Characteristics in Patients with Schizophrenia
Lena Flyckt, Jo¨rgen Borg, Kristian Borg, Tor Ansved, Gunnar Edman,
Lars Bjerkenstedt, and Frits-Axel Wiesel
Background:In a previous study of motor unit properties in patients with schizophrenia, muscle fiber histologic and electrophysiologic abnormalities were observed. The present study was designed to compare patients with schizophrenia with healthy control subjects with regard to muscle fiber histology and motor unit function. A second objective was to relate these variables to clinical characteristics.
Methods:Twelve patients with first-episode schizophre-nia and fifteen patients with chronic schizophreschizophre-nia (DSM-III-R) and 27 matched control subjects were included in the study. Muscle biopsies were performed either in m. tibialis anterior or m. vastus lateralis. Electromyographic recordings (macro EMG) were made from the m. tibialis anterior motor units. Psychiatric ratings included the PANSS and extrapyramidal side effects.
Results:Seven of the muscle biopsy specimens from the patients and one from the control subjects were classified as abnormal (p 5 .049). The most frequent abnormality was atrophic muscle fibers. Eight patients and no control subjects exhibited pathological macro EMG (p5 .032). The findings were present in chronic as well as in first-episode patients with schizophrenia.
Conclusions:In approximately 50% of the patients, neu-romuscular abnormalities were found either in the muscle biopsy or the macro EMG investigations. The results indicate that either a common pathologic process or different pathological processes are at hand in the neuro-muscular system in patients with schizophrenia. The find-ings are compatible with a disturbed cell membrane function. Biol Psychiatry 2000;47:991–999 ©2000 So-ciety of Biological Psychiatry
Key Words:Muscle biopsy, macro EMG, schizophrenia, muscle fiber atrophy, fiber density, motor unit
Introduction
T
he evidence of neuromuscular abnormalities inschizophrenia is extensive. The type of abnormalities range from skeletal muscle fiber changes (Meltzer and Crayton 1975; Ross-Stanton et al 1980), alterations of
a-motoneuron excitability (Crayton et al 1977a), increased motor unit fiber densities (Crayton et al 1977b; Crayton and Meltzer 1979), and increased branching of terminal motor nerves (Meltzer and Crayton 1974b; Ross-Stanton and Meltzer 1981) to elevated levels of muscular enzymes (Meltzer and Crayton 1974a; Zweig et al 1981). The nature of the neuromuscular changes in patients with schizophrenia is far from clear, but most studies have concluded that they are secondary to neurogenic processes of either central or peripheral origin (Borg et al 1987; Crayton et al 1977a, 1977b; Crayton and Meltzer 1979; Meltzer and Crayton 1974b, 1975; Ross-Stanton and Meltzer 1981; Ross-Stanton et al 1980).
In the studies by Meltzer and coworkers, muscle fiber abnormalities including atrophic fibers, “type-grouping,” “central core fibers,” and “ring fibers” were found in about half of the psychotic patients (Meltzer 1972; Meltzer and Crayton 1975). In healthy controls subjects, muscle fiber abnormalities were found in two of 34 cases (6%) (Meltzer et al 1976). Similar abnormalities have been found in patients with psychotic mood disorders, indicating that they are not specific to schizophrenia (Meltzer 1973). Thus, these findings may be associated with the occur-rence of psychotic symptoms rather than a categorical diagnosis. In a study of schizophrenic patients by Borg and coworkers (1987), the muscle biopsy findings included atrophic fibers, central nuclei, “moth-eaten fibers,” “ring fibers” fiber splitting, and subsarcolemmal glucogen drop-lets. Neuroleptic medication could not explain the ob-served findings because they were also present in neuro-leptic-free patients. The histologic changes observed have been described in neurogenic as well as myogenic disor-ders. Because of the nonspecific nature of the muscular changes, other concurrent findings must be considered to better understand its causal connections. Thus, electro-physiologic investigations of single motor unit properties From the Department of Psychiatry, Karolinska Institutet, Danderyds Hospital (LF,
GE, LB) and the Department of Clinical Neuroscience, Neurology, Karolinska Hospital (JB, KB, TA), Stockholm, and the Department of Neuroscience, Psychiatry, Uppsala University Hospital, Uppsala (F-AW), Sweden. Address reprint requests to Lena Flyckt, M.D., FoUU, Department of Psychiatry,
Danderyds Hospital, S-182 88 Danderyd, Sweden.
