found an association between cancer and vitamin K given by whatever route; in 1970 most doses would probably have been by injection. In the second study they distinguished between oral and intramuscular vitamin K. Only injected vitamin K was found to be significantly associated with cancer. This may be a dose effect. The preparation used is Konakion, which is phytomenadione (vitamin K,) in a vehicle of polyethoxylated castor oil, propylene glycol, and phenol. Though it was designed for intramuscular injection, it is also given orally in the United Kingdom. After a single injection blood concentrations in the newborn reach peak values at 12 hours of 350-1900 ng/l, while after a single oral dose the peak is reached at four hours and ranges from 20 to 330 ng/l.4 Vitamin K given orally is not as effective as injected vitamin K in preventing late onset haemorrhagic disease caused by vitamin K deficiency. It is breast fed infants who suffer from bleeding caused by vitamin K deficiency; the condition is rare in formula fed infants. Most infant formulas are now fortified with vitamin K, but even before they were they were reported to be protective. There is no evidence that formula fed babies are more likely to develop cancer than those who are breast fed. In the first 30 days of life infants fed on formula may receive as much as 1-5 mg of vitamin K, whereas breast fed infants receive a tenth of that amount.6 While we await confirmation or otherwise of the association between vitamin K and childhood cancer it seems reasonable to develop a programme of oral administration for all infants at birth, followed by vitamin K supplementation of breast fed infants over subsequent weeks. When the oral route is not available a small dose
(100 Mtg) could be given by injection. There are, however, problems with the products available, and the British Paediatric Association and the Department of Health recently convened a meeting of experts to consider the issues in detail and make recommendations. Vitamin K is given to prevent haemorrhagic disease of the newborn (which occurs in infants aged up to 7 days) and late onset bleeding due to vitamin K deficiency (which occurs from 7 days to 6 months). Late onset bleeding is rare (4 to 8 cases per 100 000),7 but 30-50% of those affected die or are left with severe brain damage. This is certain; the risk of cancer is not. Before we change our practice we must be confident that any programme of oral administration will be as effective as injected vitamin K and that all infants who need it will receive it. DAVI1HULL
'professor of Child Health, (Queen's Medical Centre, NottinghanrNG7 2UH I Gilman EA, Kinnear-Wilson LAS, Kneale GW., Waterhouse JAH. Childhood cancers and their association with prcgnanicy, drugs and illnesses. Pacdiatr Perinat EpidemioIl 1989;3:66-94. 2 Goldinig J, Patcrsoni MNl, Kinlen LJ. Factors associated with childhood cancer in a national cohort study. Brj Cancer 1990;62:304-8. 3 Golding J, Greenwood R, Birmiigham K, lott M. Childhood cancer, intramuscular vitamin K, and pethidine given duritig labour. BMJ 1992;305:341-6. 4 M1cNinch AW, Upton C, Samuels M, Shearer MJ, McCarthy P, Tripp JH, et al. Plasma concentrations after oral or intramuscular vitamin K1 in neonates. Arch Dis Child 1985;60:814-8. 5 von Kries R. Vitamin K prophvlaxis-a useful public health measure? Paediatr Perinat Eptdemiol 1992;6:7-13. 6 Greer FR, Miarshall S, Cherry J, Suttie JW. Vitamin K status of lactating mothers, human milk, and breast feeding infants. Pediatrics 1991;88:751-6. 7 McNinch AW, 'Iripp JH. Haemorrhagic disease of the newborn in the British Isles: two year prospective study. B.RI7 1991;303:1105-8.
