Correlation between tremor parameters

Transcription

1 Correlation between tremor parameters Ivan Milanov St Naum University Neurological Hospital Sofia, Bulgaria Reprint requests to: Prof. Ivan Milanov, St Naum University Neurological Hospital, B l v d. Tzarigradsko shosse - IV kilometer Sofia 1113, Bulgaria milanovivan@hotmail.com Accepted for publication: September 14, S u m m a r y The relationship between clinically scored tremor and electromyographic measurements of tremor amplitude is not clear and few reports have examined correlations between different tremor parameters. The aim of this study was to evaluate the correlations between different tremor parameters, clinical tremor scoring, age and disease duration. A total of 525 patients with different types of tremor were included in this study. Tremor in various limb positions was graded according to the Webster Tremor Scale. Surface electromyographic recordings of the limb found, in each position, to be most involved were performed in all the patients. The Spearman rank correlation coefficient was used to analyse the relationship between clinical and electromyographic parameters, while the Pearson correlation coefficient was used to analyse the relationship between the electromyographic parameters. Our results revealed moderate correlation between tremor amplitude as measured by electromyography and Webster tremor scores. The correlation between different tremor parameters revealed that low frequency tremors have higher amplitude, and longer burst duration, interval between bursts and missing interval. As the disease progresses, the tremor amplitude increases, while the frequency decreases. A strong positive correlation emerged between corresponding tremor parameters in different limb positions. This suggests that the tremor in all limb positions is generated by the same oscillator. In conclusion, a good correlation was found between clinically scored tremor and EMG measurements of tremor amplitude. The electromyographic examination is a useful tool for researching tremor mechan i s m s. KEY WORDS: Clinical tremor scoring, correlation, electromyogra - phy, tremor. Introduction Electromyography is a tool used in scientific studies of different types of tremor and in tremor treatment trials. Various scales for the clinical assessment of tremor are also widely used. However, the correlation between clinically scored tremor and electromyographic (EMG) measurements of tremor amplitude is not clear. It is known that the amplitude of the EMG activity depends on the distance between the motor units and the recording electrode (1) and that it is proportional to the muscle force and the number of motor units producing the tremor (2). It has been reported that clinical rating scales are even more reliable than accelerometry for tremor assessment (3). Thus it is important to establish whether EMG tremor amplitude is a reliable measure of clinically assessed tremor. Different tremor parameters may be evaluated using EMG examination: tremor amplitude and frequency are the ones usually measured. Rarely, burst duration and the interval between bursts are also measured. Few reports have focused on correlations between the different EMG parameters. A negative correlation has also been reported between tremor amplitude and frequency (4-7), and appears to be most clearly expressed in high amplitude tremors. A negative correlation has also been reported between tremor frequency and burst duration (6,8,9). No studies have looked at possible correlations between tremor amplitude and burst duration or between amplitude, frequency, burst duration and the interval between bursts. In patients with an alternating pattern, the tremor bursts in antagonistic muscles do not coincide. The interval between the end of the agonist and the beginning of the antagonist tremor burst (defined as the missing interval) has never been examined in detail and its correlation with the amplitude, frequency, burst duration and the interval between bursts has not been reported. Establishing the correlations between tremor parameters, age and tremor duration is also important for the prognosis of the disease. It has been reported that, with increasing age, the frequency of physiological tremor decreases (10). This is supposedly attributable to the changes in tissues and mechanical limb properties as well as to a decrease in the motoneuron firing frequency (10). Tremor amplitude, on the other hand, increases (10). The frequency of essential tremor has also been found to decrease with age, while the amplitude increases (11-17). These changes can also be attributed to changes in the tissues and mechanical limb properties (14,18). These correlations between tremor parameters and age were found to be more strongly expressed in type II essential tremors (19). The aim of this study was to evaluate correlations between different tremor parameters, clinical tremor scoring, age and disease duration. Functional Neurology 2002; 17(1):

