Proton magnetic resonance spectroscopy (1H-MRS) of the brain in patients with tick-borne encephalitis

Tick-borne encephalitis (TBE) is a disease caused by a tick-borne encephalitis virus (TBEV) belonging to the Flaviviridae family. The aforementioned virus is transmitted by the bite of infected ticks. In the recent years, TBEV has become a serious public health problem with a steady increase in its incidence, mainly due to the climate changes and spreading the infected ticks into new territories. The standard protocol of TBE diagnosis involves the serological laboratory test with a minor role of imaging techniques such as magnetic resonance imaging. Long-term complications affecting patients daily activities are reported in about 40–50% of the cases. However, no changes are revealed in the laboratory tests or the imaging examination. The development of new imaging techniques such as proton magnetic resonance spectroscopy (1H-MRS) can broaden the knowledge about TBE, contributing to its prevention. The aim of this study was to assess the usefulness of 1H-MRS of the brain in patients with TBE. Compared to controls, a statistically significant decrease in the N-acetylaspartate /creatine ratio was found bilaterally in the right and left thalamus as well as a statistically significant increase in the choline/creatine ratio in the right and left thalamus.

of leukocytes in the second encephalic phase 16 . The key role in diagnosing TBE plays determination of TBEV specific IgM and IgG serum antibodies obtained by means of a highly specific immunosorbent assay (ELISA), while polymerase chain reaction (PCR) helps in the virus detection only in the first viral phase of the infection. Additionally, the cerebrospinal fluid (CSF) analysis may reveal pleocytosis.
Due to the limited spatial resolution, CT is considered to be less sensitive than MR and has an auxiliary role before the lumbar puncture to exclude brain oedema. To this time there has been a limited number of reports involving macroscopic CNS changes exhibiting hypodense areas in the CNS [17][18][19] .
The purpose of this study was to evaluate whether patients with TBE may develop brain metabolite alterations within the brain regions without any macroscopic changes on the conventional MR examination, especially, considering the fact that in about 40-50% of the cases, long-term complications affecting daily activity are reported without any changes in the laboratory test or the imaging examination. It is important to note that to this date and to our knowledge no major studies describing the features of TBE in 1H-MRS have been conducted in people. We hypothesized that since TBEV infects the brain, it may lead to alterations in the cerebral metabolism measured by 1 H-MRS.

Materials and Methods
study subjects and protocol of proton magnetic resonance spectroscopy. The present study was carried out in 25 patients diagnosed with TBE and 25 healthy controls. Only patients with meningoencephalitis and meningoencephalomyelitis, were included in the study excluding those with meningitis, since they presented no neurological symptoms,thus no 1H-MRS abnormalities were expected. The clinical course of the disease, age and sex distribution are presented in Table 1. No statistically significant difference was found in age and gender distribution between the two groups. The inclusion criteria were based on the European Centre for Disease Prevention and Control (ECDC) guidelines and consisted of a typical two phase clinical course of TBE, detection of IgM and/or IgG antibodies in blood serum and CSF by ELISA as well as granulacytosis in the CSF examination.
protocol of MRI and proton magnetic resonance spectroscopy ( 1 H-MRs). All patients underwent a routine MRI protocol with a single 3.0 Tesla MRI scanner (Biograph mMR, Siemens Healthineers, Erlangen, Germany), using a 16-channel head matrix coil.
Scan was performed between 10 th and 14 th day of the disease. Routine imaging was carried out prior to spectroscopy to exclude any structural abnormalities and included T1W (TR, 180 ms; TE, 2,5 ms; section thickness The preparation phase of 1 H-MRS included automatic procedures for water suppression, shimming, and tuning of the radiofrequency and gradient system and an acquisition without water suppression for correction of magnetic field distortions. 1 H-MRS planning was acquired on T1W sequences (TR; 18 ms; TE; 4,97 ms, section thickness 1 mm; intersection gap 0,2 mm; 160 saggital slices; FOV [230 × 230]; and matrix size 256 × 256) from this sequence, coronal and axial planes were reconstructed with the same technical parameters. 1 H-MRS examinations were performed with the following acquisition parameters: (TR/TE: 2000/135 ms, bandwidth 1200 Hz, number of excitation, 128). A point-resolved spectroscopy sequence (PRESS) was used for the rectangular volume of interest (VOI) selection. Single VOIs were manually located bilaterally within the normal-appearing brain tissue in the basal ganglia, thalami and in the hemispheres of the cerebellum with the size of VOI of 15 × 15 × 15 mm (Fig. 1). VOIs were positioned aiming to cover the regions that had been investigated for the www.nature.com/scientificreports www.nature.com/scientificreports/ greatest viral accumulation shown in the previous studies 35 . The localization of VOIs was confirmed by three orthogonal MR images in the axial, sagittal, and coronal planes. Metabolite ratios were calculated based on the peak integrals of N-acetylaspartate (NAA, 2.01 ppm), creatine (Cr, 3.03), choline-containing compounds, (Cho, 3.23 ppm). The metabolite concentrations (Fig. 2) were analysed using LCmodel (Provencher, 1993) quantification algorithm 36 .
The MR images were interpreted by a senior MR radiologists (with 20 years of experience in MRI).
statistical analysis. The significance of differences in the metabolite ratios between TBE and control group were estimated with the unpaired Welch's t test with unequal variances. Since the total of TBE patients and control group is limited, the p-values were bootstrapped for the two tailed t test with 10000 replicates. A value of p < 0.05 was considered statistically significant. The robustness of the results was confirmed by means of the nonparametric methods (Kruskal-Wallis test, Wilcoxon rank sum test). The analysis was performed with Stata 15 software.