Understanding schizophrenia Structural and functional abnormalities of the brain are present in the condition As we know little of the aetiology of the condition and have failed to identify any consistent structural or physiological abnormalities in those possessing the familiar symptoms, the only defining characteristic available to us is the syndrome itself. Kendall
In the 20 years since this was written neuroscientific research has put the brain back into schizophrenia. Research is proliferating and providing ever more evidence of organic dysfunction in schizophrenia. What are the relevant findings, and what do they tell us about neural functioning and aetiology in schizophrenia? Neuropathology had little to offer in the early part of this century. The findings were so conflicting as to be uninterpretable, and research on the subject was nearly abandoned because of the many methodological difficulties in clinical, epidemiological, and technical research.2 Modern techniques of neuroimaging, better diagnostic methods, and rigorous neuropathological studies have combined to resurrect the topic,' and an extraordinary research effort is yielding some consistent necropsy findings. Brains from some (though not all) schizophrenic patients are lighter and smaller than those from controls.4" Enlargement of the lateral ventricles has been described repeatedly, particularly in the anterior and temporal horns.4" Reductions in the volume of the medial temporal cortex structures (hippocampus and parahippocampal gyrus) have been reported in studies comparing schizophrenic patients with non-schizophrenic controls and normal conBMJ
VOLUME 305
8 AUGUST 1992
trols.47 Neuronal loss has been found in the hippocampus and entorhinal cortex89 and in other cortical areas, including the prefrontal cortex and the cingulate and motor cortices.'° Disarray in cerebral cytoarchitecture has also been noted. 2 The evidence that the medial temporal cortex is affected is consistent enough for one authority to conclude that "it is probable that all schizophrenics have abnormalities in the medial temporal lobe which differ in degree but not in kind,"'" but the aetiological importance of this finding is unclear. Cellular loss and disarray in the absence of gliosis strongly implicate aberrant neurodevelopment, but reports of widespread focal cerebral lesions in some cases is more suggestive of postnatal neurodegenerative processes.6 Imaging techniques that show the brain's structure (computed tomography and magnetic resonance imaging) and function (positron emission tomography and single photon emission tomography) have revolutionised investigation of these questions. Johnstone et al first used computed tomography to substantiate early pneumoencephalographic findings of enlarged ventricles in schizophrenic patients,' and their findings renewed interest in postmortem research. A flood of imaging studies has provided further evidence of subtle non-progressive ventricular enlargement and reduced cortical volume particularly affecting the temporal lobe and the hippocampus.'4'6 The structural abnormalities have been correlated with various neuropsychological deficits and poor clinical outcome in some patients.'7 Interestingly, however, studies of identical twins discordant for schizophrenia have 327
shown similar structural anomalies only in the schizophrenic
twins" 1 and indicate that further research must account for
both environmental and genetic influences. Functional imaging examines dynamic brain-behaviour relations by regional localisation and quantification of injected radiolabelled compounds. Cerebral metabolism, blood flow, and brain neurochemical variables may be measured. These are "state" rather than "trait" indices and are sensitive to both technical and biological artefacts. Metabolism and blood flow may be assessed at rest, when clinical symptoms such as hallucinations are active, or during a neuropsychological challenge task. An early finding from resting studies was frontal hypoperfusion,"9 but technical advances