Articles

Patterns of sensory abnormality incortical stroke

Evidence for a dichotomized sensory systemJong S. Kim, MD

Abstract—Objective: To characterize sensory symptoms in patients with stroke occurring in the cerebral cortex. Methods:We studied clinical and imaging findings of 24 patients who had prominent sensory symptoms without definitive motordysfunction. Results: According to the sensory manifestations, patients were divided into dominant impairment of primi-tive sensation (DIPS) group, dominant impairment of cortical sensation (DICS) group, and paresthesia-only group. DIPSwas related to lesions involving the parietal operculum and the insular cortex, whereas DICS was related to the lesionsaffecting the postcentral gyrus. Patients with paresthesia only had smaller lesions located in the postcentral gyrus. DIPSgroup patients were more often women (p � 0.013), more often had dysarthria (n � 0.043), and more often developedcentral poststroke pain or paresthesia (n � 0.005) than the DICS group patients. Restricted sensory changes are common,predominantly involving the perioral or finger areas. Conclusions: Sensory patterns in these patients are generallyconsistent with the dichotomized (SI and SII) sensory system in the cerebral cortex. Involvement of insular and opercularareas is related to primitive sensory impairment and development of central poststroke pain, whereas postcentral gyrusinvolvement is related to cortical sensory loss without poststroke pain. The pattern of restricted sensory changes isgenerally consistent with the Penfield sensory topography.

NEUROLOGY 2007;68:174–180

Pure or predominant sensory symptoms are pro-duced by strokes occurring anywhere in the sensorytract.1 They most often occur in patients with tha-lamic,2,3 pontine,4 or lenticulocapsular5 strokes. Cor-tical strokes producing pure or predominant sensorysymptoms are rare, and clinical–radiologic correla-tion has not been properly conducted.2,6,7 Moreover,there is debate regarding the occurrence of centralpoststroke pain or paresthesia (CPSP), a distressingsequela of sensory strokes,8 in patients with corticalstroke.6,9 In this study, we sought to characterize thesensory syndrome in patients with stroke occurringin the cerebral cortex.

Methods. Between April 1997 and December 2005, I personallyexamined 24 consecutive patients who were admitted to the AsanMedical Center presenting with sensory disturbances secondary tostrokes occurring in the cerebral cortex. Excluded were patientswith 1) definitive (�IV/V) limb weakness, 2) a previous strokethat could have affected the current neurologic symptoms, 3) con-comitant infarcts in the other location such as the thalamus or the

pons, 4) unreliable sensory examination due to any reasons suchas severe aphasia, dementia, or confusion, and 5) lesions primarilylocated at the subcortical area. However, other symptoms such asdysarthria, facial palsy, aphasia, or visual field defects were al-lowed as long as these symptoms did not result in inaccuratesensory testings.

Superficial pain (pinprick), cold (using cold tuning fork), andvibration (128-Hz tuning fork) sensations were assessed andgraded as mild (�70% of the normal side), moderate (30 to 70%),and severe (�30%). Position sense was assessed by identificationof up and down passive movements of fingers or toes and forwardand backward movements of the earlobe. Deficits in position sen-sation was graded as “mild” when patients correctly identified themovements, but with a feeling of “unclear” sensation; “moderate”when there was an occasional error; and “severe” when most ofthe movements could not be identified. Stereognosis (identificationof coins, ball-pen, or key), graphesthesia (recognition of letterswritten on the palm), and two-point discrimination were per-formed and compared with the intact side.

Dominant impairment of primitive sensation (DIPS) was de-fined when 1) there was impaired pain or temperature sensoryperception with intact perception of position sense, 2) degree ofpain and temperature sensory deficit was definitely severe thanthat of position sense, or 3) position and other cortical sensory

Commentary, see page 167

From the Department of Neurology, University of Ulsan, Asan Medical Center, Seoul, Korea.Supported by a research fund from the Korean Ministry of Health and Welfare (03-PJ1-PG1-CH06-0001) and a grant from Brain Research Center of the 21stCentury Frontier Research Program funded by the Ministry of Science and Technology of Korea (M103KV010010 06K2201 01010).Disclosure: The author reports no conflicts of interest.Received April 17, 2006. Accepted in final form October 4, 2006.Address correspondence and reprint requests to Dr. J.S. Kim, Department of Neurology, Asan Medical Center, Song Pa PO Box 145, Seoul 138-600, Korea;e-mail: jongskim@amc.seoul.kr

174 Copyright © 2007 by AAN Enterprises, Inc.

deficits improved out of proportion to pain and temperaturesenses during follow-up even if both senses were similarly affectedinitially. Dominant impairment of cortical sensation (DICS) wasdefined when the opposite was the case.

