Journal of Neurology

, 258:1708

4-Aminopyridine improves gait variability in cerebellar ataxia due to CACNA 1A mutation


    • Department of NeurologyUniversity Hospital Munich
    • Integrated Research and Treatment Center for Vertigo, Balance and Ocular Motor Disorders: IFB(LMU)University of Munich
  • Max Wuehr
    • Graduate School of Systemic NeurosciencesUniversity of Munich
  • Nibal Ackl
    • Department of NeurologyUniversity Hospital Munich
  • Adrian Danek
    • Department of NeurologyUniversity Hospital Munich
  • Thomas Brandt
    • Integrated Research and Treatment Center for Vertigo, Balance and Ocular Motor Disorders: IFB(LMU)University of Munich
    • Institute of Clinical NeurosciencesUniversity of Munich
  • Michael Strupp
    • Department of NeurologyUniversity Hospital Munich
    • Integrated Research and Treatment Center for Vertigo, Balance and Ocular Motor Disorders: IFB(LMU)University of Munich
  • Klaus Jahn
    • Department of NeurologyUniversity Hospital Munich
    • Integrated Research and Treatment Center for Vertigo, Balance and Ocular Motor Disorders: IFB(LMU)University of Munich
Letter to the Editors

DOI: 10.1007/s00415-011-5987-z

Cite this article as:
Schniepp, R., Wuehr, M., Ackl, N. et al. J Neurol (2011) 258: 1708. doi:10.1007/s00415-011-5987-z

Dear Sirs,

The fall risk of patients with cerebellar ataxia is markedly increased and these falls often result in serious injuries or a fear of falling [13]. Patients with cerebellar ataxia typically walk with a broadened base of support, decreased step length and increased gait variability [4]. The latter is a critical parameter in the assessment of patients with gait disorders since increased variability is associated with a higher risk of falls [2].

Here we report on the effect of 4-aminopyridine on temporal gait variability in two patients with cerebellar ataxia caused by mutations of the calcium-channel gene CACNA 1A on chromosome 19p13. The gene CACNA 1A encodes the CA(v)2.1 subunit of the P/Q-type calcium channel, which is particularly expressed in Purkinje cells. CACNA 1A mutations are known to cause episodic ataxia type 2 (EA2), spinocerebellar ataxia type 6 (SCA6), as well as familial hemiplegic migraine type 1. The clinical phenotype of patients with CACNA 1A mutations is highly variable. Cerebellar symptoms range from mild intermittent ataxic episodes to severe permanent ataxia in some patients [6]. Animal models have shown that a dysfunction of the calcium channels results in irregular pacemaker activity of Purkinje cells, which contributes to the motor symptoms associated with EA2 [15].

Our two patients (patient 1: male, 50 years; patient 2: male, 73 years) presented with a progressive, permanent gait ataxia. The time course of the symptoms was 5 years in patient 1 and over 20 years in patient 2. Patient 2 also reported monthly episodes of migraine like headache attacks lasting for 2–5 h during which a worsening of his ataxia occurs. Clinical examination of both patients revealed a moderate cerebellar ataxia [Scale for Assessment and Rating of Ataxia (SARA), Table 1] with gait impairment [Functional Gait Assessment (FGA), Table 1]. Their subjective ambulatory scores were abnormal [Falls Efficacy Scale I (FES-I), Activity-specific Balance Confidence scale (ABC), Table 1]. Molecular testing of both patients revealed a mutation of the CACNA 1A gene (Table 1).
Table 1

Clinical evaluation and subjective ambulatory function of the enrolled subjects


Patient 1

Patient 2

CACNA 1A mutation

c.1741G > C

c.5404-13G > A

Age in years/sex



Duration of symptoms

5 years

>20 years






FGA [points; norm: 27.9 ± 2.3]





SARA [points; norm: 3.4 ± 1.2]





FES-I [points, norm: 28.7 ± 12.0]





ABC [%; norm: 94 ± 3]





Treatment with 4-aminopyridine improved subjective ambulatory function, whereas clinical ataxia and gait scores were unchanged

m male, FGA Functional Gait Assessment [14], SARA Scale for the Assessment and Rating of Ataxia [10], FES-I Falls Efficacy Scale-International [16], ABC Activity-specific Balance Confidence Scale [9]

