Radiculopathy is sometimes evaluated with needle electromyography as additional examination to radiographic imaging. The purpose of this study was to determine the optimal needle electromyography protocol and the use of needle electromyography in the diagnosis of radiculopathy. The authors analysed English language writings from years 1981-2000. The selected studies reported a wide range of needle EMG abnormalities--from about 30 to 100% of pathological EMGs. That is caused by wide variety of patient populations and various EMG protocols. EMG of leg muscles in radiculopathy can consist of the assessment of the spontaneous activity only or be completed with the assessment of motor unit potential (MUP). The EMG of paraspinal muscles assesses spontaneous activity only. The number of abnormalities in needle EMG grows with the presence of characteristic radiculopathy symptoms, but positive EMG examinations are of importance only in cases where diagnosis is complicated. The rate of pathological results is higher when needle electromyography is used in the muscles of paraspinal region and extremities at the same time. The assessment of the spontaneous activity in extremity muscles alone minimizes the influence of error- causing factors on the EMG result. Nevertheless lack of assessment of motor units potentials makes electromyographic investigation of chronic radiculopathy impossible. Sensitivity and specificity of the method have to be determined before the clinical use of the assessment of MUP in chronic radiculopathy, especially in patients without motor weakness. Needle EMG abnormalities correlate with radiological findings in limited range as would be expected from the occurrence of radiological findings in asymptomatic patients and overlapping innervations of myotomes.