Published ahead of print on April 22, 2010
doi: 10.3174/ajnr.A2113

American Journal of Neuroradiology 31:E54, June-July 2010
© 2010 American Society of Neuroradiology

L.C. Hygino da Cruz, Jr^a, R.C. Domingues^a, I. Vilanova^b and E.L. Gasparetto^c

^aClínica de Diagnóstico Por Imagem
^bClinic IRM
^cDepartment of Radiology
University Federal of Rio de Janeiro
Rio de Janeiro, Brazil

We read with special interest the article of Zuccoli et al¹ entitled “MR Imaging Findings in 56 Patients with Wernicke Encephalopathy: Nonalcoholics May Differ From Alcoholics.” The authors reported the typical and atypical imaging findings of Wernicke encephalopathy (WE). They suggested that alcoholic and nonalcoholic patients present different MR imaging patterns due to differences in the pathophysiologic process. Typical imaging findings, which were more common in alcoholic patients, consist of symmetric signal-intensity abnormalities in the thalami, mamillary bodies, tectal plate, and periaqueductal area, as well as contrast enhancement in the mamillary bodies and thalami, whereas signal-intensity alterations in the cerebellum, cerebral vermis, cranial nerve nuclei, red nuclei, dentate nuclei, caudate nuclei, splenium, and cerebral cortex represent atypical MR imaging findings and are more frequent in nonalcoholic patients. The authors speculated that alcohol might have a protective effect on the brain areas that show atypical lesions in WE, which could explain why atypical findings are seen only in nonalcoholic patients.

We report 2 cases of nonalcoholic WE: 1 presenting with typical and another with atypical MR imaging findings. The first case is a 21-year-old woman with non-Hodgkin lymphoma, having undergone chemotherapy and bone marrow transplant (BMT). Thirty-one days after BMT and receiving parenteral nutrition, she presented neurologic and visual disturbances first attributed to cyclosporine. However, MR imaging findings revealed, on T2-weighted images, high signal intensity symmetrically and bilaterally in the medialvestibular nuclei, facial nuclei, medial and pulvinar thalami (Fig 1A), and periventricular region of the third ventricle, as well as in the mamillary bodies. After gadolinium administration, there was enhancement in the mamillary bodies and in the facial nuclei (Fig 1B).

View larger version (131K): [in this window] [in a new window] Fig 1. Case 1. Coronal T2-weighted image (A) shows high signal intensity in the thalami (arrows) and facial nuclei (arrowheads), which demonstrates contrast enhancement (arrowheads) in the T1-weighted image (B) after gadolinium administration. Case 2. FLAIR images (C and D) show bilateral and symmetric areas of high signal intensity on the mamillary bodies (arrows), periaqueductal gray matter, thalami, and putamina.

The additional case was a 13-year-old boy with suppurative appendicitis, who underwent laparotomy complicated with mesenteric infarction, adhesion, and hematoma of the rectum. Thus, he underwent a colostomy, and after a month of parenteral nutrition, he developed a sixth-nerve palsy, seizures, and coma. The MR imaging demonstrated well-definedsymmetric and bilateral lesions hyperintense on T2-weighted and fluid-attenuated inversion recovery (FLAIR) images in thetectum of the midbrain, the periaqueductal gray matter, mamillary bodies, thalami, putamina, and periventricular region of the third ventricle, as well as in the frontal cortex. The lesions also presented restricted diffusion and no contrast enhancement. In both cases, the diagnosis of WE was suggested and appropriate therapy was initiated. However, the patients did not respond and died in the following days.

We report these cases of nonalcoholic WE demonstrating that both typical and atypical MR imaging findings may coexist. Although Zuccoli et al¹ suggested that contrast enhancement of the mamillary bodies and thalami indicate alcoholic WE findings, these findings were already demonstrated in nonalcoholic patients,² corroborating our cases. Our findings indicate that the WE physiopathology might not differ between alcoholic and nonalcoholic patients. Alcohol may not play as important a role as thiamine deficiency does in the whole process. Many clinical conditions can directly or indirectly cause thiamine deficiency, which will lead to WE.

Finally, as suggested by Zuccoli et al,^1,3 the correct diagnosis is the main challenge in the clinical approach to these patients. The initiation of the required treatment can avoid or delay the appearance of severe neurologic deficits and even death, no matter the cause of thiamine deficiency.

