664 J Neurol Neurosurg Psychiatry 1999;67:664–667 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.67.5.664 on 1 November 1999. Downloaded from

SHORT REPORT

Frontal lobe syndrome reassessed: comparison of patients with lateral or medial frontal brain damage

Sergio Paradiso, Eran Chemerinski, Kazim M Yazici, Armando Tartaro, Robert G Robinson

Abstract show , indiVerence, speech poverty, and Examination of mood and behaviour poor executive abilities,4 and with acute left changes after frontal damage may con- hemispheric injuries, major .9 Dam- tribute to understanding the functional age to the orbital prefrontal cortex may lead to role of distinct prefrontal areas in depres- poor impulse control, puerility, , sion and . Depression and anxiety increased energy, aggression, and violence.10 disorders, symptoms, and behaviour were The orbital cortex can be divided into medial compared in eight patients with single lat- and lateral subregions. The medial subregion eral and eight patients with single medial connects with limbic areas whereas the lateral frontal lesions matched for age, sex, race, subregion connects with the primary and education, socioeconomic status, side, and sensory association cortex.11 For its limbic aetiology of lesion 2 weeks and 3 months connections,11 the medial prefrontal wall re- after brain injury. DSM IV major depres- sembles the medial subregion of the orbital sive and generalised anxiety disorders cortex. Large lesions of the frontal medial walls copyright. were more frequent in patients with (and anterior cingulate) have been associated Department of lateral compared with medial lesions at 2 with apathy12 but emotional and instinctual Psychiatry, University disorders as well as sociopathy have been asso- of Iowa Iowa City, IA, weeks but not at 3 months. At 3 months, however, patients with lateral damage ciated with damage to the confluence of the USA 13 S Paradiso showed greater severity of depressive medial and orbital cortex. R G Robinson symptoms, and greater impairment in Much of our knowledge of the psychiatric both activities of daily living and social consequences of lesions involving the medial Department of functioning. At initial evaluation de- prefrontal cortex in humans is based on case Neuropsychiatry, Raul report studies.8 This literature tends to select Carrea Institute of pressed mood and slowness were more Neurological Research, frequent, whereas at 3 months slowness, positive findings. We therefore report on the first systematic examination of depression, FLENI Buenos Aires, lack of energy, and social unease were http://jnnp.bmj.com/ Argentina more frequent in the lateral than the anxiety, apathy, disinhibition, and psychosocial E Chemerinski medial group. Patients with lateral lesions functioning at 2–4 weeks and at 3 months after showed greater reduction of and injury in patients with lateral or medial damage Department of to the prefrontal cortex. We hypothesised that Psychiatry, Hacettepe motivation (apathy) during both exami- University, Faculty of nations. Medial frontal injury may fail to medial frontal injury would lead to euphoria Medicine, Ankara, produce emotional dysregulation or may and agitation, whereas lateral injuries would Turkey inhibit experience of mood changes, anxi- lead to depression. No selective association of K M Yazici either medial or lateral lesions with apathy was ety, or apathy. Lateral prefrontal damage on September 26, 2021 by guest. Protected 812 may disrupt mood regulation and drive predicted. Istituto di scienze radiologiche, while leaving intact the ability to experi- Universita G ence (negative) . Methods D’annunzio, Chieti, (J Neurol Neurosurg Psychiatry 1999;67:664–667) Data from about 360 CT scans were reviewed Italy A Tartaro Keywords: frontal lobe; depression; anxiety; apathy; dis- by a neurologist, a neuroradiologist (AT), and inhibition a psychiatrist with experience in neuroanatomy Correspondence to: (KY) blind to the other measures. Scans were Dr Sergio Paradiso, from a consecutive series of prospectively University of Iowa College of Medicine Psychiatry Prefrontal dysfunction has been found in mood evaluated patients admitted to the Shock Research, MEB, Iowa City, disorders and anxiety. Neuroimaging, however, Trauma Center of the Maryland Institute of IA 52241, USA. Telephone has failed to consistently identify specific loca- Emergency Medical Services System 001 319 353 4446; fax 001 319 353 3003; email tions and types of functional changes within (MIEMSS) for traumatic brain injury, or the [email protected] the prefrontal cortex associated with University of Maryland Hospital for acute depression1–5 or anxiety.67 stroke, or the Younker’s Rehabilitation Hospi- Received 10 November 1998 The prefrontal lobe has been divided into tal in Des Moines, Iowa following acute stroke. and in revised form 8 5 May 1999 dorsal lateral, medial, and orbital cortices. The inclusion criterion was a single lateral or Accepted 21 May 1999 Patients with lateral prefrontal damage may medial (including anterior cingulate gyrus) Medial and lateral frontal lobe damage 665 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.67.5.664 on 1 November 1999. Downloaded from