Received May 28, 1999; revised November 2, 1999; accepted November 10, 1999.
© 2000 Society of Biological Psychiatry 0006-3223/00/$20.00
demonstrated impaired distal impulse propagation, whereas the axonal conduction velocity and refractory period were normal, indicating a peripheral nerve involve-ment in the patients with schizophrenia (Borg et al 1987). The findings of neuromuscular changes in patients with schizophrenia are intriguing and deserve further explora-tion to find possible pathophysiologic mechanisms. The aim of this study was to evaluate the hypothesis that muscle fiber abnormalities are secondary to loss of distal neural influence in patients with schizophrenia. The novel approach to this question was the inclusion of patients with first-episode schizophrenia, as well as the use of the macro electromyographic (EMG) method because of its ability to discern collateral sprouting, a sign of neurop-athy, and thereby disentangling primary or myogenic muscle fiber abnormalities from those secondary to distal neuropathy. Another aim was to investigate the
relation-ship between motor unit properties and clinical
characteristics.
Methods and Materials
Subjects
Patients with schizophrenia were consecutively recruited on admission to a psychiatric clinic in Stockholm.To be included, the patients had to meet the DSM III-R (American Psychiatric Association 1987) criteria for schizophrenia and be between 18 to 45 years of age. A clinical diagnostic interview and the Structured Clinical Interview for the DSM-III-R (SCID; Spitzer et al 1987) were performed independently by two clinicians (GE and LF). Twenty-seven patients gave their informed consent to enter the study. Twelve of these were experiencing their first-episode of schizophrenia, and 15 were patients with chronic schizophrenia. The two clinicians (GE and LF) agreed in their diagnosis in 26 of 27 cases (96%). In the case of disagreement, consensus was reached after consultation. If a first-episode patient had less than a 6-month history of illness, he or she was followed up to confirm the diagnosis of schizophrenia. For each of the included patients, one healthy control subject with match-ing age and gender was selected from a group of 55 normal individuals (Table 1). The exclusion criteria for patients and control subjects were a history of drug abuse, head injury, or any neurological or serious somatic disease. All subjects were asked about their physical activity and dietary habits, and those considered aberrant were not included (e.g., top-level athletes, binge or anorectic eating habits, vegetarians, allergy to certain foods). Control subjects were excluded if they filled the criteria for a psychiatric diagnosis (DSM-III-R) or had a family history in first- or second-degree relatives of psychiatric illnesses. There was no overlap between the subjects in the present study and the study by Borg et al (1987).
Clinical Measures
The symptoms of illness were rated according to the Positive and Negative Syndrome Scale (PANSS; Kay et al 1987). The Global
Table 1. Sociodemographic and Clinical Characteristics of Schizophrenic Patients and Healthy Control Subjects
Variable
Duration of medication (months)
M 58.2 0.0
Md 30.0 0.0
Range 4 –192 0 – 0
Neuroleptic medication (daily dosage in equivalents of chlorpromazine)
Assessment of Functioning scale (GAF; American Psychiatric Association 1987) was used to rate the level of functioning. The Extrapyramidal Symptom Rating Scale (ESRS; Chouinard et al 1980) was used to rate side effects of neuroleptic medication. All clinical ratings were made when the subject entered the study. Because it was not always possible to perform the muscle biopsy in the acute phase, the clinical ratings occasionally preceded the biopsy by some weeks.Responsiveness to neuroleptic medica-tion was based on a global rating of recovery (“yes, no, partly”) made by a trained clinician (LF) about 6 months after the initiation or adjustment of neuroleptic medication.
Neuroleptic Medication
Two of the first-episode schizophrenic patients had never taken medication for their illness, and one of the chronic patients had been off neuroleptic medication for 6 months. Twenty patients were on conventional neuroleptics, and four were on clozapine. The mean dosage (equivalents of chlorpromazine; Beckman and Laux 1990) and mean duration of neuroleptic medication at the time for the muscle biopsy are shown in Table 1. Additional medication for day and night sedation (benzodiazepines) was allowed and a few patients were on low doses of oxazepam and nitrazepam.
