BIOL PSYCHIATRY 1 05 1 1991;29:1051-1054 BRIEF REPORT Cortisol Responses to Mental Arithmetic in Acute and Remitted Depression Robert L. Trestman, Emil F. Coccaro, David Bernstein, Timothy Lawrence, Steven M. Gabriel, Thomas B. Horvath, and Larry J. Siever Introduction Strategies are needed to model the effect of nat- urally occumng stressful events that may con- tribute to the manifestation of depression in sus- ceptible individuals (Gold et al 1988). A paradigm of a dysregulated system that will not respond appropriately to specific stimuli implies that re- sponses to naturalistic challenges or stressors may unmask significant alterations in pt:.ysio- logical function that may be associated with the diathesis to depression (Siever and Davis 1985). The present study u,;ilizes a mental arithmetic task (MAT) (Esler et al 1989) to test the hy- pothesis that the response of plasma cortisol to a voluntary, effortful task would be state-inde- pendent in depression and would be decreased in both acute and remitted .,leprc~d eatients. Subjects and Methods The study comprised 9 acute and 9 remitted depressed male patients and 12 male normal controls (Table 1). All were medically healthy~ From the Bronx Veterans Administration MedicalCenter (RLT, DB, TL, LJS), Mount Sinai School of Medicine (RLT, SMG, LJS), Medical College of Pe-nsylvania (EFC), and Brooklyn Veterans Administration Medical Center (TBH). Address reprint requests to Robert L. Trestman, Ph.D., M.D.. Psy- chiatry Service, 116A, Bronx VA Medical Center, Bronx, NY 10468. Received June 8, 1990; revised November 19, 1990. Acutely depressed patients met Research Diagnostic Criteria (RDC) for current major depressive disorder at the time of assessment by the Schedule for Affective Disorders and Schizophrenia-Lifetime Version (SADS-L) (Spitzer et al 1987). Remitted depressed pa- tients had a prior episode of major depressive disorder as determined from the SADS-L, had not been hospitalized for at least 6 month% and had a 23-item Hamilton Depression Rating Scale (HDRS) (Guy 1976) score below 12. Controls were recruited by advertisement, and had no personal or family history of major psychiatric disorder. All subjects were medicauon-free for at least 2 weeks. Protocols began at 8 r~.~,l following an overnight fast with the placement of an in- travenous catheter. The MAT consisted of two 12-rain blocks of arithmetic problems with concomitant auditory distraction. Two plasma samples fo- _ortisol were drawn for baseline determination. The subjects began the MAT at 2 r'M ar'd plasma samples for cortisol were drz.wn then and at 15-rain intervals until 4 eM. Assessment of diurnal plasma cortisol concen- trations were made on a separate control day under identical conditions but without the MAT. Plasma concentrations of cortisol were de- termined by radioimmunoassay using reagents ~:m ~CN/Micromedic (Carson, CA) with intra- © 1991 Society of Biological PsycLiatD' 0006-3223191/$03.50