- Email: [email protected]
Neurocognitive Functioning in Recently Abstinent, Cocaine-Abusing Schizophrenic Patients
NEUROCOGNITIVE FUNCTIONING IN RECENTLY ABSTINENT, COCAINE-ABUSING SCHIZOPHRENIC PATIENTS MARK R. SERPER* MARC L. COPERSINO DANIELLE RICHARME NEHAL VADHAN Hofstra University, Hempstead, NY, USA ROBERT CANCRO New York University School of Medicine, New York, NY, USA ABSTRACT: Purpose. This report examined a broad range of cognitive functioning in a group of recently abstinent, cocaine-abusing schizophrenic patients (CA + SZ). Methods. Measures of selective and sustained attention, learning and memory, and executive functioning were administered to CA + SZ patients within 72 h of last cocaine use. A comparison group of non-substance-abusing schizophrenic patients (SZ) presenting for inpatient psychiatric treatment were also examined in an identical time frame. We hypothesized that the neurobiological impact of cocaine abuse and acute abstinence would cause CA + SZ to manifest deficits in all domains of cognitive functioning relative to non-abusing SZ patients. Results. Results revealed that CA + SZ displayed significant memory impairment relative to their non-abuser SZ counterparts. No group differences, however, were detected on any other neurocognitive measure. CA + SZ were able to selectively process digit strings during the presence and absence of distracting stimuli, sustain attention, and perform executive functions at performance levels equal to their non-abuser SZ counterparts. Implications. These results are consistent with many past studies that have found CA + SZ patients to manifest memory impairment but have relatively well perserved functioning in other cognitive domains. The results are discussed in terms of the biological concomitants of cocaine abuse and acute abstinence in schizophrenia. *Direct all correspondence to: Dr. Mark R. Serper, Department of Psychology, Hofstra University, 222 Hauser Hall, 1000 Fulton Street, Hempstead, NY 11549-1270, USA; E-mail: [email protected] JOURNAL OF SUBSTANCE ABUSE, Volume 11, Number 2, pages 205 ±213. Copyright # 2000 by Elsevier Science Inc. All rights of reproduction in any form reserved. ISSN: 0899-3289
206
JOURNAL OF SUBSTANCE ABUSE
Vol. 11/No. 2/2000
INTRODUCTION The effects of cocaine on the cognitive functioning of schizophrenic (SZ) patients remain relatively unknown.1 There is some evidence from experimental and naturalistic studies that the acute effects of psychostimulants may enhance SZ symptomatic and neurocognitive functioning by activation of prefrontal cortical brain regions (e.g., Goldberg et al., 1991; Serper et al., 1995). Goldberg et al. (1991), for example, found positive behavioral and cognitive effects in chronic SZ patients after receiving a single low dose of dextroamphetamine (0.25 mg/kg) in combination with haloperidol (0.4 mg/kg per day). In this study, 6 out of 25 patients manifested improved conceptual performance on the Wisconsin Card Sorting Test (WCST), as well as a beneficial increase in psychomotor speed. The authors suggest that their findings may be attributable to amphetamine-induced enhancement of prefrontal cortical dopamine receptor activity. In addition, others have found that psychostimulant-induced cognitive enhancement endures longer than druginduced euphoria (Stillman et al., 1993). Yet, the psychostimulant-induced symptomatic or cognitive enhancement appears to be transitory. Cocaine abstinence following binge use is associated with dopaminergic (DA) and serotoninergic (5-HT) depletion (e.g., Dackis and Gold, 1985; McDougle et al., 1992, Parsons et al., 1995; Volkow et al., 1997) and some investigations have found cocaine-abstinent non-SZ patients present with memory and other cognitive impairments (e.g., Berry et al., 1993; Beatty et al., 1995). Similarly, one study found cocaine-abusing schizophrenic patients manifested impaired memory but intact attention compared to their non-abusing SZ counterparts. No study to date, however, has examined a broad range of neurocognitive functioning in SZ patients who present comorobid for cocaine abuse. Consequently, the neurocognitive effects of cocaine abuse on schizophrenic patients remains relatively unknown. In the present report, we examined a wide range of cognitive functions in CA + SZ patients. In addition to learning and memory, we chose to examine selective attention, sustained attention, and executive functioning tasks because schizophrenics have repeatedly been found to show deficits in these cognitive areas (for a recent review, see Keefe, 1995; Rund, 1998). Non-cocaine-abusing schizophrenic patients (SZ), entering Bellevue Hospital Psychiatric Emergency Service (PES), were also tested in an identical time frame and served as the comparison group. We hypothesized that CA + SZ subjects would show performance deficits across all cognitive domains relative their non-abusing SZ counterparts.