Muscle Biopsy Procedure
Muscle biopsy was performed in patients (n526) and control subjects (n 5 26) either in m. tibialis anterior (TA) or in m. vastus lateralis (VL) using the percutaneous conchotome method (Radner 1962). The reason for the choice of two muscles instead of one was to increase the sites of investigation and thereby increase the representativity. The muscle biopsies from the patients were performed in the same muscle as the matching control subject. The biopsy material was immediately frozen in Freon 22, which was kept at its melting point (2190°C) by liquid nitrogen, and then placed in a freezer at 275°C until further processed. Sections of 10 –15mm were cut in a cryostat operating at 225°C. One patient dropped out of the study because of noncompliance.
Histochemical and Morphometrical Techniques
Cross-sections were stained with haematoxylin-eosin and modi-fied tricrome (Engel and Cunningham 1963) for myofibrillar ATPase (mATPase; Padykula and Herman 1955) and NADH-TR (Scarpelli et al 1958). The classification of muscle fiber types was based on their mATPase staining characteristics, as de-scribed by Brooke and Kaiser (1970).
The cross-sectional areas of the muscle fibers were measured directly from the microscope via a CCD camera (Hamamatsu C3077, Hamamatsu Photonica KK, Japan) connected to an image-analysis processor (Vidas, Kontron Bildanalyse, GmbH, Munich). Measurements were made on 100 type I and type II fibers from each biopsy specimen. The muscle fibers were selected from a central part of the biopsy specimen considered not to contain artifacts. If the total number of fibers of respective type was smaller than these numbers, then all fibers of that type were measured.
Classification of Histopathologic Changes in Muscle Biopsies
The muscle biopsies were blindly classified as normal and abnormal by two neurologists (KB and TA). The biopsy was classified as abnormal if the specimen contained any of the following changes: central nuclei in more than 3% of the fibers, more than eight atrophic muscle fibers, splitting phenomena, inclusion bodies, more than three irregular staining of formazan granules, and the presence of type grouping. The criteria of the classification are based on the work of Mastaglia and Detchaut (1992). A neurologist (TA), blinded to the diagnosis, estimated modal distributions of fiber areas by inspection of fiber area histograms made for each subject.
Macro EMG
The first 22 patients and 10 control subjects included in the study were investigated with macro EMG. Electromyographic record-ings were made from the m. tibialis anterior motor units by macro EMG needle electrodes (Medelec 17915) and displayed on a Medelec oscilloscope (MS92B no 67127) connected to a microcomputer (Victor PCII). The needle electrode is a modified single fiber EMG electrode. One single fiber recording was registered from a side port on the needle electrode with an uptake area of one or two muscle fibers. The single fiber action potential was used to trigger the recordings of simultaneous electrical activities picked up by the electrode shafts from “all” muscle fibers belonging to the same motor unit. By averaging during slight voluntary activation, the electric activity was obtained as the macro EMG action potential.
Recordings were made from motor units recruited at slight voluntary activation, a prerequisite for the analysis. Because one of the patients was unable to keep a slight and stable activation, the subject was excluded. Macro EMG recordings were analyzed according to Stålberg (1990) using Stålberg Intersoft program Macro 4.1 1992. Data collected from the patients were compared with published age-related reference data from a previous study. (Stålberg and Fawcets 1982) and 10 age-and gender-matched control subjects from the present study. Two or more motor unit potentials with an amplitude, area, or median value outside the defined range (62 SDs) was considered pathological (Stålberg 1990; Stålberg and Fawcets 1982). The fiber density was defined as the mean number of single muscle fiber action potentials at each of the 20 macro EMG recording sites according to estab-lished criteria (Stålberg and Trontelj 1979).
Statistics
correlation for relationships between variables. A value ofp,
.05 was considered to be statistically significant. Because of the low power of the comparisons, no correction of multiple com-parisons was made so as not to increase the risk of committing type II errors. By making planned comparisons, the increased risk of type I errors was taken into account.
