Sensitivity and specificity of spinal cord Magnetic Resonance Imaging in the diagnosis of HTLV-1 associated myelopathy

Vestibular Evoked Myogenic Potential on Ocular, Cervical, and Soleus Muscles to Assess the Extent of Neurological Impairment in HTLV-1 Infection

Introduction: Vestibular Evoked Myogenic Potential (VEMP) can be used to test central vestibular pathways from the midbrain to the lumbar spine, according to the muscle tested.

Purpose: to compare the spinal cord alteration in individuals with HTLV-1-associated myelopathy (HAM) and with HTLV-1-asymptomatic infection using the VEMP recorded from different muscles.

Methods: VEMP was recorded in 90 individuals of whom 30 had HAM, 30 were HTLV-1 asymptomatic carriers, and 30 negative controls. VEMP was recorded in the oculomotor muscle (oVEMP), testing the vestibulo-ocular reflex, and in the cervical muscle (cVEMP) and soleus muscle (sVEMP), testing the vestibulospinal reflex, respectively, in the cervical and in the lumbar spinal level. The type of stimulation was auditory for oVEMP and cVEMP, and galvanic for sVEMP. The compared variables were the latencies of the electrophysiological waves.

Results: HTLV-1-asymptomatic group was similar to the controls regarding oVEMP (p = 0.461), but different regarding cVEMP (p < 0.001) and sVEMP (p < 0.001). HAM group has presented the worst latencies and was different from the HTLV-1-asymptomatic group in the VEMP of all the tested muscles (p < 0.001). The concomitant occurrence of VEMP alterations in the three recorded muscles of the same individual was found in 2 (6.7%) asymptomatic carriers and in 20 (66.7%) patients with HAM (p = 0.001). The analysis of VEMP alteration per group and per muscle has showed that, in HTLV-1-asymptomatic group, oVEMP was altered in 3 (10.0%) individuals, cVEMP in 10 (33.3%) and sVEMP in 13 (43.3%). In HAM group, oVEMP was altered in 23 (76.6%) individuals, cVEMP in 27 (90%), and sVEMP in 30 (100%).

Conclusion: HTLV-1-neurological damage has followed an ascendant progression beginning at the lumbar spine in the stage of a clinically asymptomatic infection, whereas HAM has affected not only the spine, but also the midbrain.

Spinal cord hypometabolism associated with infection by human T-cell lymphotropic virus type 1(HTLV-1)

Background

HTLV-1 infection is endemic in Brazil. About 1 to 2% of the Brazilian population is estimated to be infected, but most infected HTLV-1 individuals do not know about their own infection, which favors the continuity of sexual and vertical virus transmission. In addition, HTLV-1 associated central nervous system diseases and their pathophysiologic mechanisms are not fully understood. This study aimed to evaluate the correlation of spinal cord metabolism, viral and inflammatory profiles with features of neurological presentation in HTLV-1 infected individuals.

Methodology

This is a cross-sectional study of a cohort including 48 HTLV-1 infected individuals clinically classified as asymptomatic-AG (N = 21), symptomatic-SG (N = 11) and HAM/TSP-HG (N = 16) and a nested case-control study with HTLV-1 infected individuals-HIG (N = 48) and HTLV-1 non infected controls-CG (N = 30) that had their spinal cord analysed by Positron Emission Tomography with 18F-Fluordeoxyglucose (18F-FDG PET/CT). HTLV-1 infected individuals had 18F-FDG PET/CT results analyzed with clinical and demographic data, proviral load, cytokines and chemokines in the blood and cerebrospinal fluid (CSF).

Principal Findings

18F-FDG PET/CT showed hypometabolism in the thoracic spinal cord in HTLV-1 infected individuals. The method had an accuracy of 94.4% to identify HAM/TSP. A greater involvement of the thoracic spinal cord was observed, although hypometabolism was also observed in the cervical spinal cord segment in HTLV-1 infected individuals. Individuals with HAM/TSP showed a pro-inflammatory profile in comparison to asymptomatic and symptomatic groups, with a higher level of Interferon-inducible T-cell alpha chemoattractant (ITAC/CXCL₁₁), IL-6, IL-12p70 in the plasma; and ITAC, IL-4, IL-5, IL-8 (CXCL₈) and TNF-alpha in the CSF. Using regression, thoracic spinal cord SUV (standardized uptake value) and CSF ITAC level were identified as the HAM/TSP predictors in the multivariate model.

Conclusions

18F-FDG PET/CT imaging showed spinal cord hypometabolism in most HTLV-1 infected individuals, even in the asymptomatic HTLV-1 group. Thoracic spinal cord hypometabolism and CSF-ITAC levels were identified predictors of HAM/TSP.

Significance

Our findings suggested that in most HTLV-1 infected individuals there was compromise of central nervous system (CNS) structures despite of the lack of clinical symptoms. To explain the found hypometabolism, the role of microcirculatory and metabolic factors in the pathogenesis of neurological diseases associated with HTLV-1 infection must be further investigated. It is paramount to evaluate the central nervous function and to compare the performance among HTLV-1 infected individuals considered asymptomatic to the uninfected controls.

For the past 30 years, human T-cell lymphotropic virus type-1 (HTLV-1) has been isolated and associated with neoplastic, inflammatory, and infectious diseases. It is known that the neurological disorder associated with HTLV-1 comprises HTLV-1-associated myelopathy (HAM/TSP) or any other isolated signals and symptoms. Despite all the knowledge accumulated so far, the association of neurological diseases to HTLV-1 infection remains difficult and neglected. We designed this study in order to assess the degree of neurological impairment associated with HTLV-1 infection through a metabolic evaluation with 18F-FDG PET/CT. Our results evidenced a more pronounced hypometabolism in the spinal cord of individuals with neurological impairment, but also evidenced hypometabolism in asymptomatic HTLV-1 infected individuals. We believe that areas of the CNS with lower circulatory and perfusional balance are more vulnerable to HTLV-1 infection. Mechanisms of cellular entry of the virus may be associated with loss of microcirculatory homeostasis and predisposition to a breakdown of the blood-brain barrier in these areas. Further studies are still necessary to shed light on the mechanisms associated with brain and spinal hypometabolism.