Vol. 16, No. 4 July/August 1992

0 145-6008/92/ 1604-0726$3.00/0 ALCOHOLISM: CLINICAL AND EXPERIMENTAL RESEARCH

Systolic Blood Pressure and Neuropsychological Test Performance of Alcoholics John T. Braggio and Vladimir Pishkin

This study examined resting systolic blood pressure (SBP) as a mediator of neuropsychological (NP) test performance in nonhypertensive alcoholics and controls. A median-split was used to assign alcoholics and controls to low and high SBP groups. Results showed that: (1) When SBP level was not considered, alcoholics only performed worse than controls on the WAlS Comprehension subtest. (2) Alcoholics and controls in the high SBP groups had fewer correct items on the WAlS Comprehension and Raven's, Set I tests than subjects in the low SBP groups. (3) Significant Diagnostic Group by SBP Group interaction was found for the Sentence Writing test. For this test only controls in the High SBP Group did worse than controls in the Low SBP Group. (4) Individual group comparisons for all NP tests showed that alcoholics in the High SBP Group were more impaired than controls in the Low SBP Group on the WAlS Comprehension, Shipley Abstraction Age and Stark Visual-Spatial tests; but alcoholics in the Low SBP Group did not differ from, or outperformed, controls in the High SBP Group on the WAlS Comprehension, Shipley Abstraction Age, Raven's, Set I,and Stark Visual-Spatialtests. These data demonstrate that both alcoholism and high SBP adversely and differentially affect the NP test performance of alcoholics and controls. Key Words: Systolic Blood Pressure, Neuropsychological Tests, Alcoholics.

OME INVESTIGATORS have reported that alcoholS ics show greater deficits on neuropsychological (NP) tests than controls. Others have found fewer differences 1-5

between alcoholics and controls on NP tests6 Grant et aL6 suggested that, under some circumstances, differences between alcoholics and controls on NP tests can occur if alcoholics are improperly screened so that those with secondary damage to the central nervous system (CNS) are included in a study. In a later review paper, Grant7 identified three factors, test characteristics (i.e., difficulty and complexity), brain structure and function, and motivation and affect, as determinants of the final NP outcome in alcoholism. Parsonss showed that in a given study about From the Veterans Affairs Medical Center, University of Oklahoma Health Sciences Center, and Willow View Mental Health System. Oklahoma City, Oklahoma Received for publication July 23, 1990: accepted March 17, 1992 Paper presented at the Twenty-Ninth annual meeting of the Society for Psychophysiological Research, New Orleans, Louisiana, October 18- 4 22, 1989. This work was supported by VA medical researchfunds and Oklahoma Center for the Advancement of Science and Technology Health Research Contract No. 3608. Reprint requests: John T. Braggio, Ph.D., Behavioral Sciences Laboratories (15IA), Veterans Affairs Medical Center, 921 N.E. 13th Street, Oklahoma City, OK 73104. Copyright 0 1992 by The Research Society on Alcoholism. 726

70% of the alcoholics have NP deficits. For both the Wechsler Adult Intelligence Scale (WAIS) and the Halstead-Reitan (HR) NP test battery, Parsons and Farr3 concluded that alcoholics showed a pattern of impairment characterized by relatively poor performance in visualspatial (i.e., Block Design, Object Assembly, Picture Arrangement) and tactual-spatial (Tactual Performance Test (TPT) -Time and -Location), constructional tasks, nonverbal abstracting, and set flexibility tasks (i.e., Halstead Category Test, Trails B), and visuomotor speed (i.e., Digit Symbol). While careful subject selection is indeed important, there could be other factors that can influence the alcoholics' NP test performance. One factor which could affect NP test performance is hypertension or high blood Several studies found that nonalcoholic hypertensives are more impaired than nonalcoholic normotensives on NP and on the WAIS.15 These studies indicate that hypertensives are more impaired than normotensives on tests of verbal functioning (i.e., WAIS-Information, Comprehension, Similarities and Vo~abulary),'~ abstracting (i.e., Halstead learning and memory (i.e., TPT-time and location, immediate and delayed memory),' 1 , 1 2 ~ 1and 4 perceptual-motor tasks (i.e., finger tapping,".l3 Digit Substitution testI3). Hypertensives show reduced performance speed and psychomotor coordination. Between 9%and 39% of alcoholics who undergo detoxification have elevated blood pressures of at least 140/90 mmHg16,17 or 160/95 mmHg.ls King et al.I9 used a handgrip dynamometer to test alcoholics 4 to 5 days and 3 to 4 weeks after admission to an alcohol rehabilitation program. They reported significant correlations between diastolic blood pressure and admission systolic blood pressure (SBP) and diastolic blood pressure response to handgrip at both testings. Also, SBP at admission correlated with handgrip-produced increases in SBP at 4 to 5 days. This study showed that alcoholics could have blood pressure dysregulation, even after 1 month of abstinence. The relationship between SBP and NP test performance could explain why in some studies alcoholics are more impaired than control^,^ and in other reports alcoholics do not differ from controk6We suggest that SBP mediates NP test performance in alcoholics. In this experiment we tested alcoholics and controls with Low and High resting SBP on NP tests that have been used in other studies to detect NP impairments in alcoholics only (i.e., visualspatial), hypertensives only (i.e., verbal), and for both (i.e., Alcohol Clin Exp Res, Vol 16, NO 4, 1992: pp 726-733