During follow-up visits, the presence of distressing pain orparesthesia was assessed. The severity of the sensory symptomswas assessed using visual numeric scale (1 � slight, 10 � mostsevere). CPSP was defined when either the visual numeric scaleswas �5 or the patients required pain-relieving medication such asamitryptiline, carbamazepine, or gabapentine. The CPSP was re-corded as burning (including sensations of heat, warmth, flush-ing), cold (including sensations of cool, icy, freezing), numb(including sensations of tingling and tight), and painful (includingsensations of shooting, lancinating, squeezing, stabbing).10

All patients except for one (Patient 12) underwent conven-tional T2- and T1-weighted MRI as well as fluid-attenuated inver-sion recovery images within 10 days after the onset of symptoms.Diffusion-weighted MRI (single-shot echo planar spin echo se-quence, b � 1,000 s/mm2, repetition/echo time 5,000/139 millisec-onds) was additionally performed in 14 patients. Patient 12 whohad intracerebral hemorrhage underwent CT only. Imaging anal-ysis was done by a neuroradiologist who was blinded to the clini-cal data. Angiograms were performed in 23 patients: MRangiogram in 22 and conventional angiogram in 1 patient. EKGs

were performed in all patients and echocardiograms in selectedpatients.

The pathogenic mechanism of cerebral infarction was catego-rized as follows: 1) Cardiogenic embolism: there was an emboli-genic heart disease such as atrial fibrillation, valvular disease,dilated cardiomyopathy, and acute myocardial infarction. Patentforamen ovale with a right-to-left shunt was considered an embo-ligenic cause when the patient was age �50 and did not have anyvascular risk factors or angiographic evidence of atherosclerosis.2) Large vessel atherothrombotic disease: There was angiographicevidence of stenosis or occlusion of the arteries that was attribut-able for the stroke without the presence of emboligenic heartdiseases. 3) Unknown: infarction that fits with both cardiogenicembolism and large vessel diseases or did not fit with any of theabove criteria.

For statistics, Mann–Whitney test was used for comparison ofage and Fisher exact test for other variables. A two-tailed p valueof �0.05 was considered significant. SPSS for Windows (version12.0) was used.

Results. Demography, risk factors, and general neuro-logic features of patients. There were 15 men and 9women with ages ranging from 39 to 83 (mean 58.6 years)

Table 1 Characteristics of patients

No./sex/age, y Objective sensory abnormality Topography Other neurologic deficits

Dominantly impaired primitive sensation

1/F/46 TPVPo (m) ¡ TP (m) FATL Dysarthria

2/F/39 TPVPo (mo) ¡ TP (m) ¡ inc. T FATL ¡ F, A ¡ F Dysarthria

3/M/52 TP (s), Po (mo) ¡ TP (m) FATL Facial palsy, dysarthria, dysphagia

4/F/67 TP (s), VPo (mo) A � L Aphasia

5/F/65 P (m) A Dysarthria, facial palsy

6/F/48 TP (mo), VPo (m) A � L Dysarthria

Dominantly impaired positional and other cortical sensation

7/M/56 Po (s), TP (m) FA

8/M/55 Po (s), dec. Cor ATL Gerstmann syndrome

9/M/62 Po (mo), TPV (m) FATL Dysarthria

10/M/71 Po,V (S), dec. Cor FATL Gerstmann syndrome

11/M/71 Po,V (s), TP (mo) ¡Po(s),PV (m) FATL Visual field defect, hemineglect

12/M/60 Po (s), TP (m) FATL Visual field defect, hemineglect

13/M/62 Po (s), TP (m) Periroral � A (ulnar side)

14/M/65 Po (mo), dec. Cor, TP (m) FATL Cortical hearing deficit

15/M/57 TPVPo (s) ¡ Po(mo) FATL Visual field defect, heminegect

16/M/51 Po (m) Rt hand

17/F/44 Po (mo), TP (m) Lower lip � 1st, 2nd fingers Dysarthria

18/F/51 Po (s), dec. Cor, TP (mo) 1–4th fingers

Paresthesia only

19/M/58 None Perioral � thumb Dysarthria, facial palsy

20/M/47 None Rt hand ¡ thumb Dysarthria

21/F/59 None A (ulnar side)

