Psychiatric Morbidity Following Implantation of the Automatic Implantable Cardioverter Defibrillator

Psychiatric Morbidity Following Implantation of the Automatic Implantable Cardioverter Defibrillator

Psychiatric Morbidity Following Implantation of the Automatic Implantable Cardioverter Defibrillator PHILIP L. MORRIS. M.D., PH.D., JAMES BADGER. R...

813KB Sizes 3 Downloads 32 Views

Psychiatric Morbidity Following Implantation of the Automatic Implantable Cardioverter Defibrillator PHILIP

L.

MORRIS. M.D., PH.D., JAMES BADGER. R.N., M.S.

CHESTER CHMIELEWSKI. M.D., ERIC BERGER. M.D. RICHARD J. GOLDBERG. M.D.

The automatic implantable cardiOl'erter defibrillator (A/CD) has found a place in the treatment ofsustained ventricular tachycardia and in survil'Ors of sudden arrhythmic death. However, little is known aboUl the psychological impact of this del'ice on patients and their families. In a group of20 A/CD recipients, examined between 3 and 21 months post-implantation, the prevalence ofpsychiatric disorder was 50%. Following implantation. six patients sufferedfrom adjustment disorder; three sufferedfrom major depression; and one del'eloped panic disorder. Psychiatric morbidity in patients was associated with psychopathology infamily members, peri-operatil'e A/CD shocks, and social support that was perceived to be inadequate.

T

he automatic implantable cardioverter defibrillator (AICD) has become an important treatment for sustained ventricular arrhythmia refractory to pharmacologic control. When a suitably effective drug regimen cannot be identified by programmed electrophysiologic stimulation (PES) testing. implantation of an AICD may be considered. The device consists of a pulse generator and two shock-delivering patches that are surgically placed on the heart. together with two rate-sensing electrodes. The patches and electrodes are connected to the pulse generator. which is implanted subcutaneously in the abdominal wall. When the heart rate exceeds the rate Received August 24. 1989; revised November 13. 1989; accepted November 27. 1989. From the Depanment of Psychiatry. Brown University; and the Depanment of Psychiatry and the Division of Cardiology. Rhode Island Hospital. Address correspondence to Dr. Morris. Maryland Psychiatric Research Center. P.O. Bolt 21247. Catonsville. MD 21228. Copyright © 1991 The Academy of Psychosomatic Medicine.

58

threshold of the device for any reason (as would occur during sustained ventricular tachycardia [SVT] or ventricular fibrillation IVF], and, occasionally, when there is a very rapid ventricular response to atrial fibrillation [AF», the AICD delivers defibrillating shocks to the heart in an attempt to restore normal rhythm. After AICD implantation and just prior to hospital discharge. the AICD is tested in the conscious patient, who experiences an actual discharge from the device. The device has been extremely effective in reducing mortality from recurrent cardiac arrest and ventricular tachycardia. The I-year mortality from these causes has ranged from 2%-5% with the AICD. compared to 10% with PES guided anti-arrhythmic drug therapy and 10%-40% with empiric drug therapy.u The introduction of new technology. while clinically effective, can have unpredictable effects on patient and family psychosocial functioning. The impact of the AICD on psychological adjustment and quality of life has PSYCHOSOMATICS

Morris et al.

been the focus of interest in only a few studies..'-l> There are no systematic studies of the prevalence of clinical psychiatric morbidity in this population. Further. we have noted that patients receiving unplanned. unexpected. defibrillating shocks (in response to YF. SYT. or occasionally AF) early in the post-operative phase of recovery seem to have more psychological problems adjusting to the device. The study reported here had two main aims. The first was to describe the nature of psychiatric disorder subsequent to AICD implantation. The second was to examine the influence of perioperative AICD shocks on psychiatric morbidity. METHODS Subjects The sample consisted of all AICD patients (N=20) attending the Rhode Island Hospital arrhythmia clinic at the time of the study (19871988). Patients underwent AICD implantation after serial PES testing revealed no anti-arrhythmic drug fully effective in controlling their arrhythmias. Each patient was interviewed at least 3 months after AICD implantation. Three months were allowed to elapse before interview in order for a stable pattern of psychological response to the device to be established. In practice. some patients were seen a considerable time (up to 21 months) after implantation because their AICD was implanted well before the study was initiated. Where possible. a spouse or another family member was interviewed to corroborate the medical. psychiatric. and family history. No patient refused interview. and an informed consent was obtained in all cases. Measures A semi-structured psychiatric interview by a consultation-liaison psychiatrist (PM) formed the basis of data collection. Information concerning the presence of psychiatric problems prior to AICD implantation. the appearance of psychiatric morbidity subsequent to the AICD. and the presence of psychopathology in family members VOLUME .'2· NUMBER I· WINTER 1991