have uncovered a more complex picture. Alterations in cerebral blood flow in patients with schizophrenia may be found in the left medial temporal region, the frontal areas (especially the dorsolateral prefrontal cortex), and subcortical regions.202' Lateralised changes have been found with greater flow to the left globus pallidus and increased metabolism in the left hemisphere overall.22 23 During tasks that activate the frontal lobes the patients (unlike controls) fail to increase flow to left frontal cortical regions.21 24 Fronto-limbic-striatal areas are critically concerned in such higher functions as attention, cognition, willed intention, executive planning, and memory-all of which are disrupted in schizophrenia. Recent research with neurochemical functional imaging research has been focused on the dopaminergic system. An initial study seemed to confirm the prediction of the dopamine hypothesis of schizophrenia by finding raised numbers-of D2 receptors in patients25- but this finding has not been replicated.2128 Other studies, however, have shown a greater left sided striatal asymmetry of D2 receptor binding in patients than in controls (L S Pilowsky et al, international congress on schizophrenia research, Tucson, US, 1991).26 I Kendall RE. Schizophrenia: the remedy for diagnostic confusion. Brj Hosp Mfed 1972;8:383-90. 2 Plum F. Prospects for research on schizophrenia 3. Neuropsvchology: neuropathological findings. Neurosciences Research Program Bulletin 1972;10:384-8. 3 Johnstone EC, Crow TJ, Frith CD, Husband J, Kreel L. Cerebral ventricular size and cognitive impairment in chronic schizophrenia. Lancet 1976;ii:924-6. 4 Brown R, Colter N, Corsellis JAN, Crow TJ, Frith CD, Jagoe R, et al. Postmortem evidence of structural brain changes in schizophrenia. Arch Gen Psvchiatry 1986;43:36-42. 5 Plakkenberg B. Post mortem sttudy of schizophrenic brains. Br Psychiatry 1987;151:744-52. 6 Bruton CJ, Crow TJ, Frith CD, Johnstone EC, Owens DG, Roberts GW, et al. Schizophrenia and the brain. Psvchol Med 1990;20:285-304. 7 Bogerts B, Mveertz E, Schonfeldt-Bausch R. Basal ganglia and limbic system pathology in schizophrenia: a morphometric study of brain volume and shrinkage. Arch Gen Psychiatry 1985;42:784-91. 8 Falkai P, Bogerts B, Rozumek N\. Limbic pathology in schizophrenia. The entorhinal region-a morphometric study. Btol Psvchiatrn 1988;24:515-21. 9 Jeste DV, Lohr JB. Hippocampal pathologic findings in schizophrenia: a morphometric study. Arch Gen Psvchiatr' 1989;46:1019-24. 10 Benes FM, Davison J, Bird ED. Quantitive cyto-architectural studies of the cerebral cortex in schizophrenics. Arch C,en I'sychiatrv 1986;43:31-5. 11 Benes FM, McSparren J, Bird ED. Deficits in small interneurons in prefrontal and cingulate cortices of schizophrenic and schizoaffective patients. Arch Gen Psychiatrv 1991;48:996-1001. 12 Kovelman JA, Schcibel AB. A neurohistological correlate of schizophrenia. Biol Psychiatry 1984;19:1601-21. 13 Roberts GW. Schizophrenia: a neuropathological perspective. Brj Psychiatrv 1991;158:8-17. 14 Illowskv B, Juliano DM, Bigclow LB, Wleinberger DR. Stability of CT scan findings. j Neurol
Neurosurg Psvchiatry 1988;51:209-13. 15 Vita A, Sacchettt E, %'alvassori G, Cazzullo Cl. Brain morphology in schizophrenia-a 2-5 year CT scan follow-up studv. Aua PsychiatrScand 1988;78:618-21. 16 Suddath RL, Christison GX', 'Forrev EF, Casanova MF, Weiberger DR. Anatomic abnormalities in the brains of monozygotic twins discordant for schizophrenia. N Englj Med 1990;322:789-94. 17 Johnstone EC, Owens DGC, Crow TJ, Frith CD, Alexandropolis K, Bydoer, et al. Temporal lobe structure as determined bv nuclear magnetic resonance in schizophrenia and bipolar affective disorder. J Neurol Neurosurg Psychiatry 1989;52:736-41. 18 Revelrv AM, Rcseles M\1A, Clifford CA, Murray RM. Cerebral sentricular size in twins discordant for schizophrenia ancet 1982;i:540-1. 19 Ingvar DH, Franzen (i. Abnormalities of cerebral blood distribution in patients with chronic schizophrenia. Acta Psvchiatr Scand 1974;50:425-62.
