sign in sign up
<<
Home , Bee sting

Letters, Correspondence, Book reviews 551

Vascular complications of and Correspondence to: Dr Fred Schon, Consultant We examined five patients with the appro- stings are rare. Cerebral infarction has only Neurologist, Atkinson Morley’s Hospital, London priate hyperintensity on T1 weighted MRI, been reported in three other people.23In one SW20, UK aged from 31 to 72 years (mean 55.8 (SD three wasp stings were followed by collapse 16.9) years); two with parenteral nutrition and a tonic-clonic seizure. Hypotension was containing Mn (patients 1 and 2), two with 1 Ewan PW.ABC of : . BMJ not recorded. He was treated with adrenaline, 1998;316:1365–8. Child’s grade B cirrhosis (patients 3 and 5), barbiturates, and . It is unclear 2 Starr JC, Brasher GW. Wasp sting and one without any specific factors relating whether the development of a hemiparesis with cerebral infarction. Ann Allergy 1977;39: to Mn or hepatic failure (patient 4, who had preceded or followed this treatment. Brain 431–3. 3 Barnard JH. Studies of 400 hypenoptera sting parkinsonism). In addition, we investigated CT confirmed cerebral infarction. Both other deaths in the United States. J Allergy Clin 10 age matched control subjects without patients died after bee or wasp stings. At Immunol 1973;52:259–64. hyperintensity, aged from 28 to 78 years postmortem cerebral infarction was found in 4 Wagdi P, Mehan VK, Burgi H, et al. Acute myo- (mean 54.2 (SD 15.9) years) (table). The both.3 The mechanism of cerebral infarction cardial infarction after wasp sting in a patiient with normal coronary arteries. Am Heart J MRI was performed on a 1.5 Tesla magnet. was not alluded to. 1994;128:820–3. In all five patients, T1 weighted MRI in the Acute myocardial infarction has been 5 Jones E, Joy M. Acute myocardial infarction patients showed hyperintensity in the bilat- reported four times.4 It has been suggested after a wasp sting. Br Heart J 1988;59:506–8. 6 Wallace JF. Disorders caused by , bites eral globus pallidus and in the region of the that this may be due to a combination of cor- and stings. In: Wilson JD, Braunwald E, substantia nigra or the quadrigeminal plate, onary vasoconstriction secondary to media- Isselbacher KJ, eds. Harrison’s principles of although T2 weighted MRI and brain CT tors released after wasp sting, aggravated by internal medicine. New York: Mc Graw Hill, showed no abnormalities. Ten control sub- 1992:2187–94. exogenous adrenaline given as part of the jects from the neurology and psychiatry serv- 45 7 Waller BF. Nonatherosclerotic coronary heart treatment and by platelet aggregation. It is disease. In: Schlant RC, Alexander RW, eds. ice with no history of parenteral nutrition likely that the mechanism of cerebral infarc- The Heart. New York: McGraw-Hill, 1994: containing Mn, or hepatic failure, showed no tion in this patient was similar. Wasp venom 1239–61. abnormal findings on T1 weighted MRI. We 8 Handin RI, Loscalzo J. Hemostasis, thrombosis, contains vasoactive, inflammatory, and fibrinolysis, and cardiovascular disease. In: obtained blood and CSF samples from the thrombogenic peptides and amines, includ- Braunwald E, ed. Heart disease. Philadelphia: five patients and 10 control subjects with ing , leucotrienes, and thrombox- WB Saunders, 1998:1767–88. informed consent. The serum, whole blood, ane. The venom also contains allergenic pro- 9 Brumlik J. Myasthenia gravis associated with wasp sting. JAMA 1976;235:2120–1. and CSF Mn concentrations were measured teins such as phospholipases which elicit an 10 Goldstein NP, Rucker CW, Woltman HW. Neu- by a standard method using graphite furnace IgE response, resulting in mast cell ritis occurring after stings. JAMA 1960; atomic absorption spectrometry (Model activation.6 Mast cell activation results in 173:181–4. VARIAN SPECTRA A-40) within 1 month release of preformed substances such as 11 Goldstein NP, Rucker CW, Klass DW.Encepha- lopathy and papilledema after bee sting. JAMA after recognition of the symmetric pallidal histamine as well as de novo synthesis of other 1964;188:1083–4. hyperintensities. The CSF Mn concentra- mediators. Constriction of coronary arteries tions were measured by diluting the sample has been shown to occur in