METHODS Subjects Two groups of subjects were studied: 35 schizophrenic patients with no current substance abuse diagnoses (SZ); and 21 cocaine-abusing schizophrenia patients with or without other substance use (CA + SZ) tested within 72 h of last cocaine use.2 The subjects were initially selected and tested from patients successively entering the PES at Bellevue Hospital who had an initial chart diagnosis of schizophrenia or
Neurocognitive Functioning in Recently Abstinent CA + SZ Patients
207
TABLE 1 Demographic Data on All Subjects Group SZ Variable Age % Male % African American % Caucasian % Hispanic Years of education Number of previous hospitalizations Amount $$ spent on cocaine 48 h prior to ER arrival CPZ dose Equivalent
M 37.62 82.8 68 23 9 11.82 7.90 ± 416.61 (8 subjects)
CA + SZ SD 10.83 ±
2.70 4.66 ± 298.58
M
SD
34.40 76.1 66 24 10 10.95 7.30
6.06 ±
2.09 3.47
194.56
425.79
590.50 (6 subjects)
305.78
schizophrenia and cocaine abuse/dependence, a positive cocaine urine analysis, and reported last cocaine use within the last 2 days. All subjects were formally interviewed 2 weeks later on the psychiatry inpatient service in order to make a formal DSM-IV diagnoses while limiting the effects of any acute drug-induced psychosis or abstinence symptoms on the initial interview and test results. At this time, the Structured Clinical Interview for the DSM III-R (SCID; Spitzer et al., 1990) was administered and diagnoses for all psychoactive substance abuse/ dependence and for schizophrenia were generated independently by trained SCID raters using DSM-IV criteria for substance abuse/dependence and schizophrenia. All SCID interviews were then subjected to consensus diagnosis with a senior level psychiatrist. The subjects in the current study were a subgroup of our larger sample of 116 patients (Serper et al., 1999). Based on the total sample, the diagnostic reliability between the raters and the psychiatrist was high for all subjects: kappa = 0.87 for schizophrenia; and kappa = 0.82 for cocaine abuse/dependence for all cases. The diagnostic disagreements were settled by discussion. Also, because inconsistencies may exist between the self-report of drug use and urine analysis (e.g., Wilkins et al., 1991), positive urine toxicology for cocaine was required in order to be entered into the CA + SZ patient group. No CA + SZ subject tested urine positive for opiates, methadone, barbiturates or benzodiazepines. SCID diagnostic and substance abuse profiles were also supplemented by the Bellevue Substance Abuse Questionnaire (BSAQ; Munsey et al., 1992) administered to all CA + SZ patients. In addition, the medical chart for all subjects was available and examined for a history of brain illness or injury, mental retardation or any disorder of the central nervous system (CNS) including HIV/AIDS. No subject was entered into the protocol if they had any CNS disorder, mental retardation, HIV/AIDS, or head injury. All subjects were between the ages of 18 and 65 years of age. Demographic information for both subject groups is presented in Table 1.
208
JOURNAL OF SUBSTANCE ABUSE
Vol. 11/No. 2/2000
Although every effort was made to assess the patients in a medication-free state at hospital admission, 14 subjects (eight SZ and six CA + SZ) received neuroleptic medication administration prior to initiation of the protocol. There were no significant differences on any demographic or cognitive variable for patients who received medication versus those who were medication-free, while for those patients receiving neuroleptics, there were no significant group differences in the amount of medication received using chlorpromazine (CPZ) dose equivalents (Davis, 1974).
Assessment Battery CONTINUOUS PERFORMANCE TEST (CPT) A 3±7 version of the CPT was administered. Subjects were given instructions to monitor a series of digits presented in the center of the screen. Subjects were then asked to press the space bar of the computer keyboard with their preferred hand as quickly as possible whenever the 3±7 target sequence appeared. A series of 10 practice trials preceded testing. Using a Taxon monitor with an IBM AT-compatible computer, a series of 500 test trials were presented at a 1,000-ms ISI for 50 ms. In each 100 trials there were 10 3±7 target sequences and 10 non-target occurrences of 3's and 7's. The stimuli faded rapidly without a mask after presentation. The computer-collected data on task performance, which were the number of errors of omission (i.e., a target was presented and the subject failed to respond) and the number of errors of commission (i.e., the subject responds in the absence of a target stimulus) were used to calculate d prime. DIGIT SPAN DISTRACTION TASK (DSDT) In the DSDT (Oltmanns and Neale, 1975), subjects are asked to recall a series of digits presented in a female voice while ignoring competing information presented in a male voice. The DSDT has been used in previous studies that examined selective attention in schizophrenia (e.g., Oltmanns et al., 1978; Serper et al., 1994) with the distraction and non-distraction conditions matched at the time of the tasks' creation for difficulty, reliability, and variability. Two versions of the task were created by randomly selecting and intermixing the distraction trials form the origin task, with the two versions found to be essentially identical in psychometric characteristics. Each version of the task consists of seven distraction and seven non-distraction trials presented to the subject in a taperecorded format. Every other subject received the alternate version of the task. The distraction trials contained five target digits presented in a female voice with four irrelevant digits presented in a male voice in the 2-s interval between the presentation of each target digit. In the non-distraction condition, six target digits were presented at the same rate in a female voice, the distraction and non-distraction trials were intermixed and presented in a fixed random order. Each tape contained a distraction and a non-distraction practice trial. The dependent measures were the percentage of digits correctly recalled in the distraction and non-distraction conditions. WISCONSIN CARD SORTING TEST (WCST) A computerized version of the WCST was administered (Jones et al., 1989). Four stimulus cards were displayed at the top of the computer screen, each labeled with a
Neurocognitive Functioning in Recently Abstinent CA + SZ Patients
209
FIGURE 1 CVLT Memory Performance for Each Group
numerical position. The cards to be sorted were presented one at a time at the bottom of the screen and subjects were instructed to depress the numerical key that corresponds with the desired position for the card. Feedback for each response was provided by a visual display: a correct response was followed by the word ``right'' and an incorrect sort was followed by the word ``wrong'' flashed on the screen. The computer program recorded the responses, and computed all dependent variables. The dependent measures typically reported for the WCST and used in the present report were perseverative errors, number of categories successfully sorted, and number of perseverative responses. CALIFORNIA VERBAL LEARNING TEST (CVLT) Verbal learning and memory was assessed with the CVLT (Delis et al., 1987). The CVLT involves the repeated administration of a word list consisting of common items, which are semantically related to four common conceptual categories (food, clothing, spices, or tools). The list (List A) is presented in a fixed order for five times, with each presentation followed by a free recall for the list. A second word list (List B) taken from the same conceptual categories is then presented, and the subject is asked to recall the items from that list. Free and cued recall of the list is measured again after a long (20 min) delay. The dependent measures included in the present report were the rate of acquisition of List A (Trial 1 vs. Trial 5) and 20-min delayed recall of List A (Trial 5 vs. Delayed Recall).
