Regional cerebral blood flow at the time of diagnosis of chronic toxic encephalopathy induced by organic-solvent exposure and after the cessation of exposure.

LINDGREN M. Regional cerebral blood flow at the time of diagnosis of chronictoxic encephalopathy induced by organic-solvent exposure and after the cessation of exposure. Scand J Work Environ Health 1989;15:130- 5. Regional cerebral blood flow was measured in patients with the diagnosis of chronic toxic encephalopathy induced by exposure to organic solvents. Measurements were made at the time of diagnosis and 24-84 months after the cessation of exposure. Duringthe follow-upthe patients werecarefullyexaminedfor other possiblecausesof brain dysfunction. Comparisons were made to unexposed and solvent-exposed referents. At the first examination the pa tients had a 7 070 lower mean flow levelthan the unexposed referents and a 5 % lower levelthan the ex posed workers. The largest flow differenceswereseenin the frontotemporal areas. At the follow-up, the differencein the mean flowlevelbetweenthe patients and referents wasno longer significant. Regionally, the flow had increased, especially in areas whichinitially showed the most pronounced decreases. There wasa significant negative correlation betweenthe initial cerebral blood flow leveland the degreeof nor malization of the flow level at follow-up.

Several stu dies ha ve provided evidenc e of chro nic effects of long-term exposure to or ganic solvents on the central ner vous system . Expo sed subjects have been shown to display memory disturbances, tiredness , irritability, emotional labilit y, difficulties to concentrate, and loss of initiative, a clu ster of symptoms referred to as the psychoorganic or neurasthenic syndrome (1-4). Neuropsychological investigat io ns on exposed workers ha ve shown su bno rma l scores on tests measuri ng memory fun ctions, visual perceptual abilities, and psychomotor fun ctions (2,(5)(6)(7)(8). This type of brain dysfunction, named chronic tox ic encephalopathy , has been recognized as a diagnostic entity for more than a decade in the Scandinavian countries.
In the further study of brain dysfunction associated with solvent exposure, objective methods for evaluating central nervous system fun ction are valuable complements to clinical and neuropsychological techniques. Measurements of the regional cerebral blood flow (regional CBF) by the 133Xenon inh alation method (9, 10) ha ve revealed subn ormal cerebra l blood flow levels and specific pattern s of regional flo w pathology in patients with brain disorders such as organic dementia. For example, the method ha s proved to be highl y d iscriminative between Alz heime r 's and Pi ck 's diseas es and multiinfarct dem entia (11).
T he influence of alcohol on br ain blood flow ha s also been studied . Chronic alcoholics have been found to ha ve reduced regional CBF, espe ciall y in th e frontal lobe s (12). Some normalization ha s been seen after a per iod of abstinenc e (13). Patients with alcohol dementia have, ho wever , a per sistent regional C BF redu ction (12). Thus speci fic patterns of regional CBF abnormalities have been demonstrated in pat ient gro ups with clinical sympto ms somewhat similar to those observed in ch ronic toxic enceph alopath y.
House painters with sus pected chronic toxic encepha lopa thy and computerized tomographic findings indicating brain atrophy have been shown to have a 19 070 lowe r CBF level than referents (14). In a study of 32 exposed workers, Maximilian et al (15) found a 6 % decrea se in the mean flow level in comparison to estim ated control values . We have pr eviously reported a reduction of the CBF level in 50 expo sed paint factory worker s as compared to une xposed re fer ent s mat ched for age and education (16) . Reg ionally, the expos ed wo rkers sho wed th e largest decr ea ses in the fro ntotempo ral areas. The difference in flow level between th e gro ups increased at higher acc umulated expo su re levels.
In the pr esent study patients with clinica lly and neuropsychoiogicall y set diagnoses of chronic to xic encephalopathy were investigated with measur ement s of th e regional C BF at the time o f dia gno sis and afte r 24-84 months without exposure. Th e ob je ctives o f the study were to co n firm clini cal and neuropsychological indications o f brain dysfunction and to determine whether brain function had norm alized a fter the cessation of exposure.

