Brief Reports BIOL PSYCHIATRY 1485 1987:22:1481-1485

test of olfactory function. Physiol Behav 32:498- histories of non-insulin dependent diabetes mel- 502. litus. Chem Senses 6:435-443.

Hamilton M (1960): A rating scale for depression. J Settle RG (1986): Diabetes and chemical senses. In Neurol Neurosurg Psychiatry 23:56-62. Meiselman HL, Rivlin R (eds), Clinical Mea- Harris B, Young J, Hughes B (1984): Appetite and surement of Taste and Smell. New York: Mac- weight changes in patients presenting with de- millan, pp 487-5 13. pressive illness. .I AfSect Dis 6:331-339. Steiner JE, Rosenthal-Zifroni A, Edelstein EL (1969): Hopkinson G (1981): A neurochemical theory of ap- Taste perception in depressive illness. Israel Ann petite and weight changes in depressive states. Psychiatry Rel Discip 71223-232. Acta Psychiatr Stand 6412 1l-225. Mezzich JE, Raab E (1980): Depressive symptom- Winokur A, Amsterdam JD, Maislin G (1985): Al- atology across the Americas. Arch Gen Psychiatry terations in glucose utilization and insulin secre- 37:818-823. tion in depression. Annual Meeting of the Society of Biological Psychiatry, Dallas, Texas (abstr). Paykel ES (1977): Depression and appetite. J Psy- chosom Res 2 1:40 l-407. Wright JH, Jacisin JJ, Radin NS, Bell RA (1978): Settle RG (1981): Suprathreshold glucose and fruc- Glucose metabolism in unipolar depression. Br J tose sensitivity in individuals with different family Psychiatry 132:386-393.

Decrease in Core Temperature as an Indication of Overdrive during Withdrawal

Steven C. D&aver and Mark J. Majchrzak

Introduction like symptomatology, (2) excessive, vivid Dilsaver et al. (1983a,b, 1984) observed that dreaming and initial, middle, and terminal in- the withdrawal of tricyclic antidepressants (TCAs) somnia, (3) parkinsonism or akathisia, and results in the development of an array of symp- (4) hypomania. The first three syndromes may toms that can be categorized to define four syn- be due to withdrawal-induced cholinergic ov- dromes: (1) general somatic distress with flu- erdrive at peripheral and central sites. The hypothesis that TCA withdrawal pro-

From the Mental Health Research Institute (S.C.D.). Department of duces cholinergic overdrive is supported by re- Psychiatry (S.C.D.. M.J.M.), University of Michigan, Ann Ar- ports that antimuscarinic agents produce toler- bor, MI Supported by Physician Scientist Career Development Award (Mus- ance (Friedman et al. 1969), all TCAs block carinic Receptor Abnormalities in Affective Illness) SRC IKI I physiological (Dilsaver 1986) and biochemical MHO055302 and NIH 2507RR0538300025. Address reprint requests to Dr. S.C. Dilsaver. Depanment of Psy- responses mediated by muscarinic receptors chiatry, Division of Neuroscience, Ohio State University Col- (mAchRs) (Richelson and Dininetz-Romero lege of Medicine, Ohio State University, 473 West 12th Avenue, Columbus OH 43201.1228. 1977), and TCAs competitively displace mAchR Received March 10. 1987; revised May 11,1987. receptor radioantagonists in vitro (Snyder and