Results
Muscle Biopsy Findings
Histologic changes were found in seven patients and in one control subject (Fisher’s exact test50.049). The most frequent abnormality in the patients was atrophic muscle fibers scattered or in groups (Figure 1 and Table 2). On ATPase staining, the atrophic fibers were of both type I (slow twitch) and type II (fast twitch). Regarding other abnormalities in modified trichrome and NADH-TR stained sections such as vacuols, ring fibers, tubular aggregates, subsarcolemmal glucogen droplets, and irreg-ular staining of formazan granules, there were no signifi-cant differences between the biopsies from patients and those of control subjects.
Mean muscle fiber areas of type I, type II a, or type II b did not differ significantly between patients and control subjects. No differences between patients and control subjects in modal distributions of fiber area histograms was observed. There was no significant difference in mean fiber area between the patients classified as normal or pathological in muscle fiber morphology.
Macro EMG
Macro EMG recordings were available from 20 motor units in each of the 21 patients. According to the used criteria (see Methods and Materials), eight of the schizo-phrenic patients and none of the control subjects exhibited
pathological macro EMG recordings (Fisher’s exact test5
0.032; Table 3). Figure 2 illustrates the recordings from two motor units, one with a larger amplitude and area (Figure 2A) and the other with an amplitude and area within the normal range (Figure 2B). The upper traces show the averaged single fiber action potentials, and the lower traces show the averaged macro EMG potentials. Figure 3 depicts the amplitudes and areas from all 20 motor units recordings in two patients (A and B). Figure 3A shows the data from a patient with a shift of the motor unit potentials toward the upper limit for amplitude and area. Six motor unit potentials were outside the normal range (indicated by the outer square border), and the median value also was outside the normal range (indicated by the inner square border). Figure 3B shows the corre-sponding data from a patient with ordinary motor unit potential amplitudes and areas.
Fiber density (FD; i.e., the mean of the number of single muscle fiber potentials at each recording site) was calcu-lated for all patients and control subjects. In the eight patients with pathological macro EMG motor unit poten-tials, fiber densities ranged from 1.1 to 2.0 (mean51.4), in the 13 patients with normal macro EMG from 1.1 to 1.9.(M51.3), and in the control subjects from 1.1 to 1.7 (M51.4). There were no significant differences in fiber density among these groups.
Relationships among Muscle Biopsy, Macro EMG, and Clinical Characteristics
Seven of twenty-six patients exhibited muscle biopsy abnormalities and 8 of 21 patients a pathological macro EMG. Only three patients showed abnormalities in both investigations [x2(1)5 0.36,p5 .550].
No relationship was found between symptoms of illness and muscle biopsy abnormalities except in patients with positive symptoms, especially delusions (Z51.6657,p5
.09). There were no significant differences between pa-tients with and without muscle fiber abnormalities in the daily dosage of neuroleptic medication, lifetime exposure (months) to neuroleptic medication, GAF, or any of the ESRS items. Patients who were experiencing their first episode of schizophrenia (n512) or who had a chronic course (n514) did not differ in muscle fiber morphology. Three chronic and four first-episode schizophrenic pa-tients had muscle fiber abnormalities. There was no relationship between increased age or responsiveness to medication and muscle biopsy findings (Table 2).
Patients with pathological and normal macro EMG did not differ significantly in daily dosage (equivalents of chlorpromazine), lifetime exposure (months) to neurolep-tic medication, or responsiveness to neurolepneurolep-tic
tion. There were no significant differences between these patient groups in age; in the PANSS, ESRS, or GAF scores; or in whether the patients were first episode (n5
7) or had a chronic course (n514). Two first episode and six chronic patients had pathological macro EMG.
Discussion
Neuromuscular abnormalities were found in 13 patients (48%), either in the muscle biopsy or macro EMG inves-tigations, in both chronic and first-episode schizophrenic patients. The muscle biopsy findings in seven of the patients included angulated atrophic muscle fibers, scat-tered or in groups. Atrophy is the most common response of a muscle fiber to loss of neural influence. Angulated
small fibers, previously normal in size, are considered to be the result of a neural degeneration and subsequent atrophy. Scattered or randomly distributed fiber atrophy may be present in the early phase of neural loss. These findings are in accordance with the results from earlier studies (Crayton et al 1977a, 1977b; Crayton and Meltzer 1979; Meltzer and Crayton 1975; Ross-Stanton et al 1980; Ross-Stanton and Meltzer 1981).