2 I. Milanov Materials and methods A total of 525 patients with different types of tremor were included in this study. Two hundred and twenty patients (126 female, 94 male) fulfilling the criteria for definite or probable essential tremor (20) were examined. Their mean age was 62.1±14.3 years, and they had a mean disease duration of 13 years. The mean age at onset was 49.1±18.2 years. One hundred and ten patients (44 female and 66 male) with clinically definite or probable Parkinson s disease were also included in this study (21). The mean age of these subjects was 63.1±10.2 years, and they had a mean disease duration of 47 months. Their mean age at tremor onset was 60.3±9.0 years. The rubral tremor group was made up of 24 patients (12 female and 12 male) with a mean age of 39.4±19.3 years. Their mean disease duration was 64.2 months. This group included nine patients with multiple sclerosis, 5 with Wilson s disease, 5 with head trauma and 5 with cerebrovascular disease. The cerebellar tremor group included 22 patients (12 female and 10 male) with a mean age of 45.8±20.6 years and a mean disease duration of 24.4 months. The group included 10 patients with cerebellar degeneration, 4 with multiple sclerosis, 4 with Wilson s disease and 4 with cerebrovascular disease. One hundred and twenty patients (65 female, 55 male) who met DSM-III-R criteria for generalised anxiety disorder also took part in the investigation. They all had a history of anxiety disorder ranging from 1 to 10 months and been affected by enhanced physiological tremor since shortly after the onset of the disease. Patients with associated depression, psychotic features or mania were not included. The mean age of this patient group was 52.8±15.6 years. A further twenty-nine patients (21 female, 8 male) fulfilling the criteria for clinically established and probable psychogenic tremor (22) were included in this study. They had a mean age of 53.5±13.6 years and a mean disease duration of 3.7 years. Their mean age at disease onset was at 49.5±13.9 years. The historical details of the disorder (relating to the onset and progression of the tremor) were noted. A detailed neurological examination was performed. The tremor was studied clinically with the limbs, head and body fully supported against gravity (resting tremor), and with the hands outstretched (postural tremor). Postural tremor in the legs, head and body was, instead, tested in an upright position. Kinetic tremor was investigated during limb (using finger-nose or heel-shin testing) movements, and intention tremor was examined as the limb approached a target. In each position, tremor of the body part found to be most involved was graded according to the Webster Tremor Scale (23), a scale of 0-3. All the patients underwent surface EMG recordings, in each position, on the limb most involved. Tremor activity was recorded, using a pair of electrodes, in the antagonistic groups of muscles of the wrist. The electrodes were positioned 3 cm apart in a longitudinal direction over the extensor carpi radialis and flexor carpi ulnaris muscles (24). Each recording session for a given limb position was followed by a 5-minute rest. The mean tremor frequency (Hz), amplitude (mv), burst duration (msec), and interval between bursts (msec), as well as the tremor pattern (synchronous or alternating), were assessed and the missing interval (msec) was measured. Amplitude was measured peak-to-peak for each burst and the mean value was considered. The burst duration was measured for each burst and the mean value was determined. The examination was performed at the same time (11 a.m.) in all patients (25). None of the patients, apart from the parkinsonian subjects, were taking drugs known to influence tremor They were still taking their regular medication, which included levodopa and/or anticholinergic drugs, but were asked to abstain from alcohol the previous evening and to avoid caffeine-containing beverages and tea on the morning of the examination. The examination was performed in a quiet room after the patients had been allowed to become familiar with their surroundings. The Spearman rank correlation coefficient (r) was used to analyse the relationship between clinical and EMG parameters. The Pearson correlation coefficient (r) was used to analyse the relationship between the EMG parameters. The correlation was considered as poor if r<0.3 and moderate if 0.3<r<0.5. The correlation analyses were performed considering both the single groups of patients and the different types of tremor. Results The progression of the tremor was fast in the parkinsonian, rubral and cerebellar patients. These subjects reported tremor exacerbation after a mean of 6.1 months and involvement of a new anatomical region after a mean of 13.2 months. The progression of the disease was slow in the essential tremor patients, who reported tremor exacerbation after a mean of 4.6 years and involvement of a new anatomical region after a mean of 5.8 years. No tremor progression was reported by the patients with psychogenic and enhanced physiological tremor (patients with anxiety). The correlation between