22/M/83 None A

23/M/67 None Thumb Dysarthria

Combined

24/M/45 TPVPo (mo) FATL Visual field defect, hemineglect

T � temperature; P � pinprick; V � vibration; Po � position; Cor � other cortical senses; M � mild; mo � moderate; s � severe; inc �increased; dec � decreased; F � face; A � arm; T � trunk; L � leg; A � L � arm more severe than leg; CPSP � central poststrokepain; B� burning; C � cold; Nb � numbness; P � painful; ICH � intracerebral hemorrhage.

January 16, 2007 NEUROLOGY 68 175

(table 1). Vascular risk factors included hypertension in13, diabetes mellitus in 2, current cigarette smoking in 4,hypercholesterolemia (fasting serum cholesterol level�200 mg/dL) in 5, and emboligenic heart diseases in 12,which included atrial fibrillation in 7 (1 had concomitantmitral valve disease), dilated cardiomyopathy in 1, andpatent foramen ovale with right-to-left shunt in 4 patients.Three patients had a past history of stroke and two had ahistory of coronary heart disease.

The neurologic deficits were summarized in table 1.Sensory disturbances were the most important manifesta-tions. Other symptoms or signs included dysarthria in 11,transient or mild visual field defect in 6, hemineglect (as-sessed by line-bisection and letter cancellation tests) in 4,mild aphasia in 1, and Gerstmann syndrome (right-to-leftdisorientation, finger agnosia, acalculia) in 2 patients.

Twenty-two patients had infarcts and two had intrace-rebral hemorrhages. The lesions were on the right in 9 andon the left in 15 patients. The patients were followed upfrom 1 to 75 (mean 28.8) months. Two patients developedgeneralized seizures at 1 (Patient 8) and 2 (Patient 7)years after the onset. One (Patient 13) developed anteriorcerebral artery territory infarction 6 years later. Nonedied.

Pathogenic mechanism and treatment. Presumedpathogenic mechanisms included cardiogenic embolism in11, artery-to-artery embolism in 7 (6 due to carotid athero-sclerosis, 1 due to probable dissection), and hypertensivehemorrhage in 2 (table 1). The mechanism remained un-known in four patients. Anticoagulation (warfarin) wasused in 13 patients, whereas 6 received antiplatelets. Ca-

rotid endarterectomy was performed in two patients (Pa-tients 7 and 9).

Patterns of sensory manifestations. According to thesensory abnormalities, the patients were categorized intothree groups.

DIPS group (Patients 1 to 6). The impaired sensationwas solely (Patient 5) or predominantly (Patients 1 to 4, 6)the primitive sensations (pinprick and cold). Positionalsensory impairment was less severe and shorter-lastingthan primitive senses and eventually resolved in all thepatients. The topographies of sensory disturbances werewhole hemibody including the face in three, arm and leg intwo, and arm only in one patient. Accompanying neuro-logic deficits included dysarthria in five (83%) and facialpalsy in two (33%).

On follow-up, four patients (67%) developed CPSP usu-ally several weeks or months later and required pain-relieving medication such as amitryptiline. Two of themhad allodynia. On the longer follow-up, the CPSP gradu-ally diminished in its intensity, although there remainedpersistent sensory discomfort. As shown in figure 1, thelesions were generally located in the lower part of theparietal or frontal area. Parietal opercular area andthe insular cortex were always involved. The lower part ofthe frontal cortex was also involved in Patients 1 to 3 and 5.

DICS group (Patients 7 to 18). The impaired sensationwas solely (Patients 8, 10, and 16) or predominantly theposition and other cortical sensations. Sensory topographyincluded whole hemibody including the face in six patientsand arm and leg in one (Patient 8). In others, the sensorysymptoms were limited to the face and hand (Patient 7),

Figure 1. Lesions producing predominant primitive sensory impairment. Numbers indicate patients’ numbers.