In these earlier studies, distal muscle groups were investigated (m peroneus brevis for the muscle biopsy and m flexor digitorum longus for EMG studies), but in our study, more proximal muscles were studied (m tibialis anterior and m vastus lateralis for the muscle biopsy and m tibialis anterior for the macro EMG). The findings seem to be present regardless of the site of investigation, indicating
Table 2. List of Patient Characteristics
Gender
Male No Pathologic Normalb 31 (24) 65 18 100 Yes
Male No Pathologic Normalc 168 (4) 40 144 0d Partly
Male No Pathologic Pathologicb Moth-eaten fibers, I—atrophy
192 (48) 45 192 175 No
Male No Pathologic Normalc 49 (60) 50 48 325e Partly
Male No Normal Normalb 173 (24) 37 36 150 Partly
Male No Normal Normalc 120 (96) 40 120 800e No
Male No Normal Normalc 67 (12) 45 63 350 No
Male No Normal Normalc 23 (8) 45 12 200 Yes
Male No Pathologic Normalc 47 (18) 50 27 200 Partly
Male No Normal Normalc 132 (48) 65 120 375e No
Male Yes Pathologic Normalc 7 (12) 53 7 200 Yes
Male Yes Normal Pathologicc Scattered atrophic fibers
16 (9) 20 16 100 Yes
Male Yes Normal Not done 15 (12) 50 4 200 Yes
Male Yes Not done Normalc 17 (12) 65 0 0f No
Male Yes Not done Pathologicc I—atrophy 10 (9) 70 10 150 No
Female No Normal Normalc 108 (24) 60 88 200 Yes
Female No Pathologic Pathologicb II–atrophy 305 (2) 50 180 350e No
Female No Normal Normalc 94 (12) 45 74 325 Yes
Female No Not done Normalc 108 (12) 67 69 200 Yes
Female No Normal Pathologicc II b—atrophy, central nuclei
180 (6) 50 120 225e Partly
Female Yes Normal Normalc 19 (12) 70 16 150 Yes
Female Yes Normal Pathologicc Scattered atrophic fibers
56 (12) 50 12 250 Yes
Female Yes Normal Pathologicc II b—atrophy 8 (10) 35 8 200 Yes
Female Yes Not done Normalc 127 (6) 35 20 500e Partly
Female Yes Not done Normalc 12 (6) 40 12 200 Partly
Female Yes Pathologic Normalc
7 (7) 65 4 300 Yes
Female Yes Not done Normalc 2 (48) 45 0 0f n/a
EMG, electromyogram; GAF, Global Assessment of Functioning scale.
aThe figures indicate the time from the first psychiatric contact to the investigation. Figures within parentheses show the time from the onset of symptoms to the first psychiatric contact.
that the peripheral motor neurons and skeletal muscles are affected throughout the body.
Pronounced physical inactivity may result in muscle fiber atrophies (Edstro¨m and Grimby 1986; Houston et al 1979; Lindboe and Platou 1982). If the muscle fiber atrophies among the patients in our study were caused by inactivity, primarily the type 2 b fibers would have been affected (Houston et al 1979), but atrophies were found in all types of fibers.Thus, inactivity did not seem to be the reason for the findings. In a study by Borg et al (1987) the muscle biopsy findings included atrophic fibers, central nuclei, and “moth-eaten fibers,” indicating loss of lower motor neurons in accordance with our study. We did not replicate the findings of vacuols, ring fibers, tubular aggregates, subsarcolemmal glucogen droplets, and irreg-ular staining of formazan granules, however. In our study, about half of the patients were women and half were experiencing their first episode of schizophrenia; in the study by Borg et al (1987), there were no women and only one patient was first episode. Studies have indicated gender-related differences in muscle fiber morphology, although this aspect was not observed in our study (Glenmark et al 1992). Thus, the large proportion of first-episode schizophrenic patients in our study seems to be the most likely reason for the different findings.
