Cognitive and motor improvement after retesting in normal-pressure hydrocephalus: a real change or merely a learning effect?

J Neurosurg 112: 399–409, 2010. (DOI: 10.3171/2009.4.JNS081664)

Object. The test-retest method is commonly used in the management of patients with normal-pressure hydrocephalus (NPH). One of the most widely used techniques in the diagnosis of this condition is evaluation of the
patient’s response to CSF evacuation by lumbar puncture (a so-called tap test or spinal tap). However, interpretation of improved results in subsequent evaluations is controversial because higher scores could reflect a real change in specific abilities or could be simply the result of a learning effect.
Methods. To determine the effect of testing-retesting in patients with NPH, the authors analyzed changes documented on 5 neuropsychological tests (the Toulouse-Pieron, Trail Making Test A, Grooved Pegboard, Word Fluency, and Bingley Memory tests) and several motor ability scales (motor performance test, length of step, and walking speed tests) in a series of 32 patients with NPH who underwent the same battery on 4 consecutive days. The same tests were also applied in 30 healthy volunteers. In both groups, the authors used the generalized least-squares regression method with random effects to test for learning effects. To evaluate possible differences in response depending on the degree of cognitive impairment at baseline, the results were adjusted by using the Mini-Mental State Examination scores of patients and controls when these scores were significant in the model.
Results. In patients with NPH there were no statistically significant differences in any of the neuropsychological
or motor tests performed over the 4 consecutive days, except in the results of the Toulouse-Pieron test, which
were significantly improved on Day 3. In contrast, healthy volunteers had statistically significant improvement in the results of the Toulouse-Pieron test, Trail Making Test A, and Grooved Pegboard test but not in the remaining neuropsychological tests. Patients in the healthy volunteer group also exhibited statistically significant improvement in the motor performance test but not in step length or walking speed.
Conclusions. No learning effect was found in patients with NPH on any of the neuropsychological or motor tests. Clinical improvement after retesting in these patients reflects real changes, and this strategy can therefore be used in both the diagnosis and evaluation of surgical outcomes.