721

ALCOHOLIC’S RESTING SBP

tactual spatial/learning and memory, constructional tasks, nonverbal abstracting, and set flexibility/abstracting) and visuomotor speed. For the purposes of this paper, NP tests that were used to demonstrate impairments due to alcohol abuse only, resting SBP only, and both factors were grouped into the NP clusters of Verbal, Perceptual-Motor, Abstracting, and Learning-and-Memory. The Stark Verbal and Visual-Spatial test was used as a measure of Learningand-Memory. It was hypothesized that controls with high resting SBP should be impaired on the Verbal subtest, while alcoholics should make more errors on the VisualSpatial subtest. Further, alcoholics with high resting SBPs should do worse on NP tests than alcoholics with low SBPs. In addition, the difference between alcoholics with high and low SBPs should be greater on a Visual-Spatial test than on a Verbal test.20This prediction is based on the fact that alcoholics demonstrate greater impairments on visual-spatial tests than on verbal tests.3 It is possible that for alcoholics visual-spatial tests are functionally more difficult than verbal tests2’ METHOD Subjects We tested 54 adult alcoholics and 26 controls. The alcoholics were inpatients in the 28-day program of the Alcohol Rehabilitation Unit, Oklahoma City Veterans Affairs Medical Center. All alcoholics had a DSM-I11 diagnosis of alcohol dependence, continuous.22 The patients were assigned this diagnosis by the staff psychiatrist. All alcoholic patients showed impairments in social and occupational functioning due to alcohol abuse, which usually included absence from work, loss of job, legal problems, and arguments with family and friends. The exclusion criteria which we used were similar to those reported by Grant et a1.,6 except that the upper age limit of our patients was higher (59 versus 46), and we only selected right handed subjects. All incoming patients were screened by using questionnaires, tests, and a personal interview. Patients who had a medical disorder that could possibly affect CNS functioning, had been unconscious for 1 hr or longer, had cirrhosis of the liver, diabetes, respiratory disease such as COPD, affective disorders like major depression, schizophrenia, nonalcohol produced seizures, a resting blood pressure of 150/90 mmHg or higher were excluded. Hypertensive alcoholics were not included to avoid the adverse effects of hypertension on NP test performance. Patients who had abused drugs, other than marijuana, at least once per month during the 6 months prior to testing, or had received treatment because of drug abuse in the last year prior to testing did not participate. Also, alcoholics with another diagnosis, in addition to alcoholism, were not selected. A newspaper advertisement was used to recruit controls; they came from the Oklahoma City area. Potential volunteers were initially screened on the telephone. Certain individuals were selected only if they met the inclusion criteria used to pick the patients. In addition, controls were selected if they drank 1 ounce or less of ethanol per day during the 6 months prior to the start of the study, and never had a medical or psychosocial problem because of drinking, or received inpatient or‘ outpatient treatment because of alcohol abuse. All subjects were paid for participating in the study.

Procedure The subjects were tested using six NP tests. The tests provided one or two measures of NP functioning for each of the Verbal, Abstracting, Learning-and-Memory, and Perceptual-Motor cluster^.^.^ The WAIS

