The Study Of Turns-Amplitude Cloud Analysis Of The Electromyographic Interference Pattern In Neuromyopathy

Electrophysiologic detections play an important role in nerve and muscle disorders.Electromyography(EMG)is one of the most important electrophysiologic techniques.It can detect the position and severity of neuromuscular lesions,and be helpful for convalescence assessment.Because of the rapid development of electronic technology,EMG detection tends to be quantitative.The earliest quantative method of needle EMG is the quantitative evaluation of mean motor unit potential(QMUP),which was developed since 1940s.This method requires selecting at least 20 different MUPs,in order to analyse the average time,average amplitude,as well as the percentage of polyphasic potentials.The disadvantage of QMUP is that all these different MUPs can only be captured at a slight muscle contraction,the strength of which is only about 4%of that of the maximal voluntary contraction(maximal voluntary contraction,MVC).With the contraction becoming stronger,more and more motor unit are activated and develop action potentials' interference and summation,which is namely interference pattern(IP),and it is impossible to identify a single different MUP.Therefore,the QMUP can only assess the motor unit potentials when the muscles mildly contract.Theoretically,interference pattern analysis(IPA)contains information not only related to the amplitude and complexity of the MUPs of which it is composed,but also related to their recruitment pattern,which may differ from the normal pattern in neuropathy and myopathy.As IP analysis can provide more information about activatied motor unit at various levels,it is necessary to establish methods to analyze IP.As electromyographic interference pattern could not be recognized by human eyes,it is necessary to develop automatic methods to analyze IP.Before applying this technique to clinical practice,reliable and discriminating normative data are needed.Since 1960s,many automatic analysis methods were developed to evaluate IP,including measuring the number of turns(NT)and the mean amplitude(MA)change between successive IP signal.Other analysis methods such as zero rate,peak ratio,power spectrum and cloud analysis are also developed.Of them,cloud analysis is the most well-established.This technique was first developed by Stalberg et al in 1983.A plot of mean amplitude and turns can be obtained at different contraction levels but the measurements of muscle force are not needed.By collecting enough values form normal volunteers and calculating the upper limit,lower limit,max turns,max amplitude with a special algorithm,a 'cloud' picture can be established.If one subject has two points out of ten outside this cloud,it is considered pathological.In patients with myopathy,mean amplitude/number of turns values were found to be lower than normal,whereas patients with neurogenic lesions had higher values.Many studies had confirmed that this method had good sensitivity and specificity for the identification of neuromuscular disease.And it is convenient and fast as it does not need to measure the force of muscles.It is difficult to establish the normal limit values of the cloud,because it is not clear whether the normal limit values are affected by race,age,gender,muscle,or limb dominance.Thus,it has not been widely applied domestically and there is few domestic research in related fields.The normal range of cloud and related influcing factors has not been established.It is not known whether the diagnostic accuracy of cloud analysis varies according to patient characteristics.Thus,we performed this study in order to answer the questions above.Objective:We aimed to evaluate the cloud analysis of normal adults and patients with neuromuscular disease.We wanted to find the potential factors influencing cloud analysis.Then,we aimed to establish the normal range of cloud and investigate the potential factors influencing cloud and identify the cloud-criterion for neuromuscular disorders in an Chinese population.Method:(1)The included subjects are as follows:normal volunteers from our hospital physical examination center,patients with local nerve disease but without involvement of the muscle examinated.Patients with typical neuromuscular disease were recruited in the same period in outpatient department and wards,according to the inclusion criteria and exclusion criteria.IP cloud analysis was used to detect the IP signal of the muscle of normal subjects.The IP and MUP signals of the selected muscles of the patients and normal control were detected by two methods simultaneously:the traditional needle electrode EMG and cloud analysis.And the age,sex,and muscle name of the subjects were recorded at the same time.(2)The distribution of turns and amplitude of the muscles were analyzed.The data were analyzed after logarithm.To assess whether the analysis on sampling obeys the law of normal distribution and linear regression or not.(3)Factors that affect the shape of normal cloud were investigated.The turns and amplitude measurements of the electromyographic interference pattern were collected to make regression analysis,and parameters were calculated to determine the range of normal values,and normal cloud boundaries were formed.(4)The IP records and MUPs in the patients and healthy controls were analyzed using the clouds developed in this study and QMUP respectively.Then,the diagnosis yield by these two methods were evaluated.We compared the difference of our clouds with thoses previously published abroad.Results:(1)In total,60 healthy volunteers,27 patients with neuromuscular disorders and 20 normal controls were included in the study.(2)There was a correlation between NT/s and mean amplitude.The data of NT/s and amplitude/NT were both in normal distribution after log transformation.Linear regression analysis showed a linear correlation between these two log transformed parameters.(3)The shape of cloud was influenced by age and sex.According to the age and sex,different linear regression lines(with 95%confidence intervals)were calculated from the log transformed "turns/second" and "amplitude/turn".These confidence intervals were then transformed back into the original parameters to yield scatter plots with confidence curves,to establish normal cloud.(4)IPA had a comparable sensitivity and specificity with quantitative evaluation of QMUP.Conclusion:The cloud analysis is a feasible method to automatically assess the IP signal with great convenience.The normal cloud are correlated with sex and age.We calculated the normal values of cloud for five muscles(biceps brachii muscle,deltoid muscle,common extensor digitorum muscle,tibialis anterior muscle and quadriceps femoris muscle)according to sex and age respectively.IPA had a comparable sensitivity and specificity with QMUP.The maximal amplitude value of the Chinese people were lower than the values reported in foreign researches.Thus,ethnic differences should be noted when using the normal cloud.It was necessary to establish diagnostic criteria for different races.