176 NEUROLOGY 68 January 16, 2007

perioral area � ulnar-sided arm (Patient 13), fingers (Pa-tient 16), lower lip and first two fingers (Patient 17), andthe forearm and the first four fingers (Patient 18). Accom-panying neurologic deficits included visual field defect infive, hemineglect in three, dysarthria in three, Gerstmannsyndrome in two, and cortical hearing impairment in onepatient.

During follow-up, none developed CPSP, although sevenhad mild residual numb or tight sensation in the affectedbody parts. However, none required medication for CPSP.

Transient choreoathetotic movements (Patients 10 and 14)or dystonic postures (Patients 11 and 15) were observed.The responsible lesions were generally located in the upperpart of the parietal lobe (figure 2). The posterior centralgyrus was always involved, while the posterior insula wasalso involved in two patients (Patients 14 and 15).

Paresthesia without objective sensory impairment(Patients 19 to 23). Patients presented with paresthesiawithout objective sensory deficits. In all of them, the sen-sory symptoms were restricted to certain body parts: upper

Figure 2. Lesions producing predominant positional and cortical sensory impairment. Numbers indicate patients’ num-bers. Patients 8 and 12 had intracerebral hemorrhage, whereas the others had infarcts.

January 16, 2007 NEUROLOGY 68 177

and lower lips � half of the tongue � thumb (Patient 19),ulnar-sided arm (Patient 21), forearm (Patient 22), andthumb only (Patient 23). In Patient 20, sensory symptomsinitially involved the right hand and then localized to thetip of the thumb in 1 day. In this patient, mild choreoath-etosis was observed for 1 day. Accompanying neurologicdeficits included mild dysarthria in three patients. Thelesions were small and usually located at the postcentralgyrus (figure 3). On follow-up, sensory symptoms disap-peared in all the patients except for Patient 19, who hadmild residual sensory symptoms in the thumb.

Both sensations involved to a similar degree (Patient 24).Patient 24 had impaired cortical and primitive sensoryperception to a similar degree. The patient developedCPSP and required medication.

Comparison between DIPS and DICS groups. Asshown in table 2, DIPS group patients were more oftenwomen, more often had associated dysarthria, and moreoften had residual CPSP than DICS group patients. Ageand the side of the lesion were not different between thetwo groups. Because the lesion location in the paresthesiagroup patients was similar to that of DICS group, we thencombined these patients and the DICS group and com-pared the DIPS vs DICS � paresthesia group. The resultswere identical except that the difference in dysarthria lostits significance. Because of the small number of the DIPS

group, we did not perform multiple logistic regressionanalysis.

Discussion. We identified three main patterns ofsensory disturbances due to cortical stroke: DIPS,DICS, and paresthesia only. The lesions responsiblefor the DICS were located at the upper part of theparietal lobe including mainly the postcentral gyrus,whereas DIPS was related to lesions involving thelower part of the cortex including the parietal oper-cular area and the insular cortex. Patients with par-esthesia only showed small lesions at the areasimilar to the DICS group. Therefore, these casesmay be considered a milder form of DICS.

Our results are consistent with previous re-ports,6,11,12 in that positional and other cortical sen-sory impairment occurred in patients with parietalcortical stroke involving the postcentral gyrus corre-sponding to the primary sensory (SI) area.13 The ob-servation that lesions at the parietal operculum orthe insular cortex were related to DIPS was alsoconsistent with previous report that removal of tu-mor occurring in the parietal operculum abolishedepisodic pain.14 A study using PET showed that nox-ious heat stimuli of the forearm evoked significant

Figure 3. Lesions producing paresthe-sia only. Numbers indicate patients’numbers. Small lesions are indicatedby arrows.

Table 2 Comparison between patients with different sensory characteristics

DIPS, n � 6 DICS, n � 12 p Value* DICS � paresthesia, n � 18 p Value†

Age, y, mean � SD 52 � 11.0 58.8 � 8.1 0.279 59.9 � 9.7 0.15

Female, no. (%) 5 (83.3) 2 (16.7) 0.013 3 (17.6) 0.009

Dysarthria, no. (%) 4 (66.7) 3 (25) 0.043 6 (35.3) 0.069

Central poststroke pain, no. (%) 4 (66.7) 0 0.005 0 0.002

Involuntary movement, no. (%) 0 4 (33.3) 0.245 5 (29.4) 0.272

Cardiogenic infarction, no. (%) 4 (66.7) 5 (41.7) 0.62 7 (41.2) 0.371

Arterial origin infarction, no. (%) 1 (16.7) 4 (33.3) 0.615 6 (35.3) 0.621

* Comparison between DIPS and DICS.† Comparison between DIPS and DIPS � paresthesia.