Comprehension subtest served as the measure of Verbal ability. Items consisted of statements and questions that were used to evaluate an individual’s understanding of ideas and social norms. Answers were scored using the guidelines specified by Wech~ler.’~ The Shipley Institute of Living Scale was used to derive Vocabulary and Abstraction Age scores.24The four subgroups of alcoholics and controls in the Low and High SBP Groups were equated on premorbid intelligence by using the Shipley Vocabulary Age, another measure of verbal functioning (see below). This test uses a multiple choice format to evaluate the breadth of a person’s vocabulary. For each item the subject has to select one of four options given which means the same thing as the reference word. Each correct response receives 1 point; maximum score is 40. Then, the total correct score is converted to a normed score known as Vocabulary Age. The first abstracting measure was obtained by using another Shipley subtest. The items include sequences of numbers, letters, and words which end with one or more blank spaces. The subject had to look at each item, analyze the information given, determine the pattern, and complete the missing space@). The total raw score was converted to an Abstracting Age value using the normed table in the test manual. The second abstracting measure was obtained using the Raven’s, Set I, t e ~ t . 2 ~ This nonverbal test required the subject to form comparisons between figures, and develop a logical method of reasoning. Items consist of incomplete geometric patterns. The subject had to look at each design, determine what the pattern should be in the incomplete section of the figure, and select a pattern from among those given which completes the figure. The Stark test was used to evaluate learning and memory. It consists of pairs of verbal items for one test and two-dimensional visualspatial items for the other. Each test has an easy and hard version. The purpose of the test is to establish a perseverative response tendency with the easy test, and then evaluate performance on the hard test. On the Verbal test the subject had to associate the same set of stimulus words to both easy (is., dark-light) and difficult (i.e., dark-salt) response items.26 On the Visual-Spatial test the subject had to associate the same set of two dimensional designs by drawing easy and difficult geometric stimuli. Since the verbal and visual-spatial associations which the subject was required to learn for the easy tests had a high probability of occurrence, most volunteers make no errors. For this reason, errors and trials to criterion for the easy tests were not scored. The learning criterion on the difficult Verbal and Visual-Spatial tests was two consecutive errorless trials in a maximum of three or six trials, respectively. Dependent measures included determining if a subject reached the preestablished learning criterion, total trials, and total errors. It has been demonstrated that alcoholics make more errors on the Stark Visual-Spatial test than the Verbal test.’” This result is consistent with the literature that found impaired visual-spatial deficits in alcoholic^.^ Perceptual-Motor functioning was assessed using the Digit Vigilance test.’ It consisted of two pages of 0-9 digits. The subject had to cross out each “6.” Omissions were counted as errors. Total error score was used as the dependent measure. The other measure of Perceptual-Motor functioning was obtained using the “Sentence Writing” test. On this test the subject had to write the same sentence up to 5 times, as quickly as possible, so that it was legible, and the words were spelled correctly. The dependent measure was the average amount of time needed to write the sentence. SBP was recorded using a Dinamap Adult/Pediatric Vital Signs Monitor (Critikon, Model 845), and a Dinamap Trend Recorder (Critikon, Model 950).27SBP was measured while each subject sat in a comfortable, straight-back padded chair. Alcoholics had been abstinent at least 21 days when SBP was recorded. Following a 15-min baseline, three SBP readings were taken during the subsequent 5-min period. The SBP readings were averaged, and this value was used to rank, in ascending order, subjects in each of the two groups. Alcoholics with SBP above the median were assigned to the High SBP Group, and those with values below the median were placed in the Low SBP Group. A median split analysis was also used to assign controls to the Low and High SBP Groups.

728

BRAGGIO AND PlSHKlN

RESULTS

Matching Variables The alcoholics and controls in the Low and High SBP Groups were compared on resting SBP and selected matching variables. These data are shown in Table 1. Alcoholics did not differ from controls on mean resting SBP ( p > 0.05). However, as expected, alcoholics in the Low SBP Group had significantly lower SBP values than patients in the High SBP Group ( p < 0.000 1). Similarly, controls in the Low SBP Group had significantly lower SBP values than the controls in the High SBP Group ( p < 0.0001). The four subgroups did not differ in age, education, and Shipley Vocabulary Age (all p’s > 0.05). The alcoholics had a significantly higher daily intake of ethanol than controls ( p < 0.0001). However, alcoholics in the Low and High SBP Groups did not differ in daily ethanol intake ( p > 0.05). Controls in the two SBP Groups did not differ in daily ethanol intake ( p > 0.05), either. The alcoholics had a chronicity of 14.22 years, and had been abstinent for 44.44 days (median = 37.5) before testing. The alcoholics in the two groups did not differ in either chronicity or total days abstinent (both p’s > 0.05). There was a significant negative correlation between days abstinent and SBP for alcoholics in the Low SBP Group (rxy(25)= -0.454, p < 0.05), but not for alcoholics in the High SBP Group (rxy(25)= 0.087, p > 0.05). For all alcoholics the correlation between days abstinent and SBP was not significant (rxy(52)= -0.2 18, p > 0.05). NP Tests The NP results are presented in Table 2. Each test was analyzed using a Diagnostic Group (Alcoholics, Controls) by SBP Group (Low, High) MANOVA.28The dependent measures were scores from the five NP tests listed in Table 2. A Diagnostic Group by SBP Group ANOVA design was used to analyze each dependent mea~ure.’~ Additional analyses were carried out using t and Chi square tests, and correlation coefficients. The MANOVA main effects were not significant. Nonetheless, each dependent measure was analyzed using ANOVA because the SBP effect could be test specific. ANOVAs on each NP test revealed no significant main effect