Procedure All of the subjects were assessed on cognitive tasks in a fixed random order. Informed written consent for study participation was obtained before initiation of the cognitive battery.
210
JOURNAL OF SUBSTANCE ABUSE
Vol. 11/No. 2/2000
TABLE 2 Means and Standard Deviations of Cognitive Performance for Each Group Group SZ Test
CA + SZ
M
SD
M
SD
15.51 21.83 532.49
13.08 32.61 144.63
15.90 46.00 534.70
10.50 48.56 155.79
DSDT Non-distraction trials (% correct) Distraction trials (% correct)
65.94 54.11
20.88 27.94
58.76 55.52
22.14 21.23
WCST Perseverative responses Perseverative errors Categories achieved
30.00 26.34 2.66
23.44 17.66 2.32
38.10 33.05 1.71
25.01 18.71 1.52
CPT Errors of omission Errors of commission Reaction time
Notes.
CA + SZ = Cocaine Abuse and Schizophrenia diagnoses. SZ = Schizophrenia diagnosis.
Data Analysis The demographic data and performance on the WCST, DSDT and the CPT were analyzed using one-way analyses of variance (ANOVA) with group (SZ and CA + SZ) as the independent variable. For the CVLT, two repeated measures ANOVAS were conducted. The first repeated measures ANOVA examined group differences in the extent of List A learning over the task with Group (SZ and CA + SZ) and Learning Trial (Trial 1 and Trial 5) as the independent variables and number of words recalled as the dependent variables. To examine group differences in memory performance, a second repeated measures ANOVA was conducted. The independent variables were group (SZ and CA + SZ) and Learning Trial (Trial 5 and Delayed Recall) and the dependent variables were number of words recalled.
RESULTS The first set analyses examined the demographic characteristics of the groups. There were no significant differences in the age of the subjects, or number of previous hospitalizations, years of education, racial make-up, or CPZ dose medication equivalents for the two groups ( p > 0.05). The means and standard deviations (SD) for participants CVLT is presented in Fig. 1 and participants CPT, WCST, and DSDT scores are presented in Table 2. On the CPT, the one-way ANOVA for d prime was not significant, F(1,54) = 0.01, p = .90. On the DSDT, no significant differences were detected in either distraction, F(1,54) = 0.04, p = .84, or non-distraction, F(1,54) = 1.48, p = 0.23.
Neurocognitive Functioning in Recently Abstinent CA + SZ Patients
211
Similarly, on the WCST, no group differences were detected for perseverative responses, F(1,54) = 1.43, p = 0.23, perseverative errors, F(1,54) = 0.40, p = .52 and number of categories achieved, F(1,54) = 2.68, p = 0.10. On the CVLT, repeated measures ANOVA for the amount of learning revealed a significant main effect for Learning Trial, F(1,28) = 51.23, p < 0.001, indicating that both groups displayed learning over the trials. The main effect for diagnosis, F(1,28) = 1.68, p = 0.21, and the learning by diagnosis interaction, F(1,28) = 1.59, p = 0.21, were not significant. For recall performance, the main effect for Learning Trial was significant, F(1,28) = 13.12, p < 0.001. The main effect for diagnosis was not significant, F(1,28) = 1.42, p = 0.23. The recall by diagnosis interaction, however, was significant, F(1,28) = 4.53, p < 0.05, revealing that the CA + SZ patients retained significantly fewer words during the Delayed Recall Trial then the SZ patients.
DISCUSSION The results of the present investigation found that the majority of CA + SZ patients' neurocognitive functioning was relatively well preserved compared to non-abusing SZ patients. Recent cocaine abuse, specifically, did not appear to worsen the CA + SZ patients' ability to sustain their attention, encode, and immediately recall digit strings during the presence and absence of distracting stimuli or perform executive function operations. While it is clear that both groups' cognitive performance is impaired relative to normal subjects, the CA + SZ subjects performed attentional and executive tasks at compatible levels to their non-abuser SZ counterparts. CA + SZ patients did, however, display significant memory impairment relative to non-abusing SZ patients. Past studies, which examined the effects of cocaine on cognitive functioning in non-SZ patients, are complex. The pattern of results we found, however, are consistent with some past reports examining neuropsychological performance in non-SZ cocaine-abusing subjects. Many past studies have found, for example, that cocaine abusers demonstrate significant memory impairment but retain intact functioning in other cognitive domains (e.g., Mittenberg and Motta, 1988; Ardila et al., 1991; O'Malley et al., 1992; Berry et al., 1993) and the present results are consistent with this finding. Animal studies have found that impaired memory consolidation occurs after the alteration of DA or 5-HT neurotransmitter systems (Dubrovina and Ilyutchenok, 1996; Meneses and Hong, 1997). Reduced DA tone has also been associated with memory impairment and other cognitive impairment in humans (e.g., Volkow et al., 1998). Since acute cocaine abstinence also results in DA and 5-HT depletion, the current results are consistent with the notion that acute cocaine abstinence primarily affects memory functioning in individuals with schizophrenia. Cocaine has also been found to block the induction of long-term potentiation (LTP) of the hippocampus, which has also been implicated in memory impairment (Smith et al., 1993). If the results of past reports can be generalized to the present findings, they suggest that recent cocaine abuse by schizophrenic patients is specifically associated with significant memory impairment.