DIPS � dominant impairment of primitive sensation; DICS � dominant impairment of cortical sensation.

178 NEUROLOGY 68 January 16, 2007

increase in regional cerebral blood flow in the pari-etal opercular area.15 Thus, the insular or opercularregions that roughly correspond to the secondarysensory (SII) area13 seem to mediate or modulatepain or thermal sensation.16

Nevertheless, the lesion–sensory symptoms corre-lation was not perfect; there was impaired positionalsensation in most DIPS group patients and impairedprimitive sensation in most DICS group patients,although these were milder and shorter-lasting thanthe main sensory disturbances. Perhaps, the mildconcomitant impairment in other sensory modalitiesmay be attributed to the damage of the corticocorti-cal connection between the SI and SII areas.

Although a previous study did not observe CPSPdue to parietal cortical strokes,6 we found that CPSPdid occur in DIPS group patients. On the other hand,DICS group patients had mild residual numb sensa-tion only. Our results agree with the previous notionthat the pathogenesis of the CPSP is related to spi-nothalamic sensory impairment8 and confirms thatCPSP can be produced by lesions involving the oper-culum or insular cortex.9 A previous study showedthat unpleasantness perception in skin and musclestimulation was associated with bilateral insular ac-tivation.17 Another study showed that a parietalopercular lesion was related to decreased ability toidentify noxious stimuli as painful, whereas motiva-tional and affective responses to noxious stimuliwere associated with insular lesions.18 Therefore, dis-connection between the thalamus and the SII areaand subsequent reorganization process seem to beresponsible for the CPSP in patients with DIPS.Whatever the mechanism, the nature of CPSP doesnot appear to be different from CPSP from subcorti-cal lesions.

The reason why women occurred more frequentlyin the DIPS than in DICS group remains unclear.This could in part be related to the fact that cardio-genic embolism was relatively prevalent in the DIPSas compared with the DICS group. Cardiogenic em-bolism as an etiology of cerebral infarction is morecommon in women than in men in our stroke registry(unpublished data). Whatever the explanation, be-cause there has been no evidence of gender differ-ence in the incidence of CPSP,8 the gender differencebetween the two groups does not appear to influenceour result that CPSP results primarily from opercu-lar and insular lesions.

Dysarthria was more often found in the DIPSthan DICS group, probably because the lesions sit-uated in the lower part of the frontoparietal areainvolved the lower part of the motor cortex corre-sponding to the tongue or face area (Patients 1 to3). On the other hand, hemineglect, visual fielddefect, and Gerstmann syndrome were presentusually in the DICS group, owing probably to theconcurrent involvement of the angular gyrus or theupper temporal lobe. Chorea or dystonic postureswere also found mainly in the DICS group, whichare related to positional sensory loss.19 Embolism

was the main stroke mechanism in both groups; itseems that DIPS may be related to an occlusion ofthe anterior parietal artery, whereas DISC is re-lated to the posterior parietal artery occlusion withoccasional involvement of the angular branch.

The sensory change was frequently restricted tocertain body parts. According to previous studies,sensory representation of the human hand is somato-topically arranged in the postcentral gyrus (SIarea),20,21 the area representing the thumb beinglarger than that of other fingers.21 The lesion respon-sible for the sensory changes in the first to fourthfingers (Patient 18) was located adjacent to the lat-eral border of the central knob, whereas lesions re-sponsible for perioral–thumb (Patient 19) andperioral–first and second fingers (Patient 17) werelocated slightly lower than that of Patient 18. Thesefindings agree with previous reports that sensoryfibers subserving the thumb are located at the lat-eral shoulder of the central knob of the motor cor-tex,22 and those related to lips are located below thethumb area.20 In Patient 20, sensory changes ini-tially noted on the hand were soon restricted to thethumb, whereas paresthesia on the tip of the thumbwas the only clinical manifestation in Patient 23. Wealso found sensory changes restricted in the lateral(ulnar) areas of the lower arm (Patients 13 and 21).However, the small lesions detected by diffusion-weighted imaging did not allow us to analyze therelationship between the lesion and the gyral mark-ing. Nevertheless, our result seems generally consis-tent with the Penfield somatotopic localization ofsensory fibers in the SI area.