EMG-measured tremor amplitude and Webster tremor scores was moderate (r=0.6) in the patients with parkinsonian, essential, rubral, cerebellar and psychogenic tremor, and poor (r=0.3) in the subjects with enhanced physiological tremor. A moderate positive correlation was found between tremor amplitude and burst duration, as well as between burst duration, interval between bursts and missing interval (Tables I-VI); examination of the relationship of the interval between bursts to the missing interval also revealed a moderate positive correlation. Moderate negative correlations were found between tremor frequency and amplitude, burst duration, interval between bursts and missing interval and between the interval between bursts, missing interval and tremor amplitude. The correlation between the various tremor parameters was not different (p>0.1) in the single patient groups. The correlation also remained constant (p>0.1) when considering the tremor in the various limb positions (resting, postural, kinetic or intention). Moderate negative correlations emerged between tremor frequency, disease duration and patient age (Tables I-IV), while a moderate positive correlation was 20 Functional Neurology 2002; 17(1): 19-23

3 Tremor correlation Table I - Correlations between different electromyographic parameters of essential tremor Amplitude Burst duration Missing interval Interval between bursts Disease duration Age Table II - Correlations between different electromyographic parameters of parkinsonian tremor Amplitude Burst duration Missing interval Interval between bursts Disease duration Age Table III - Correlations between different electromyographic parameters of rubral tremor Amplitude Burst duration Missing interval Interval between bursts Disease duration Age Table IV - Correlations between different electromyographic parameters of cerebellar tremor Amplitude Burst duration Missing interval Interval between bursts Disease duration Age Table V - Correlations between different electromyographic parameters of enhanced physiological tremor Amplitude Burst duration Missing interval Interval between bursts Disease duration Age Functional Neurology 2002; 17(1):

4 I. Milanov Table VI - Correlations between different electromyographic parameters of psychogenic tremor Amplitude Burst duration Missing interval Interval between bursts Disease duration Age found between tremor amplitude and burst duration, disease duration and patient age. These findings were not seen in the enhanced physiological and psychogenic tremor groups in which no correlation emerged between tremor parameters, patient age and tremor duration (Tables V, VI). Neither were we able to detect any correlation between interval between bursts, missing interval and disease duration or patient age. In all the patient groups, a strong correlation (0.7<r<0.9) was found between corresponding tremor parameters in different limb positions. Discussion Our results confirm previous findings of a negative correlation between tremor amplitude and frequency (4-7,18) as well as between tremor frequency and burst duration (6,8,9). This means that low frequency tremors have a higher amplitude and longer burst duration and vice versa. Although not previously reported, it can be assumed that a positive correlation exists between tremor amplitude and burst duration. This assumption was confirmed by our data. In addition, we found other more complicated correlations between tremor parameters: a positive correlation was found between the interval between bursts and the missing interval. Both the interval between bursts and the missing interval correlate positively with burst duration and negatively with tremor frequency and amplitude. Thus, the correlation between tremor parameters becomes more complicated. It may be concluded that low-frequency tremors have a higher amplitude, and a longer burst duration, interval between bursts and missing interval. However, upon further increase of the tremor amplitude, the interval between bursts and the missing interval decrease. It has been suggested that, with advancing age, tremor frequency decreases, while amplitude increases (26). Our results support this suggestion, but we also found that tremor frequency decreased and amplitude increased with longer disease duration. These findings show that tremor amplitude increases, while its frequency decreases, as the disease progresses, indicating that limb tissue changes cannot be regarded as the sole cause of the well-known changes in tremor amplitude and frequency. Moreover, these correlations were not found in non-progressing psychogenic and enhanced physiological tremors. We were not able to find published reports on the strong positive correlation that we found between corresponding tremor parameters in different limb positions a correlation that suggests that the different types of tremor are interconnected and determined by each other. It is known that tremor depends partially on mechanical properties of the limb and that it can, in a single patient, present different features in