following implantation was obtained. Psychiatric diagnoses were made using the criteria of the Diagnostic and Statistical Manllal. 3rd Edition. Rel'ised (1987) of the American Psychiatric Association. The following information was obtained from the patient. the cardiologist. and the medical record: demographics. cardiac diagnoses. arrhythmia classification. the number of PES tests prior to implantation. medical complications and hospitalizations post-AICD. functional status (measured by assignment to New York Heart Association Functional Class). left ventricular ejection fraction (LYEF). cardiac and other medications. and the number of unexpected. unplanned (yet therapeutic) AICD shocks post-implantation. both peri-operative (within the period of hospitalization for the implant procedureusually 7-10 days) and delayed (following discharge from hospital). Because the perceived level of social support has been shown to modulate the presence of psychiatric morbidity. 7 it was assessed in the study. Adequacy of social support was measured with the Perceived Social Support Scale" (higher scores indicate social support that is perceived to be more adequate). Cognitive performance was evaluated with the Mini-Mental State Examination (MMSE)" in order to assess any effect of sudden death episodes (cardiac arrest) on mental functioning. Data Analysis Data were summarized with descriptive statistics (mean±SD). Comparisons between groups were made by t test or analysis of variance for continuous measures. and the chi-square statistic or Fisher's exact test (where expected cell frequencies were less than five) was used for nominal data. Tests were two-tailed. and a statistically significant result was defined as a p value of <0.05. RESULTS Sample Characteristics Twenty patients were interviewed between 3 and 21 months post-AICD implantation (mean 59

Psychiatric Morbidity

7.5±5.6 months). A description of the demographics, clinical characteristics, psychiatric morbidity, and family psychopathology of the patients is presented in Table I. Peri-operative and delayed medical complications were experienced by 10 patients. Complications included pericarditis. pneumonia, thyrotoxicosis, pancreTABLE I.

atitis, pleural effusion, congestive hean failure and atrial fibrillation. Since interview. two patients died; one from progressive congestive hean failure, and the other from sudden death, despite repeated discharges from the AICD. Six patients received unplanned AICD shocks during the peri-implantation period; shocks in this group

Patient characteristics

Months or Follow-up #/Age/ at Interview Sex

Cardiac Diagnosis

1/49/F

5

Sarcoidosis

Arrhythmia Type SVT

NYHA LVEF Functional % Class

Antiarrhythmic Family Drugs at Psychiatric Member Time or Diagnosis Psychopathology Interview Post-AICD Post-AICD

67

2

Tocainide. Oilantin Oilantin

2/48/M

3

Ischemic

SVT

21

3

3/61/M

5

Ischemic

SVT

30

I

4/69/M

12

Cardiomyopathy SO

56

I

5/62/M

II

Ischemic

SO

35

I

6{72/F

17

Ischemic

SO

III

2

7/63/M

3

Ischemic

SO

26

8/65/M

3

Ischemic

SO

25

2

Inderal. encainide

9/62/M

3

Ischemic

SVT

30

2

Verapamil. procainamide Encainide. amiodarone

1Of73/M

3 4

Ischemic

SO

42

2

Mexiletine

11/59/F

Sarcoidosis

SO

35

2

12/62/M

15

Ischemic

SO

12

2

Quinidine. diltiazem

13/60/M

II

Ischemic

SVT

54

2

Procainamide Adjustment disorder

14n4/M

3

Ischemic

SVT

20

15/63/F

4

Ischemic

SVT

29

16/41/F

4

Sclerodenna

SVT

51

17I56/M

10

Cardiomyopathy SVT

15

4

Amiodarone

Major depression

18/54/M

3

Ischemic

SO

38

2

Encainide

Major depression

Marital problem

19nI/M

21

Ischemic

SO

34

Major depression

Spouse depressed

20/54/M

13

Cardiomyopathy SO

26

Panic disorder

Son developed panic anacks

Inderal Mexiletine Metoprolol

2

Adjustment disorder Adjustment disorder

Encainide

Adjustment disorder

Encainide. diltiazem

Adjustment disorder

Procainamide Adjustment disorder

Inderal. Oilantin

Spouse/daug.hter anxious Marital problem

Son developed aSlhma

Note: SVT=suslained ventricular lachycardia; SD=sudden death episode; LVEF=left ventricular ejeclion fraclion: NYHA=New York Hean Association; AICO=automalic implantable cardiovener defibrillator.