The preferential involvement of the perioral areaand the fingers in our series and in the previousliterature23-25 may also be due to the vulnerability ofthese areas compared to other body parts.25 Thus,the various topography of restricted sensory distur-bances may be determined by a combined influenceof somatotopic lesion location and different vulnera-bility of individual body parts.26

References1. Kim JS. Sensory abnormality. In: Bogousslavsky J, Caplan LR, eds.

Stroke syndrome. 2nd ed. Cambridge, UK: Cambridge University Press,2001:34–47.

2. Fisher CM. Thalamic pure sensory stroke: a pathological study. Neurol-ogy 1978;28:1141–1144.

3. Kim JS. Pure sensory stroke: clinical–radiological correlates of 21 cases.Stroke 1992;23:983–987.

4. Kim JS, Bae YH. Pure or predominant sensory stroke due to brainstemlesion. Stroke 1997;28:1761–1764.

5. Kim JS. Lenticulocapsular hemorrhages presenting as pure sensorystroke. Eur Neurol 1999;42:128–131.

6. Bassetti C, Bogousslavsky J, Regli F. Sensory syndromes in parietalstroke. Neurology 1993;43:1942–1949.

7. Birklein F, Rolke R, Muller-Forell W. Isolated insular infarction elimi-nates contralateral cold, cold pain, and pinprick perception. Neurology2005;65:1381.

8. Bowsher D. Central pain: clinical and physiological characteristics. JNeurol Neurosurg Psychiatry 1996;61:62–69.

9. Schmahmann JD, Leifer D. Parietal pseudothalamic pain syndrome:clinical features and anatomic correlates. Arch Neurol 1992;49:1032–1037.

10. Kim JS, Choi-Kwon S. Sensory sequelae of medullary infarction: differ-ences between lateral and medial medullary syndrome. Stroke 1999;30:2697–2703.

January 16, 2007 NEUROLOGY 68 179

11. Verger H. Sur les troubles de la sensibilite generale consecutifs auxlesions des hemispheres cerebraux chez l’homme. Arch Gen Med 1900;6:641–713.

12. Dejerine J, Mouzon J. Deux cas de syndrome sensitif cortical. RevNeurol (Paris) 1915;28:388–392.

13. Parent A. Carpenter’s human neuroanatomy. 9th ed. Baltimore: Wil-liams & Wilkins, 1996:886–909.

14. Potagas C, Avdelidis D, Singounas E, et al. Episodic pain associatedwith a tumor in the parietal operculum: a case report and literaturereview. Pain 1997;72:201–208.

15. Coghill RC, Talbot JD, Evans AC, et al. Distributed processing of painand vibration by the human brain. J Neurosci 1994;14:4095–4108.

16. Craig AD, Chen K, Bandy D, Reiman EM. Thermosensory activation ofinsular cortex. Nat Neurosci 2000;3:184–190.

17. Schreckenberger M, Siessmeier T, Viertmann A, et al. The unpleasant-ness of tonic pain is encoded by the insular cortex. Neurology 2005;64:1175–1183.

18. Greenspan JD, Lee RR, Lenz FA. Pain sensitivity alterations as afunction of lesion location in the parasylvian cortex. Pain 1999;81:273–282.

19. Kim JS. Delayed onset mixed involuntary movements after thalamicstroke: clinical, radiological and pathophysiological findings. Brain2001;124:299–309.

20. Penfield WJ, Boldrey E. Somatic motor and sensory representation inthe cerebral cortex of man as studied by electrical stimulation. Brain1937;60:389–443.

21. Sutherling W, Levesque MF, Baumgartner C. Cortical sensory repre-sentation of the human hand: size of finger regions and nonoverlappingdigit somatotopy. Neurology 1992;42:1020–1028.

22. Kumabe T, Nakasato N, Inoue T, Yoshimoto T. Primary thumb sensorycortex located at the lateral shoulder of the inverted omega-shape onthe axial images of the central sulcus. Neurol Med Chir (Tokyo) 2000;40:393–403.

23. Bogousslasvky J, Dizerens K, Regli F, Despland PA. Opercular cheiro-oral syndrome. Arch Neurol 1991;48:658–661.

24. Shiga K, Makino M, Ueda Y, Nakajima K, Hirata T. Cheiro-oral syn-drome following a cortical brain infarction in the precentral gyrus andat the base of the central sulcus. A case report. Clin Neurol (Tokyo)1996;36:1104–1106.

