Search

Abstract Background Lumbo-sacral radiculopathy (LSR) is a common musculoskeletal disorder for which patients seek medical care and referrals for advanced imaging. However, accurate diagnosis remains challenging. Neuropathic pain screening questionnaires, clinical neurological examination and magnetic resonance imaging (MRI) are used in the initial diagnosis. The utility of these tools in diagnosing LSR varies and their correlation has not been reported. Methods A cross-sectional, multicentre, blinded design was used in six physiotherapy departments in Kenya. Each participant was blindly examined by three independent clinicians using the Self-Reported Leeds Assessment of Neuropathic Symptoms and Signs (S-LANSS) score, clinical neurological examination (CNE) and MRI. Spearman’s rank coefficient (r) was used to examine the correlation between the three tests. Linear regression and odds ratios were used to establish correlations between socio-demographic, clinical and diagnostic parameters. The diagnostic accuracy of individual or combined sets of CNE tests in diagnosing LSR, with reference to MRI, was determined using Receiver Operating Characteristics (ROC) curves. Results We enrolled 102 participants (44 males, 58 females; mean age: 44.7 years). Results indicated a significant positive correlation (r = 0.36, P = 0.01) between S-LANSS, CNE and MRI among patients with low back and radiating leg symptoms. Positive agreement existed between combined neuro-conduction tests (sensory, motor and reflex) and neuro-dynamic tests (NDT). The NDT component of CNE (Straight Leg Raise Test [SLRT] and Femoral Nerve Stretch Test [FNST]) was significantly associated (P = 0.05) with MRI: patients who had positive NDT results had higher odds (8.3) for positive nerve root compromise on MRI versus those who had negative NDT results. Conclusion This was the first study to investigate the correlation between S-LANSS, CNE and MRI in patients presenting with low back and radiating leg symptoms. Results indicated a significant positive correlation. The strongest correlations to MRI findings of LSR were firstly, NDT (SLRT and FNST); secondly, the S-LANSS score; and thirdly, the CNE components of motor power and deep tendon reflex. The clinical implication is that clinicians can confidently use the S-LANSS score and CNE to diagnose and make therapeutic decisions in LSR, when MRI is medically contra-indicated, unaffordable or unavailable.