Clinical And Experimental Investigation Of Nonalcoholic Thiamine Deficiency And Wernicke's Encephalopathy

Part 1. Clinical Characteristics and MRI Features of Nonalcoholic Wernicke's EncephalopathyObjective: Nonalcoholic Wernicke's encephalopathy (WE) is prone to be underestimated in clinical practice and this study is to improve its awareness and early accurate diagnosis.Patients and Methods: A retrospective review and analysis was conducted of twelve cases of nonalcoholic Wernicke's encephalopathy, consisting of clinical characteristics, MR imaging features, and outcomes after thiamine administration. All patients exhibited neurological manifestations. Initial symptoms of central nervous system (CNS) appeared generally within 14-34 days after fasting in 11 patients with therapeutic fasting and 42 days after taking less food and being injected intravenously glycose saline discontinuously in 1 patient. Initial CNS symptoms included dizziness (5 patients), drowsiness (2 patients), apathy (3 patients), memory disturbance (3 patients), walking unsteadiness (3 patients), psychotic symptom (2 patients), diplopia (1 patients), and decreased visual acuity (1 patient). When acute WE was suspected, there were 1 patient with decreased visual acuity binocularly, 1 patients with diplopia, 2 patients with psychotic symptom, 3 patients with ataxia, 3 patients with memory disturbance, 4 patients with abnormal ocular movement, 7 patients with nystagmus, and 10 patients with conscious disturbance consisted of 1 patients with drowsiness, 3 patients with lethargy, 4 patients with mild coma, and 2 patients with deep coma.Of all patients, only 2 patients had a typical triad (neurophthalmological manifestation, ataxia, and global confusion) and 3 patients had conscious disturbance only without focal neurological signs.All patients were immediately examined with T1- and T2-weighted imaging when acute Wernicke's encephalopathy was clinically suspected, 10 patients with FLAIR imaging, 4 patients with DWI imaging, and 3 patients (1 patient with drowsiness, 2 patients with mild coma) with enhanced scanning by gadolinium (Gd) contrast medium.The patients were immediately administered with 100mg of intravenous thiamine and, thereafter, with 100mg per day for 5-15 consecutive days intravenously or intramuscularly when acute WE was suspected. All patients were followed up for 1-2 years and were administered with 150mg Fursultiamine per day via the gastrointestinal tract during the follow-up with the exception of one patient who died and two patients who missed.The follow-up was performed once for every three months.Results: All 7 patients with mild coma or lethargy exhibited typical MR imaging of acute Wernicke's encephalopathy that showed areas of increased T2-weighted and/or FLAIR signals symmetrically surrounding the aqueduct and the third ventricle, at the floor of the fourth ventricle, and in the medial thalami. Only increased signal intensities on T2-weight and/or FLAIR images surrounding the aqueduct were observed in 3 patients without conscious disturbance or only with drowsiness. Two patients with deep coma exhibited increased T2-weight and FLAIR signal intensities not only in bilateral medial thalami but also in motor cortex. Four patients (2 patients with mild coma and 2 with deep coma) companied abnormal signal intensity in the capita of the caudate nuclei. Atrophy of the cerebellar vermis and mammillary bodies was not observed in all patients.Among 4 patients with DWI imaging, 3 patients with mild coma exhibited only slightly increased signal intensities within bilateral thalami. No abnormal signal intensity was found in other brain regions and in the other patient without conscious disturbance. Among 3 patients with enhanced scanning by Gd contrast medium, 2 patients with mild coma were observed enhancing signal of mammillary bodies, which indicated mammillary bodies underwent acute disruption of the blood brain barrier during acute WE, and non-enhancing signal of the other brain zones. The other patient with drowsiness exhibited slight enhancing signal at the floor of the fourth ventricle.Only slight abnormal signal intensity on DWI images in 3/4 patients was seen, which showed DWI imaging may be insensitive to WE lesion.The follow-up showed 8 patients without deep coma and cortex damage recovered fully both in clinical and cranial MRI aspects within the duration of twelve months follow-up. It was worthy notice that the follow-up MRI of all 8 patients did not exhibit atrophy of mammillary bodies and cerebellar vermis. Two patients with deep coma and cortex damage had bad prognosis. One patient died 15 days after initial neurological manifestations, and the other, whose MRI follow-up exhibited progressive atrophy of whole brain, entered permanent vegetative state. It showed deep coma and cortex damage might be indicative of irreversible damage and poor prognosis.Conclusion: 1. Most patients with nonalcoholic