25. Chen WH, Lan MY, Chang YY, Lui CC, Chen SS, Liu JS. Corticalcheiro-oral syndrome: a revisit of clinical significance and pathogenesis.Clin Neurol Neurosurg 2006;108:446–450.

26. Kim JS. Restricted acral sensory syndrome following minorstroke: further observation with special reference to the differentialseverity of symptoms among individual digits. Stroke 1994;25:2497–2502.

NeuroImagesAcute reversible cerebellar lesionsassociated with metronidazole therapyJoshua L. Bonkowsky, MD, PhD; Cole Sondrup, BS; andSusan L. Benedict, MD, Salt Lake City, UT

A 27-year-old man with acute lymphoblastic leukemia developedataxia, dysarthria, and confusion 2 weeks after initiating metroni-dazole therapy for Clostridium difficile colitis.

Brain MRI revealed cytotoxic edema, without contrast en-hancement, in the cerebellar dentate nuclei (figure, A and B),dorsal pons, and medulla. Metronidazole was discontinued, andhis symptoms resolved over the following 3 weeks. Repeat brainMRI 6 weeks later showed resolution of the imaging changes(figure, C and D).

Metronidazole, a widely used antibiotic, has only recently beendescribed to have CNS toxicity including encephalopathy, sei-zures, and ataxia.1 MR spectroscopy results suggest that mito-chondrial dysfunction may underlie the cytotoxic edema.2

Copyright © 2007 by AAN Enterprises, Inc.

1. Woodruff BK, Wijdicks EFM, Marshall WF. Reversible metronidazole-induced lesions of the cerebellar dentate nuclei. N Engl J Med 2002;346:68–69.

2. Cecil KM, Halsted MJ, Schapiro M, Dinopoulos A, Jones BV. ReversibleMR imaging and MR spectroscopy abnormalities in association withmetronidazole therapy. J Comput Assist Tomogr 2002;26:948–951.

Disclosure: The authors report no conflicts of interest.

Address correspondence and reprint requests to Dr. Josh Bonkowsky, Divi-sion of Pediatric Neurology, Department of Pediatrics, University of UtahSchool of Medicine, 295 Chipeta Way/Williams Building, Salt Lake City, UT84108; e-mail: joshua.bonkowsky@hsc.utah.edu

Figure. Axial fluid-attenuated inversion recovery (FLAIR)and diffusion-weighted (DWI) MR images. (A, B) FLAIRand DWI MRI obtained on day 20 of metronidazole ther-apy, showing increased signal and diffusion restriction inthe cerebellar dentate nuclei. (C, D) Resolution of theFLAIR and DWI changes, 6 weeks after cessation of metro-nidazole therapy.

180 NEUROLOGY 68 January 16, 2007

DOI 10.1212/01.wnl.0000242579.37851.262007;68;180 Neurology 

Joshua L. Bonkowsky, Cole Sondrup and Susan L. BenedictAcute reversible cerebellar lesions associated with metronidazole therapy

This information is current as of January 15, 2007

ServicesUpdated Information &

http://n.neurology.org/content/68/3/180.fullincluding high resolution figures, can be found at:

Supplementary Material http://n.neurology.org/content/suppl/2007/07/27/68.3.180.DC1

Supplementary material can be found at:

References http://n.neurology.org/content/68/3/180.full#ref-list-1

This article cites 2 articles, 0 of which you can access for free at:

Subspecialty Collections

http://n.neurology.org/cgi/collection/mriMRI

http://n.neurology.org/cgi/collection/gait_disorders_ataxiaGait disorders/ataxiafollowing collection(s): This article, along with others on similar topics, appears in the

  Permissions & Licensing

http://www.neurology.org/about/about_the_journal#permissionsits entirety can be found online at:Information about reproducing this article in parts (figures,tables) or in

  Reprints

http://n.neurology.org/subscribers/advertiseInformation about ordering reprints can be found online:

Online ISSN: 1526-632X.1951, it is now a weekly with 48 issues per year. Copyright . All rights reserved. Print ISSN: 0028-3878.

® is the official journal of the American Academy of Neurology. Published continuously sinceNeurology