References

1. Zuccoli G, Santa Cruz D, Bertolini M, et al. MR imaging findings in 56 patients with Wernicke encephalopathy: nonalcoholics may differ from alcoholics. AJNR Am J Neuroradiol 2009;30:171–76. Epub 2008 Oct 22[Abstract/Free Full Text]
2. Fei GQ, Zhong C, Jin L, et al. Clinical characteristics and MR imaging features of nonalcoholic Wernicke encephalopathy. AJNR Am J Neuroradiol 2008;29:164–69[Abstract/Free Full Text]
3. Zuccoli G, Gallucci M, Capellades J, et al. Wernicke encephalopathy: MR findings in twenty-six alcoholic and non-alcoholic patients. AJNR Am J Neuroradiol 2007;28:1328–31[Abstract/Free Full Text]

Reply

Published ahead of print on April 22, 2010
doi: 10.3174/ajnr.A2116

American Journal of Neuroradiology 31:E56, June-July 2010
© 2010 American Society of Neuroradiology

G. Zuccoli^a

^aChildren’s Hospital of Pittsburgh
University of Pittsburgh Medical Center
Pittsburgh, Pennsylvania

We thank Dr L. Celso Hygino da Cruz, Jr and colleagues for their interest in our article.¹ We evaluated 56 alcoholic (43%) and nonalcoholic (57%) patients affected by Wernicke encephalopathy (WE),¹ showing that signal-intensity alterations in areas considered atypical for the disease were noted only in nonalcoholic patients and always in association with the typical findings. Furthermore, we noted that contrast enhancement of the thalamus and mamillary bodies was significantly associated with alcohol abuse. However,as described in our article, contrast enhancement was also present in nonalcoholic patients.

Thiamine deficiency leads to brain lesions selectively involving brain regions with high thiamine turnover within 14–21 days.² The blood-brain barrier breakdown as depicted by contrast enhancement begins 7–10 days after thiamine deficiency.³ We postulated that the higher incidence of contrast enhancement in the mamillary bodies and thalami observed in alcoholic compared with nonalcoholic patients underlines a selective anatomic susceptibility to the toxic effects of alcohol.¹ In alcoholism, the low thiamineabsorption rate at the mucosal level, the impaired hepatic function, and the alcohol-related raised thiamine metabolism together may lead to the development of chronic thiamine deficiency. Thus, in alcoholism, a pre-existing status of chronic subclinical thiamine deficiency may accelerate the blood-brain barrier breakdown, resulting in a higher incidence of contrast enhancement at the time the patient is imaged.¹

Most interesting, each of the 2 patients presented by Dr Hygino da Cruz, Jr and colleagues had peculiar patterns for WE. Infact, the first patient showed involvement of the pulvinar of the thalami associated with lesions typical of the disease,confirming that not only the medial nuclei but also the posterior (pulvinar) nuclei of the thalami may be involved in WE.⁴ Thesecond patient, a 13-year-old boy, demonstrated involvement of the putamina, which is a typical finding in pediatric patientsaffected by WE.⁵ To explain basal ganglia selective involvement in pediatric patients with WE, we speculated that during development, the rate of thiamine-dependent metabolism is increased.⁵

In conclusion, WE shows a wide spectrum of neuroradiologic presentations, with the constant presence of typical alterations involving the mammillary bodies, thalami, and periaqueductal gray matter. At least 1 of the typical findings is invariably present in alcoholic and nonalcoholic patients.¹ Although a prevalence of contrast enhancement is observed in the alcoholic population,¹ this may be related to a pre-existing state of chronic thiamine deficiency, not just to the toxic effects of alcohol. In the clinical setting of WE, basal ganglia involvement does represent a typical finding of the disease in the pediatric age group.⁵

References

1. Zuccoli G, Santa Cruz D, Bertolini M, et al. MR imaging findings in 56 patients with Wernicke encephalopathy: non-alcoholics may differ from alcoholics. AJNR Am J Neurordiol2009;30:171–76. Epub 2008 Oct 22
2. Tanphaichitr V. Thiamin. In: Shils ME, Olson JA, Shike M, et al eds. Modern Nutrition in Health and Disease, 9th ed. Baltimore: Williams & Wilkins; 1999:381–89
3. Sechi G, Serra A. Wernicke’s encephalopathy: new clinical settings and recent advances in diagnosis and management. Lancet Neurol 2007;6:442–55[CrossRef][Medline]
4. Stone R, Archer JS, Kiernan M.J. Wernicke’s encephalopathy mimicking variant Creutzfeldt-Jakob disease. Clin Neurosci 2008;15:1308–10[CrossRef]
5. Zuccoli G, Siddiqui N, Bailey A, et al. Neuroimaging findings in pediatric Wernicke encephalopathy: a review.Neuroradiology 2009 10 21. [Epub ahead of print]