60

PSYCHOSOMATICS

Morris et al.

ranged from one to six. By the time of interview, 18 patients had experienced unplanned shocks. The number of shocks ranged from I to 66. Psychiatric Morbidity Pre-AICD implantation psychiatric morbidity included two patients with organic mental disorder (mild anoxic encephalopathies), which resulted from arrhythmic sudden death episodes, four patients with past alcohol dependence. one patient with generalized anxiety disorder and past alcohol dependence, and one patient with a past history of recurrent depression. Following implantation. 10 patients (50%) had no identifiable psychiatric disorder. Ten patients (50%) developed psychiatric disorder: there were six (30%) with transient adjustment disorder (mixed, anxious or depressive in type), three (\5%) with major depression (one associated with iodide-induced thyrotoxicosis). and one (5%) with panic disorder. Following AICD implantation, the members TABLE 2.

of 14 families (70%) did not manifest any obvious adjustment difficulties. However, ofthe other six families, three had a spouse or child become psychologically distressed (three cases of adjustment disorder: two anxious, one depressed) to the point of requiring medical attention. In two families, marital problems developed; in one the spouse (wife) threatened to leave the patient. In another two families, a child developed a stressrelated illness (panic disorder and asthma, respectively). Correlates of Psychiatric Morbidity In order to identify potential risk factors for the presence of psychiatric morbidity, a comparison of patients divided into three groups was undertaken. The three groups included the following: I) no psychiatric morbidity, 2) adjustment disorder, and 3) major psychiatric morbidity (which included the three patients with major depression and the patient with panic disorder). Table 2 presents this comparison of the

Comparison of no psychiatric morbidity (NPM I, adjustment disorder (AD), and major psychiatric morbidity (MPM)

Factor Months follow-up (mean±SD) Age: yrs (mean±SD) Sex: rk female LVEF. % (mean±SD)

NPM (N=IOI

AD (N=6)

MPM (N=4)

65±5.0

6.3±45

11.7±7.4

62.4±1l.4

59.11±10.6

511.7±1l.2

20

50

0

35.0±15.7

335±16.6

2ll.2±\O.\

90

100

75

3.7±\.9

5.1±\.9

4.7±2.8

Unplanned peri-operative AICD shocks: q. shocked

10

33

75

Delayed medical complications/ month of follow-up

0.04±O.1

O.15±O.I

0.04±O.1

Post AICD hospitalizations/ month of follow-up

0.16±O.2

0.15±O.2

0.22±O.1

1.3 ±I. 6

1.3±O.7

2.2±2.9 29.I±05

NYHA functional class: ':f 1-2 PES Tests (mean±SD)

Delayed AICD shocks/ month of follow-up Past psychiatric history: Ok Positive

40

50

Mini-Mental Slate exam (mean±SD)

211.6±\.4

33 29.I±O.8

Social suppon (mean±SD)

60.8±3.7

60.6±3.6

54.0±5.3

0

50

75

Family member Psychopathology: Ok Positive

NOIt': LVEF=left ventricular ejection fraction: NYHA=New York Hean Association; AICD=automalic implantable cardiovener defibrillator.

