Early presentation of gait impairment in Wolfram Syndrome

Diabetes insipidus, diabetes mellitus, and optic atrophy (DIDMOAD) or Wolfram syndrome (WFS) is associated with a host of symptoms [1‐4] including diabetic, psychiatric [5], neurologic, opthalmologic [6] and urologic [7] complications. An autosomal recessive disease with a reported prevalence between 1 in 100,000 [8] and 1 in 700,000 [2], the diagnosis of WFS is confirmed via presence of one or both of the identified WFS genes [9‐12]. Complications of the disorder result in reduced quality of life and premature death, with a reported median life expectancy of 30 years [2]. Currently, effective interventions are lacking and there is no known cure for WFS.

The brainstem and cerebellum are thought to be particularly vulnerable to the neurodegenerative process in WFS [2]. In a single case of WFS, Pakdemirli et al. [13] reported atrophy of the brainstem and cerebellum as well as the cerebral cortex. Clinical evaluation of magnetic resonance images (MRIs) in a small sample of young individuals (mean age 17.5 years) with WFS also revealed atrophy in the brainstem [14]. This finding was confirmed and extended by a recent, objectively quantified analysis of a larger and younger sample of people with WFS, which found reduced gray and white matter volumes and reduced white matter microstructural integrity within the brainstem and cerebellum as compared to age and gender comparable healthy and type 1 diabetic control groups [15]. As the cerebellum and brainstem have been commonly linked to gait and balance function [16‐19], a clinical presentation of ataxic gait may be expected in WFS. However, until recently gait and balance abnormalities in WFS have only been vaguely and qualitatively described.

“Truncal ataxia causing unsteady gait and falls” [2] was reported but not quantified in 15 of 45 individuals with WFS in the late third and throughout the forth decade of life [2]. A more recent review of individual case histories found that clinically apparent gait abnormalities can appear in the second decade of life [3]. We have previously quantified balance performance in young individuals with WFS and found that balance impairment was common in these children and related to the severity of neurological soft signs rather than chronological age [20]. Based on these findings, we hypothesized that quantifiable gait impairments may also be detectable early in life in individuals with WFS compared to typically developing (TD) individuals.

Should gait deficits be present in individuals with WFS, understanding the underlying causes of the impairments would be helpful in informing treatment options. For example, gait impairments could be a direct result of deficits in balance [20]. Alternatively, gait impairments could result from deficits in visual acuity, which are common in individuals with WFS [4, 21, 22], and have been shown to impact typical gait development in other populations [23]. Finally, it is possible that the motor system is altered at an early age and leads to both balance and gait deficits [3, 15]. Understanding this information will help characterize WFS neurological features, which may be the target of future pharmaceutical intervention. Furthermore, quantification of atypical gait patterns could be useful for developing physical therapy interventions to improve gait function in everyday life for people with WFS.

The objectives of this study were to: 1) compare spatiotemporal gait parameters of individuals with WFS to TD controls and 2) determine the relationship between visual acuity, balance, and gait impairment in individuals with WFS. We hypothesized that individuals with WFS would have decreased overground walking velocity, increased cadence, take shorter and wider steps, and spend a greater percentage of time in the double support phase as compared to TD controls. Further, we expected that both visual acuity and balance would be related to gait performance in individuals with WFS.

Demographic data for both groups are shown in Table 1. WFS and age-matched controls did not differ by age (Z = −1.08, p = 0.28), height (Z = −0.59, p = 0.56), weight (Z = 0.37, p = 0.71), or year in school (Z = 1.06, p = 0.29). Additionally, the average number of verbal responses generated during dual task walking trials for TD and WFS did not differ significantly (F = 0.408, p = 0.53) between groups. Mean blood glucose levels taken during the study for the individuals with WFS averaged 222.5 mg/dL (SD = 69.1 mg/dL, max = 312.5 mg/dL, min = 94 mg/dL). No patients experienced any acute neurological symptoms related to hypoglycemia or hyperglycemia during the study.

The goal of this study was to quantify spatiotemporal gait function in young individuals with WFS and compare these individuals to a group of TD individuals. Although ataxic gait has been noted in clinical reports of WFS [2, 3], to our knowledge, no study aside from our own prior work focused on balance deficits [20] has attempted to quantify motor deficits in the WFS population. As hypothesized, individuals with WFS walk significantly slower, take shorter and wider steps, spend more time in double support and have more difficulty performing tandem walking compared to typically developing individuals. Deficits in the WFS groups were present across all tested gait tasks. These differences do not appear to be directly associated with age, visual deficits or impaired balance in WFS patients. These findings indicate that gait is measurably and independently impaired in young individuals with WFS.

As compared to the TD group, the WFS group showed significant deficits in normalized velocity, normalized base of support, step extremity ratio and the percent of the gait cycle spent in double support across all gait tasks. The spatiotemporal gait abnormalities in WFS may indicate the use of compensatory strategies for decreased balance. Supporting this idea, individuals with WFS were more likely to experience a loss of balance resulting in a lateral deviation during tandem walking. Similarly, voluntary changes in step width and length have been shown to alter the stability of trunk motion [27]. In an examination of turning strategies of individuals with cerebellar ataxia, Mari et al. [28] showed that during a demanding gait task, such as turning, ataxic patients displayed wider and shorter steps. The authors theorized that the observed gait difference between healthy and ataxic individuals was a compensatory strategy used to reduce instability. Slower walking velocity has also been previously been linked to increased stability [29] indicating that individuals will compensate for a loss of stability via a decrease in gait velocity. Dingwell et al. [30] noted that individuals with diabetic peripheral neuropathy demonstrate slower walking speeds and increased dynamic stability as compared to healthy, age-matched individuals. They concluded that decreased walking speed is a compensatory strategy used by the patient group to maintain stability during walking. As such, we hypothesized that the observed deficits in spatiotemporal gait measures would be related to previously reported balance deficits in individuals with WFS.

