(Golden Oldie) Accuracy of clinical tests in detecting disk herniation and nerve root compression in subjects with lumbar radicular symptoms

Key points

1. 01The slump test caught most cases (sensitivity over 70%) but specificity was low (.36 and .38), so a positive result did not confirm the MRI finding.
2. 02Radiculopathy assessed as neurologic signs matching a specific nerve root (radiculopathy II) was the only test with both good sensitivity and specificity, but only for subarticular nerve compression (.71 and .73).
3. 03No neurodynamic test showed any relationship with foraminal nerve compression (foraminal stenosis) on MRI.
4. 04Diagnostic accuracy (AUC) was low for extrusion (.48 to .60) and foraminal compression (.33 to .57), and only moderate for subarticular compression (.63 to .82).
5. 05Overall the tests lacked enough accuracy to stand in for MRI when localizing the problem.

How it was conducted

Design

Prospective cohort validity study with MRI as the reference standard, in secondary care

Participants

99 adults (mean age 58y, 54% women) referred for transforaminal epidural steroid injection for lumbar radicular pain, with positive clinical and MRI findings

Index tests

Slump test, straight-leg raise (SLR) test, femoral neurodynamic test, and radiculopathy I (a neurologic sign) and radiculopathy II (neurologic signs matching a specific nerve root)

Reference standard

1.5-tesla MRI graded by a blinded radiologist for disk extrusion, subarticular nerve root compression, and foraminal nerve root compression

Primary outcome

Sensitivity, specificity, and ROC area under the curve (AUC) of each clinical test for detecting MRI findings

What they found

• Positive clinical findings: slump test n=67, SLR test n=50, femoral neurodynamic test n=7, radiculopathy I n=70, radiculopathy II n=33.
• Positive MRI findings: extrusion n=27, high-grade subarticular nerve compression n=14, high-grade foraminal nerve compression n=25.
• The slump test was the most sensitive test (over 70%) but had low specificity (.36 and .38).
• Radiculopathy I was most sensitive for foraminal nerve compression (sensitivity .80) but with low specificity (.34).
• Radiculopathy II had concurrent high sensitivity and specificity for subarticular nerve compression (.71 and .73, respectively).
• AUC ranges across tests were .48 to .60 for extrusion, .63 to .82 for subarticular nerve compression, and .33 to .57 for foraminal nerve compression.

Limitations

• Single-center sample of only 99 patients already selected for steroid injection, which limits how widely the results apply.
• All participants had positive clinical and MRI findings, so the study cannot estimate how the tests perform in people who test negative.
• Interrater reliability was not assessed in this study; the authors relied on reliability shown in prior work.
• Motor tests for the L4 and S1 myotomes were omitted to avoid provoking pain, leaving the neurologic exam incomplete.

Why it matters

For patients

A positive sciatica stretch test does not confirm exactly what an MRI will show, so imaging may still be needed to pinpoint the cause.

For clinicians

These bedside tests, especially neurodynamic ones, cannot reliably localize disk extrusion or foraminal compression and should not replace MRI for that purpose.

For readers

Physical exam tests for lumbar radicular pain agree only modestly with MRI findings, with the weakest agreement for foraminal stenosis.