Schematic templates of CT slices showing cross sectional areas of lesion for patients with medial (left) and lateral (right) frontal lesions. Darker areas indicate greater across subject lesion overlap.

frontal lesion as ascertained by all raters. Scans psychosocial measures including the Hamilton copyright. were performed using GE 9900, AS and E 500, rating scale for depression (Ham-D), the mini and EMI 1010 scanners (stroke patients from mental state examination (MMSE), and the Baltimore), GE 1010 (trauma patients from Johns Hopkins functioning inventory (JHFI) Baltimore), and Picker international 1200 or measuring physical independence in activities Somaton plus 24 B30B (stroke patients from of daily living (ADL, higher scores=greater Iowa) without contrast at 0 degrees angle from impairment). Social functioning was assessed the canthomeatal line. Assessment of lesion using the social ties checklist (STC) and the location involved transposing the area of social functioning examination (SFE). The ischaemia as visualised on CT onto standard STC quantifies the number of social connec- 14 templates of axial brain slices (figure). tions available to the patient (higher Exclusion criteria were: history of other scores=less social support). The SFE assesses http://jnnp.bmj.com/ neurological disorders or life threatening physi- patients’ satisfaction with their social function- cal illness, CT evidence of previous brain ing (higher scores=less satisfaction). A modi- injury, decreased level of consciousness or fied present state examination (PSE),15 was medical instability at the time of interview, administered by a psychiatrist. A diagnosis of aphasic disorders (not able to follow a three major or minor depression, or generalised stage command or complete the first part of the anxiety disorder (GAD) was made by using token test), penetrating head injury, associated DSM-IV criteria. spinal cord injury, multiple non-CNS involve- on September 26, 2021 by guest. Protected ment, or secondary brain damage due to severe Seventeen PSE symptom clusters were calculated using the method described by Wing hypotension or hypoxia. 15 Thirty two patients with lateral and 14 et al. The presence and severity of disinhibi- patients with medial lesions met these criteria. tion (agitation//violence, embarrass- Eight of fourteen patients with medial frontal ing behaviour/loss of social restraint, irreverent/ lesions were matched as closely as possible for histrionic behaviour, hypomanic aVect, age (±5 years), sex, education (±2 years), race, pressured/non-social speech), and reduction in socioeconomic status (±1 social class), side, emotion and motivation (self-, and aetiology of lesion with eight patients with slowness/underactivity, mannerism/posturing, lateral lesions. The two groups constituted the stereotypies/ catatonic movements, blunted study sample. Six medial patients could not be aVect, slow/restricted quantity of speech/ mute- matched or had incomplete data. A patient ness) were assessed using the PSE. (medical group) had a small thalamic lesion.