0 I987 Society of Biological Psychiatry otJO6-3223/X7/$03 50 1486 BIOL PSYCHIATRY Brief Reports 1’)87:22:1485-1487

Yamamura 1977; Batra and Biorklund 1986). ment with AMI, 10 mg/kg at -12-hr intervals. Finally, TCA withdrawal symptoms respond to Methylscopolamine nitrate, 1 mg/kg ip, was ad- centrally active antimuscarinic agents (Dilsaver ministered 30 min prior to the measurement of et al. 1983a). baseline temperature. Baseline temperature was Dilsaver et al. (1987) recently demonstrated measured once again after 17 days of treatment. that chronic treatment with amitriptyline hydro- However, this measurement was preceded by (AMI) results in supersensitivity to the the administration of hydrobrom- hypothermic effects of the ide, 2.0 mg/kg ip. The actual number of animals . The capacity of this TCA to pro- yielding useful data varied due to occasional duce this effect demonstrates dose dependence Mini-Mitter failure. Instrument failure is gen- (Dilsaver and Snider 1987). However, these erally the consequence of poor contact between pharmacological paradigms deviate from the a battery inside the device and a connecting clinical situation. In human subjects, the de- wire. It is not uncommon for these two elements velopment of TCA withdrawal symptoms pre- to spontaneously reunite and the instrument to sumably results in cholinergic rebound, the begin transmitting. Thus, we have data on 10, symptomatic effects of which are evident with- 12, 10, and 10 animals after 14, 15, 16, and 17 out the administration of a cholinomimetic agent. days of treatment. We recently conducted two experiments in which we treated rats with AMI, 10 mg/kg twice daily, for 7-2 1 days and then measured the fall in core Statistical Analysis temperature relative to baseline (prior to the first Data were analyzed using EPISTAT, a statis- injection of AMI) 12 or 18 hr after the previous tical package, on an IBM PC. Significance was injection of AMI. assessed using Student’s paired t-test. Measures of variance refer to the standard error of the Methods mean (SEM). Experiment I Eight adult male Sprague-Dawley rats (11 weeks Results old) received AMI, 10 mg/kg ip, at 9:00 AM and 5:00 PM. Core temperature of each animal Experiment 1 was telemetrically measured before and after 1 Mean baseline core temperature prior to treat- and 3 weeks of treatment with AM1 using the ment with AM1 was 36.6 + 0.13”C versus Model VM Mini-Muter (Mini-Muter Corp., Sun 36.0 + 0.27”C after 1 week (p < 0.001, River, OR). The second and third measurements t = 5.44, df = 7) and 35.8 + 0.25”C (p < were made 18 hr after the previous dose of AMI. 0.035, t = 2.71, df = 7) after 3 weeks of Core temperature was measured 1 and 3 weeks treatment. after treatment. The reliability and validity of temperature measurements using this device was Experiment 2 described elsewhere (Tocco-Bradley et al. 1985). Animals were injected with methylscopolamine Mean baseline temperature prior to treatment nitrate, 1 mg/kg ip, 30 min prior to the mea- with AM1 was 36.8 2 0.05”C. After 14 days surement of core temperature to block peripheral of treatment, mean baseline temperature was cholinergic effects. 36.4 r O.l3”C@ < 0.006, t = 3.61, df = 9). Mean baseline temperature of those animals re- ceiving 15 and 16 days of treatment was Experiment 2 36.4 ? 0.13”C @ < 0.006, t = 3.43, df = 11) Core temperature was measured in up to 12 adult and 36.4 + 0.19”C (p < 0.09, t = 1.88, male Sprague-Dawley rats (age circa 8 weeks) df = lo), respectively. prior to and after 14, 15, and 16 days of treat- The mean baseline temperature of those an- Brief Reports BIOLPSYCHIATRY 1487 1987;22:1485-1487