212
JOURNAL OF SUBSTANCE ABUSE
Vol. 11/No. 2/2000
A limitation to the above discussion is that our study design did not control for all potential determinants of cognitive performance. It may be the case that baseline differences between the groups accounted for the present findings. It should be noted, however, that the two groups, were equivalent on all demographic variables, and hospitalization historyÐsupporting the notion that group differences in memory functioning were the result of cocaine abuse status. Future studies may wish to examine the stability of learning and memory deficits in CA + SZ over periods of prolonged abstinence. ACKNOWLEDGMENTS: We acknowledge the New York City Health and Hospitals Corporation
and Bellevue Hospital for their cooperation. We would also like to thank the Bellevue Emergency Psychiatry Nursing Staff for their cooperation and help in completing this study. This research was supported by a Young Investigator Award from the National Alliance for Research on Schizophrenia and Depression (NARSAD) to Dr. Serper.
NOTES 1. Portions of this article were presented at the 1998 Meeting of the Society of Biological Psychiatry. 2. For the California Verbal Learning Test, a subset of this total sample, consisting of 16 SZ and 14 CA + SZ patients were used because this test was added to the neurocognitive battery at a later date. All cognitive and demographic data were re-analyzed using this subset in order to rule out any potential subsample biases when interpreting the CVLT results. When the data were re-analyzed using the subsample, an identical pattern of results were obtained for all demographic and cognitive test data.
REFERENCES Ardila A, Rossselli M, Strumwasser S Neuropsychological deficits in chronic cocaine abusers. Int J Neurosci 57: 73 ± 9, 1991. Beatty WW, Katzung VM, Moreland VJ, Nixon SJ: Neuropsychological performance of recently abstinent alcoholics and cocaine abusers. Drug Alcohol Depend 37: 247 ± 53, 1995. Berry J, Van-Gorp W, Herzberg DS, Hinkin C: Neuropsychological deficits in abstinent cocaine abusers. Drug Alcohol Depend 32: 231 ± 37, 1993. Delis DC, Kramer JH, Kaplan E, Ober BA: California Verbal Learning Test, Form II, San Antonio, TX, The Psychological Corporation Research ed., 1987. Dubrovina NI, Ilyutchenok RY: Dopamine and opioid regulation of the memory retrieval recovery in mice. Behav Brain Res 79: 23 ± 9, 1996. Goldberg TE, Bigelow LB, Weinberger DR, Daniel DG, Kleinman JE: Cognitive and behavioral effects on coadministration of dextroamphetamine and halperidol in schizophrenia. Am J Psychiatry 148: 78 ± 84, 1991. Keefe RSE: The contribution of neuropsychology to psychiatry. Am J Psychiatry 152: 6 ± 15, 1995. Meneses A, Hong E: A pharmacological analysis of serotonergic receptors: Effects of their activation of blockade in learning. Prog Neuropsychopharmacol Biol Psychiatry 21: 273 ± 96, 1997. Mittenberg W, Motta S: Effects of chronic cocaine abuse on memory and learning. Arch Clin Neuropsychol 8: 477 ± 83, 1988. Oltmanns TF, Neale JM, Ohayon J: The effect of anti-psychotic medication and diagnostic criteria on distractibility in schizophrenia. J Psychiatr Res 14: 81 ± 91, 1978.
Neurocognitive Functioning in Recently Abstinent CA + SZ Patients
213
O'Malley S, Adamse M, Heaton RK, Gawin FH: Neuropsychological impairment in chronic cocaine abusers. Am J Drug Alcohol Abuse 18: 131 ± 44, 1992. Rund BR: A review of the longitudinal studies of cognitive functions in schizophrenic patients. Schizophr Bull 24: 425 ± 36, 1998. Serper MR, Alpert M, Richardson NA, Dickson S, Allen M, Werner A: Clinical effects of recent cocaine use in acute schizophrenia. Am J Psychiatry 152: 1464 ± 69, 1995. Serper MR, Harvey PD, Davidson M: Attentional predictors of clinical change during neuroleptic treatment for schizophrenia. Schizophr Res 9: 55 ± 61, 1994. Serper MR, Chou JC-Y, Allen MH, Czobor P, Cancro R: Symptomatic overlap of cocaine intoxication and acute schizophrenia at emergency presentation. Schizophrenia Bull, 25: 387±94, 1999. Sevy S, Kay SR, Opler LA, van Praag HM: Significance of cocaine history in schizophrenia. J Nerv Ment Dis 178: 642 ± 47, 1990. Smith DA, Browning M, Dunwiddie TV: Cocaine inhibits hippocampal long-term potentiation. Brain Res 608: 259 ± 65, 1993. Spitzer RL, Williams JBW, Gibbon M, First MB: Structured Clinical Interview for the DSM-III-R, Washington, DC, American Psychiatric Press, 1990. Stillman R, Jones RT, Moore D, Walker J, Welm S: Improved performance 4 hours after cocaine. Psychopharmacology 110: 415 ± 20, 1993. Volkow ND, Gur R, Wang GJ, et. al: Association between decline in brain dopamine activity with age and cognitive and motor impairment in healthy individuals. Am J Psychiatry 155: 344 ± 49, 1998. Volkow ND, Wang GJ, Fowler JS, et. al: Decreased striatal dopaminergic responsiveness in detoxified cocaine-dependent subjects. Nature 386: 830 ± 33, 1997. Wilkins JN, Shaner AL, Patterson CM, et. al: Discrepancies between patient report clinical assessment and urine analysis in psychiatric patients during inpatient admission. Psychopharmacol Bull 27: 149 ± 54, 1991.