STATISTICAL ANALYSIS PSYCHIATRIC EXAMINATION Data were analysed using means (SD), F tests, After giving written informed consent, patients and likelihood ratio (LR) ÷2 for frequency dis- were administered quantitative psychiatric and tributions. Values of p are two tailed unless 666 Paradiso, Chemerinski, Yazici,et al J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.67.5.664 on 1 November 1999. Downloaded from

Severity of psychiatric symptoms, cognitive impairment, activities of daily living, and social Disinhibition or functioning During the initial evaluation, four patients in the lateral (three displayed agitation and hostile Medial Lateral EVect size Power irritability; one histrionic and hypomanic Initial evaluation: behaviour) and two in the medial group (one Ham D 8.1 (8.1) 14.0 (5.3)* 0.86 0.49 PSE 10.5 (13.2) 24.1 (17.6)** 0.87 0.37 hypomanic aVect, another agitation) had MSE 25.9 (3.7) 21.1 (10.0) 0.63 0.22 symptoms of disinhibition/hypomania. Total JHFI 2.9 (3.4) 7.1 (8.1) 0.67 0.24 number of symptoms was fourfold greater STC 3.0 (1.6) 4.0 (2.1) 0.53 0.17 SFE 0.1 (0.05) 0.3 (0.2)*** 1.13 0.72 (nine) in the lateral than in the medial group 3 months: (two). The mean number of symptoms was Ham D 5.7 (3.7) 9.6 (3.6)† 1.06 0.51 1.12 (SD 1.35) in the lateral and 0.25 (SD PSE 6.0 (6.8) 10.2 (6.5) 0.63 0.22 MMSE 23.8 (5.8) 23.9 (8.9) 0.01 0.05 0.46) in the medial group (t (14)=1.27, JHFI 1.6 (2.3) 5.2 (4.1)† 1.09 0.52 p=0.10, power=0.36, ES=0.86). At 3 months, STC 2.4 (1.0) 4.0 (2.0)‡ 0.71 0.27 SFE 0.1 (.10) 0.24 (.16)‡ 1.07 0.51 three patients in the lateral group and one patient in the medial group showed *t(14)=1.72, p=0.05 one tailed; **t(14)=1.75, p=0.10; ***t(14)=2.78, p<0.015; other compari- disinhibition/hypomania (NS, power=0.07, sons p>0.10. ES=0.12). †t(14)=2.1, p<0.06; ‡t(14)=1.9, p<0.09; other comparisons p>0.1. otherwise specified. Power and eVect sizes (ES) Reduction in emotion and motivation are reported. Five patients (62.5%) in the lateral and one (12.5%) in the medial group showed decreased emotion and motivation {LR ÷2 (1)= 4.55, p<0.04, power=0.98, ES=1.0}. These patients Results showed a combination of severe self neglect, BACKGROUND CHARACTERISTICS slowness, blunted aVect, and decreased speech. Each group was composed of three women and One patient was almost mute at both evalua- five men, five right and three left hemispheric tions. No patient in either group showed man- lesions, and three head injuries and five strokes. nerisms, posturing, or stereotypies. At 3 Mean age of the medial group was 39.5 years months, five patients in the lateral group (SD 17, range 27–74), and that of the lateral showed decreased emotion and motivation group was 44.1 (SD 17, range 20–61). Five compared with none in the medial group {LR white, two African-American, and one His- ÷2(1)= 9.29, p<0.003, power=0.99, ES=1.3}.