VOLUME]2· NUMBER I • WINTER 1'191

61

Psychiatric Morbidity

three groups. As length of follow-up varied between the groups, factors dependent on this characteristic (number of delayed AICD shocks, the frequency of delayed medical complications, and hospitalizations) were divided by length of follow-up (in months) to generate a frequency rate per unit of time. There were no statistically significant differences between the three groups with respect to sociodemographic factors, clinical characteristics. premorbid psychiatric history. or cognitive functioning. There was a statistically significant association between psychiatric morbidity and family problems. Of the 10 patients without psychiatric morbidity, none had family members with psychological problems, compared to 6 of the 10 patients with psychiatric morbidity (p=0.OO5. Fisher's exact test). Social support was perceived as more inadequate by patients with major psychiatric morbidity than by those with no morbidity or with adjustment disorder (F=4.47; df=2. 17; p=O.02). There was a trend for unplanned peri-operative shocks to be associated with psychiatric disorder, particularly major morbidity (X 2 =5.97; df=2; p=O.055). Although the result just failed to reach statistical significance, it tended to support the hypothesis that peri-operative shocks predispose to psychiatric morbidity. In order to determine whether the relationship between peri-operative shocks and psychiatric morbidity might be explained by other factors, patients with (n=6) and without (n=14) peri-operative shocks were compared across the variables shown in Table 2. There were no significant differences noted between the two groups; in particular. mean length of follow-up was similar for both (8.1±7.3 months for the peri-operatively shocked group, and 7.2±4.9 months for the nonshocked group; t=O.77; df=18; p=NS). As the presence of family member psychopathology was strongly associated with psychiatric morbidity. the relationship between peri-operative shocks and family member disorder was examined. No significant relationship was found: three of six patients (50%) with peri-operative shocks had family members with psychopathology vs. five of 14 (36%) without peri-operative shocks (p=NS, Fisher's exact test). 62

DISCUSSION In our sample of 20 AICD recipients observed over a 3- to 2 I-month period. the prevalence of psychiatric disorder was 50%. although the majority of these conditions were of a mi Id transient nature (adjustment disorder). Psychiatric disorder in patients was associated with psychological distress in family members, social support that was perceived by the patient to be inadequate. and unplanned peri-operative AICD shocks. Before further discussion. the limitations of the study should be pointed out. First. the small sample size limits the study's statistical power to discern significant associations between psychiatric morbidity and the independent variables that may. in fact. exist. This sample-size limitation also may have precluded the detection of significant differences between the peri-operatively shocked and nonshocked patients that could have explained the difference in psychiatric morbidity between these two groups. Second. the comparison of groups with and without psychiatric morbidity involved repeated statistical tests. and there is a possibility that statistically significant differences could have been found by chance. Third. the sample may not be representative of the wider AICD population. However. the nature of the patients' cardiac disease. the severity of their illness. and the AICD implantation procedure are similar to those in other reports. I-I> and. as our sample included all AICD recipients attending the Rhode Island Hospital. selection bias is less likely. The rate of psychopathology described falls within the range documented for other patients with chronic medical conditions. Ill. I I and the pattern of psychiatric diagnoses is also consistent with other samples of medically ill patients.12-1~ More difficult to ascertain is whether the implantation of the AICD device precipitates psychological disturbance. The partial confirmation of our suspicion that unplanned peri-operative shocks predicted psychopathology provides a clue to the answerto this question. We have noted that when shocks occur very early post-implant. patient confidence in the device is markedly shaken. confirming other reports~5 of the perPSYCHOSOMATICS

Morris et al.

ceived threatening nature of spontaneous discharges. While it might be expected that the demonstrated ability of the AICD to fire in response to an arrhythmia would enhance patient confidence, this has not been our experience. Rather. the palpable reminder that the patient has a lethal illness may make the individual more vulnerable to psychiatric complications, particularly since AICD shocks coincide with lifethreatening events that clearly match the anxious concerns and preoccupations of the AICD patien!.? The lack of any relationship between perioperative shocks and family problems suggests that this factor has particular salience for the patient only. Because peri-operative shocks predated psychiatric disorder. and because the association between peri-operative shocks and psychiatric morbidity was not explained by other confounding factors, this characteristic of AICD implantation would seem to have an independent intluence. Indeed, because there was a trend in our patients for post-operative AF to precipitate peri-operative AICD shocks (three of four patients with AF were shocked vs. three of sixteen without AF: p=O.06. Fisher's exact test). we have now instituted more aggressive management of post-operative supraventricular arrhythmias. The association between psychiatric morbidity and family psychopathology has been noted previousll and might be explained in two ways. First. emotional distress is a burden for family members of patients with psychiatric disorder.'~ and physical illness that may result in sudden death heightens family member anxiety .16 Both situations may lead to frank disturbance in family members. Second, preexisting family dysfunction may act to make an individual patient more prone to develop psychological distress

after stressful life events.'~ including AICD implantation. While our study was not able to clarify the direction of association between psychiatric and family psychopathology. the results suggest that family members need assessment if a patient's psychological functioning deteriorates. Social support that was perceived to be inadequate was associated with major psychiatric morbidity. This finding retlects recent literature that postulates that the perception of social support mediates reactions to stressful life events.I?IK Furthermore. while the presence of physical illness may serve to mobilize meaningful support, those with fewer key supports at this time may be particularly vulnerable to psychological distress.l~

In conclusion. psychiatric morbidity is a risk faced by many AICD recipients. When it occurs, it is highly likely it will be accompanied by distress in family members. Management plans need to address both these aspects of patient care. Patients at high risk of psychiatric morbidity should be monitored over time, and the provision of adequate social support for patient and family may ameliorate psychological distress. As a way of addressing this concern, we have shown 20 that a support group is a practical and effective method of monitoring progress and of providing social support for AICD recipients and their families.