response to with 0.5% (v/v) nitric acid to yield absorb- histamine.7 Both thromboxane and leucot- Cerebrospinal fluid manganese ance values within the linear range and rienes have been shown to be concentrations in patients with injecting 200 µl into the furnace. The mean vasoconstrictors.8 The adrenaline that the symmetric pallidal hyperintensities on serum, whole blood and CSF Mn concentra- patient was given may also have been T1 weighted MRI tions were calculated for the patients and implicated in vasoconstriction, resulting in control subjects. The non-parametric Mann- her cerebral infarct. Many of the factors Recently, there have been some reports that Whitney U test was used to assess the signifi- released, including thromboxane and leucot- MRI shows characteristic brain lesions in cance of diVerences between the two groups. rienes, cause platelet aggregation resulting in patients with parenteral nutrition containing The serum, whole blood and CSF Mn a prothrombotic state. manganese (Mn), or hepatic failure, and that concentrations of the patients and control The other neurological complications of the serum or whole blood Mn concentration subjects are listed in the table. All the serum stings which have been reported are indi- is often increased.1–3 T1 weighted MRI in and whole blood concentrations of the vidual cases of ocular myasthenia gravis,9 these patients has shown hyperintensity, control group were within the normal range, optic neuritis, limb numbness, and trigeminal always in the bilateral globus pallidus and and their CSF concentrations were mean neuralgia10 and three cases of encephalopa- sometimes in part of the brainstem, although 0.47 (SD 0.25)µg/l, a relatively narrow range. thy, one of which was fatal.11 Postulated no abnormalities have been found on T2 The CSF Mn concentration (2.1 µg/l) of mechanisms include both a toxic eVect of weighted MRI. The Mn concentrations of patient 4, which was the lowest in the patient venom9 and hypersensitivity to venom.10 11 CSF in these patients, however, have not group, was much higher than 2 SD above the FRANCESCA CRAWLEY been previously measured, because values in mean of the control group, but the serum Mn FRED SCHON control subjects were previously undeter- concentration of patient 4 and the whole Department of Neurology, Atkinson Morley’s Hospital, mined. The present study was designed to blood Mn concentrations of patients 1, 3, and London, UK investigate the CSF Mn concentrations in 4 were all within the normal range. The MARTIN M BROWN control subjects, and to evaluate the concen- serum and CSF Mn concentrations of the Division of Clinical Neuroscience, St George’s Hospital trations in patients with symmetric pallidal patient group were significantly higher Medical School, London, UK hyperintensities on T1 weighted MRI. (p=0.023 and p=0.002 respectively) than

Serum, whole blood, and CSF Mn concentrations in five patients with hyperintensity on T1 weighted MRI and 10 control subjects without hyperintensity

Primary disease (parenteral Mn dose Serum Mn Whole blood Mn CSF Mn Age (y) Sex (mg/day)/duration (days)) (µg/l) (µg/dl) (µg/l) Patients: 1 31 F Pylonephritis (1.1/20) 1.7 1.6 6.7 2 46 F Wernicke’s encephalopathy (1.1/51) 2.1 4.4 3.8 2 64 M Child’s grade B cirrhosis 2.2 2.4 3.0 4 66 M Parkinsonian syndrome 1.2 1.6 2.1 5 72 F Child’s grade B cirrhosis 2.3 4.2 3.1 Mean (SD) 55.8 (16.9) 1.90 (0.45) 2.84 (1.37) 3.74 (1.76) Control subjects: 1 28 F Acute disseminated encephalomyelitis 1.5 1.7 0.7 2 40 M Chorea-acanthocytosis 1.5 2.0 0.9 3 43 M 1.3 1.7 0.2 4 45 F Neurosis 0.9 1.0 0.6 5 49 M Guillain-Barré syndrome 1.9 2.0 0.4 6 57 F Parkinsonian syndrome 1.0 1.3 0.6 7 61 F Malignant syndrome 0.8 1.4 0.2 8 70 M Progressive supranuclear palsy 0.9 1.1 0.3 9 71 M Progressive supranuclear palsy 1.2 1.5 0.6 10 78 M Parkinson’s disease 1.3 2.8 0.2 Mean (SD) 54.2 (15.9) 1.23 (0.34) 1.64 (0.53) 0.47 (0.25)