206
JOURNAL OF SUBSTANCE ABUSE
Vol. 11/No. 2/2000
INTRODUCTION The effects of cocaine on the cognitive functioning of schizophrenic (SZ) patients remain relatively unknown.1 There is some evidence from experimental and naturalistic studies that the acute effects of psychostimulants may enhance SZ symptomatic and neurocognitive functioning by activation of prefrontal cortical brain regions (e.g., Goldberg et al., 1991; Serper et al., 1995). Goldberg et al. (1991), for example, found positive behavioral and cognitive effects in chronic SZ patients after receiving a single low dose of dextroamphetamine (0.25 mg/kg) in combination with haloperidol (0.4 mg/kg per day). In this study, 6 out of 25 patients manifested improved conceptual performance on the Wisconsin Card Sorting Test (WCST), as well as a beneficial increase in psychomotor speed. The authors suggest that their findings may be attributable to amphetamine-induced enhancement of prefrontal cortical dopamine receptor activity. In addition, others have found that psychostimulant-induced cognitive enhancement endures longer than druginduced euphoria (Stillman et al., 1993). Yet, the psychostimulant-induced symptomatic or cognitive enhancement appears to be transitory. Cocaine abstinence following binge use is associated with dopaminergic (DA) and serotoninergic (5-HT) depletion (e.g., Dackis and Gold, 1985; McDougle et al., 1992, Parsons et al., 1995; Volkow et al., 1997) and some investigations have found cocaine-abstinent non-SZ patients present with memory and other cognitive impairments (e.g., Berry et al., 1993; Beatty et al., 1995). Similarly, one study found cocaine-abusing schizophrenic patients manifested impaired memory but intact attention compared to their non-abusing SZ counterparts. No study to date, however, has examined a broad range of neurocognitive functioning in SZ patients who present comorobid for cocaine abuse. Consequently, the neurocognitive effects of cocaine abuse on schizophrenic patients remains relatively unknown. In the present report, we examined a wide range of cognitive functions in CA + SZ patients. In addition to learning and memory, we chose to examine selective attention, sustained attention, and executive functioning tasks because schizophrenics have repeatedly been found to show deficits in these cognitive areas (for a recent review, see Keefe, 1995; Rund, 1998). Non-cocaine-abusing schizophrenic patients (SZ), entering Bellevue Hospital Psychiatric Emergency Service (PES), were also tested in an identical time frame and served as the comparison group. We hypothesized that CA + SZ subjects would show performance deficits across all cognitive domains relative their non-abusing SZ counterparts.
METHODS Subjects Two groups of subjects were studied: 35 schizophrenic patients with no current substance abuse diagnoses (SZ); and 21 cocaine-abusing schizophrenia patients with or without other substance use (CA + SZ) tested within 72 h of last cocaine use.2 The subjects were initially selected and tested from patients successively entering the PES at Bellevue Hospital who had an initial chart diagnosis of schizophrenia or
Neurocognitive Functioning in Recently Abstinent CA + SZ Patients
207
TABLE 1 Demographic Data on All Subjects Group SZ Variable Age % Male % African American % Caucasian % Hispanic Years of education Number of previous hospitalizations Amount $$ spent on cocaine 48 h prior to ER arrival CPZ dose Equivalent
M 37.62 82.8 68 23 9 11.82 7.90 ± 416.61 (8 subjects)
CA + SZ SD 10.83 ±
2.70 4.66 ± 298.58
M
SD
34.40 76.1 66 24 10 10.95 7.30
6.06 ±
2.09 3.47
194.56
425.79
590.50 (6 subjects)
305.78
schizophrenia and cocaine abuse/dependence, a positive cocaine urine analysis, and reported last cocaine use within the last 2 days. All subjects were formally interviewed 2 weeks later on the psychiatry inpatient service in order to make a formal DSM-IV diagnoses while limiting the effects of any acute drug-induced psychosis or abstinence symptoms on the initial interview and test results. At this time, the Structured Clinical Interview for the DSM III-R (SCID; Spitzer et al., 1990) was administered and diagnoses for all psychoactive substance abuse/ dependence and for schizophrenia were generated independently by trained SCID raters using DSM-IV criteria for substance abuse/dependence and schizophrenia. All SCID interviews were then subjected to consensus diagnosis with a senior level psychiatrist. The subjects in the current study were a subgroup of our larger sample of 116 patients (Serper et al., 1999). Based on the total sample, the diagnostic reliability between the raters and the psychiatrist was high for all subjects: kappa = 0.87 for schizophrenia; and kappa = 0.82 for cocaine abuse/dependence for all cases. The diagnostic disagreements were settled by discussion. Also, because inconsistencies may exist between the self-report of drug use and urine analysis (e.g., Wilkins et al., 1991), positive urine toxicology for cocaine was required in order to be entered into the CA + SZ patient group. No CA + SZ subject tested urine positive for opiates, methadone, barbiturates or benzodiazepines. SCID diagnostic and substance abuse profiles were also supplemented by the Bellevue Substance Abuse Questionnaire (BSAQ; Munsey et al., 1992) administered to all CA + SZ patients. In addition, the medical chart for all subjects was available and examined for a history of brain illness or injury, mental retardation or any disorder of the central nervous system (CNS) including HIV/AIDS. No subject was entered into the protocol if they had any CNS disorder, mental retardation, HIV/AIDS, or head injury. All subjects were between the ages of 18 and 65 years of age. Demographic information for both subject groups is presented in Table 1.