The authors thank Donna Badger. R.N.. and Nancy Davis for help with data collection and processing. Joan Petteruti for secretarial serI'ices. Ray Niaura. Ph.D.Jor statistical consultation. and Duane Bishop. M.D .. for adl'ice concerning the interpretation ofthe family data.

References I. Kelly P. Cannon D. Gaan H. et al: The automatic implantable cardioverter defibrillalOr: efficacy. complications. and survival in patients with malignant ventricular arrhylhmias. J Am Call Cardiol II: 1278-1281. 1988 2. Mereando A. Furman S. Johnston D. el al: Survival of patients with the aUlomalic implantable cardioverter defibrillalor. PACE II :2059-2062. 1988 3. Cooper DK. Luceri RM. Thurer RJ. el al: The impact of

VOLUME 32· NUMBER I • WINTER 1991

the automatic implantable cardiovener defibrillator on qualily of life. Clinical Progress in ElectrophysiolollY and Pacing 4:306-308. 1986 4. Pycha C. Gulleage AD. Hutzler J. et al: Psychological responses to the implantable defibrillator: preliminary observations. Psychosomatics 27:841-845.1986 5. Brodsky AM. Cannom DS: "High-Tech" Devices That Can Save Lives May Also Heighten Fears (abstract

63

Psychiatric Morbidity

#621). Proceedings of the American Hean Association National Meeting. Washington. DC. November 14-17. 1988 6. Vlay SC: The automatic internal cardiovener defibrillator: comprehensive clinical follow-up. economic and social impact: the Stony Brook experience. Am Heart J 112:189-193.1986 7. Brown GW. Bifulco A: Social suppon. life events and depression. in Social Support: Theory. Research and Applications. Edited by Sarason (G. Samson DR. The Netherlands. Kluwer Academia Publisher. 1985 8. Blumenthal TA. Burg MM. Barefoot S. et al: Social suppon. type A behavior and coronary anery disease. Psychosom Med 49:331-340.1987 9. Folstein MF. Folstein SE. McHugh PR: The Mini-Mental State: a pmctical method for grdding the cognitive state of patients forrhe clinician. J Psychiatr Res 12: 189198.1976 10. Goldberg OP: Identifying psychiatric illness among geneml medicine patients. Br Med J 289:635-636. 1984 II. Morris PLP. Jones B: Tmnsplantation versus dialysis: a study of quality of life. Transplant Proc 20:23-27. 1988 12. Fulop A. Strain JJ. Vita J. et al: Impact of psychiatric co-morbidity on length of hospital stay for medical/surgical patients: a preliminary repon. Am J Psychiatry

64

144:878-882. 1987 13. Menza M. Stem TA. Cassem NH: Treatment of anxiety associated with electrophysiologic studies. Heart Lllng 17:555-559. 1988 14. Morris PLP. Goldberg RJ: The validity of the 28-ltem General Health Questionnaire in gastroenterology inpatients. Psychosomatics 30:290-295.1989 15. Bishop OS. Miller IW: Traumatic brain injury: empirical family assessment techniques. JOllrnal of HI,atl Trallma Rehabilitation 3: 16-30. 1988 16. Rolland JS: Toward a psychosocial topology of chronic and life-threatening illness. Family Systems Medicine 2:245-262. 1984 17. O'Connor P. Brown GW: Supponive relationships: fact or fancy'! JOllrnal of Social and Pa.wnal Rl'laliollships 1:159-175.1984 18. Steinglass P. Weisstub E. Oe-Nour AK: Perceived personal networks as mediator of stress reaction. Am J Psychiatry 145: 1259-1264. 1988 19. Grant I. Pallerson TL. Yager J: Social suppons in relation to physical health and symptoms of depression in the elderly. Am J Psychiatry 145: 1254-1258. 1988 20. Badger JM. Morris PLP: Observations of a suppon group for AICO recipients and their spouses. Heart Lllng 18:238-243.1989

PSYCHOSOMATICS