imals challenged with scopolamine at baseline References was 36.7 2 0.07”C versus 36.7 f. 0.17”C af- Barta S, Biorklund A (1986): Binding affinities of terwards (NS, I = 0.00, df = 9). four drugs to muscarinic cholinergic receptors in human parotid gland. Psy- chopharmacology 40: I 4. Dilsaver SC (1986): Pharmacologic induction of cho- Discussion linergic system up-regulation and supersensitivity in affective disorders research. J C/in Psycho- Core temperature was significantly decreased after pharmacol6:65-74. 7, 14, and 2 1 days in Experiment 1 and after Dilsaver SC, Greden JF (1984): Antidepressant with- 15 days of treatment in Experiment 2. Further- drawal phenomena. Biol Psychiatry 19:237-256. more, the decrease in core temperature on day Dilsaver SC, Snider RM (1987): Amitriptyline pro- 16 in Experiment 2 constituted a trend toward duces dose-dependent supersensitivity of a central significance. The fact that pretreatment with cholinergic mechanism. J C/in Psychopharmacol (in press). scopolamine completely eliminated the differ- ence between the pretreatment and posttreat- Dilsaver SC, Feinberg M, Greden JF (1983a): An- tidepressant withdrawal symptoms treated with ment baseline temperatures suggests that a 12- agents. Am J Psychiatv 140:249- 18 hr lapse between injections of AMI may pro- 251. mote the development of a mild cholinergic Dilsaver SC, Kronfol Z, Greden JF, Sackellares JC overdrive state that is sufficient to produce a (I 983b): Antidepressant withdrawal syndromes: decrease in core temperature. Evidence supporting the cholinergic overdrive hy- The data suggest that a spontaneous reduction pothesis. J Clin Psychopharmacol 3: I57- 164. in core temperature provides an indication of Dilsaver SC, Snider RM, Alessi NE (1987): Ami- AM1 withdrawal-induced cholinergic overdrive. triptyline supersensitizes a central cholinergic mechanism. Biol Psychiatry 22:495-507. The data are also consistent with the capacity Friedman MJ, Jaffe JH, Sharpless SK (1969): Central of AM1 to supersensitize a central muscarinic nervous system supersensitivity to pilocarpine af- mechanism (Dilsaver et al. 1987), and the as- ter withdrawal of chronically administered sco- sociation between its withdrawal and the oc- polamine. .I Pharmacol Exp Ther 167:45-55. currence of symptoms is suggestive of cholin- Lee HK, Chaic Y, Wayner MJ, et al (1977): Mech- ergic rebound (Dilsaver et al. 1983a,b; Dilsaver anism of amitriptyline induced hypothermia in the and Greden 1984). rat. Pharmacol Biochem Behav 7: 159- 18.5 Our data raise the question of the effect of Lomax P, Jenden DJ (1986): Hypothermia following AMI on core temperature. Lee et al. (1977) systematic and intracerebral injection of oxotre- studied the effect of a single injection of AMI marine in the rat. Neuropharmacolog? 5:353-359. on rectal temperature. Analysis disclosed a sig- Pawlowski L (1983): Amitriptyline and femoxstine, nificant decrease lasting 3-4 hr for AMI, 20 and but not clomiprimine or citalopram. antagonize hyperthermia induced by directly acting 5-hy- 50 mg/kg, but not for 10 mg/kg. Pawlowski droxy-tryptamine-like drugs in heat adapted rats. ( 1983) observed that AMI, 10 mg/kg ip, did not J Pharm Pharmacol66: 197- 199, alter basal temperature. Other reports of the ef- Richelson E, Dininetz-Romero S (1977): Blockade fect of AMI on core temperature relate to spe- by psychotropic drugs of the muscarinic acetyl- cial circumstances or unusually high doses. In receptor in cultured nerve cells. Biol Psy- summary, there is evidence that doses of AM1 chiatry 12:771-785. much higher than 10 mg/kg lower core temper- Snyder SH, Yamamura HI (1977): Antidepressants ature. However, the dose employed in the ex- and the muscarinic receptor. Arch Gen Psychiaty 34:236-239. periments presented here is not known to de- Tocco-Bradley R, Kluger MJ, Kauffman CA ( 1985): crease core temperature. Thus, the effect re- Effect of age on fever: An acute-phase response ported here is most likely due to effects of drug of rats to endotoxin and Salmonella typhimurium. withdrawal. Infect lmmun 47: 1OC- 1 I 1.