Brief ReportThis section will carry communications of work in progress, preliminary researchreports, or interesting and unusual vignettes. Such reports will be considered fortheir practical clinical relevance or heuristic value.

Could Implantable Cardioverter DefibrillatorsProvide a Human Model Supporting the LearnedHelplessness Theory of Depression?

Mark Goodman, Ph.D., and Brian Hess, M.A.

Abstract: Affective symptoms were examined retrospectivelyin 25 patients following placement of implantable cardioverterdefibrillators (ICD) which can produce intermittent shockswithout warning in response to cardiac ventricular arrhyth-mias. The number of ICD random, uncontrollable dischargeshocks and pre-ICD history of psychological distress (i.e., de-pression and/or anxiety) were documented in all patientsusing a demographics questionnaire and a standardized behav-ioral/psychological symptoms questionnaire (i.e., SymptomChecklist-90 Revised). ICD patients were dichotomized intotwo groups: those without a history of psychological distressprior to ICD (n 5 18) and those with a history of psychologicaldistress prior to ICD (n 5 7). In ICD patients without a priorhistory, results indicated that quantity of ICD discharge shockswas significantly predictive of current reported depression (r 50.45, p 5 0.03) and current reported anxiety (r 5 0.51, p 50.02). Conversely, in patients with a reported history of psy-chological distress, there was no significant relationship foundbetween quantity of discharge shocks and current reporteddepression or anxiety. This study may provide evidence insupport of a human model of learned helplessness in that itsupports the notion that exposure to an unavoidable and ines-capable aversive stimulus was found to be related to patients’reported depression. Further studies may wish to prospectivelyconsider a larger sample as well as a more comprehensiveassessment of premorbid psychological symptoms. © 1999Elsevier Science Inc.

Mirowski et al. [1–3] developed the automatic im-plantable cardioverter defibrillator (ICD), which

has been shown to cause a significant decline inmortality rates due to sudden cardiac death frompotentially fatal ventricular tachyarrhythmias[4–10]. Consequently, the Agency for Health CarePolicy Research (AHCPR) has broadened its patientselection criteria for ICD recipients instead of justtraditional daily administered drug antiarrhythmictherapy to reduce this sudden cardiac death phe-nomenon [11]. Moreover, the superiority of the ICDin preserving patients’ “Quality-of-Life” has alsobeen recently reported [12] as well as the psycho-social effects of this device [13].

However, other investigations have shown in-creased depression and anxiety among ICD pa-tients [14–16]. Specifically, patients with the ICDassociate the experience of negative emotions withfear of discharge and device failure and concernover possible negative lifestyle changes such as de-creases in physical activity and concerns about be-ing a burden to their families. Thus, the unpredict-ability associated with 75–150 joule ICD dischargesand concern over possible negative lifestylechanges appear to place ICD patient at an increasedrisk for affective symptoms such as depression [17].

In an attempt to explain certain forms of depres-sion, the learned helplessness theory of depressionwas developed by Martin Seligman et al. [18,19].Seligman found that laboratory dogs could be clas-sically conditioned to accept a painful electric shockwhen restrained. Such restraint eventually taughtthe animal that it had no way to avoid the aversivestimulus. The dogs eventually gave up trying toavoid the shock, making no attempt to escape newshocks. The apparent giving up generalized to

University of Medicine and Dentistry of New Jersey—SOM,West Orange, New Jersey, (M.G.) and Department of Educa-tional Psychology, Research and Measurement, The Universityof Georgia, Athens, Georgia (B.H.).

Address reprint requests to: Dr. Mark Goodman, BehavioralMedicine Center, 187 Forrest Hill Road, West Orange, NJ 07052,USA.

General Hospital Psychiatry 21, 382–385, 1999382© 1999 Elsevier Science Inc. All rights reserved.ISSN 0163-8343/99/$–see front matter

655 Avenue of the Americas, New York, NY 10010PII S0163-8343(99)00030-4

other situations, and eventually the dogs alwaysappeared to be helpless. In short, learned helpless-ness develops when the animal learns that no be-havioral pattern can influence the environment.