The macro EMG recordings were pathological in 8 of 21 patients. The motor unit potentials had increased amplitude and area compared with data from age- and
gender-matched control subjects, whereas fiber densities did not differ. The interpretation of the abnormal macro EMG data must consider factors affecting the muscle fibers, as well as the peripheral and central neurons (Stålberg 1990).
Muscle fiber hypertrophy may cause an increase of the motor unit potential (Stålberg 1990); however, no hyper-trophic fibers were found in the biopsies of the schizo-phrenic patients. Other factors have to be considered, including those influencing the muscle fiber action poten-tial (E. Stålberg, personal communication). Thus, dis-turbed muscle fiber membrane function may be consistent with an increase of the macro EMG potential and retained fiber density.
Increased fiber density, a sign of denervation and reinnervation by collateral sprouting, has been described previously in studies of psychotic patients (Crayton et al 1977b; Meltzer and Crayton 1974a; Ross-Stanton and Meltzer 1981). The patients in these studies exhibited different psychotic conditions, however; our study in-cluded a large number of first-episode schizophrenic patients, a variable not previously investigated in relation to fiber density. Therefore, it is possible that the differ-ences in fiber density between the studies may indicate a progress in motor unit pathology with the duration of illness. We did not find any relationship between the disease duration and motor unit abnormalities, however. Moreover, muscle fiber abnormalities were found in four
Table 3. Macro Electromyographic Data in Schizophrenic Patients and Matched Control Subjects
Subject
Amplitude (mV)
Area (mV3ms)
Fiber density
(n) Classification
Gender P C P C P C P C
Malea 254 186 1709 1207 1.3 1.6 Pathol Normal
Female 278 190 1287 1164 1.1 1.1 Pathol Normal
Male 239 190 1554 837 1.2 1.6 Pathol Normal
Male 298 1345 1.2 Pathol
Male 479 1988 1.1 Pathol
Male 478 1936 2.2 Pathol
Male 532 268 2331 1450 1.5 1.7 Pathol Normal
Femalea 307 257 2079 968 1.8 1.2 Pathol Normal
Femalea 254 199 1279 972 1.3 1.3 Normal Normal
Malea 239 1020 1.3 Normal
Female 181 186 906 1018 1.3 1.4 Normal Normal
Male 176 684 1.1 Normal
Malea 181 960 1.3 Normal
Male 187 955 1.4 Normal
Male 195 215 1191 1001 1.4 1.6 Normal Normal
Femalea 181 156 1154 810 1.4 1.5 Normal Normal
Female 185 1130 1.4 Normal
Femalea 210 184 1258 1041 1.9 1.4 Normal Normal
Male 141 663 1.5 Normal
Female 170 748 1.4 Normal
P, patient; C, control subject.
first-episode schizophrenic patients, and pathological macro EMG were found in two first-episode schizophrenic patients, demonstrating that the changes in motor unit properties begin in the early course of schizophrenia and are not the result of a chronic course. Differences between the results of the two studies may be attributablex to contrasting methods (single fiber EMG vs. macro EMG), different patient samples (other psychotic conditions vs. schizophrenia), or both.
Loss of lower motor neurons with a compensatory, collateral reinnervation might cause an enlargement of the macro EMG motor unit potentials; however, there was no relationship between the macro EMG and the muscle biopsy findings. Further, the fiber density did not differ between patients with and without increased action poten-tials. This argues against a loss of lower motor neurons. A selective loss of low-threshold, type I motor units may be consistent with increased macro EMG potentials and normal fiber density, but the muscle biopsy data did not support that interpretation. Thus, the present electromyo-graphical data do not permit any conclusion about a loss of lower motor neurons.
The possibility of a disturbed central activation must also be considered. In normal subjects, low threshold motor units have smaller motor unit action potentials than do high-threshold motor units (Milner-Brown et al 1973; Stålberg 1990). If there is a changed central activation pattern with an earlier recruitment of larger motor units in schizophrenic patients than in normal subjects, it might explain the observed macro EMG changes. This might then be attributable to a primary disease process affecting both the central and peripheral nervous system.