Subjects and methods
Twenty-eight men with the diagnosis of chronic toxic encephalopathy induced by solvent exposure were examined . The diagnosis (17) was based on clinical evidence of neurasthenic symptoms and subnormal performance in the neuropsychological testing, preceded by extensive exposure to organic solvents [range 7-50 (median 26) years]. According to criteria of the World Health Organization (18), the patients all had mild toxic encephalopathy. The regional CBP results were used in the diagnostic evaluation onl y to exclude patients with other organic brain disorder s causing neura sthenic symptoms. The patients were 33-63 (mean 50, SD 10) years of age at the time of the first examination . The clinical and psychometric findings ha ve been described elsewhere (17).
The reference group consisted of 72 male volunteers from the same socioeconomic level as the stud y group . They were without occupational exposure to organic solvents and were selected on the basis of strict criteria for good health and the results of phy sical and neurological examinations and an evaluation of drinking and smoking habits. The referents were 28-68 (mean 48, SD 12) years old.
Patients or referents with a history of severe head trauma, cerebrovascular disease, other neurological disorders, diffuse arteriosclerotic signs, alcohol/drug abuse, and psychiatric or somatic disease were excluded from the study.
Comparisons were also made with a group of 50 male paint factory workers exposed to mixtures of organic solvents (16). The age range of this group was 26-62 (mean 41, SD 10) years, and they had been exposed to solvents for at least 10 (mean 18, SD 10) years.
After the diagnosis of solvent-induced chronic toxic encephalopathy was established, the patients were granted compensation for occupational disease and were withdrawn from exposure . They have been clinically and neuropsychologically reexamined, and before inclusion in the present study they were subjected to a renewed careful differential diagnostic penetration to exclude other possible causes of brain dysfunction than solvent exposure (17). They were also investigated by computerized tomography (CT) , which showed no deviations in brain morphology from healthy referents (19).
The patients had been exposed daily to solvents for 7-50 (median 26) years at the time of the first clinical assessment. An exposure index was calculated from the multiplication of the years of exposure with an intensity factor (20), yielding an exposure index for the group in the range 12-100 (mean 48, SD 23). Eighteen of the patients were still exposed at the time of the initial regional CBP examination [mean age 48 (SD II) years, exposure index 50 (SD 20)], while the remaining 10 patients had already been removed from ex-posure [mean age 52 (SD 9) years, exposure index 45 (SD 28)].
The noninvasive I33Xenon inhalation method was used for measuring regional CBP (10). The technique was outlined by Mallet & Veall (21) and further developed by Obrist et al (22) and Risberg et al (9). A mixture of 133Xenon and air (70-100 MBq /l) is inhaled by the subject during 1 min via a face mask and a rebreathing spirometer system. This procedure is followed by 10 min of normal air breathing. The inert gamma-emitting tracer diffuses into brain tissue from arterial blood and is cleared by venous blood. The gamma radiation emitted by the tracer is continuously monitored by external scintillation detectors. The rate of washout of the tracer forms the basis for flow calculations according to principles described by Obrist et al (22) and Risberg et al (9). In our system (NOVO Diagnostic Systems Inhalation Cerebrograph), 32 detectors covered 16 homologous regions of each hemisphere. The detectors were placed in parallel at a right angle to each lateral surface of the head. The results presented are based on the initial slope index (ISI), which is an index dominated by the blood flow through grey matter (9).
The partial pressure of carbon dioxide in arterial blood (apC0 2 ) was estimated from recordings of endtidal carbon dioxide concentrations (Beckman LB2 anal yzer). Correction of flow values for variations in apC0 2 was performed with the use of correction factors described by Maximilian et al (23).

Design
The regional CBF was measured with the subject resting supine on a comfortable bed. The subject' s eyes were closed , and he was asked to relax without falling asleep. The measurement was preceded by some time of adjustment to the measurement situation. The subject's eyes were covered with eye pads, and ambient noise was kept at a minimum . Resting measurements were made at the time of diagnosis and 24-84 (mean 38.1) months after the cessation of exposure. All the subjects included in the study were examined under identical experimental conditions and with the same equipment.

