Proton Nuclear Magnetic Resonance (NMR) Metabolomics: Future in Etiological Diagnosis of Encephalomyelitis.

Abstract

Encephalomyelitis is relatively rare (10.5–13.8/100,000 children) but may be fatal, and frequently causes permanent disability. The cause remains unknown in more than half the cases [1,2]. Cardinal symptoms of encephalitis include fever, headache, neck stiffness, photophobia, confusion and seizures. Clinical examination will reveal altered mental status ranging from somnolence to lethargy and coma. Delirium, cranial nerve palsies, ataxia, rash and sign of increased intracranial pressure may be seen. It is not always possible to ascertain the specific cause of encephalitis from clinical presentation alone. However there are features that can give some clue to possible etiology. Arbovirus encephalitis often heralds with a flu-like syndrome followed by increasing confusion and stupor. Herpes simplex encephalitis often presents with an abrupt change in behavior, memory loss, focal or generalized seizures, and speech concerns. West Nile encephalitis may present with flaccid, asymmetric motor weakness with altered sensorium of variable degree. Varicella may manifest with cerebellar features as Japanese encephalitis presenting with basal ganglia symptoms. The current evaluation of encephalomyelitis requires a battery of tests including neuroimaging, expensive and often limited by the inability to obtain enough specimens in pediatric population along with unacceptable turnaround time. It is warranted to explore newer methodologies to establish early diagnosis, which is detrimental to favorable clinical outcome. Proton NMR Metabolomics The concept that biological fluids reflect the health of an individual has existed for a long time. Nuclear Magnetic Resonance (NMR) spectroscopy is based on measuring the absorption of light (radio waves) due to changes in nuclear spin orientation of molecules of different metabolites. Proton nuclear magnetic resonance (NMR) metabolomics can be used to study metabolic profile of cerebrospinal fluid and urine. NMR is fully quantitative, highly reproducible, and detects all metabolites simultaneously in one snapshot. The samples are completely recoverable. Distinct Cerebro Spinal Fluid (CSF) metabolomics profile for normal controls, human rabies,6 West Nile encephalitis, and Lyme meningitis have been well described. One can correlate CSF and urine metabolomics with clinical course, imaging, and laboratory findings, to develop a rapid screen to differentiate infectious from auto-inflammatory and autoimmune causes of encephalomyelitis by cluster analysis. We have already used proton nuclear magnetic resonance (H+-NMR) to identify and quantify 56 metabolites from normal and diseased CSF using 0.5 ml of CSF in under 2 hours, and discriminate a CSF metabolomics profile by unsupervised (unbiased) cluster analysis. Contrast this with the standard diagnostic approach –almost a century old– of quantifying two (2) substances (glucose and protein) from the same CSF volume with similar turn-around time. Preliminarily, we can with high accuracy discriminate 6 central nervous system (CNS) diseases using NMR metabolomics. Contrast this with turn-around time of 4–7+ days for conventional encephalitis testing for oligoclonal bands, serology for EBV, Varicella Zoster Virus (VZV) or Lyme disease, and N-methyl-D-aspartate receptor (NMDAR), Voltage –gated potassium channel (VGKC) or aquaporin-4 autoantibodies. We conducted a study of CSF metabolomics comparing persons under treatment for rabies encephalitis, to normal controls. We were able to describe a metabolomic profile for human rabies across a number of weeks of illness. We also identified metabolic changes that correlated with clinical worsening or, alternatively, with survival. More recently, we compared CSF metabolomics profiles from patients without infection, rabies encephalitis, West Nile encephalitis, Lyme meningitis, fungal meningitis, malaria encephalopathy, and multiple sclerosis. CSF profiles clustered well and were surprisingly distinct between diseases. We hypothesize that these same findings may hold true for other forms of infectious encephalitis and will cleanly distinguish these from (ADEM) acute disseminated encephalomyelitis (approximating MS) We do not intend to supplant highly accurate and definitive testing for specific viruses or autoantibodies, but NMR might accelerate and focus initiation of effective therapies, to improve outcomes, and improve patient safety and financial risk by limiting expensive, CSF- or blood-consuming diagnostics.