208
JOURNAL OF SUBSTANCE ABUSE
Vol. 11/No. 2/2000
Although every effort was made to assess the patients in a medication-free state at hospital admission, 14 subjects (eight SZ and six CA + SZ) received neuroleptic medication administration prior to initiation of the protocol. There were no significant differences on any demographic or cognitive variable for patients who received medication versus those who were medication-free, while for those patients receiving neuroleptics, there were no significant group differences in the amount of medication received using chlorpromazine (CPZ) dose equivalents (Davis, 1974).
Assessment Battery CONTINUOUS PERFORMANCE TEST (CPT) A 3±7 version of the CPT was administered. Subjects were given instructions to monitor a series of digits presented in the center of the screen. Subjects were then asked to press the space bar of the computer keyboard with their preferred hand as quickly as possible whenever the 3±7 target sequence appeared. A series of 10 practice trials preceded testing. Using a Taxon monitor with an IBM AT-compatible computer, a series of 500 test trials were presented at a 1,000-ms ISI for 50 ms. In each 100 trials there were 10 3±7 target sequences and 10 non-target occurrences of 3's and 7's. The stimuli faded rapidly without a mask after presentation. The computer-collected data on task performance, which were the number of errors of omission (i.e., a target was presented and the subject failed to respond) and the number of errors of commission (i.e., the subject responds in the absence of a target stimulus) were used to calculate d prime. DIGIT SPAN DISTRACTION TASK (DSDT) In the DSDT (Oltmanns and Neale, 1975), subjects are asked to recall a series of digits presented in a female voice while ignoring competing information presented in a male voice. The DSDT has been used in previous studies that examined selective attention in schizophrenia (e.g., Oltmanns et al., 1978; Serper et al., 1994) with the distraction and non-distraction conditions matched at the time of the tasks' creation for difficulty, reliability, and variability. Two versions of the task were created by randomly selecting and intermixing the distraction trials form the origin task, with the two versions found to be essentially identical in psychometric characteristics. Each version of the task consists of seven distraction and seven non-distraction trials presented to the subject in a taperecorded format. Every other subject received the alternate version of the task. The distraction trials contained five target digits presented in a female voice with four irrelevant digits presented in a male voice in the 2-s interval between the presentation of each target digit. In the non-distraction condition, six target digits were presented at the same rate in a female voice, the distraction and non-distraction trials were intermixed and presented in a fixed random order. Each tape contained a distraction and a non-distraction practice trial. The dependent measures were the percentage of digits correctly recalled in the distraction and non-distraction conditions. WISCONSIN CARD SORTING TEST (WCST) A computerized version of the WCST was administered (Jones et al., 1989). Four stimulus cards were displayed at the top of the computer screen, each labeled with a
Neurocognitive Functioning in Recently Abstinent CA + SZ Patients
209
FIGURE 1 CVLT Memory Performance for Each Group
numerical position. The cards to be sorted were presented one at a time at the bottom of the screen and subjects were instructed to depress the numerical key that corresponds with the desired position for the card. Feedback for each response was provided by a visual display: a correct response was followed by the word ``right'' and an incorrect sort was followed by the word ``wrong'' flashed on the screen. The computer program recorded the responses, and computed all dependent variables. The dependent measures typically reported for the WCST and used in the present report were perseverative errors, number of categories successfully sorted, and number of perseverative responses. CALIFORNIA VERBAL LEARNING TEST (CVLT) Verbal learning and memory was assessed with the CVLT (Delis et al., 1987). The CVLT involves the repeated administration of a word list consisting of common items, which are semantically related to four common conceptual categories (food, clothing, spices, or tools). The list (List A) is presented in a fixed order for five times, with each presentation followed by a free recall for the list. A second word list (List B) taken from the same conceptual categories is then presented, and the subject is asked to recall the items from that list. Free and cued recall of the list is measured again after a long (20 min) delay. The dependent measures included in the present report were the rate of acquisition of List A (Trial 1 vs. Trial 5) and 20-min delayed recall of List A (Trial 5 vs. Delayed Recall).
Procedure All of the subjects were assessed on cognitive tasks in a fixed random order. Informed written consent for study participation was obtained before initiation of the cognitive battery.