The fact that the histologic abnormalities were found in some but not all patients, as well as the low concordance between the muscle biopsy and macro EMG findings may be caused by a high rate of false negative muscle biopsy specimens. In our study, the site of investigation included
two different muscle groups; however, it is not possible to eliminate the false-negative problem because there are ethical issues associated with multiple biopsies in the same patient. If the lack of concordance is genuine, however, the interpretation may be that the neuromuscular systems— both central and peripheral—are affected at different levels by one common or several underlying factors, but the processes have reached different stages.
Neuromuscular abnormalities may be connected to disease-related factors, such as clinical characteristics and neuroleptic medication; however, no relationship to symp-toms, duration of illness, or function were found. Further-more, the duration, side effects, and dosage of neuroleptic medication were not related to the neuromuscular results. In a previous study, of the m. tibialis anterior, drug-free patients exhibited the same neuromuscular pathology as those on neuroleptic drugs (Borg et al 1987). This indi-cates the neuromuscular changes are not related to the symptoms, a chronic course or neuroleptic medication.
In conclusion, our data add to the accumulating number of observations regarding motor-system disturbances on multiple levels in schizophrenia, corresponding to previ-ous suggestions by Meltzer and Crayton (1974a, 1975). It also provides new insights, however. The neuromuscular findings seem to be present in first-episode schizophrenic patients, possibly in skeletal muscle groups throughout the body. The combination of findings, that is, an increased Figure 2. Examples of electromyogram (EMG) recordings
showing the average single muscle fiber potential (upper trac-ings) and macro EMG potential (lower tractrac-ings) of two motor units (AandB). The macro EMG potential amplitude and area were(A) above the normal range for one motor unit and (B)
within the normal range for the other.
amplitude and area of the muscle fiber action potential together with a retained fiber density, point to changes in the electrophysiological properties of the cell membrane. Furthermore, the lack of concordance between findings indicates that the pathophysiologic processes may operate independently at different levels. Looking for a possible factor in common, one may speculate about relation to the observations of increased phospholipase A2 activity in
blood cells (Ross et al 1997), regarded as evidence of an increased breakdown of cell membranes. This also seems to take place in the brain, as indicated by reduced levels of phosphomonesters and increased levels of phosphodiesters in schizophrenic patients, determined by magnetic reso-nance spectroscopy (Pettegrew et al 1991). An altered cell membrane composition and function in schizophrenia, as
previously suggested (Horrobin 1998; Wiesel and
Bjerkenstedt 1996), may be consistent with the neuromus-cular findings in the present study.
This study was supported financially by the North-East Health Care district, Stockholm, Sweden. The study was also supported by the Swedish Medical Research Council (Diary Nos. 8318 and 3875). The authors thank Lise Cederstro¨m, nurse, and Elisabeth Ottenhall, secretary, for their skillful assistance.
References
American Psychiatric Association (1987):Diagnostic and Sta-tistical Manual of Mental Disorders, 3rd ed. Washington, DC: American Psychiatric Press.
Beckman H, Laux G (1990): Guidelines for the dosage of antipsychotic drugs.Acta Psychiatr Scand358(suppl):63– 66. Borg J, Edstro¨m L, Bjerkenstedt L, Wiesel FA, Farde L, Hagenfeldt L (1987): Muscle biopsy findings, conduction velocity and refractory period of single motor nerve fibers in schizophrenia. J Neurol Neurosurg Psychiatry 50:1655– 1664.
Brooke M, Kaiser K (1970): Muscle fiber types: How many and what kind?Arch Neurol23:369 –379.
Chouinard G, Ross-Chouinard A, Annable L, Jones BD (1980): The extrapyramidal symptoms rating scale.Can J Neurol Sci
7:233–244.
Crayton J, Stålberg E, Hilton Brown P (1977b): The motor unit in psychotic patients: A single fiber EMG study.J Neurol Neurosurg Psychiatry40:455– 463.
Crayton JW, Meltzer HY (1979): Degeneration and regeneration of motor neurons in psychotic patients. Biol Psychiatry
14:803– 819.
Crayton JW, Meltzer HY, Goode DJ (1977a): Motoneuron excitability in psychiatric patients.Biol Psychiatry12:545– 561.
Edstro¨m L, Grimby L (1986): Effects of exercise on the motor unit.Muscle Nerve9:104 –126.
Engel W, Cunningham G (1963): Rapid examinatin of muscle tissue—an improved trichrome method for fresh-frosen sec-tions.Neurology13:919.