Normal ranges: serum Mn 0.2-1.6µg/l; whole blood Mn 1.3-3.1µg/dl; CSF Mn concentrations have not been determined. 552 Letters, Correspondence, Book reviews those of the control group, whereas the whole 6 Mena I, Horiuchi K, Burke K, et al. Chronic second order linear diVerential equation.5 blood Mn concentrations were not signifi- manganese : individual susceptibility This is an approximation for horizontal and absorption of iron. Neurology 1969;19: cantly diVerent (p=0.086) between the two 1000–6. movements and a greater approximation for groups. vertical eye movements. It also rests on the This is the first study to evaluate the CSF assumption that the anatomical connections Mn concentrations in patients with symmet- are more simple than in reality. The variabil- ric pallidal hyperintensities on T1 weighted ity of decay constants is consistent with the MRI. Hyperintensity on T1 weighted MRI is findings of Hirose et al.3 This may reflect the associated with many factors, including CORRESPONDENCE varying activity of other aVerents to the peri- calcification, lipid, haemorrhage, and Mn, hypoglossal nuclei. Nevertheless, the approxi- but the pattern of MRI abnormalities in all mate linearity of the plots suggest that part of the five patients was identical with that seen the function of the nucleus intercalatus, the because of Mn deposition and the serum or most caudal of the perihypoglossal nuclei, is whole blood Mn concentrations were often The role of the nucleus intercalatus in integration. Perhaps the reason that such a increased.1–3 In the blood, Mn can bind to vertical gaze holding caudal structure may be involved in vertical transferrin in the trivalent state and to albu- integration is the need to coordinate vertical min or 2-macroglobulin or low weight I was interested to read the report of Janssen head position signals from cervical a erents á 1 V solutes in the divalent state. The blood Mn is et al of a patient with upbeat nystagmus who with integrated head velocity signals from transferred to the brain through the blood- had had a medial medullary infarct. By vestibular nuclei. 2 brain barrier by several transport systems.4 contrast with our patient, their patient had It would be of interest to know whether Recently, increased Mn concentrations were slow phases with a constant velocity, a velocity–position plots of any of the slow recognised in postmortem tissue samples “vestibular” type of nystagmus of central ori- phases of the patient of Janssen 1 show a 3 et al from the brains of patients with long term gin. As noted by Janssen et al, Hirose et al non-zero gradient. parenteral nutrition containing Mn, or se- have reported on a patient with a medial vere cirrhosis.25 From our study, the CSF medullary infarct and upbeat nystagmus; N A R MUNRO Department of Neurology, King’s Healthcare Trust, Mn concentrations were increased more some slow phases were exponential, some of constant velocity. Mapother House, Decrispigny Park, than those in serum and whole blood in all London SE5 8AZ, UK patients and, in particular, patient 4, who In the analysis of slow phases it is useful to had no parenteral nutrition or hepatic plot eye velocity against position rather than failure, showed an increase only in CSF Mn plot eye position against time. In this 1 Janssen JC, Larner AJ, Morris H, et al. Upbeat representation, the plot of a vestibular type of nystagmus: clinicoanatomical correlation.J concentration. The CSF Mn concentrations Neurol Neurosurg Psychiatry 1998;65:380–1. are considered to reflect the accumulation of slow phase with constant velocity is a 2 Munro NAR, Gaymard B, Rivaud S, et al. Mn in the brain tissue most directly of the horizontal straight line. However, when posi- Upbeat nystagmus in a patient with a small three Mn concentrations. The reason why tion varies exponentially with time, velocity is medullary infarct. J Neurol Neurosurg Psychiatry a linear function of position: 1993;56:1126–8. patient 4 is connected with Mn may be indi- 3 Hirose G, Ogasawara T, Shirakawa T, et al.Pri- 6 mary position upbeat nystagmus due to unilat- vidual susceptibility. Moreover, Mn dx/dt = −kx, x = x0exp(−kt) neurotoxicity caused neuropsychiatric eral medial medullary infarction. Ann Neurol The gradient k is the decay constant. A 1998;43:403–6. symptoms including pyramidal and extra- regression line may be fitted and confidence 4 Munro NAR. Upbeat nystagmus in a patient 1–3 pyramidal signs. The findings suggest that limits for k established. A more detailed with a small medullary infarct. In: The control of a high degree