DIAGNOSTICS E712 www.spinejournal.com May 2011 SPINE Volume 36, Number 11, pp E712–E719 ©2011, Lippincott Williams & Wilkins Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Diagnostic Accuracy of the Clinical Examination in Identifying the Level of Herniation in Patients with Sciatica Mark J. Hancock, PhD,* Bart Koes, PhD,† Raymond Ostelo, PhD,‡§ and Wilco Peul, PhD¶# Study Design. Cross sectional Objective. To investigate the ability of the neurological examination to identify the specific level of a disc herniation in patients with sciatica and confirmed disc herniation. Summary of Background Data. Tests included in a neurological examination theoretically provide accurate diagnostic information about the level of the herniated disc. However, there is currently very little evidence about the diagnostic accuracy of individual tests or combinations of tests. Methods. The study included 283 patients with sciatica and confirmed disc herniation from a previous randomized controlled trial. The reference test for the current study was the MRI scan, reported for level of disc herniation. Index tests investigated were a neurologist’s overall impression of the level of disc herniation, individual neurological tests ( e.g., sensation testing) and multiple test findings ( i.e., the number of positive tests). The index tests were performed blinded to the MRI results. The diagnostic accuracy of the index tests in predicting herniations at the lower three lumbar discs was investigated using area under the curve (AUC), sensitivity and specificity. Results. None of the individual neurological tests from the clinical examination were highly accurate for identifying the level of disc herniation (AUC  0.75). The outcome of multiple test findings was slightly more accurate but did not produce high sensitivity and specificity. The dermatomal pain location was generally the most T he typical clinical examination of a patient with sciatica includes test procedures to determine if a disc herniation is the likely source of symptoms. 1,2 Some of the informa- tion gathered during this clinical examination is nonspecific, in that it may help indicate the likelihood of a disc herniation but does not provide information about the spinal level of the herniation or nerve root causing the symptoms. 3 Common examples of such tests include the straight leg raise (SLR) test, and questions about pain with activities like coughing that increase intra-abdominal pressure. 4,5 Other tests commonly included in the clinical examination potentially provide in- formation about the specific level of the disc herniation likely to be responsible for the symptoms. 3,6 Examples of these tests include dermatomal tests of sensation, myotomal tests of muscle strength and reflex testing. Magnetic resonance imaging (MRI) is widely used to assess patients with sciatica to determine if a disc herniation exists. An essential part of clinical decision-making for neurologists, rheumatologists and spinal surgeons is to decide if the find- ings from the clinical examination are likely to be the result of a herniation observed on MRI. It is the combination and correlation of the clinical examination findings and MRI find- ings that is essential for successful selection of patients for surgical management of sciatica. If a herniation seen on MRI is not likely to be responsible for the patients symptoms (for example, because the level of the herniated disc and com- pressed nerve root do not correspond with the pain distri- bution and/or neurologic deficit) then is not reasonable to expect that surgical removal or other treatment aimed at that disc will be beneficial to the patient. Even when surgical From the *Faculty of Health Sciences, University of Sydney, Australia; †Department of General Practice, Erasmus Medical Centre, Rotterdam, The Netherlands; ‡EMGO Institute VU University Medical Centre, Amsterdam, Netherlands; §Department of Health Economics & Health Technology Assess- ment, Institute of Health Sciences, VU University, Amsterdam, Netherlands; ¶Department of Neurosurgery, Leiden University Medical Center/Medical Center Haaglanden, The Hague, The Netherlands; and #Department of Neurosurgery, Medical Center Haaglanden, The Hague, The Netherlands. Supported by a grant from the Netherlands Organisation for Health Research and Development (ZonMW) and the Hoelen Foundation, The Hague. Acknowledgement date: February 15, 2010. Revised date: May 19, 2010. Accepted date: July 15, 2010. The manuscript submitted does not contain information about medical device(s)/drug(s). No funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript. Address correspondence and reprint requests to Mark Hancock, PhD, University of Sydney PO Box 170, Lidcombe 1825, NSW, Australia; E-mail: M.Hancock@usyd.edu.au informative individual neurological test. The overall suspected level of disc herniation rated by the neurologist after a full examination of the patient was more accurate than individual tests. At L4/5 and L5/S1 herniations the AUC for neurologist ratings was 0.79 and 0.80 respectively. Conclusion. The current study did not find evidence to support the accuracy of individual tests from the neurological examination in identifying the level of disc herniation demonstrated on MRI. A neurologist’s overall impression was moderately accurate in identifying the level of disc herniation. Key words: low back pain, sciatica, clinical examination, disc herniation, MRI. Spine: 2011;36:E712-E719 DOI: 10.1097/BRS.0b013e3181ee7f78