Effectiveness of Diagnostics with Neurodynamic Tests
How effective can neurodynamic testing really be in diagnosis?
Two key aspects of diagnosis of neurodynamic problems, as with all diagnosis, are sensitivity and specificity.
Sensitivity of neurodynamic tests ANSWERS the question: in those who HAVE a neural problem, how many of them show an abnormal neurodynamic test? So for 100 subjects with, say, carpal tunnel syndrome (CTS), an abnormal MNT1 in 80 subjects would produce a sensitivity rating of 80% or .80.
Specificity ASKS the question: in those who DO NOT HAVE the problem in question (ie. subjects do NOT have CTS), how many show a normal result with neurodynamic testing? For a test, for instance the MNT1, to show a high specificity rating, it should produce a normal result in all subjects who do not have the problem. In this case, the normal test should eliminate the presence of CTS.
Taking this further, if we did a blinded MNT1 on 100 subjects with clinical presentation of CTS. Fifty of them had abnormal nerve conduction and the other 50 had normal conduction tests, then it is possible to make some comparisons with electrophysiological testing. For instance, when the MNT1 is compared to nerve conduction studies in CTS sufferers, for the MNT1 to have sensitivity and specificity ratings of 100%, all those who have an abnormal MNT1 should also have abnormal nerve conduction and all those who show a normal MNT1 should have normal nerve conduction (specificity). A key point here is that abnormal impulses due to mechanosensitivity (hence, possibly neurogenic pain) may come from nerves with normal conduction. Eliav et al 2001) showed that inflamed and mechanosensitive nerves can conduct normally. But what we are considering here is the relationship between nerve conduction failure (ie. reduced nerve conduction) as the gold standard for diagnosis of 'neuropathy' compared with responses to physical testing eg. mechanosensitivity or impairment of nerve movement.
Another rating is the positive predictive value. This ANSWERS the question: in those who SHOW an abnormal neurodynamic test (eg. MNT1), how many also have abnormal electrophysiology? Clearly the aim is to have a 100% correlation but, in physical medicine and manual therapy, this is rarely, if ever, the case.
For the following study, Coveney et al (1997) won the prize for the best paper at the Biennial Conference of the Manipulative Physiotherapists' Association of Australia, Melbourne.
Coveney et al (1997) investigated the relationships mentioned above. The MNT1 was performed on patients with CTS where electrophysiological function of the median nerve was measured by a neurologist. Reliability in performance of the MNT1 (in relation to the Overt Abnormal Response category) prior to testing was excellent with test-retest agreements on the following:
- 91% - presence or absence of the patient's symptoms
- 93% - area of most intense symptoms
- 91% - effect of contralateral lateral flexion
- 100% - effect of ipsilateral lateral flexion
- 82% - change in elbow extension ROM with ipsilateral lateral flexion
The examiner was blinded to the results of the electrophysiology tests. The MNT1 was performed and then rated against the electrophysiology results. The outcomes were as follows:
- sensitivity rating 82%
- specificity rating 75%
- positive predictive value 93%
Sample size - 21
What is particularly satisfying is that these results are very good, not just because the efficacy ratings are high but also because another group (Selvaratnam et al 1997) had performed a similar study independently of our group, with the similar results. This helps the cause of neurodynamics in diagnosis and is consistent with other research of diagnostic efficacy in neural disorders with neurodynamic testing, such as the cervical nerve root. Wainner et al (2003) showed that the ULNT1/MNT1 produced a sensitivity rating of .97 (CI 95) in patients with electrophysiological evidence of cervical radiculopathy and the MNT1 was the single best test in eliminating the problem when compared with other physical tests for cervical radiculopathy (eg. abduction test and Spurling’s manoeuvre).
Excellent reliability of performance of the MNT1 can be achieved, using the Overt Abnormal Response category described in Shacklock 2005 (Clinical Neurodynamics).
Sample size was small which weakens the study.
Similar results were obtained by another group (Selvaratnam et al 1997) independently which strengthens the above conclusions.
'Reproduction of the patients symptoms' was used as one of the diagnostic criteria (an aspect of the overt abnormal response category). However, clinically we often observe that other more subtle abnormalities such as stiffness, altered muscle behaviour and reduced range of motion that are positive to structural differentiation often coexist with neurodynamic disorders. Possibly, diagnostic efficacy could be improved with the inclusion of some of these more subtle abnormalities. This necessitates applying the Covert Abnormal Response category. What is the covert abnormal response? See Shacklock (2005), pp 100-104.
For information on how to improve sensitivity in diagnosis see The Sporting Nerve.
Coveney B, Trott P, Grimmer K, Bell A, Hall R, Shacklock M 1997 The upper limb tension test in a group of subjects with a clinical presentation of carpal tunnel syndrome. In: Proceedings of the Manipulative Physiotherapists' Association of Australia, Melbourne: 31-33.
Eliav E, Benoliel R, Tal M 2001 Inflammation with no axonal damage of the rat saphenous nerve trunk induces ectopic discharge and mechanosensitivity in myelinated axons. Neuroscience Letters 311: 49-52
Selvaratnam P, Cook S, Matyas T 1997 Transmission of mechanical stimulation to the median nerve at the wrist during the upper limb tension test. In: Proceedings of the Manipulative Physiotherapists' Association of Australia, Melbourne: 182-188
Shacklock M 2005 Clinical Neurodynamics: a new system of musculoskeletal treatment. Elsevier, Oxford
Wainner R, Fritz J, Irrgang J, Boninger M, Delitto A, Allison S 2003 Reliability and diagnostic accuracy of the clinical examination and patient self-report measures for cervical radiculopathy. Spine 28 (1): 52–62