patient were in the medial, and six white copyright. and two African-American patients were in the Discussion lateral group. Five patients in the medial and During the inhospital evaluation, patients with four in the lateral group were of socioeconomic single lesions of the lateral prefrontal cortex (Hollingshead) class IV and V. Both groups showed more frequent depression and anxiety had 11.8 (SD 3) mean years of education. No disorders than patients with medial frontal statistically significant diVerences were found. lesions and a greater frequency of symptoms of depression and anxiety at 3 months postinjury. Patients with lateral lesions also showed greater PSYCHIATRIC EVALUATION reduction of emotion and motivation/apathy at Depression or anxiety both evaluations. At initial evaluation, depression was more Before discussing these findings, the limita- frequent in lateral (six or 75%, three major, and tions of our methods should be acknowledged. http://jnnp.bmj.com/ three minor depression) than medial lesion Due to a relatively few patients some important patients (two or 25%, one major, and one diVerences may have been missed. Although minor depression) {LR ÷2 (1)=4.2, p<0.05, laterality questions were not examined, our power=0.99, ES=1.1}. Five (62.5%) patients findings were unlikely to be influenced by had GAD in the lateral and one (12.5%) in the laterality eVects because of the side matched medial group {LR ÷2 (1)=4.5, p<0.04, group design. The aetiology of the brain injury power=0.98, ES=1.0}. Patients with lateral may have influenced some of our findings such damage showed less satisfaction with their as duration of depression (trauma has been on September 26, 2021 by guest. Protected social functioning (table). At 3 months, seven associated with shorter duration of depression patients (87.5%) in both groups no longer ful- than stroke) or the role of non-brain injury fac- filled criteria for depression and/or GAD. tors in depression. Patients with lateral damage However, patients with lateral damage showed had larger lesions. This may have influenced greater severity of depressive symptoms, im- the greater functional impairment found in the pairment in ADL, and psychosocial function- lateral group and perhaps greater depression. ing (table). Comparing the symptom clusters However, physical impairment in activities of from the PSE at initial evaluation, the lateral daily living was not significantly correlated with group showed a greater frequency of simple depression (data available from the authors) depression {LR ÷2 (1)=4.2, p<0.05, power= and depression has not been correlated with 0.99, ES=1.1} and slowness {LR ÷2 (1)=4.5, lesion volume in the acute poststroke period.16 p<0.04, power=0.98, ES=1.0}. At 3 months, Some of the larger lateral lesions involved part patients with lateral damage showed higher of temporal lobes and this may have influenced frequency of slowness {LR ÷2 (1)=6.9, p<0.01, our findings. power=0.97, ES=0.98}, and greater lack of A most surprising finding was that medial energy and social unease {both LR ÷2 (1)=4.8, frontal lesions did not lead to higher euphoria, p<0.03, power =0.88, ES=0.79}. disinhibition, loss of emotion, or apathy than Medial and lateral frontal lobe damage 667 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.67.5.664 on 1 November 1999. Downloaded from