328
Pharmacological studies with positron emission tomography and single photon emission tomography have shown that typical antipsychotic drugs block D2 receptors in vivo,28 but the relation of this blockade to clinical efficacy is less clear. Studies in patients who do and do not respond to drugs show similar degrees of D2 occupancy in both groups,"-" and the novel antipsychotic drug clozapine produces clinical improvement with a much lower occupancy of D2 receptors."' These findings suggest that simplistic models of neurochemical dysfunction are no longer tenable. Other systems are under investigation and may yield greater opportunities for the development of new drugs. The glutamate-dopamine interaction is of particular interest in this context as loss of glutamate receptor subtypes has been reported in the temporal cortices of schizophrenic patients. This is thought to be a product of altered gene expression in the region.3233 Postmortem neurochemical studies have identified aberrations in other limbic neurotransmitter systems." Excitatory amino acids and cholecystokinin are important trophic factors in hippocampal development, and thus may indicate a pathway for both developmental pathogenesis and prophylactic intervention. Neuroscientific inquiry into this most devastating of mental illnesses is, then, bearing hard won fruit. Few would now dispute that a substantial proportion of patients with schizophrenia do indeed have consistent structural and physiological brain abnormalities. The challenges for research workers are to establish the specificity of these abnormalities to schizophrenia and to clarify the relations between biological, aetiological, and diagnostic heterogeneity. L S PILOWSKY
Wellcome Research Fellow and Lecturer, Institute of Psychiatry, London SE5 8AF 20 Liddle PF, Friston KJ, Frith CI), Hirsch SR, Joncs '', Frackowiak RSJ. Patterns of cerebral blood flow in schizophrenia. Br] Isvchluatry 1992;160:179-86. 21 Lewis SW, Ford RA, Sved GAM, Reveley ANM, 'I'o
(100 Mtg) could be given by injection. There are, however, problems with the products available, and the British Paediatric Association and the Department of Health recently convened a meeting of experts to consider the issues in detail and make recommendations. Vitamin K is given to prevent haemorrhagic disease of the newborn (which occurs in infants aged up to 7 days) and late onset bleeding due to vitamin K deficiency (which occurs from 7 days to 6 months). Late onset bleeding is rare (4 to 8 cases per 100 000),7 but 30-50% of those affected die or are left with severe brain damage. This is certain; the risk of cancer is not. Before we change our practice we must be confident that any programme of oral administration will be as effective as injected vitamin K and that all infants who need it will receive it. DAVI1HULL
'professor of Child Health, (Queen's Medical Centre, NottinghanrNG7 2UH I Gilman EA, Kinnear-Wilson LAS, Kneale GW., Waterhouse JAH. Childhood cancers and their association with prcgnanicy, drugs and illnesses. Pacdiatr Perinat EpidemioIl 1989;3:66-94. 2 Goldinig J, Patcrsoni MNl, Kinlen LJ. Factors associated with childhood cancer in a national cohort study. Brj Cancer 1990;62:304-8. 3 Golding J, Greenwood R, Birmiigham K, lott M. Childhood cancer, intramuscular vitamin K, and pethidine given duritig labour. BMJ 1992;305:341-6. 4 M1cNinch AW, Upton C, Samuels M, Shearer MJ, McCarthy P, Tripp JH, et al. Plasma concentrations after oral or intramuscular vitamin K1 in neonates. Arch Dis Child 1985;60:814-8. 5 von Kries R. Vitamin K prophvlaxis-a useful public health measure? Paediatr Perinat Eptdemiol 1992;6:7-13. 6 Greer FR, Miarshall S, Cherry J, Suttie JW. Vitamin K status of lactating mothers, human milk, and breast feeding infants. Pediatrics 1991;88:751-6. 7 McNinch AW, 'Iripp JH. Haemorrhagic disease of the newborn in the British Isles: two year prospective study. B.RI7 1991;303:1105-8.
Understanding schizophrenia Structural and functional abnormalities of the brain are present in the condition As we know little of the aetiology of the condition and have failed to identify any consistent structural or physiological abnormalities in those possessing the familiar symptoms, the only defining characteristic available to us is the syndrome itself. Kendall