The learned helplessness paradigm has beenused to explain depression in humans. However,replicating this paradigm on humans for researchpurposes has, in the past, been difficult for obviousethical reasons. Because recent studies have linkedICD discharges with depression and anxiety, it isbelieved that the ICD now permits examination oflearned helplessness theory using a human model.To explain, the human model posits that uncontrol-lability of a personal negative life event despiteresponding will gradually lead to depression.Keeping with ICD patients, uncontrollability of in-termittent ICD shocks despite behavioral attemptsto reduce their physical activity in an effort to keeptheir heart rate below the threshold in which theICD will activate may leave the patient feeling help-less. In short, no matter what behavioral patternthey use to gain control, they are unable to avoidthe aversive shock. In turn, this feeling of helpless-ness may lead to depression.

The purpose of the present study was to examinethe learned helplessness theory of depression bystudying cardiac patients whose condition necessi-tated implantation of the ICD. ICD patients withand without history of psychological distress priorto ICD implantation were assessed for currentsymptoms of mental depression and anxiety as afunction of their number of ICD electrical shockfirings. It was hypothesized that, in ICD patientswithout a history of mental depression and anxiety,number of reported intermittent shocks would bepredictive of greater depression and anxiety. Con-versely, in ICD patients with a history of mentaldepression and anxiety, there would be no signifi-cant relationship between number of reported in-termittent shocks and depression or anxiety.

Patients and Methods

Patient Sample

The present study consisted of 25 patients who hadan ICD implanted between February, 1990 and July,1994 at The Union Memorial Hospital in Baltimore,Maryland. An extensive cardiac evaluation includ-ing electrophysiological studies had been com-pleted on all patients prior to their entrance intothis study. Fifteen patients received ICD model1550; three received model 1555; two received

model 1600; three received model 1705; and theremaining patients received model 1715. All pa-tients were Caucasian men. The mean age was 65years (29–80 years). Six of the 25 patients were stillemployed and 19 were retired. Fourteen patientswere married and 11 were single.

Procedure

Institutional Review Board approval was obtainedprior to data collection. Thirty-seven consecutivepatients were mailed a packet containing a demo-graphics questionnaire, a standardized behavioral/psychological symptoms questionnaire, and apostage-prepaid envelope to increase compliancewith the study. The return rate for this study was68% (25/37) and all provided informed consent.

Demographics Questionnaire

Participants reported basic background informa-tion including premorbid medical and psychiatricself-report health histories. Additionally, informa-tion regarding the total number of ICD firings wasobtained through patient consultation and chartrecords.

Revised Symptom Check List-90 (SCL-90-R)

The SCL-90-R [20] is a self-report symptom inven-tory of 90 items used to assess psychological dis-tress, and requires 15–20 minutes to complete. It hasnine primary symptom domains (i.e., Somatization,Obsessive-Compulsive, Interpersonal Sensitivity,Depression, Anxiety, Hostility, Phobic Anxiety,Paranoid Ideation, and Psychotocism). Patients re-ported degree of distress experience during the 7days prior to completing the SCL-90-R. Responseoptions were coded based on a 5-point scale rang-ing from “not-at-all” (1) to “extremely” (5), withhigher scores indicating greater psychologicalsymptomatology. For the purpose of this study,only two of the nine primary symptom dimensions(i.e., depression and anxiety) was used (see Deroga-tis [20] for reliability and validity documentation).

Statistical Analyses

Regression analysis was used to assess the predic-tive potential of ICD discharge shocks on depres-sion and anxiety. A p-value less than 0.05 was con-sidered statistically significant.

Depression and Implantable Defibrillators

383

Results

For the total sample of 25 patients, the mean num-ber of discharge shocks was 2.20 6 1.41. The meanscore on the SCL-90-R measure of depression was1.73 6 0.59 and the mean score on the SCL-90-Rmeasure of anxiety was 1.56 6 0.56.