We performed gait analysis using a pressure-sensitive carpet (GAITRite®, CIR System, Havertown, PA, USA; length 6.70 m, sampling rate 120 Hz). The carpet system provides mean values and standard deviations for temporal and spatial gait parameters. The walking patterns of both patients were recorded at three different walking speeds (slow, preferred, maximal). Each speed was tested twice. Temporal gait variability was calculated as the coefficient of variation (CV) of stride time with the formula:
$$ {\text{CV}} = (\text {standard deviation/mean}) \times 100 $$

The CV of stride time was calculated for each single walk. Non-linear regression analysis was performed on the CV values of both patients to test for the presence of curvilinear dependency on speed. Second-order polynomial curves were fitted to the individual stride time CV values for both patients. Second-order polynomials were used since we found significant quadratic components. The adjusted r2 values indicated that the data fit well to a second-order polynomial. In all cases r2 values for linear fits were smaller.

Both patients had abnormally high CV values, predominantly during walking with slow and maximal speed. The polynomial function showed a U-shaped curve with minimal variability at medium speed, closely related to self-selected pace (Fig. 1 b, d upper panel).
Fig. 1

Speed dependency of temporal gait variability. Speed-dependent, second-order polynomial function of the coefficient of variation (CV) of stride time of both patients. Treatment with 4-aminopyridine (5 mg t.i.d.) normalized increased CV values, mainly in the high-speed section (b, d, middle panel). After a washout period of 3 weeks, temporal gait variability deteriorated to the level of the pretreatment condition in patient 1 (b, lower panel)

The patients were treated with 15 mg 4-aminopyridine (4-AP) p.o. per day (5 mg t.i.d.). 4-AP, a reversible potassium-channel blocker, prevents ataxia attacks in patients with EA2 [11] and suppresses cerebellar oculomotor symptoms [12]. Moreover, it was shown to improve interictal gait ataxia in one EA2 patient [7].

In our patients, treatment with 4-AP increased the preferred walking speed due to an increase of cadence (patient 1: 106–128 cm/s; patient 2: 87–109 cm/s). 4-AP also reduced temporal gait variability in both patients, mainly under the condition of fast walking (Fig. 1b, d, middle panel). To prove the treatment effect, patient 1 stopped 4-AP for 3 weeks. In this period, the CV of stride time deteriorated again and showed a polynomial U-shaped fit similar to the pretreatment condition (Fig. 1b, lower panel). Subsequent treatment alleviated gait variability again.

Subjective ambulatory scores (FES-I and ABC) were also improved in both patients under 4-AP therapy, whereas clinical scores did not show any significant changes (FGA and SARA) (Table 1).

In summary, 4-AP reduced gait variability in two patients with permanent cerebellar ataxia due to CACNA 1A mutation. As increased variability is related to an increased risk of falls this might be of functional relevance for the gait of both patients. Accordingly, their subjective ambulatory scores were also improved. Thus, 4-AP offers a new option in the symptomatic treatment of cerebellar gait disorders and should be investigated in a larger patient group within a placebo-controlled, randomized clinical trial.

In addition, these two cases indicate that the analysis of gait variability provides an useful marker for measuring treatment effects not visible in global clinical ataxia and gait scores.

Moreover, the two cases support the hypothesis of a permanent disturbed Purkinje cell function in patients with CACNA 1A mutations, although a wide spectrum of clinical phenotypes from intermittent ataxia to severe permanent ataxia are described [6].

For the understanding of the mode of action of 4-AP on cerebellar gait problems, the selective effect of the drug on gait variability during fast walking is of interest. This might be explained by an improvement of the function of a cerebellar locomotor region. In cats, this region is located close to the fastigial nuclei and contains a network of Purkinje cells with intrinsic pacemaker activity, thereby influencing locomotion speed and cadence [5, 8]. In animal models of EA2, it was shown that the precision of this intrinsic pacemaker activity is reduced due to malfunctioning P/Q-type calcium channels. 4-AP effectively restores pacemaker precision of these cells by increasing the duration of action potentials and the amplitude of afterhyperpolarization [1]. This mechanism may explain the described improvement in our patients, as temporal gait variability has been suggested to reflect the precision of pacemaker activity of cerebellar Purkinje cells [3].


The authors thank Judy Benson for copy editing the article. The work was supported by the German Research Foundation (Deutsche Forschungsgemeinschaft, DFG JA 1087/1-1) and the German Hertie Foundation.

Conflict of interest


Copyright information

© Springer-Verlag 2011