structural) and cognition after repeated courses of antibiotic treatment? Second, are there patient variables that predict who is more likely to respond? Third, are there neuroimaging pretreatment variables that are associated with treatment response? Fourth, can structural or functional neuroimaging be used as an objective tool to monitor response to treatment? And finally, is there a subgroup of patients with possible Lyme encephalopathy who no longer respond to treatment because of cerebrovascular disease?

These questions will be addressed by a newly funded study of persistent Lyme encephalopathy being conducted at the Columbia Presbyterian Medical Center in New York City. Under the direction of Dr. Brian Fallon, this 4-year study will enroll 60 patients with cognitive deficits secondary to Lyme disease and 20 healthy controls. The 60 patients must have well-documented Lyme disease and must have received a total of at least 8 weeks of IV antibiotic therapy previously. Patients and controls will get baseline serologic, CSF, and imaging tests at baseline. The imaging tests include MRI and PET. The PET imaging includes a test of neuronal metabolism (FDG), a test of vascular flow (O15-water), and a test of vascular flow after a hypercapnic challenge. The hypercapnic challenge involves breathing in a small amount of CO₂-enhanced air which, in a person with normal blood vessels, would result in a broad increase in perfusion. In a patient with vasculitis with ischemia at which the vasculature may already be maximally dilated, however, the O15 PET scan after the hypercapnic challenge would reveal a relative deficiency of increased perfusion in the affected areas - thus providing a relative simple way of determining whether a patient's disease process involves vascular disease.

The treatment component of this new study (conducted at the patient's home) involves a placebo-controlled treatment with 10 weeks of IV ceftriaxone in which 40 patients get randomized to IV antibiotics and 20 patients get randomized to IV placebo. Patients will then be followed off antibiotics for 14 weeks, monitored both by cognitive tests and brain imaging 2 weeks after the end of treatment and 14 weeks after end of treatment. The aim is to determine whether patients who have persistent cognitive deficits despite considerable past IV antibiotic therapy benefit from a repeated course of intensive antibiotic therapy. This study should address questions regarding the time course of improvement cognitively and via imaging as well as whether the improvement between these 2 assessment modalities are closely correlated. Collaborators at other institutions will attempt to culture Borrelia organisms from the spinal fluid and examine the spinal fluid for markers of infection, such as matrix metalloproteinase, PCR, and B burgdorferi-specific immune complex. In addition, the investigation of multiple variables at baseline (serum, CSF, imaging, clinical) may help to identify markers that would predict who responds to treatment and who does not. Dr. Fallon concluded his talk by providing the phone number for physicians interested in referring patients to the study (212-543-5367).

Neurologic Manifestations of Lyme Disease in the Pediatric Population

Involvement of CNS. According to Dr. Pietrucha, children with neurologic Lyme disease may present acutely with headache, blurry vision, double vision, confusion, irritability, fever, and/or stiff neck. Chronically, they may be encephalopathic and have lingering headache, personality change, and depression. Patients who present acutely may have an aseptic meningitis with pleocytosis and elevated protein in the spinal fluid. Occasionally, there may be lesions on the MRI, and about 20% of patients may have abnormal EEGs.

Increased intracranial pressure with an opening pressure above 200 mm/H₂O is seen much more often in children with Lyme disease than in adults. This is sometimes referred to as "pseudotumor cerebri," although it is not a classic pseudotumor picture as the children do not necessarily have to be overweight or have a problem with their menstrual cycle. The CSF pleocytosis usually improves. Frequently, it may improve even without treatment, but certainly, if the patient is treated with antibiotics, this will clear. The increased intracranial pressure responses to medications such as Diamox and seizures should be treated with anticonvulsant medications. The lesions on the MRI may remain or may disappear with time.

The most common lingering problem that patients have as a result of involvement of the CNS in Lyme is encephalopathy, which the children call "brain fog." These children complain of persistent headache and fatigue. There may be personality change, irritability, and frequently depression.

The impact academically is most significant. These children have fall-off in academic performance, difficulty learning new material, problems with short-term memory, problems with word finding, and a number of them have lost reading skills. Frequently, these children may present with a picture of ADD or may have an underlying ADD or ADHD that is made worse by the Lyme. Incidentally, Dr. Pietrucha noted that children with Tourette's may also have a worsening of their tics when they have been ill with Lyme disease symptoms. A case report published in Lancet indicated that a child with Tourette's who was found to have concurrent CNS Lyme disease experienced a remission of the Tourette's after the Lyme disease was treated.

This Lyme encephalopathy merits special attentions because it has a significant impact educationally and also economically. These children may require at-home tutoring, necessitating a parent to stay home from work to be with the child. When they return to school, many frequently need a shortened schoolday and continued home instruction. Many have to be classified as "other health impaired" and receive ongoing services, such as Resource Room. A number of these children need to be placed on medication for their short attention spans and distractibility, as any other ADD patient would require. Frequently, the depression has to be treated, both with medications and counseling.

In addition to the impact on these children educationally, there is a social burden because they cannot participate in extracurricular activities and they lose contact with peers.

Peripheral nervous system. Patients may present with a sudden onset of weakness. It may be facial weakness, weakness of an extremity, or an ascending weakness or paralysis. There may be pain, a burning sensation in the extremities, numbness, tingling, and myalgia.

Children do not have involvement of the peripheral nervous system as frequently as they have involvement of the CNS. The most common peripheral nervous manifestation in children is Bell's palsy, with a sudden onset of facial palsy. Rarely, there has been involvement of an isolated extremity or even an isolated nerve involvement, such as the peroneal nerve. Patients have presented with a picture very typical of Guillain Barré (GB), and many children do complain of a burning sensation, numbness, and tingling -- although this is a mild sensory neuropathy. There have been cases of children presenting with muscle pain, weakness, and elevated CPK.