Results
The NAA/Cr ratio was significantly lower in the left and right thalamus (bootstrapped p = 0.000 and p = 0.001, respectively). Additionally, the difference was also found in the right basal ganglia confirmed by both the one-sided t-test and nonparametric methods. A statistically significant increase in the Cho/Cr ratio was observed in the left thalamus and left cerebellar hemisphere (bootstrapped p = 0.007 and p = 0.016 respectively), as shown in Table 2. There was no statistically significant brain metabolite changes in other study areas.
The examination was performed between 10th and 14th day of the disease. No macroscopic CNS lesions were found in the study group after MRI examination, apart from one patient presenting the dimmed nonspecific lesion observed in the left frontal lobe and bilateral diffused hyperintense lesions in the posterior horns of the lateral ventricles.

Discussion
Findings presented in our study suggest that the TBEV infection may lead to metabolite alterations in the CNS, which can be in turn associated with the generation and persistency of psychiatric symptoms. The metabolite changes in the CNS of patients affected by TBEV could explain their clinical state, and can possibly affect their further loss of life quality. To the best of our knowledge, no 1 H-MRS studies carried out in patients with TBE have been reported. Only one report on the examination performed in veterinary medicine involving dog models has been found 37 . The authors of that paper showed changes from a mild to moderate decrease in NAA and Cr peaks, and a mild increase in glutamate/glutamine peaks. It is worth mentioning that the TBEV infection is characterised by the distribution of the flavivirus (FV) attachment factors and receptors (GAGs, HSPGs, CD81, claudins, DC-SIGN) promoting FV entry into the neurons, glia and non-CNS tissues, which facilitates FV entry into these cells via a multi-step process, sometimes involving one or more of these attachment factors 38 . This leads to the rapid microglial/macrophage activation, release of soluble inflammatory mediators and over expression of MHC class I molecules as well as the recruitment of cytotoxic T cell, leading to three possible mechanisms of tissue destruction: by direct neuronal injury, by induction of an inflammatory response inducing the neuronal death, or a combination of these two mechanisms. The study revealed the following principal findings: NAA/Cr levels decreased bilaterally in the thalamus and the right-sided basal ganglia, while Cho/Cr levels increased in the left thalamus and left cerebellar hemisphere 39,40 . NAA is a marker of neuronal density and integrity, so the decreased values of NAA in the examined area prove that the axonal injury has taken place, even without any macroscopic changes in a routine MR examination both in the thalami and in the basal ganglia. A statistically significant increase in Cho/Cr was observed in the left thalamus and left cerebellar hemisphere. Choline as one of the components of the phospholipid metabolism of the cell membrane. Thus, disturbances in the Cho levels can cause problems with transporting and cell proliferation within the area affected. Moreover, acetylocholine, a    www.nature.com/scientificreports www.nature.com/scientificreports/ precursor of choline, is a neurotransmitter which plays a major role in expressing emotions, whereas fluctuations in the levels of choline and compounds containing choline can explain emotional disturbances. No correlations were found between the fluctuations of metabolite levels and macroscopic changes in the study group. Only one patient had a subcortical nonspecific dimmed lesion in the left frontal lobe and bilateral diffused hyperintense lesions in the posterior horns of the lateral ventricles. This can prove that metabolite level changes occur first before macroscopic changes can be seen.

Conclusions
The reduction of NAA/Cr and increase Cho/Cr ratio observed in patients that suffered from TBE can suggest a direct neurotropic activity of the TBEV. Our study has proved that 1 H-MRS can be a sensitive tool for assessing brain metabolite changes in patients with TBE.