210
JOURNAL OF SUBSTANCE ABUSE
Vol. 11/No. 2/2000
TABLE 2 Means and Standard Deviations of Cognitive Performance for Each Group Group SZ Test
CA + SZ
M
SD
M
SD
15.51 21.83 532.49
13.08 32.61 144.63
15.90 46.00 534.70
10.50 48.56 155.79
DSDT Non-distraction trials (% correct) Distraction trials (% correct)
65.94 54.11
20.88 27.94
58.76 55.52
22.14 21.23
WCST Perseverative responses Perseverative errors Categories achieved
30.00 26.34 2.66
23.44 17.66 2.32
38.10 33.05 1.71
25.01 18.71 1.52
CPT Errors of omission Errors of commission Reaction time
Notes.
CA + SZ = Cocaine Abuse and Schizophrenia diagnoses. SZ = Schizophrenia diagnosis.
Data Analysis The demographic data and performance on the WCST, DSDT and the CPT were analyzed using one-way analyses of variance (ANOVA) with group (SZ and CA + SZ) as the independent variable. For the CVLT, two repeated measures ANOVAS were conducted. The first repeated measures ANOVA examined group differences in the extent of List A learning over the task with Group (SZ and CA + SZ) and Learning Trial (Trial 1 and Trial 5) as the independent variables and number of words recalled as the dependent variables. To examine group differences in memory performance, a second repeated measures ANOVA was conducted. The independent variables were group (SZ and CA + SZ) and Learning Trial (Trial 5 and Delayed Recall) and the dependent variables were number of words recalled.
RESULTS The first set analyses examined the demographic characteristics of the groups. There were no significant differences in the age of the subjects, or number of previous hospitalizations, years of education, racial make-up, or CPZ dose medication equivalents for the two groups ( p > 0.05). The means and standard deviations (SD) for participants CVLT is presented in Fig. 1 and participants CPT, WCST, and DSDT scores are presented in Table 2. On the CPT, the one-way ANOVA for d prime was not significant, F(1,54) = 0.01, p = .90. On the DSDT, no significant differences were detected in either distraction, F(1,54) = 0.04, p = .84, or non-distraction, F(1,54) = 1.48, p = 0.23.
Neurocognitive Functioning in Recently Abstinent CA + SZ Patients
211
Similarly, on the WCST, no group differences were detected for perseverative responses, F(1,54) = 1.43, p = 0.23, perseverative errors, F(1,54) = 0.40, p = .52 and number of categories achieved, F(1,54) = 2.68, p = 0.10. On the CVLT, repeated measures ANOVA for the amount of learning revealed a significant main effect for Learning Trial, F(1,28) = 51.23, p < 0.001, indicating that both groups displayed learning over the trials. The main effect for diagnosis, F(1,28) = 1.68, p = 0.21, and the learning by diagnosis interaction, F(1,28) = 1.59, p = 0.21, were not significant. For recall performance, the main effect for Learning Trial was significant, F(1,28) = 13.12, p < 0.001. The main effect for diagnosis was not significant, F(1,28) = 1.42, p = 0.23. The recall by diagnosis interaction, however, was significant, F(1,28) = 4.53, p < 0.05, revealing that the CA + SZ patients retained significantly fewer words during the Delayed Recall Trial then the SZ patients.
DISCUSSION The results of the present investigation found that the majority of CA + SZ patients' neurocognitive functioning was relatively well preserved compared to non-abusing SZ patients. Recent cocaine abuse, specifically, did not appear to worsen the CA + SZ patients' ability to sustain their attention, encode, and immediately recall digit strings during the presence and absence of distracting stimuli or perform executive function operations. While it is clear that both groups' cognitive performance is impaired relative to normal subjects, the CA + SZ subjects performed attentional and executive tasks at compatible levels to their non-abuser SZ counterparts. CA + SZ patients did, however, display significant memory impairment relative to non-abusing SZ patients. Past studies, which examined the effects of cocaine on cognitive functioning in non-SZ patients, are complex. The pattern of results we found, however, are consistent with some past reports examining neuropsychological performance in non-SZ cocaine-abusing subjects. Many past studies have found, for example, that cocaine abusers demonstrate significant memory impairment but retain intact functioning in other cognitive domains (e.g., Mittenberg and Motta, 1988; Ardila et al., 1991; O'Malley et al., 1992; Berry et al., 1993) and the present results are consistent with this finding. Animal studies have found that impaired memory consolidation occurs after the alteration of DA or 5-HT neurotransmitter systems (Dubrovina and Ilyutchenok, 1996; Meneses and Hong, 1997). Reduced DA tone has also been associated with memory impairment and other cognitive impairment in humans (e.g., Volkow et al., 1998). Since acute cocaine abstinence also results in DA and 5-HT depletion, the current results are consistent with the notion that acute cocaine abstinence primarily affects memory functioning in individuals with schizophrenia. Cocaine has also been found to block the induction of long-term potentiation (LTP) of the hippocampus, which has also been implicated in memory impairment (Smith et al., 1993). If the results of past reports can be generalized to the present findings, they suggest that recent cocaine abuse by schizophrenic patients is specifically associated with significant memory impairment.