We have previously reported a deficit in balance in individuals with WFS [20], particularly during tasks completely reliant on balance and on which compensatory strategies cannot be employed, such as standing on one foot or standing on an unstable surface. In contrast, during the dynamic gait subsection of the mini-BESTest which relies on gait-related balance tasks such as changing velocity and turning the head while walking, no deficits were detected in WFS. Perhaps this is because they were able to employ compensatory strategies during this group of tasks that could not be used during more static balance tasks, such as a wider stance. However, if the observed differences in gait between WFS and TD individuals are the result of strategies used to compensate for balance deficits, a clear association between balance and spatiotemporal gait measures would be expected. However, there appears to be no consistent correlation between spatiotemporal gait measures and previously reported balance scores, suggesting that other factors should be explored.

Only two spatiotemporal gait features, normalized fast forward gait velocity and double support during backward walking, were significantly correlated with balance. The significant negative correlation between mini-BESTest score and normalized velocity was unexpected. This indicates that individuals with WFS with lower mini-BESTest scores (worse balance) performed FF walking trials at higher velocity than individuals with WFS and higher mini-BESTest scores (better balance). This result is strongly driven by the performance of a single individual who scored poorly on the mini-BESTest but was able to walk at a fast velocity. We are only able to speculate at this time, but this type of high velocity gait with low overall balance may lead to an increased number of falls. Future studies should include a falls assessment to address this question and may be helpful in identifying the needs of the patient group.

In addition to balance, vision has been repeatedly linked to locomotor performance [31, 32]. Specifically, in young people, visual impairment correlates with spatiotemporal gait deficits [23, 33]. In WFS, visual deficits are well described (for review see [34]) thus we expected gait impairments in individuals with WFS to correlate with measures of visual impairment. However, we did not find any correlation between visual acuity and any spatiotemporal measures of gait function, suggesting that vision may not be a key factor in the gait differences noted in WFS compared to TD individuals. Here again, as with correlations to measures of balance, the small sample size should be noted.

Previous reports indicate that children may exhibit mature gait patterns as early as four years of age [35], though others argue that gait continues to follow a developmental trend and does not reach adult-like patterns until after the seventh year and perhaps even beyond the tenth year of age [36]. As our sample spans from five to 26 years, age-related changes in spatiotemporal gait measures cannot be overlooked. Three of the selected measures (velocity, base of support and step length) were normalized to account for anatomically related developmental changes such as increased leg length and height. Within the WFS group, no significant correlations were present between age and these normalized measures. In the TD group, only three of the twelve normalized variables examined were significantly correlated with age. However cadence, a non-normalized variable, was significantly negatively correlated with age across all gait tasks in the TD group but not in the WFS group. This matches our previous report of significant correlations between balance and age in the TD group but not in the WFS group [20]. This lack of a clear relationship between age and cadence within the WFS population may indicate that individuals with WFS are neurologically affected early in their lifespan and thus do not conform to normal gait developmental patterns. This notion is further supported by the early presence of reduced gray and white matter volumes and reduced white matter microstructural integrity within the brainstem and cerebellum as compared to age and gender comparable healthy and type 1 diabetic control groups [15]. However, caution is warranted as the sample of individuals with WFS in the present study is relatively small.

Based upon the current findings, gait impairments cannot be accounted for by the particular sensory or motor deficits we evaluated. Additionally, age does not consistently relate to spatiotemporal gait measures in WFS. The overall model of neurological symptom presentation and the corresponding treatment approach to WFS is complicated by the observed deficits in multiple sensory and motor systems as well as the unique phenotypic presentation of these symptoms for each patient diagnosed with WFS. To date, the sensory and motor characteristics of individuals with WFS have been primarily qualitatively described. Clinically noted deficits have been reported but are rarely quantified. Longitudinal data quantitatively tracking neurological symptoms in a larger sample across the lifespan will be necessary to fully address this question. However, the presentation of early gait and balance deficits, along with the presence of early abnormalities of the brainstem and cerebellum indicates that neurological symptom presentation occurs in early childhood, and could be addressed clinically at an early age.

Wolfram Syndrome is rare and debilitating with neurological motor symptom onset occurring far earlier than previously reported. We have quantified deficits in normalized velocity, stride length, normalized base of support, and the percent of the gait cycle spent in double support in a group of young individuals with WFS. These data provide evidence of the early emergence of gait abnormalities in WFS.

Washington University Wolfram Study Group Members: In addition to the authors: P. Austin, M.D., A. Bondurant, B.A., T. Hullar, M.D., R. Karzon, Ph.D., J. Lapp, B.A., J. Leey M.D., H. M. Lugar M.A., L. Manwaring, M.S., C. Nguyen, B.S., A.R. Paciorkowski M.D., J. Rutlin, J. Shimony, F. Urano, M.D., Ph.D., A. Viehoever, M.D., J. Wasson B.S., and N. H. White M.D., CDE.