To test for a human model supporting thelearned helplessness theory of depression, ICD pa-tients were dichotomized into two groups: thosewithout a self-reported history of psychological dis-tress prior to ICD (n 5 18) and those with a self-reported history of psychological distress prior toICD (n 5 7). For patients without a history ofpsychological distress, the mean number of dis-charge shocks was 2.39 6 1.54. The mean score onthe SCL-90-R measure of depression was 1.62 60.56 and the mean score on the SCL-90-R measureof anxiety was 1.44 6 0.51. For patients with ahistory of depression, the mean number of dis-charge shocks was 1.71 6 0.95. The mean score onthe SCL-90-R measure of depression was 2.01 60.52 and the mean score on the SCL-90-R measureof anxiety was 1.87 6 0.58.

Results indicated a significant positive correla-tion between the number of ICD discharge shocksand current depression among ICD patients with-out prior history of psychological distress (r 50.45, p 5 0.03). Also, results indicated a significantpositive correlation between the number of ICDdischarge shocks and current anxiety among ICDpatients without prior history of psychological dis-tress (r 5 0.51, p 5 0.02). However, no significantcorrelation was found between the number of ICDdischarge shocks and either current depression (r 50.16, p . 0.05) or current anxiety (r 5 0.29, p .0.05) among ICD patients with a prior history ofpsychological distress.

Discussion

The results of the present investigation indicate thatthe total number of ICD discharge shocks couldsignificantly predict mental depressive symptomsonly in ICD patients without a history of psycho-logical distress prior to ICD. It is believed that ICDpatients may be reporting greater depression be-cause they have learned that, with increasing num-ber of intermittent shocks, they are unable to mod-ify or control the quantity of discharges. This mayleave them in a state of helplessness and in turnreport greater depression. Our findings are consis-tent with previous research which found a signifi-

cant increase in overall psychological distress fol-lowing ICD firings [12,13,16]. However, our dataare beginning to suggest that the exposure to per-ceived and actual uncontrollability of ICD dis-charge shocks may be associated with the onset ofdepression, and may provide a human model forthe learned helplessness theory of depression.

One major criticism of the present investigation isthat it used a small sample size to compare thepremorbid history with those who had no history ofpsychological distress. The majority of subjects didnot report a history of psychological distress, leav-ing a minority of subjects (n 5 7) with a reportedhistory of psychological distress. While this mayhave made the analysis less powerful because itwas not based on an equal number in the twogroups it nevertheless may be considered a robustfinding because of the magnitude of the significantcorrelation found with the no history of psycholog-ical distress group. Also, it is possible that the un-derlying severity of heart disease may account for acomponent of mental depression. Future studiesmay wish to prospectively consider a larger sampleas well as a more comprehensive assessment ofpremorbid psychological symptoms in order to fur-ther demonstrate a human model for the learnedhelplessness theory of depression.

In conclusion, to more fully understand the af-fective consequences of ICD discharge shocks, fu-ture ICD research is needed to consider the follow-ing: 1) the influence of marital status and genderupon ICD psychiatric sequelae; 2) the cardiotoxiceffect of expressed hostility which has been shownto be predictive of future clinical cardiac pathology[21]; and 3) a neurobehaviorally mediated phobicanxiety conditioned response (i.e., Pavlovian) to theICD shock. Furthermore, considering the potentialexacerbation of depression after ICD implantation,the clinician may wish to consider prophylactic orconcomitant antidepressant therapy (see [22] for athorough review of cardiac electrophysiological ef-fects of antidepressants).

We wish to express thanks for a Union Memorial Heart Foundationgrant supporting the Principal Investigator, Dr. Mark Goodman. Aportion of this paper was presented at the American College ofPhysicians Associates Meeting on May 8, 1997, Baltimore, MD,USA.