of correlation exists between 4 saccadic and smooth pursuit eye movements in analysis of the upbeat nystagmus in our patients with lesions of the central hyperintensity and the CSF Mn concentra- patient with a medial medullary infarct [D Phil thesis]. Oxford: Oxford University, tion, and that the increase may be a most confirmed that decay constants were signifi- 1995;314–29. useful marker and predictor of Mn neurotox- cantly diVerent from zero and therefore not 5 Cannon SC, Robinson DA. Loss of the neural icity in patients with symmetric pallidal integrator of the oculomotor system from brain “vestibular”. However, the decay constants stem lesions in the monkey. J Neurophysiol hyperintensities on T1 weighted MRI. Fur- for the diVerent slow phases varied and the 1987;57:1383–409. ther studies with more patients are necessary plot of eye velocity against position seemed to to elucidate a precise correlation between be non-linear (figure). It is not surprising Mn neurotoxicity and the CSF Mn concen- slow phases attributable to “integrator fail- Bronstein et al reply: tration. ure” might not be strictly exponential. The We thank Munro for his interest in the patient T KATSURAGI model of the perihypoglossal nuclei as a pure we reported in this Journal with a low medul- E ISEKI integrator rests on the assumption that the lary lesion and upbeat nystagmus.1 The lesion K KOSAKA statics of the eye (the oculomotor plant) can probably involved the nucleus intercalatus, Department of Psychiatry, Yokohama City University be modelled by a pure “spring and dashpot”, School of Medicine, 3–9 Fukuura, Kanazawa-ku, the lowermost part of the perihypoglossal Yokohama 236–0004, Japan nucleus, a nucleus thought to perform 16 integration of ocular-motor signals. For the S KOYANO benefit of the general readership of this Jour- K IWABUCHI 15 Department of Neurology and Psychiatry, nal we should like to clarify that the integration alluded to is mathematical inte- Kanagawa-ken Rehabilitation Center, 516 Nanasawa, 14 Atsugi, Kanagawa 243–0121, Japan gration. For instance, eye or head velocity signals arriving at such an integrator emerge 13 Correspondence to: Dr T Katsuragi, Department of as approximations of eye or head position Psychiatry, Yokohama City University School of 12 signals. Currently accepted theories of

Medicine, 3–9 Fukuura, Kanazawa-ku, Yokohama Position (degrees) ocular-motor function establish that a lesion 236–0004, Japan. Telephone 0081 45 787 2667; fax 11 to a gaze holding integrator produces 0081 45 783 2540; email [email protected] 7.4 7.5 7.6 7.7 7.8 7.9 8.0 8.1 8.2 8.3 8.4 nystagmus with slow phase velocity showing cu.ac.jp Time (s) exponential decay. By contrast, peripheral vestibular lesions cause nystagmus with linear 1 Fell JME, Reynolds AP, Meadows N, et al. (constant) slow phase velocity. Manganese in children receiving long- 7.6–7.85 s term parenteral nutrition. 0 The current discussion is centred on the Lancet y = –7.318x + 89.89 1996;347:1218–21. findings in three recently reported patients 2 Krienger D, Krienger S, Jansen O, et al. Manga- –5 with lesions probably involving the nucleus nese and chronic hepatic encephalopathy. Lan- intercalatus.1–3 The patients reported by cet 1995;346:270–4. –10 2 3 3 Hauser RA, Zesiewicz TH, Rosemurgy AS, et al. Munro et al and Hirose et al, with large Manganese intoxication and chronic liver –15 medullary lesions, had predominantly expo- failure. Ann Neurol 1994;36:871–5. Velocity nentially decaying slow phase velocity. Our 4 Heilbronn E, Eriksson H. Implications of man- –20 ganese in disease, especially central nervous patient, with a small paramedian lesion, had system disorders. In: Yasui M, Strong MJ, Ota –25 nystagmus with linearly decaying slow phase K, et al,eds. and neurotoxicology. 12.5 13.0 13.5 14.0 14.5 15.0 velocity.1 Following Munro’s suggestion we Boca Raton: CRC Press, 1997:311–7. Position obtained velocity-position plots of single nys- 5 Tracqui A, Tayot J, Kintz P, et al. Determination of manganese in human brain samples. Forensic Position-time and velocity-position plots of the tagmic beats and found most of them to be Sci Int 1995;76:199–203. slow phase of a vertical nystagmoid jerk. linear (horizontal line on velocity-position