for Diagnostic Group for four of the tests [F(1,76) < 1 , p > 0.051: Shipley Abstraction Age, Raven’s, Set I, Digit Vigilance, and Sentence Writing. However, on the WAIS Comprehension subtest alcoholics had significantly lower scores than controls [F(1,76) = 5.02, p < 0.021. Consistent with our hypothesis, alcoholics did not differ from controls when resting SBP was not considered. There were significant effects for the SBP Group factor for WAIS Comprehension [F(1,76) = 4.24, p < 0.051, and Raven’s, Set I [F(1,76) = 3.94, p < 0.051 tests. The High SBP Group performed worse than the Low SBP Group on those two tests, again consistent with our prediction. A significant interaction was found for only one measure, the Sentence Writing test [F(1,76) = 5.99, p < 0.051. These data are presented in Figure 1A. Alcoholics in the Low SBP Group required more time to write the sentence than the alcoholics in the High SBP Group. However, statistical comparisons using the NewmanKeuls test indicated that the means for the two alcoholic groups were not significantly different ( p > 0.05). On the other hand, controls in the Low SBP Group took significantly less time to write the sentence than the controls in the High SBP Group ( p < 0.05). These results indicated that on this particular Perceptual-Motor test high resting SBP differentially impaired the ability of controls to quickly write the sentence, but did not significantly impede the speed of alcoholics to quickly and correctly write the sentence. Stark Verbal and Visual-Spatial Subtests SBP Group membership did not determine the number of alcoholics and controls who reached criterion on the ) p > 0.05). However, SBP Group Verbal test ( ~ ~ (=34.00, membership did affect the number of alcoholics and controls who attained the learning criterion on the VisualSpatial test ( ~ ~ ( =3 )8.33, p < 0.05). A higher percentage of controls in the Low SBP Group (77%) reached the learning criterion on the Visual-Spatialtest than alcoholics (54%).The percentage of controls in the High SBP Group who reached the learning criterion on the Visual-Spatial test was low (31%), and resembled the percentage of alcoholics in the High SBP Groups who attained the learning criterion on the Visual-Spatial test (33.5%). These

Table 1. Differences Between Alcoholics and Controls Subdivided by Low and High Systolic Blood Pressure on Selected Variables Alcoholics Controls

Variables

Low SBP group (N = 27)

High SBP group (N = 27)

Alcoholic

Low SBP group (N = 13)

High SBP group (N = 13)

Control roup (N = 86)

114.42 40.00 12.78 17.06

132.52, 41.85 13.37 17.12

i123.47 40.92 13.07 17.08

115.15 37.23 13.08 17.72

131.75’ 40.54 14.15 17.37

123.45 38.88 13.62 17.54

13.60 15.18 43.00

14.40 13.26 45.89

14.02 14.22 44.44

0.87

0.54

0.71

( t r u g p q )

~

SBP (rnmHg) Age Education Shipley vocabulary age (norrned score) EtOH ounces/day Chronicity (in years) Days (since last drink)

Comparisons are between low vs. high SBP alcoholics, low vs. high SBP controls, or alcoholics vs. COntrOlS. * Significant group differences at p < 0.0001.

ALCOHOLICS RESTING SBP

729

Table 2. Neuropsvcholoaical Blood Pressure Groups . . - Test Results and ANOVAs for Alcoholics and Controls, and Subiects in the Low and High - Systolic . Diagnostic Group (D) SBP Group (S) F's Neuropsychologicaltests grouped by cluster

Alcoholics (N = 54)

Controls

Low

(N = 26)

(N = 40)

High (N = 40)

D

S

DXS

16.17

17.96'

17.50

16.00*

5.02'

4.24'

Systolic blood pressure and neuropsychological test performance of alcoholics.

This study examined resting systolic blood pressure (SBP) as a mediator of neuropsychological (NP) test performance in nonhypertensive alcoholics and ...
903KB Sizes 0 Downloads 0 Views