Features of gait most responsive to tap test in normal pressure hydrocephalus

Abstract

Objective

To identify components of gait associated with a positive tap test (TT) in patients with idiopathic normal pressure hydrocephalus (iNPH).

Patients and methods

Thirty-three patients with iNPH underwent clinical evaluation pre- and post-TT and were classified as responders (Rs) or non-responders (NRs). Elements of gait were assessed with a formal standardized Gait Scale and compared between groups.

Results

Analysis of pre/post-TT group differences revealed an interaction for Total Gait Score and Walking Score, with improvements in responders only. Total Gait Scores improved by 29% in the Rs and 4.85% in the NRs. Rs showed significant post-TT improvements on a timed 10 m walk, turning, and balance. Tandem walking, turning, truck balance and start stop hesitation showed trends toward improvement.

Conclusions

The classic features of gait often used in determining diagnosis of NPH (wide based stride, reduced foot-floor clearance, and small steps) were not helpful in identifying responders to the TT. Walking speed, steps for turning, and tendency towards falling were most likely to improve post-TT. These straightforward measures can readily be adapted into clinical practice to assist in determination of shunt candidacy.

Introduction

Idiopathic normal pressure hydrocephalus (iNPH) is a syndrome consisting of a clinical triad of gait disturbance, cognitive dysfunction, and urinary symptoms first described by Hakim and Adams in 1965 [1]. The gait disturbance is typically the most prominent clinical feature of iNPH and is often the first symptom to develop [2], [3]. Dramatic gait improvement can be achieved in some patients with the placement of a shunt; [4], [5] however, the procedure can be associated with as much as 50% morbidity, including permanent neurological deficits, intracerebral hemorrhage, subdural hemorrhage, meningeal infection, as well as death [6], [7], [8], [9]. Given the potential risk associated with this intervention, there is a need for improved patient selection criteria for shunt surgery. The most widely used prognostic test to assess candidacy for shunt placement is the CSF tap test, also called a large volume lumbar puncture. While this test has not undergone rigorous evaluation in carefully controlled prospective evaluations and is not required for diagnosis, the tap test (TT) is commonly used to prognosticate shunt responsiveness. Clinical improvement following TT is one of the few established prognostic indicators of a positive response to shunting in patients with iNPH [10], yet there is wide variability in how response to the TT is determined [11], [12].

Standard of care in evaluating TT response typically involves subjective assessment by the clinician of whether an individual's gait has changed post-tap. This can be extremely challenging since the degree of change may be subtle in some cases. Further, the time frame in which improvement may be observed is variable across patients; therefore, determination of whether a patient has responded is sometimes partially based on the subjective impression of improvement by the family and caregivers after the patient leaves the clinician's office. A positive TT response has been shown to have good positive predictive value of shunt response [10]. However, failure to respond to TT does not necessarily indicate a negative treatment outcome. The TT has a high false negative rate, likely owing to the subtlety of the determination of post-tap changes [13], [14], [15].

The literature examining response to the TT is limited and often based on subjective clinician's ratings as opposed to objective measures of change. To date, few studies have employed quantitative gait assessment to describe the NPH gait disturbance or to examine which aspects of the gait disorder are most responsive to TT. One small study of 10 NPH patients by Stolze et al. found that only stride length improved following the TT [16]. Another study (n = 10) of NPH patients found increased speed of movement post-TT [17]. More recently, Bugalho and Guimaraes [18] assessed gait in a small sample of NPH patients and showed that hypokinesia, but not disequilibrium, responded to TT. Further, patients with frontal release signs at baseline were less likely to respond to TT. The studies are generally small and although there is some overlap, specific aspects of gait that are evaluated vary between studies. In a well regarded study, Wikkelso and colleagues defined improved gait as >5% post-tap change on the average of three walking trials which consisted of counting number of steps required to walk 18 m [11], [19]. The gait disturbance was characterized by short and variable stride length, freezing, and impaired turning in the majority of patients, but additional information was not provided on how gait was measured or which aspects specifically improved post-tap [11]. Clearly, there is a need for specific quantitative measures of gait improvement to aide in the determination of surgical candidacy and to evaluate improvement following shunt placement.

The Gait Scale, a useful quantitative measure of gait impairment in NPH patients, is a component of a larger measure developed as part of the Dutch normal pressure hydrocephalus study [20]. The Gait Scale evaluates various aspects of gait, including walking speed, stride length, stance, foot-floor clearance, balance, tandem walking ability, turning ability, and start hesitation. Although the utility of this scale to characterize pre/post-TT gait changes has not been well documented, it is an objective measure which captures the unique gait features associated with iNPH.

To aid in clinical decision-making and to avoid putting patients at unnecessary surgical risk, improved confidence in the results of a diagnostic TT is desirable. Given that the determination of TT response typically depends on the clinician's subjective assessment of change, identification of specific gait features most likely to respond to the TT would be of clinical importance. The purpose of this investigation was to identify the components of gait that respond to diagnostic tap test in patients with suspected iNPH.