WE exhibited atypical clinical process and are prone to be misdiagnosed.2. To reduce the misdiagnosis of nonalcoholic WE, all patients should be suspected as nonalcoholic WE when they present vertigo, ataxia, nystagmus, neurophthalmological manifestation, disturbed memory, and unconsciousness under the condition of potential thiamine deciency such as fasting, prolonged vomit, and anorexia.3. Cranial MRI is valuable not only to diagnose nonalcoholic WE but also to evaluate the pathologic evolution and prognosis of the disorder.4. The involvement of the caudate nucleus and cortex is not an uncommon presentation but rather the sign of pathologic evolution.5. It is different from alcoholic WE that nonalcoholic WE can be cured without any sequela apart from the patients with deep coma and cortic damage.Part 2. Establishment of Mouse Model of Nonalcoholic Thiamine Deficiency and Wernicke's Encephalopathy.Objective: To establish mouse model and to explore pathological temporal orderliness and staging of nonalcoholic thiamine deficiency (TD) and WE.Method: Two methods of thiamine-depleted diet combined with intraperitoneal injection of pyrithiamine (TDp) and only thiamine-depleted diet (TDo) were utilized to produce nonalcoholic TD mouse model. With or without intraperitoneal pyrithiamine injection daily, TD mice received a thiamine-depleted diet and distilled water ad libitum. Control animals received a thiamine-containing diet and distilled water ad libitum. To reverse TD, TDo mice were intraperitoneally re-administered a thiamine injection at 100mg/kg body weight in 0.1ml saline every day for 5 consecutive days following TDo9 and TDo 16.The learning ability of TD mice was tested with Y-maze on 2nd, 9th, 16th, and 20th day of TD. The Y-maze test was repeated five days after the first test (7th, 14th, 21st, and 25th) and the percentage of correct response in 10 consecutive tests was used to compare the memory. Simultaneously, H-E staining, Nissl staining, ultrastructural study, and immunohistochemical staining with ChAT were conducted.Results: Significant decrease of learning ability in TD mice began from TDp9 or TDo9. TDo mice manifested memory disturbance since TDo21. The number of total trials of TDo9 mice was 23.25±0.86, a significant increase compared with that of the controls (13.50±1.28) (P<0.01). Following prolonged exposure to TD, the impaired learning ability exacerbated gradually and culminated on TDo20 when the mice could not reach the learning criterion even after 100 training trials. Five days after learning training, the correct percentages of the responses were recorded to determine memory status of TDo mice. The results showed a significant difference on TDo21 from that of the controls (5.17±0.40 vs 9.17±0.40, p<0.01) and no difference was seen on and before TDol4.Pathological lesion (Nissl and H-E staining), ultrastructural alteration, and the loss of cholinergic neurons were not found both TDo9 and TDp9. The mice manifested abnormal right reflex when hemorrhage damage of medial thalamus, hypothalamus, and pons with H-E and Nissl staining and the loss of cholinergic neurons with immunohistochemical stining were observed on or after TDp13-14. However, only 10-20% of TDo mice exhibited pathological lesion of medial thalamus till TDo20.Thiamine supplement could reverse decreased learning ability at TDo9 but could not at TDo 16.Conclusion: 1. The mouse model of nonalcoholic thiamine deficiency and WE could be produced through simple thiamine-depleted diet or thiamine-depleted diet combined with intraperitoneal pyrithiamine injection daily. Thalamus and hippocampus are susceptible to TD, and cortices are not.2. TDp mouse model exhibited typical symmetrical hemorrhage damage of medial thalamus, but only 10-20% of TDo mouse model showed typical symmetrical hemorrhage damage of medial thalamus.3. TD mice manifested decreased learning ability at early pre-pathological lesion stage of TDo9 and TDp9. Existing accepted mechanism of cognitive dysfunction induced by TD could not explain this phenomenon.Part 3. The Study of Pathogenetic Mechanism of Nonalcoholic Thiamine Deficiency and Wernicke's EncephalopathySection 1. Effect of Thiamine Deficiency on Hippocampal Long-term PotentiationObjective: To explore the effect of TD on hippocampal long-term potentiation at early pre-pathological lesion stage so that to uncover the relationship between cognitive dysfunction induced by TD and sypnatical plasticity of hippocampus associated to glutaminergic neural pathway.Method: Extracellular recordings were performed in the CA1 and CA3 region of the hippocampus. The stimulating and record electrodes were placed at the Schaffer collateral pathway in the stratum radiatum in CA1 region. LTP of CA3 region was evoked by direct stimulation of the mossy fiber bundle and recorded in the stratum lucidum.Results: In CA1 and CA3 region of hippocampus, theta burst stimulation could induce a robust LTP in