different limb positions. However this strong correlation indicates that the tremor in all limb positions is probably generated by the same oscillator. This finding is particularly important for parkinsonian patients, as it has been supposed that the pathophysiology of resting tremor may be different from that of postural tremor in these subjects (27). Correlations between the various tremor parameters were not found to differ in the different types of tremor. This finding has not been previously reported and it suggests that the relations between tremor parameters are similar and not dependent on tremor type. Different clinical rating scales are currently used for tremor assessment (3,28). In this investigation, we used a 0-3 scale, which allows tremor to be scored more easily than a scale with a greater number (from 0 to 10) of increments (3,28). The correlations between EMG-measured tremor amplitude and Webster tremor scores were moderate. The correlation was poor only in the enhanced physiological tremor patients. We suppose that perfect clinical rating is precluded by the very low amplitude of this tremor. In conclusion, clinically scored tremor correlates well with EMG measurements of tremor amplitude. The EMG examination is thus a useful tool for researching tremor mechanisms. References 11. Basmajian JV, De Luca CJ. Muscles Alive: Their Functions Revealed by Electromyography. Baltimore; Williams & Wilkins Bacher M, Scholz E, Diener HC. 24 continuous EMG tremor quantification based on EMG recording. Electroencephalogr Clin Nurophysiol 1989;72: Bain PG, Findley LJ, Archison P et al. Assessing tremor severity. J Neurol Neurosurg Psychiatry 1993;56: Elble RJ, Higgins C, Hughes L. Phase resetting and frequency entrainment of essential tremor. Exp Neurol 1992;116: Freund HJ, Honberg V, Hefter H. Clinical aspects and treatment. In: Benecke R, Conrad B, Marsden 22 Functional Neurology 2002; 17(1): 19-23

5 Tremor correlation CD eds Motor Disturbances I. London; Academic Press 1987: Sabra AF, Hallett M. Action tremor with alternating activity in antagonist muscles. Neurology 1984; 34: Young RR. Summary Part I: Tremor. In: Benecke R, Conrad B, Marsden CD eds Motor Disturbances I. London; Academic Press 1987: Hsu YD, Chang MK, Sung SC, Hsein HH, Deng JC. Essential tremor: clinical, electromyographical and pharmacological studies in 146 Chinese patients. Chin Med J 1990:45: Sechi GP, Zuddas M, Piredda M et al. Treatment of cerebellar tremors with carbamazepine: a controlled trial with long-term follow-up. Neurology 1989;39: Lakie M. Is essential tremor physiological? In: Findley LJ, Koller WC eds Handbook of Tremor Disorders. New York; Marcel Dekker Inc 1995: Cleeves L, Findley LJ. Amplitude fluctuations in essential tremor: implications for clinical trials. In: Benecke R, Conrad B, Marsden CD eds Motor Disturbances I. London; Academic Press 1987: Elble RJ. Physiological and essential tremor. Neurology 1986;36: Elble RJ, Higgins C, Hughes L. Longitudinal study of essential tremor. Neurology 1992;42: Elble RJ, Higgins C, Leffler K, Hughes L. Factors influencing the amplitude and frequency of essential tremor. Mov Disord 1994;9: Kachi T, Yamada T, Igata A. Clinical and physiological studies on senile tremor. Nippon Ronen Igakkai Zasshi 1995;32: Koller WC, Busenbark K, Gray C, Hassanein RS, Dubinsky R. Classification of essential tremor. Clin Neuropharmacol 1992:15; Lou JS, Jankovic J. Essential tremor: clinical correlates in 350 patients. Neurology 1991;41: Birmingham AT, Wharrad HJ, Williams EJ. The variation of finger tremor with age in man. J Neurol Neurosurg Psychiatry 1985;48: Calzetti S, Baratti M, Gresty M, Findley L. Frequency/amplitude characteristics of postural tremor of the hands in a population of patients with bilateral essential tremor: implications for the classification and mechanism of essential tremor. J Neurol Neurosurg Psychiatry 1987;50: Findley LJ, Koller WC. Definitions and behavioral classifications. In: Findley LJ, Koller WC eds Handbook of Tremor Disorders. New York; Marcel Dekker Inc 1995: Calne DB, Snow BJ, Lee C. Criteria for diagnosing Parkinson s disease. Ann Neurol 1992;32 (Suppl): Fahn S, Williams PJ. Psychogenic dystonia. Adv Neurol 1988;50: Webster DD. Critical analysis of the disability in Parkinson s disease. Modern Treatment 1968;5: Spieker S, Boose A, Jenigens CH, Dichgans J. Long-term tremor recordings in parkinsonian and essential tremor. J Neural Transm Suppl 1995;46: Van Hilten JJ, Van Dijk JG, Dinnewold RJW et al. Diurnal variation of essential and physiological tremor. J Neurol Neurosurg Psychiatry 1991;54: Bain P. Are neurophysiological techniques useful in differentiating tremor? In: Findley LJ, Koller WC eds Handbook of Tremor Disorders. New York; Marcel Dekker Inc 1995: Zimmermann R, Deuschl G, Hornig A, Schulte- Montig J, Fuchs G, Lucking CH. Tremors in Parkinson s disease: symptom analysis and rating. Clin Neuropharmacol 1994;17: Bain PG. Tremor assessment and quality of life measurements. Neurology 2000;54(Suppl 4):S26- S29 Functional Neurology 2002; 17(1):