Glenmark B, Hedberg G, Jansson E (1992): Changes in muscle fiber type from adolescence to adulthood in women and men.
Acta Physiol Scand146:251–259.
Horrobin DF (1998): The membrane phospholipid hypothesis as a biochemical basis for the neurodevelopmental concept of schizophrenia.Schizophr Res30:193–208.
Houston ME, Benson H, Larsen H (1979): Interrelationships between skeletal muscle adaptations and performance as studied by detraining and retraining. Acta Physiol Scand
105:163–170.
Kay S, Fiszbein A, Opler L (1987): The Positive and Negative Syndrome Scale (PANSS) for schizophrenia.Schizophr Bull
13:261–276.
Lindboe CF, Platou CS (1982): Disuse atrophy of human skeletal muscle. An enzyme histochemial study. Acta Neuropathol
56:241–244.
Mastaglia F, Detchaut W (1992):Skeletal Muscle Pathology, 2nd ed. London: Timothy Horue.
Meltzer H (1972): Central core fibers in an acutely psychotic patient.Arch Gen Psychiatry27:125–132.
Meltzer H (1973): Skeletal muscle abnormalities in patients with affective disorders.J Psychiatr Res10:43–57.
Meltzer HY, Crayton JW (1974a): Muscle abnormalities in psychotic patients. II. Serum CPK activity, fiber abnormali-ties, and branching and sprouting of subterminal nerves.Biol Psychiatry8:191–208.
Meltzer HY, Crayton JW (1974b): Subterminal motor nerve abnormalities in psychotic patients.Nature249:373–375. Meltzer HY, Crayton JW (1975): Neuromuscular abnormalities
in the major mental illnesses. II. Muscle fiber and subterminal motor nerve abnormalities.Res Publ Assoc Res Nerv Ment Dis54:189 –207.
Meltzer HY, Kuncl RW, Click J, Yang V (1976): Incidence of Z band streaming and myofibrillar disruption in skeletal muscle from healthy young people.Neurology26:853– 857. Milner-Brown H, Stein R, Yemm R (1973): The orderly
recruit-ment of human motor units during voluntary contraction.
J Physiol230:259 –370.
Padykula H, Herman E (1955): The specificity of the histochem-ical method of adenosine triphosphatase.J Histochem Cyto-chem3:170 –195.
Pettegrew JW, Keshavan M, Panchalingam K, et al (1991): Alterations in brain high-energy phosphate and membrane phospholipid metabolism in first episode, drug-naive schizo-phrenics.Arch Gen Psychiatry48:563–568.
Radner S (1962): Knappnålsteknik fo¨r iterativ muskelbiopsi.
Trans Swed Soc Med Sci19:94.
Ross BM, Hudson C, Erlich J, Warsh JJ, Kish SJ (1997): Increased phospholipid breakdown in schizophrenia. Evi-dence for the involvement of a calcium-independent phos-pholipase A2.Arch Gen Psychiatry54:487– 494.
Ross-Stanton J, Meltzer HY (1981): Motor neuron branching patterns in psychotic patients.Arch Gen Psychiatry38:1097– 1103.
Scarpelli D, Hess R, Age R (1958): The cytochemical localiza-tion of oxidative enzymes.J Biophys Biochem Cytol4:747– 752.
Spitzer R, Williams J, Gibbon M (1987): Structured Clinical Interview for DSM-III-R (SCID).New York: Biometric Re-search.
Stålberg E (1990): Macro EMG. Methods Clin Neurophysiol
1:1–14.
Stålberg E, Fawcets P (1982): Macro EMG in healthy subjects of
different ages.J Neurol Neurosurg Psychiatry45:870 – 878. Stålberg E, Trontelj J (1979): Single Fiber Electromyography.
Surrey, UK: Mirvelle.
Wiesel FA, Bjerkenstedt L (1996): Tyrosine transport as an indicator of cell membrane dysfunction in schizophrenia.
Prostaglandins Leukot Essent Fatty Acids55:55–58. Zweig M, Van Steirteghem A, Torrey E (1981): Creatine kinase
isoenzymes in the serum and CSF of schizophrenic patients.