212
JOURNAL OF SUBSTANCE ABUSE
Vol. 11/No. 2/2000
A limitation to the above discussion is that our study design did not control for all potential determinants of cognitive performance. It may be the case that baseline differences between the groups accounted for the present findings. It should be noted, however, that the two groups, were equivalent on all demographic variables, and hospitalization historyÐsupporting the notion that group differences in memory functioning were the result of cocaine abuse status. Future studies may wish to examine the stability of learning and memory deficits in CA + SZ over periods of prolonged abstinence. ACKNOWLEDGMENTS: We acknowledge the New York City Health and Hospitals Corporation
and Bellevue Hospital for their cooperation. We would also like to thank the Bellevue Emergency Psychiatry Nursing Staff for their cooperation and help in completing this study. This research was supported by a Young Investigator Award from the National Alliance for Research on Schizophrenia and Depression (NARSAD) to Dr. Serper.
NOTES 1. Portions of this article were presented at the 1998 Meeting of the Society of Biological Psychiatry. 2. For the California Verbal Learning Test, a subset of this total sample, consisting of 16 SZ and 14 CA + SZ patients were used because this test was added to the neurocognitive battery at a later date. All cognitive and demographic data were re-analyzed using this subset in order to rule out any potential subsample biases when interpreting the CVLT results. When the data were re-analyzed using the subsample, an identical pattern of results were obtained for all demographic and cognitive test data.
REFERENCES Ardila A, Rossselli M, Strumwasser S Neuropsychological deficits in chronic cocaine abusers. Int J Neurosci 57: 73 ± 9, 1991. Beatty WW, Katzung VM, Moreland VJ, Nixon SJ: Neuropsychological performance of recently abstinent alcoholics and cocaine abusers. Drug Alcohol Depend 37: 247 ± 53, 1995. Berry J, Van-Gorp W, Herzberg DS, Hinkin C: Neuropsychological deficits in abstinent cocaine abusers. Drug Alcohol Depend 32: 231 ± 37, 1993. Delis DC, Kramer JH, Kaplan E, Ober BA: California Verbal Learning Test, Form II, San Antonio, TX, The Psychological Corporation Research ed., 1987. Dubrovina NI, Ilyutchenok RY: Dopamine and opioid regulation of the memory retrieval recovery in mice. Behav Brain Res 79: 23 ± 9, 1996. Goldberg TE, Bigelow LB, Weinberger DR, Daniel DG, Kleinman JE: Cognitive and behavioral effects on coadministration of dextroamphetamine and halperidol in schizophrenia. Am J Psychiatry 148: 78 ± 84, 1991. Keefe RSE: The contribution of neuropsychology to psychiatry. Am J Psychiatry 152: 6 ± 15, 1995. Meneses A, Hong E: A pharmacological analysis of serotonergic receptors: Effects of their activation of blockade in learning. Prog Neuropsychopharmacol Biol Psychiatry 21: 273 ± 96, 1997. Mittenberg W, Motta S: Effects of chronic cocaine abuse on memory and learning. Arch Clin Neuropsychol 8: 477 ± 83, 1988. Oltmanns TF, Neale JM, Ohayon J: The effect of anti-psychotic medication and diagnostic criteria on distractibility in schizophrenia. J Psychiatr Res 14: 81 ± 91, 1978.
Neurocognitive Functioning in Recently Abstinent CA + SZ Patients
213
O'Malley S, Adamse M, Heaton RK, Gawin FH: Neuropsychological impairment in chronic cocaine abusers. Am J Drug Alcohol Abuse 18: 131 ± 44, 1992. Rund BR: A review of the longitudinal studies of cognitive functions in schizophrenic patients. Schizophr Bull 24: 425 ± 36, 1998. Serper MR, Alpert M, Richardson NA, Dickson S, Allen M, Werner A: Clinical effects of recent cocaine use in acute schizophrenia. Am J Psychiatry 152: 1464 ± 69, 1995. Serper MR, Harvey PD, Davidson M: Attentional predictors of clinical change during neuroleptic treatment for schizophrenia. Schizophr Res 9: 55 ± 61, 1994. Serper MR, Chou JC-Y, Allen MH, Czobor P, Cancro R: Symptomatic overlap of cocaine intoxication and acute schizophrenia at emergency presentation. Schizophrenia Bull, 25: 387±94, 1999. Sevy S, Kay SR, Opler LA, van Praag HM: Significance of cocaine history in schizophrenia. J Nerv Ment Dis 178: 642 ± 47, 1990. Smith DA, Browning M, Dunwiddie TV: Cocaine inhibits hippocampal long-term potentiation. Brain Res 608: 259 ± 65, 1993. Spitzer RL, Williams JBW, Gibbon M, First MB: Structured Clinical Interview for the DSM-III-R, Washington, DC, American Psychiatric Press, 1990. Stillman R, Jones RT, Moore D, Walker J, Welm S: Improved performance 4 hours after cocaine. Psychopharmacology 110: 415 ± 20, 1993. Volkow ND, Gur R, Wang GJ, et. al: Association between decline in brain dopamine activity with age and cognitive and motor impairment in healthy individuals. Am J Psychiatry 155: 344 ± 49, 1998. Volkow ND, Wang GJ, Fowler JS, et. al: Decreased striatal dopaminergic responsiveness in detoxified cocaine-dependent subjects. Nature 386: 830 ± 33, 1997. Wilkins JN, Shaner AL, Patterson CM, et. al: Discrepancies between patient report clinical assessment and urine analysis in psychiatric patients during inpatient admission. Psychopharmacol Bull 27: 149 ± 54, 1991.