both control and TD9 mice without significant difference.Conclusion: No significant defference of CA1 and CA3 LTP was observed between control group and TDo group at the early pre-pathological lesion stage. The result showed that sypnatical plasticity of hippocampus associated to glutaminergic neural pathway was not involved in cognitive dysfunction induced by TD at early pre-pathological lesion stage.Section 2. Effect of Thiamine Deficiency on Hippocampal NeurogenesisObjective: To explore the effect of TD on hippocampal neurogenesis (including cell proliferation, differentiation, mature and survival) and the correlation of cognitive dysfunction induced by TD and hippocampal neurogenesis.Methods: Hippocampal neurogenesis was studied with BrdU, PCNA, Dcx and NeuN immunohistochemical staining on 7th, 9th, 14th, and 25th day of thiamine deficiency. Thiamine was injected for 5 days to the thiamine deficiency animal model that appeared down-regulated neurogenesis. In addition, BrdU and Dcx immunohistochemical staining was done to study whether decreased hippocampal neurogenesis was reversible.Results: BrdU, PCNA, and Dcx immune activity in the hippocampal dentate gyrus declined significantly from the 9th day of thiamine deficiency than the normal controls. The survival rate of BrdU-positive cells also decreased significantly, while NeuN immune activity showed no obvious change between thiamine deficiency mouse model and normal controls. BrdU and Dcx immune activity showed significant increase with five days thiamine supplement at 9th day of thiamine deficiency compared with the controls without thiamine supplement, while thiamine supplement at 16th day of the mouse model could not reverse down-regulated hippoampal neurogenesis.Conclusion: Thiamine deficiency down-regulates hippocampal neurogenesis, including cell proliferation, differentiation, and survival, while little influence is shown to the mature cells. Furthermore, impaired hippocampal neurogenesis is reversible at early stage of thiamine deficiency exposure, while the late stage is unreversible.Section 3. Effect of Thiamine Deficiency on the Activities of brain pyruvate dehydrogenase,α-ketoglutarate dehydrogenase, and transketolaseObjective: To study the effect of thiamine deficiency on the activities of brain pyruvate dehydrogenase (PDH),α-ketoglutarate dehydrogenase (KGDH), and transketolase (TK) as well as the content of NADP~+ and NADPH so that to explore the mechanism of specific brain regional damage along brain medline induced by thiamine deficiency.Method: TD mouse model was produced according to previous described methods. The mice were divided into TD9, TDM, TD21 subgroups and normal control group. There were five mice in each subgroup and normal control group. After being anesthesized and decapitated, mouse brain was taken out rapidly. The brain mitochrondria were extracted according to the method reported by Stephan. The activities of PDH, KGDH, and TK were determined according to the methods reported by Humphries, Hinman, and Hecqquet respectively. The content of NADP~+ and NADPH was determined according to the method reported by the literature.Result: Compared with normal control group, The activities of brain PDH of TD9 group, TD14 group, and TD21 group were 99.63±1.88%, 95.12±1.22%, and 95.07±0.76% respectively and no statistical difference was found (P>0.05). The activities of brain KGDH of TD9 group, TD14 group, and TD21 group were 99.20±1.91%, 91.45±1.24%, and 88.34±1.53% respectively. There was a statistical difference between normal control group and TD21 group (P<0.05) . There was no statistical difference between normal control group and TD9 group and TD14 group (P>0.05) . The activities of brain TK of normal control group, TD9 group, and TD14 group were 10.90±0.67, 10.46±0.94, and 9.19±1.84 U/g protein. There was a statistical difference between TD14 group and normal control group (P<0.05) . Ther was no statistical difference between normal control group and TD9 group. The content of brain NADP~+ of normal control group, TD9 group, and TDM group was 20.22±3.67, 17.89±3.54, and 18.45±2.54 pMol/L respectively and no statistical difference was found among them. The content of brain NADPH of normal control group, TD9 group, and TD14 group was 9.37±1.57, 7.17±0.73, and 7.43±1.11 pMol/L and the rate of NADP~+ and NADPH (NADP/NADPH) was 2.17±0.21, 2.49±0.34, and 2.49±0.10. The content of NADPH and the rate of NADP and NADPH of TD9 group and TD14 group decreased significantly compared with normal control group (P<0.05) .Conclusion: It hardly explains exactly why TD induces regionally selective damage of brain by present putative mechanism that is related to the failure of energy metabolism and oxidative stress due to decreased activities of PDH andα-KGDH. The disturbance of pentose phosphate pathway due to decreased activity of transketolase induced by TD may be involved in the mechanism of regionally selective damage.