Physical exercise and goal attainment after shunt surgery in idiopathic normal pressure hydrocephalus: a randomised clinical trial

The core clinical symptoms in idiopathic normal pressure hydrocephalus (iNPH) are impaired gait in combination with a disturbed balance [1‐3] often in conjunction with cognitive impairment and incontinence [4]. The iNPH scale is a calibrated, norm-based grading scale assessing postoperative outcome in iNPH and previously used in the European multi-centre study [5]. The scale covers four symptom domains and a total score is calculated [6]. Approximately 80% of the patients are reported to benefit from shunt surgery [5, 7]. However, patients with iNPH are more sedentary in comparison with healthy controls of the same age. Despite the alleviation of symptoms, the patients’ overall physical activity does not increase after surgery [8].

In Parkinson’s disease with a symptomatology similar to iNPH, physical exercise is an essential part of the treatment [9, 10]. Goal-oriented aerobic exercise has been suggested to promote neuroplasticity, motor and cognitive improvements [11]. Exercise programmes lasting for more than 12 weeks have long-term effects [12]. Rehabilitation in iNPH is emphasised to be an important factor to improve patients’ functions [13, 14]. A recently published non-randomised clinical trial evaluating home exercise in iNPH concluded that home exercise was easy to implement and accepted by the patients [15]. However, no randomised clinical trial has explored the effectiveness of rehabilitation in iNPH.

The aim of this study was to evaluate the effect on outcome and goal achievement of an added physical exercise programme for patients with iNPH after shunt surgery, compared to a control group of iNPH patients. We hypothesised that the high-intensity functional exercise programme (HIFE) [16] would have positive effects on the main symptom domains in iNPH evaluated with the iNPH scale [6].

This randomised clinical trial evaluates the effects of physical exercise in iNPH. Our primary hypothesis that the exercise group would show greater postoperative improvement in the total iNPH scale score than the control group was not confirmed. Secondary outcome analyses showed: (1) after 6 months, the exercise group had improved more in the balance domain scores in both ITT and PP analyses; (2) at the post-intervention follow-up, the exercise PP population achieved their goals to a greater extent.

Even if we could not show an overall effect, for reasons discussed in the limitation section, the improvement in balance after 6 months is notable. The scientific evidence of the effect of balance training in iNPH is limited. In Parkinson’s disease, different interventions improve balance and reduce the number of falls up to 12 months after the end of the interventions [12]. The HIFE programme has previously been evaluated for balance with the Berg Balance Scale as an outcome measure in a population with dementia in a nursing home. In that study there was no between-group difference compared to a low-intensive control intervention at the post-intervention follow-up but long-term effects were found with between-group differences favouring the exercise group after 6 months, [20] similar to the results from our study. Gait speed was improved in the intervention group, both at the post-intervention follow-up and at the long-term follow-up [20].

Both the ITT exercise group and the control group increased their total iNPH scale scores from the post-intervention follow-up to the six-month follow-up. Although not significant this increase may indicate a continuing improvement over time after shunt surgery and that longer follow-up periods can be valuable when evaluating outcome in iNPH patients.

The sub-group populations stratified by sex are small, which makes the gender differences in outcome less robust. However, older females are generally less physically active than males [21]. In our study the females had lower physical function at baseline (lower iNPH scale score) even if the sex differences were not significant. This may have contributed to the more pronounced intervention effect on gait among the females. However, our findings generate important questions, e.g. whether females have a lower level of function before surgery, which has been previously reported [5] or whether females wait longer with symptoms before seeking medical advice, which delays surgery, and finally whether there are any differences in rehabilitation effects.

The higher goal attainment at the post-intervention follow-up in the PP population exercise group indicates that the HIFE intervention in combination with the supervision may have influenced the goal achievements. The lack of difference in goal attainment between the exercise group and the control group at the six-month follow-up also supports this notion. Goal setting could be managed by people with mild to moderate dementia, especially when using a structured approach [22]. Impaired memory and executive difficulties in iNPH improve after shunt surgery but are still not at the same level as in healthy individuals of the same age [23]. The cognitive decline can affect the ability to self-manage the goal activities long-term. INPH patients therefore probably need continuous encouragement and support to maintain reached goals. Patient-centred care with the approach to reach self-management is fundamental in the European physiotherapy guideline for Parkinson’s disease. The guideline emphasises education and repeated support from professionals as well as carer involvement to promote the patients’ behavioural changes [9]. We believe that these recommendations are applicable to iNPH patients.

The HIFE programme is developed and feasible for persons with dementia [16, 24]. Using the programme in nursing home residences, the attendance rate was about 75% [16, 25] and about 70% of all exercise sessions were performed at high intensity [25]. Motivation and co-existing medical conditions seemed not to influence the benefit from high-intensity exercise in persons with dementia [26]. We conducted our research in primary care with the intervention performed by local physiotherapists with the aim of reflecting a clinical routine. This may be one reason for the low adherence to exercise in this study.

Finally, in the research area of iNPH there are relatively few large prospective studies exploring the outcome after shunt surgery [27‐29]. This study brings not only new information about the role of physical exercise in iNPH but also adds important knowledge about the outcome after a shunt operation. The iNPH scale is constructed to evaluate outcome after surgery as a total score [6] and its potential for use in a larger context is explored here. Our findings are in line with the results of the European iNPH multi-centre study with significant postoperative improvements in the total scale score as well as in all the separate domain scores [5].

Our study has several limitations. The most severe limitation is the large dropout rate causing low adherence to the intervention, which entails loss of statistical power. This limitation probably leads to failure to identify differences in favour of intervention, as indicated by the numerical differences seen between the groups for many of the outcome measures. Future studies should be larger and focus on better adherence to exercise intervention. The intervention was performed in clinical practice by physiotherapists with variable experience and conditions that may have caused differences in supervision. There were also logistical problems for the coordinators to initiate the intervention to start in time. Additionally, shunt complications and comorbidity influenced the ability to perform the intervention. Highly supervised physical exercise clinical trials under controlled conditions often generate a higher adherence rate and optimal performance but physical exercise interventions have to be applicable in clinical practice [30]. The dropout risk is important to be aware of when studying iNPH rehabilitation in the future.

The double interventions design may have influenced the results. Both groups increased clearly in all domain scores postoperatively and the surgical effect may have covered the smaller effect from the rehabilitation intervention. Another factor to consider is that the control group may have followed the advice to exercise more than we had expected and that the difference between the groups therefore became smaller and not significant. The magnitude of exercise in the control group was unknown to us.

The primary outcome measure can be questioned. The iNPH scale is developed to evaluate the most affected symptom domains in iNPH after shunt surgery and is not validated to evaluate rehabilitation effects. The HIFE intervention is based on functional exercises of gait and balance and the lack of effects in the neuropsychology and continence domains might have been expected. Aerobic exercise at moderate intensity has been shown to have a positive effect on cognition functions in patients with mild cognitive impairment [31]. However, the HIFE programme has not been shown to have beneficial effects on cognition in individuals with dementia [32]. The individual goals in the present study were set in the enrolment process, months before the surgery. This may have influenced the effect and may have given an advantage to the exercise group overseen by the supervising physiotherapists.

An additional high-intensity exercise intervention did not influence the total iNPH scale scores after shunt surgery for patients with iNPH. The exercise group improved more in the balance domain scores at the long-term follow-up and reached their set goals more frequently. Our results indicate that physical exercise and goal setting can be effective for individuals with iNPH. Future studies are required to understand the effect of different interventions and the feasibility in clinical practice. Research is also required to illustrate whether there are sex differences in the iNPH condition as well as in aspects of rehabilitation.