References

1. Mirowski M, Mower MM, Langer A, Heilman MS,Schreibman J: A chronically implanted system forautomatic defibrillation in active to conscious dogs:experimental model for treatment of sudden death

M. Goodman and B. Hess

384

from ventricular fibrillation. Circulation 58:90–94,1978

2. Mirowski M, Mower MM, Staewen WS, Tabaznik B,Mendeloff AI: Standby automatic defibrillator. ArchIntern Med 126:158–61, 1970

3. Mirowski M, Reid PR, Mower MM, et al: Terminationof malignant ventricular arrhythmias with an im-plantable ventricular defibrillator in human beings.N Engl J Med 303:322–324, 1980

4. Echt DS, Wing RA: Management of patients with theautomatic implantable cardioverter defibrillator. ClinProgress 3:4–16, 1985

5. Greene HL: Antiarrhythmic drugs versus implant-able defibrillators. The need for a randomized con-trolled study. Am Heart J 127:1171–1178, 1994

6. Kelly PA, Cannom DS, Garan H, Mirabal GS,Harthorne JW, Hurvitz RJ, et al: The automatic im-plantable cardioverter-defibrillator. Efficacy, compli-cations and survival in 26 patients with malignantventricular arrhythmias. J Am Coll Cardiol 11:1278–1286, 1988

7. Nissan S, Barold S: Historical evolution of the auto-matic implantable cardioverter defibrillator in thetreatment of malignant ventricular tachyarrhythmias.In Alt E, Klein H, Griffin JC (eds), The ImplantableCardioverter Defibrillator. Berlin, Springer Verlag,1992

8. Powell AC, Fuchs T, Finkelstein DM, et al: Influenceof implantable cardioverter defibrilators on the long-term prognosis of survivors of out-of-state cardiacarrest. Circulation 88:1083–1092, 1993

9. Tchou PJ, Kadri N, Anderson J, Caceres JA, JazayeriM, Akhtar M: Automatic implantable cardioverterdefibrillators and survival of patients with left ven-tricular dysfunction and malignant ventricular ar-rhythmias. Ann Int Med 109:529–534, 1988

10. Winkle RA, Stinson EG, Echt DS: Practical aspects ofautomatic cardioverter defibrillator implantation.Am Heart J 108:1335–1346, 1984

11. Agency for Health Care Policy and Research. Re-search Activities. Washington, DC: November, 1991

12. Herbst JH, Goodman M, Feldstein S, Reilly JM:Health-related quality-of-life assessment of patientswith life-threatening ventricular arrhythmias. PACE22:915–926, 1999

13. Burgess ES, Quigley JF, Moran G, Sutton FJ, Good-man M: Predictors of psychosocial adjustment in pa-tients with implantable cardioverter defibrillators.PACE 20:1790–1795, 1997

14. Hegel MT, Griegel L, Goulden L, Ozahowski T: Psy-chosocial impact of the implantable cardioverter de-fibrillator (ICD). Paper presented at the Associationfor the Advancement of Behavior Therapy Conven-tion, San Diego, CA, November, 1994

15. Pycha C, Calabrese JR, Gulledge AD, Maloney JD:Patient and spouse acceptance and adaptation to im-plantable cardioverter defibrillators. Cleve ClinJ Med 57:441–444, 1990

16. Luderitz B, Jung W, Deister A, Manz M: Patient ac-ceptance of the implantable cardioverter defibrillatorin ventricular tachyarrhythmias. PACE 16:1815–1821,1993

17. Fricchione GL, Vlay SC: Psychiatric aspects of pa-tients with malignant ventricular arrhythmias. Am JPsychiatry 143:1518–1526, 1986

18. Seligman M: Helplessness: On Depression, Develop-ment and Death. San Francisco, CA, W.H. Freeman,1975

19. Miller WR, Roselini RA, Seligman M: Psycho-pathology: Experimental models. San Francisco, CA,W.H. Freeman, 1977

20. Derogatis LR: SCL-90 R (Revised) Version Manual-I.Baltimore, MD, Johns Hopkins University School ofMedicine, 1983

21. Goodman M, Quigley J, Moran G, Meilman H, Sher-man MF: Hostility predicts restenosis after percuta-neous transluminal coronary angioplasty. Mayo ClinProc 71:729–734, 1996

22. Saxena PR, Villalon CH: Cardiovascular effects ofserotonin agonists and antagonists. J CardiovascPharmacol 15:S17–S34, 1990

Depression and Implantable Defibrillators

385