Data analysis
Intragroup comparisons were made with paired-samples t-tests, and intergroup comparisons were done with two-sample t-tests . Pearson correlation coefficient s were used for the investigation of covariation between the flow variables and exposure-free time, age, psychometric performance, and exposure index. Table 1 shows the mean right and left hemisphere CBF values of the patients at the initial measurement and Table 1. Mea n rig ht and left hemisphere blo od flow fo r the pat ients at the ini tia l measurement (I) and at I oll ow-up (II), and the corresponding values fo r the referent s. Flow values were corrected for vari ati ons in th e partial pressu re of carbon dioxide in arterial blood (apCO z) (two-sample t-test cornparisons) .

Regional cerebral blood flow
Figure I shows regional flow differences between the patients and referents and the patients and the exposed workers at the initial regional CBF assessment. Compared to those of the referents, the regional flows were significantly lower in the patients at all detector sites except two (fr ontal-central) in the left hemisphere. The largest differences were seen in the prefrontal, frontotemporal, and inferior temporal areas bilaterally.
at the follow-up in comparison with those of the reference group . Initially, the patients had a significantly lower mean flow . At the follow-up, the flow difference had diminished and was not statistically significant. Thus some normalization of the mean flow level was seen , but the flow increase did not reac h statistical significance. Compared to the gro up of exposed workers, the pa tient group had a somewhat lower mean flow in both hemispheres on both mea surement occasions. Th e decrease was so mewhat larger in the left occipital ar eas than in th e sa me areas on the right side . Figure 2 shows inc reases in regional CBF in the patient group fro m the time of diagnosis to follow-up . The flow increased in the majority of the areas measured . Th e largest increases were seen in the prefronta l ar eas bilaterally an d in the temporooccipital areas, particularly in the left hemisphere.

Covariation
We did not find an y significant correlation between the age of the patients and the flow increase during the follow -up per iod (r = 0 .26) . The covariation between the length of the follow-up period and the degree of mean flow normalization was also found to be nonsigni ficant (r=0.03) . We found, howe ver, a statistically significant negati ve correlation (r = -0.52, P < 0.005) between th e initial mean flo w level and the degree of mean flow normalization during the followup period . Thus patients with an initially lower mean regional CBF level showed more marked normalization . Whether patients were expo sed or not at the time of the first exam ination had no infl uence on the improvement in the follow-up regional CBF value. The onl y psychometric te st which show ed significant impro vement in the follow-up examination was Benton 's Revised Visual Retention Test (17) . No significant correlation was, however, found between the improvement of the Benton error scores and the blood flow increase.

Discussion
We have pre viously reported reduced CBF in paint factory workers expo sed long-term to organic so lvents , especially in the prefrontal and frontotemporal areas Sign if ica nce s by the two-sampl e t-test. I ncreases of rCBF from initial measurement to follow-up Figure 2. Increases of mean (lSI) and regional cereb ral blood flow (rCBF) in the right (Rt) and left (Lt) hem ispheres fro m the initial measu rement to the follow-up in the patient group. Regional increases are shown as shadowed clock symbols in the circles. Shadowing of 180· co rresponds to a 25 % flow increase. Sign ificance s by the paired-sample t-test .