In addition to treating their Lyme disease with appropriate antibiotic therapy, these patients may require physical therapy, anti-inflammatory medication, and analgesia. Patients presenting with a picture that is typical of GB should be treated like a GB patient, keeping in mind that if the underlying cause is Lyme disease, then that too must be treated. Overall, the prognosis for children to show a complete recovery from involvement of the peripheral nervous system in Lyme disease is very good, probably better than in the adult population. Bilateral Bell's palsy has certainly been seen in Lyme disease in children. The incidence of bilateral Bell's palsy in Lyme is greater than the incidence of bilateral Bell's palsy from other causes.

Conclusion. Dr. Pietrucha's talk was based on years of clinical experience treating hundreds of children with mild-severe neurologic Lyme disease. In the question-and-answer session, she described one child with refractory generalized seizures not responsive to anticonvulsants who was also experiencing joint pain. This child tested positive for Lyme disease and responded with a remission of the seizures. The treatment course was prolonged and led to an overall dramatic improvement in clinical symptoms. Dr. Pietrucha also noted that complex partial seizures have been described in the medical literature among children with chronic Lyme disease.

Psychological Evaluation of Pediatric, Neurologic Lyme Disease

Methods. In this pilot study, Dr. Rissenberg evaluated 8 children aged 7 to 13 years (mean, 9.1) with physical, cognitive, and emotional symptoms related to Lyme disease who had neuropsychological evaluation, including academic testing early and again later in the course of their antibiotic treatment. Physical symptoms reported at the initial evaluation (E₁) included fatigue, joint pain headaches and irritability. Also reported were difficulties with schoolwork, concentration, and memory; sleep disturbance; sensory sensitivity; mood swings; impulsivity; depressed mood; anxiety; motor tics; word retrieval difficulty; balance problems; and temper outbursts.

Results. Dr. Rissenberg's results indicated that at the time of E₁, even after completing from 1 to 5 months of high-dose antibiotic treatment, children had significant cognitive deficits. As a group, the subjects had a significant discrepancy between Verbal and Performance IQ, and a significantly deficient Performance IQ. There was a significant degree of inter-subtest variability on the WISC-III, with scores ranging from the 20^th to 93^rd percentile. Scores were lowest on tests sensitive to speed of processing, visual scanning, sequencing, and causal reasoning. Deficits were noted on 2 attention tasks, one sensitive to visual scanning and sustained attention and the other to auditory tracking. While there were no statistically significant memory deficits evident at E₁, the data suggest that delayed recall of both verbal and visual material is deficient. On tests of language function, performance was deficient on a task requiring production of sentences containing a given word. On academic measures, half the S's were behind grade expectation in 2 measures of reading comprehension, as well as spelling. Most S's were above grade expectation in Basic Reading, Mathematics Reasoning and Numerical Operations. Reading skills were more advanced than math skills.

At E₂, following 10 to 32 months (mean, 17) of additional antibiotic therapy, all subjects reported significant improvement of physical and emotional symptoms, with only 1 having continued headaches and another having sleep disturbance. Five experienced improvement in cognitive and academic difficulties, though 4 continued to have some cognitive complaints and 5 continued to have some emotional issues. Two continued to have both cognitive and emotional symptoms, and 1 had both physical and emotional symptoms.

Results from repeat administration of the WISC-III at E₂ revealed significant improvement in Verbal and Performance IQ, with less of a spread between the two. Full Scale IQ and the Perceptual Organization Index also showed significant improvement. Marked improvement in performance was shown on those subtests that were deficient at E₁, specifically Picture Arrangement, Comprehension, Object Assembly, and Coding, as well as Arithmetic. This strongly supports the notion that these deficits were secondary to Lyme disease and that their improvement is attributable to antibiotic treatment. Visual scanning and sustained attention improved, while auditory tracking showed less improvement. Performance on sentence production improved. Significant improvement was noted on the Verbal Immediate Memory Index and the General Memory Index of the CMS. Short-term memory impairment is no longer apparent. Gains were made in academic achievement in all areas, with the exception of Numerical Operations (paper and pencil calculations). However, even with 6 S's dropping an average of 1 year and 3 months, only 2 fell below their expected GE. Gains were demonstrated on 3 separate measures of reading comprehension and on reading speed and accuracy. Scores on tests of mathematical calculations and fund of general and word knowledge declined. This is interpreted as reflecting a decreased rate of learning and a widening of the gap between children with Lyme disease and their healthy peers over the course of the study.

Implications. Dr. Rissenberg noted that the results provide preliminary support for broadening the CDC diagnostic criteria, extending antibiotic treatment in children, and conducting careful neuropsychological evaluation and educational monitoring. Development of educational programs for the identification, accommodation, and remediation of Lyme disease-related academic difficulties is critical as the numbers of children with the disease increases. Lyme disease-related cognitive deficits represent acquired, as opposed to developmental, learning disabilities and attentional disorders. Educational services and modifications should include, when necessary, reduction of homework, extended time for tests, provision of classroom notes and course outlines, instruction in organizational, time management, and study skills strategies, availability of abridged or tape-recorded books, shortened schoolday, and home instruction. Support may be necessary even after treatment for Lyme disease has been completed. Education of teachers and other school personnel regarding the educational impact of Lyme disease, as well as resources for parents, must be available. Dr. Rissenberg noted that further study is needed, using larger groups and more stringent controls, of the cognitive and academic functioning, physical and psychiatric symptomatology, and treatment response in children with Lyme disease.

References

Copyright © 2000 Medscape Portals, Inc.