[Clinical examination and electromyographic abnormalities in patients with lower back pain].

INTRODUCTION

Lower back pain (LBP) is one of the most frequent reasons for visits to basic and specialised health care centres and one of the commonest causes of referral for electromyographic (EMG) explorations. AIMS. The aim of this study is to determine the sensitivity and specificity of the clinical examination in patients with LBP and abnormal EMG findings.

PATIENTS AND METHODS

A cross sectional study was carried out on 364 patients with LBP who had been referred to both basic and specialised clinics, and who were attended in the Electromyography Service of a hospital belonging to the Instituto de Seguros Sociales, throughout the years 2001 and 2002. The same clinical evaluation protocol (interview and physical examination) was applied to all the patients. They were also evaluated using the same electrophysiological protocol, which in all cases included a study of the H waves and the adductor longus, tensor fasciae latae, medial vastus, anterior tibial, extensor digitorum longus, the medial and lateral heads of gastrocnemius, gluteus maximus and paraspinal muscles. The abnormal EMG examination was taken as the reference criterion and we determined the sensitivity, specificity, probability ratio, pre test probability and post test probability of the following clinical categories: referred pain (RP), pain less than three months, muscular strength of dorsiflexors, plantarflexors and knee extensor mechanisms, hypaesthesia in the L4, L5, S1 dermatomes, patellar areflexia and Achilles areflexia.

RESULTS

From the clinical history and physical examination, the pain referred to the extremity showed the highest sensitivity (80.89%), but its specificity was the lowest. This contrasted with the other clinical findings which, in general, showed high specificity (between 82.6% for hypaesthesia in the L5 dermatome and 97.5% for the weakness of the plantarflexors), but low sensitivity (from 6.37% for the weakness of the quadriceps to 35.67% for hypaesthesia in the L5 dermatome). Achilles areflexia showed the highest positive predictive value (85%) and the highest probability ratio (7.47), while the highest negative predictive value was found in pain referred to the extremity (68.75%).

CONCLUSION

Although the clinical history and findings from the physical examination do not allow the prediction of a normal or abnormal electrophysiological result in patients with LBP, the interview and some of the clinical tests do make it easier to select which patients should be referred for an EMG exploration, the purpose of which is to aid the diagnosis of a lumbar radiculopathy.