Regional CBF differen ces
-25%. . . . 25% * P < 0.05 Rt u influence might thus be the cause of the frontal accentuation of regional CBF pathology. The significant negative correlation found between the initial mean flow level and the change at followup indicates that patients with a low initial mean flow level normalize their CBF to a greater extent than patients with a high initialleveI. It is not likely that this phenomenon is caused by regression towards the mean since a low CBF level was not used as a selection criteria for the group . The results are in agreement with our recent finding of a normalization of regional CBF in chronic alcohol abusers during abstinence (13). A subgroup of older alcoholics with a low mean flow level showed more marked normalization during seven weeks of abstinence than younger alcoholics with less reduced regional CBF .
We had difficulties in establishing dose-response relationships in the present study, which illustrates the problems of assessing individual long-term exposure levels reliably. There is probably also a high degree of variation in individual susceptibility due to factor s like the intensity, duration, and profile of exposure. Factors, like work load, protective measures, and previous health status may also vary and result in marked variability in the exposure-re sponse relationships. This probability is further illustrated by the fact that some of our patients were less exposed than the fully active paint factory workers previously examined.
(16). In the present study of patients with a diagnosis of chronic toxic encephalopathy, based on neurasthenic symptoms and subnormal neuropsychological performance, the regional CBF reduction was larger and regionally more generalized. The largest flow reductions were, however, seen in prefrontal and frontotemporal areas also in this group . Thus, the flow reductions were more extensive in the present patient group and indicated more pronounced dysfunction.
Arlien -Seborg et al (14) reported a 19 070 decrease in the mean CBF level of nine house painters when they were compared with II unexposed referents; this decrease is a larger flow reduction than that recorded for the present group. The discrepancy is probably due to subject selection criteria and sample size differences. For example, our patients had normal CT results, while Arlien-Seborg et al included patients with abnormal CT findings. These patients may thus have been more organically affected than those included in the present study. Furthermore, we have found that there is a considerable variance of the CBF level of normal subjects at all ages, and thus , for reliable control data to be obtained, the reference group must be large (24).
Regionally, we found the largest flow decreases in frontal and frontotemporal areas. We have previou sly reported similar frontal regional CBF reductions in chronic alcohol abusers (12,13).
The personality changes, the difficulties in concentration, and the lack of initiative displayed by our patients are symptoms also shown by patients with frontal lobe disorders, like frontal lobe dementia (II) . Thus the symptoms displayed by the exposed patients agree well with what can be expected in frontal lobe disorders.
Growing experimental evidence suggests that longterm exposure to many different organic solvents results in changes in the nervous system which are based on structural breakdown and cellular alterations (18). Chronic neurotoxicity is believed to involve the oxidative metabolism of solvents to reactive intermediates in the neurons and glial cells, which then bind to vital intracellular macromolecules, proteins, ribonucleic acid (RNA) and deoxyribonucleic acid (DNA). This occurrence results in prolonged and progressive cellular dysfunction and eventual cell death (25). Such effects are likely to cause the regional CBF decreases seen in toxic encephalopathy.
We have no definite explanation for the frontal accentuation of the flow reduction in patients with toxic encephalopathy. Experimental evidence suggests, however, that certain organic solvents could induce irreversible astrogliotic alterations in some brain areas known to be sensitive also to chronic ethanol intake (26). Chronic ethanol abuse in man has been reported to result in histopathological glial cell changes (gliosis), especially in the frontal cerebral cortex, hippocampus, and central parts of the cerebellum (26). A higher vulnerability of the frontal cerebral cortex to such an The regional CBF findings agree well with the results of the electroencephalographic power-spectrum analysis of the present patient group (20), which showed an increa se in total power that was more marked than the increase ob served in the previou sly exam ined exposed work ers (16). Some normalization was seen in the follow-up, but the power did not decrease to the level of the unexpo sed referents.
Previous follow-up studies have indicated that, once chronic to xic encephalopathy has develop ed, reversibility or further pro gression is not seen aft er exposure has been discontinued. Our regional CBF data indicate, however, some normalizat ion of bra in function . As has been observed in chro nic alcoholics (13), normalizat ion was seen primaril y in patients with a low initial mean flow level, and thi s findin g possibly indicates that patients with a high CBF level are less organically affected .
In conclu sion, the regional CBF data confirm organic brain involvement in chroni c toxic encephalopathy. The follow-up results suggest some improvemen t in brain function after the discontinuation of exposure, a finding which indicates tha t active rehabilitation for pati ent s with chronic toxic enceph alop ath y may be meanin gful.