In the 20 years since this was written neuroscientific research has put the brain back into schizophrenia. Research is proliferating and providing ever more evidence of organic dysfunction in schizophrenia. What are the relevant findings, and what do they tell us about neural functioning and aetiology in schizophrenia? Neuropathology had little to offer in the early part of this century. The findings were so conflicting as to be uninterpretable, and research on the subject was nearly abandoned because of the many methodological difficulties in clinical, epidemiological, and technical research.2 Modern techniques of neuroimaging, better diagnostic methods, and rigorous neuropathological studies have combined to resurrect the topic,' and an extraordinary research effort is yielding some consistent necropsy findings. Brains from some (though not all) schizophrenic patients are lighter and smaller than those from controls.4" Enlargement of the lateral ventricles has been described repeatedly, particularly in the anterior and temporal horns.4" Reductions in the volume of the medial temporal cortex structures (hippocampus and parahippocampal gyrus) have been reported in studies comparing schizophrenic patients with non-schizophrenic controls and normal conBMJ
VOLUME 305
8 AUGUST 1992
trols.47 Neuronal loss has been found in the hippocampus and entorhinal cortex89 and in other cortical areas, including the prefrontal cortex and the cingulate and motor cortices.'° Disarray in cerebral cytoarchitecture has also been noted. 2 The evidence that the medial temporal cortex is affected is consistent enough for one authority to conclude that "it is probable that all schizophrenics have abnormalities in the medial temporal lobe which differ in degree but not in kind,"'" but the aetiological importance of this finding is unclear. Cellular loss and disarray in the absence of gliosis strongly implicate aberrant neurodevelopment, but reports of widespread focal cerebral lesions in some cases is more suggestive of postnatal neurodegenerative processes.6 Imaging techniques that show the brain's structure (computed tomography and magnetic resonance imaging) and function (positron emission tomography and single photon emission tomography) have revolutionised investigation of these questions. Johnstone et al first used computed tomography to substantiate early pneumoencephalographic findings of enlarged ventricles in schizophrenic patients,' and their findings renewed interest in postmortem research. A flood of imaging studies has provided further evidence of subtle non-progressive ventricular enlargement and reduced cortical volume particularly affecting the temporal lobe and the hippocampus.'4'6 The structural abnormalities have been correlated with various neuropsychological deficits and poor clinical outcome in some patients.'7 Interestingly, however, studies of identical twins discordant for schizophrenia have 327
shown similar structural anomalies only in the schizophrenic
twins" 1 and indicate that further research must account for
both environmental and genetic influences. Functional imaging examines dynamic brain-behaviour relations by regional localisation and quantification of injected radiolabelled compounds. Cerebral metabolism, blood flow, and brain neurochemical variables may be measured. These are "state" rather than "trait" indices and are sensitive to both technical and biological artefacts. Metabolism and blood flow may be assessed at rest, when clinical symptoms such as hallucinations are active, or during a neuropsychological challenge task. An early finding from resting studies was frontal hypoperfusion,"9 but technical advances have uncovered a more complex picture. Alterations in cerebral blood flow in patients with schizophrenia may be found in the left medial temporal region, the frontal areas (especially the dorsolateral prefrontal cortex), and subcortical regions.202' Lateralised changes have been found with greater flow to the left globus pallidus and increased metabolism in the left hemisphere overall.22 23 During tasks that activate the frontal lobes the patients (unlike controls) fail to increase flow to left frontal cortical regions.21 24 Fronto-limbic-striatal areas are critically concerned in such higher functions as attention, cognition, willed intention, executive planning, and memory-all of which are disrupted in schizophrenia. Recent research with neurochemical functional imaging research has been focused on the dopaminergic system. An initial study seemed to confirm the prediction of the dopamine hypothesis of schizophrenia by finding raised numbers-of D2 receptors in patients25- but this finding has not been replicated.2128 Other studies, however, have shown a greater left sided striatal asymmetry of D2 receptor binding in patients than in controls (L S Pilowsky et al, international congress on schizophrenia research, Tucson, US, 1991).26 I Kendall RE. Schizophrenia: the remedy for diagnostic confusion. Brj Hosp Mfed 1972;8:383-90. 