Multivariate logistic regression techniques were used to develop a composite nerve conduction measurement that detects lumbosacral (L5, S1, or both) nerve root compression.
To evaluate the diagnostic efficacy of a composite nerve conduction measurement for detection of lumbosacral nerve root compression.
Nerve root involvement is characterized by clinical abnormalities and confirmed by radiologic and electrodiagnostic studies. Imaging studies visualize structural abnormalities; however, they are associated with high false-positive rates. Electrodiagnostic methods assess the physiologic integrity of the nerve roots. One form of electrodiagnostic testing, nerve conduction studies, is widely used for evaluation of musculoskeletal and neuromuscular complaints. Although similar clinical value is expected for the evaluation of nerve root compromise, prior applications of nerve conduction studies have yielded widely varying results.
Two groups of subjects were compared. The L5–S1 compression group was composed of 25 patients with magnetic resonance imaging-confirmed lumbosacral (L5, S1, or both) nerve root compression and symptoms in the appropriate segmental distribution. The majority of subjects (22) had at least one of the following findings on physical examination: positive straight-leg raise test, diminished ankle reflexes, sensory loss, or weakness. The control group consisted of 35 asymptomatic individuals with no history of radiculopathy or potentially confounding neuropathology. The posterior tibial and deep peroneal nerves were evaluated bilaterally in all study subjects using standard nerve conduction procedures, which consisted of the measurement of distal motor latencies and F-wave latencies that assess nerve root pathophysiology. A composite nerve conduction measurement was determined using multivariate logistic regression analysis. The efficacy of the composite measurement was assessed by receiver operating characteristic curve analysis and by the diagnostic sensitivity and specificity.
Five F-wave latency parameters (peroneal mean F-wave latency, odds ratio = 0.42; peroneal seventh F-wave latency decile, odds ratio = 2.71; tibial mean F-wave latency, odds ratio = 8.90; tibial first F-wave latency decile, odds ratio = 0.47; tibial maximum F-wave latency, odds ratio = 0.44) were found to be predictive of nerve root compression. A composite nerve conduction measurement, NC composite, constructed from these five parameters (NC composite = exp(φ)/(1 + exp(φ)), φ = −31.2 + 1.0 • Per7thDecile − 0.88 • PerMean + 2.2 • TibMean − 0.88 • Tib1stDecile − 0.83 • TibMax) yielded an area under the receiver operating characteristic curve of 0.91. At a threshold of 0.20, NC composite had a diagnostic specificity of 84.3% and a sensitivity of 83.3%.
This preliminary study suggests that a novel composite nerve conduction measurement, based on F-wave latency parameters, may be highly effective at detecting magnetic resonance imaging-confirmed lumbosacral nerve root compression. Because these measurements provide objective evidence of functional nerve root compromise and are noninvasive, they may be of diagnostic value to clinicians evaluating patients presenting with low back and leg pain.
From *NeuroMetrix, Inc., Waltham,
†MRI Centers of New England, Newton,
and ‡Harvard-M.I.T., Division of Health Sciences and Technology, Cambridge, Massachusetts.
Supported by NeuroMetrix, Inc.
Acknowledgment date: November 7, 2001.
First revision date: January 10, 2002. Second revision date: March 14, 2002.
Acceptance date: May 6, 2002.
The device(s)/drug(s) is/are FDA-approved or approved by corresponding national agency for this indication.
Corporate/industry funds were received in support of this work. One or more of the author(s) has/have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this manuscript, e.g., honoraria, gifts, consultancies, royalties, stocks, stock options, decision-making position.
Address correspondence to Shai N. Gozani, MD, PhD, NeuroMetrix, Inc., 62 Fourth Avenue, Waltham, MA 02451, USA; E-mail: email@example.com