lateral lesions. The same instruments used in 4 Dolan RJ, Bench CJ, Liddle PF, et al. Dorsolateral prefron- tal cortex dysfunction in the major psychoses; symptom or this study, however, were sensitive to increased disease specificity? J Neurol Neurosurg Psychiatry 1993;56: cheerfulness and hypomania associated with 1290–4. right frontal compared with right temporal 5 Drevets WC, Videen TO, Price JL, et al. A functional 17 anatomical study of unipolar depression. J Neurosci lesions. Apathy has been associated with 1992;12:3628–41. extensive damage to the cingulum and mesial 6 Reiman EM. The application of positron emission tomogra- 12 18 phy to the study of normal and pathologic emotions. J Clin frontal areas as well as other structures. Psychiatry 1997;16:4–12. Decreased blood flow in the right dorsal lateral 7 Fredrikson M, Wik G, Annas P, et al. Functional frontal and left frontotemporal cortices was neuroanatomy of visually elicited simple phobic : addi- tional data and theoretical analysis. Psychophysiology 1995; found in 40 patients with apathy after stroke to 32:43–8. diverse brain areas compared with non- 8 Fuster JM. The prefrontal cortex. Anatomy, physiology, and 19 neuropsychology of the frontal lobe. 3rd ed. Philadelphia: apathetic stroke controls. Lippincot-Raven, 1997. The human prefrontal cortex has been 9 Robinson R, Kubos K, Starr L, et al. Mood disorders in 20 stroke patients: importance of lesion location. Brain 1984; shown to be involved in the experience and 107:81–93. appraisal of emotion21 22 as well as in primary 10 Grafman J, Schwab K, Warden D, et al. Frontal lobe and secondary mood disorders. No consensus injuries, violence, and aggression: a report of the Vietnam head injury study. Neurology 1996;46:1231–8. has been achieved, however, regarding the spe- 11 Carmichael ST, Price JL. Limbic connections of the orbital cific location of the prefrontal dysfunction in and medial prefrontal cortex in macaque monkeys. J Comp 1–4 23 24 Neurol 1995;363:615–41. mood disorders. 12 Cummings JL. Frontal-subcortical circuits and human Cummings found personality but not mood behavior. Arch Neurol 1993;50:873–80. 13 Damasio AR. On some functions of the human prefrontal changes after damage to the frontal- cingulate cortex. In: Grafman J, Holyoak KJ, Boller F, eds. Structure circuit.12 Medial and ventral medial frontal and functions of the human prefrontal cortex. New York: cortex participates in the response to emotion AnnNYAcadSci1995;769:241–51. 13 14 Levine DN, Grek A. The anatomic basis of after evoking stimuli and selective to sub- right cerebral infarction. Neurology 1984;34:577–82. jective emotion.25 Therefore, medial frontal 15 Wing J, Cooper E, Sartorius N, eds. Measurement and classification of psychiatric symptoms. Cambridge: Cam- activity may be necessary to experience emo- bridge University Press, 1974. tion including depressed mood, anxiety, or 16 Dam H, Pedersen HE, Ahigren P. Depression among patients with stroke. Acta Psychiatr Scand 1989;80:118–24. apathy. Medial frontal injury may either not 17 Starkstein SE, Robinson RG, Honig MA, et al. Mood lead to emotional changes or may inhibit the changes after right-hemisphere lesions. Br J Psychiatry perception of mood changes, anxiety, or 1989;155:79–85. 8 18 Marin RS. Apathy: a neuropsychiatric syndrome. J Neu- apathy. Unilateral damage to the lateral ropsychiatry Clin Neurosci 1991;3:243–54. prefrontal cortex may disrupt mood regulation 19 Okada K, Kobayashi S, Yamagata S, et al. Post-stroke apathy and drive while leaving intact the ability to and regional cerebral blood flow. Stroke 1997;28:2437–41. 20 George MS, Ketter TA, Parekh PI, et al. Brain activity dur- copyright. experience (disturbed) emotions. ing transient and in healthy women. Am J Psychiatry 1995;152:341–51. 21 Paradiso S, Robinson RG, Andreasen NC, et al. Emotional This study was presented in part at the 9th Annual Meeting of activation of limbic circuitry in elderly normal subjects in a the American Neuropsychiatry Association (Honolulu, Hawaii, PET study. Am J Psychiatry 1997;154:384–9. 1–3 February 1998). The work has been supported by NIH 22 Paradiso S, Johnson DL, Andreasen NC, et al. Cerebral grants MH53592 and MH52879. blood flow changes associated with attribution of emotional valence to pleasant, unpleasant, and neutral visual stimuli. 1999 (in press). 1 Uytdenhoef P, Portelange P, Jacquy J, et al. Regional cerebral Am J Psychiatry blood flow and lateralized hemispheric dysfunction in 23 Drevets WC, Price JL, Simpson JR, Jr, et al. Subgenual pre- depression. Br J Psychiatry 1983;143:128–32. frontal cortex abnormalities in mood disorders. Nature 2 Baxter LR Jr, Phelps ME, Mazziotta JC, et al. Cerebral 1997;386:824–7. metabolic rates for glucose in mood disorders. Studies with 24 Mayberg HS, Lewis PJ, Regenold W, et al. Paralimbic positron emission tomography and fluorodeoxyglucose F hypoperfusion in unipolar depression. J Nucl Med 1994;35: 18. Arch Gen Psychiatry 1985;42:441–7. 929–34. 3 Martinot JL, Hardy P, Feline A, . Left prefrontal glucose 25 Lane RD, Fink GR, Chau P M-L, et al. Neural activation

et al http://jnnp.bmj.com/ hypometabolism in the depressed state: a confirmation. Am during selective attention to emotional responses. Neurore- J Psychiatry 1990;147:1313–7. port 1997;8:3969–72. on September 26, 2021 by guest. Protected