2 Plum F. Prospects for research on schizophrenia 3. Neuropsvchology: neuropathological findings. Neurosciences Research Program Bulletin 1972;10:384-8. 3 Johnstone EC, Crow TJ, Frith CD, Husband J, Kreel L. Cerebral ventricular size and cognitive impairment in chronic schizophrenia. Lancet 1976;ii:924-6. 4 Brown R, Colter N, Corsellis JAN, Crow TJ, Frith CD, Jagoe R, et al. Postmortem evidence of structural brain changes in schizophrenia. Arch Gen Psvchiatry 1986;43:36-42. 5 Plakkenberg B. Post mortem sttudy of schizophrenic brains. Br Psychiatry 1987;151:744-52. 6 Bruton CJ, Crow TJ, Frith CD, Johnstone EC, Owens DG, Roberts GW, et al. Schizophrenia and the brain. Psvchol Med 1990;20:285-304. 7 Bogerts B, Mveertz E, Schonfeldt-Bausch R. Basal ganglia and limbic system pathology in schizophrenia: a morphometric study of brain volume and shrinkage. Arch Gen Psychiatry 1985;42:784-91. 8 Falkai P, Bogerts B, Rozumek N\. Limbic pathology in schizophrenia. The entorhinal region-a morphometric study. Btol Psvchiatrn 1988;24:515-21. 9 Jeste DV, Lohr JB. Hippocampal pathologic findings in schizophrenia: a morphometric study. Arch Gen Psvchiatr' 1989;46:1019-24. 10 Benes FM, Davison J, Bird ED. Quantitive cyto-architectural studies of the cerebral cortex in schizophrenics. Arch C,en I'sychiatrv 1986;43:31-5. 11 Benes FM, McSparren J, Bird ED. Deficits in small interneurons in prefrontal and cingulate cortices of schizophrenic and schizoaffective patients. Arch Gen Psychiatrv 1991;48:996-1001. 12 Kovelman JA, Schcibel AB. A neurohistological correlate of schizophrenia. Biol Psychiatry 1984;19:1601-21. 13 Roberts GW. Schizophrenia: a neuropathological perspective. Brj Psychiatrv 1991;158:8-17. 14 Illowskv B, Juliano DM, Bigclow LB, Wleinberger DR. Stability of CT scan findings. j Neurol
Neurosurg Psvchiatry 1988;51:209-13. 15 Vita A, Sacchettt E, %'alvassori G, Cazzullo Cl. Brain morphology in schizophrenia-a 2-5 year CT scan follow-up studv. Aua PsychiatrScand 1988;78:618-21. 16 Suddath RL, Christison GX', 'Forrev EF, Casanova MF, Weiberger DR. Anatomic abnormalities in the brains of monozygotic twins discordant for schizophrenia. N Englj Med 1990;322:789-94. 17 Johnstone EC, Owens DGC, Crow TJ, Frith CD, Alexandropolis K, Bydoer, et al. Temporal lobe structure as determined bv nuclear magnetic resonance in schizophrenia and bipolar affective disorder. J Neurol Neurosurg Psychiatry 1989;52:736-41. 18 Revelrv AM, Rcseles M\1A, Clifford CA, Murray RM. Cerebral sentricular size in twins discordant for schizophrenia ancet 1982;i:540-1. 19 Ingvar DH, Franzen (i. Abnormalities of cerebral blood distribution in patients with chronic schizophrenia. Acta Psvchiatr Scand 1974;50:425-62.
328
Pharmacological studies with positron emission tomography and single photon emission tomography have shown that typical antipsychotic drugs block D2 receptors in vivo,28 but the relation of this blockade to clinical efficacy is less clear. Studies in patients who do and do not respond to drugs show similar degrees of D2 occupancy in both groups,"-" and the novel antipsychotic drug clozapine produces clinical improvement with a much lower occupancy of D2 receptors."' These findings suggest that simplistic models of neurochemical dysfunction are no longer tenable. Other systems are under investigation and may yield greater opportunities for the development of new drugs. The glutamate-dopamine interaction is of particular interest in this context as loss of glutamate receptor subtypes has been reported in the temporal cortices of schizophrenic patients. This is thought to be a product of altered gene expression in the region.3233 Postmortem neurochemical studies have identified aberrations in other limbic neurotransmitter systems." Excitatory amino acids and cholecystokinin are important trophic factors in hippocampal development, and thus may indicate a pathway for both developmental pathogenesis and prophylactic intervention. Neuroscientific inquiry into this most devastating of mental illnesses is, then, bearing hard won fruit. Few would now dispute that a substantial proportion of patients with schizophrenia do indeed have consistent structural and physiological brain abnormalities. The challenges for research workers are to establish the specificity of these abnormalities to schizophrenia and to clarify the relations between biological, aetiological, and diagnostic heterogeneity. L S PILOWSKY
Wellcome Research Fellow and Lecturer, Institute of Psychiatry, London SE5 8AF 20 Liddle PF, Friston KJ, Frith CI), Hirsch SR, Joncs '', Frackowiak RSJ. Patterns of cerebral blood flow in schizophrenia. Br] Isvchluatry 1992;160:179-86. 21 Lewis SW, Ford RA, Sved GAM, Reveley ANM, 'I'o
