Background
Hereditary proximal Spinal Muscular Atrophy (SMA) is an autosomal recessive neurodegenerative disease caused by homozygous loss of function of the survival motor neuron 1 (
SMN1) gene [
1]. SMA is characterised by a wide range of disease severity ranging from neonatal respiratory insufficiency and death (SMA type 1), ability to sit without support but inability to walk independently (SMA type 2), problems with or the loss of ambulation (SMA type 3a-b) to relatively mild impairments due to proximal muscle weakness in patients with adult onset disease (SMA type 4) [
2]. All four SMA types are characterised by progressive muscle weakness and secondary loss of motor abilities over time [
3]. In addition to muscle weakness, fatigability has emerged as a rather common but often overlooked complaint among patients with SMA [
4,
5]. The current taxonomy defines ‘fatigability’ as the magnitude or rate of change in a performance criterion relative to a reference over a given time of task performance or measure of mechanical output and is the opposite of ´endurance´, which involves the prolonged maintenance of constant or self-regulated power or velocity [
6,
7]. Patients with SMA refer that they easily fatigue during repetitive activities of daily living such as lifting an arm during eating or walking even short distances. A possible explanation comes from SMA animal models and post-mortem studies that showed abnormal development and maturation of the neuromuscular junction. Neuromuscular dysfunction has been found in at least half of the patients with SMA, suggesting that this may contribute to complaints of fatigability [
5,
8‐
12]. Since outcome measures sensitive to change in fatigability are lacking, their development is a pivotal step in a better understandig of fatigabillity in SMA [
13,
14]. This study aimed to provide the framework for the development of novel clinical outcome measures for fatigability in patients with SMA type across the range of severity. We determined content validity following the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) -guidelines and recommendations by European and American regulatory authorities [
15,
16].
Methods
A set of outcome measures for fatigability was developed according to five methodological steps as recommended by COSMIN (Table
1) [
17,
18]. In this iterative process data from multiple sources were triangulated. Sources included a scoping review of scientific literature, input from a scientific and clinical multidisciplinary expert panel and three pilot studies including healthy persons (
N = 9), paediatric patients with chronic disorders (
N = 10) and patients with SMA (
N = 15). The expert panel consisted of ten clinicians and researchers including paediatric physical therapists (BB, MS), clinical exercise physiologists and movement scientists (JG, LH, HH,TT) and neurologists or neurology residents with ample experience in caring for children and adults with SMA (MS, CW, RW, WP). Three round table discussions took place with different group compositions.
Table 1
Methodological steps COSMINa
Step 1: Definition and elaboration of the construct | 1) Definition of fatigability? 2) Target population? 3) Purpose of the outcome measure? | • Key papers on fatigability assessment |
Step 2: Choice of measurement method | 1) Existing measurement that responds closely to construct to be measured? 2) Level of measurement? 3) Single or multiple measures? | • Scoping review of scientific literature • Expert panel (round table discussion 1) |
Step 3: Selecting and formulating items | 1) Which activities cause most problems? 2) Which available measures reflect these activities? | • Patient report outcome (pilot sample 3) • Expert panel (round table discussion 2) |
Step 4: Scoring issues | 1) Application in research or clinical practice? 2) Measurement level? | • Expert panel (round table discussion 3) |
Step 5: Pilot testing | 1) Comprehensibility? 2) Feasibility? 3) Relevance? | • Pilot sample 1 (healthy subjects) • Pilot sample 2 (pediatric patients with chronic diseases) • Pilot sample 3 (patients with SMA) |
Definition and elaboration of the construct intended to be measured
The first step in the development of a new outcome measure consisted of the operationalization of the theoretical construct in SMA. This included a clear definition of fatigability, a description of the target population and the purpose of the outcome measure and the composition of specific test criteria. The taxonomy for fatigue and fatigability as proposed by Kluger et al. was used as a starting point from which a construct for fatigability assessment in SMA was described. Fatigability was defined as the magnitude or rate of change in a performance criterion relative to a reference over a given time of task performance or measure of mechanical output [
7]. Several other key papers on fatigability that used a similar definition and described test methodology were selected to complement the framework [
19‐
25].
Choice of measurement method
During the second step we combined the results from a scoping review on available measures for fatigability in patients with SMA with the experiences with fatigability testing by our research group.
Scoping review of the literature
Given the fact that SMA has been associated with fatigability only recently, it was anticipated that a systematic review would not generate significantly more information than a scoping literature search. Peer-reviewed experimental articles written in English were retrieved from Pubmed and
Trial.gov up to the first of October 2014. The following search strings was used
: ((“muscular atrophy, spinal”[MeSH Terms] OR (“muscular”[All Fields] AND “atrophy”[All Fields] AND “spinal”[All Fields]) OR “spinal muscular atrophy”[All Fields] OR (“spinal”[All Fields] AND “muscular”[All Fields] AND “atrophy”[All Fields])) OR (“Stat Methods Appt”[Journal] OR “sma”[All Fields])) AND (((Fatigability [All Fields] OR Endurance [All Fields]) OR Stamina [All Fields]) OR (“fatigue”[MeSH Terms] OR “fatigue”[All Fields])). At first, papers were selected that described the measurement of fatigability or endurance in patients with SMA. Secondly, outcome measures were assessed to what extent they complied with the definition and test criteria defined within this study. In the case that no suitable outcome measure were retrieved, the expert panel discussed in the first round table discussion whether other appropriate outcome measure were available that met the clinimetric requirements and could be validated for SMA.
During the third step, questionnaires were taken from the pilot sample of patients with SMA to determine which activities of daily living (ADLs) provoked fatigability. In adults, the questionnaire by Straver et al. was used which was originally validated for peripheral nervous system disorders [
26]. A similar questionnaire was developed for children based on clinical experience from the expert panel and items from the Child Health Assessment Questionnaire, a validated questionnaire for ADLs in other clinical populations [
27]. Patient-reported activities that caused fatigability were clustered into three different functional domains, namely leg function, upper arm function and hand function. The expert panel assessed in the second round table discussion whether all domains were relevant to SMA and should be included in the development of the set of outcome measures for fatigability.
Scoring issues
During the fourth step, the expert panel discussed about the composition of the tests, taking into account the application setting (research, clinical practise) and the patient group, and selecting primary outcome parameters. For example, tests are usually shorter in clinical practise, due to time constraints [
17].
Pilot testing
Patients with SMA were recruited from the Dutch SMA registry (
http://www.treat-nmd.eu/registry/310/) [
28]. This registry contains detailed clinical information of over 300 children and adults with SMA. To minimize selection bias, all eligible patients listed in this register were offered the possibility to participate. All patients had a confirmed homozygous deletion of the
SMN1 gene or a heterozygous
SMN1 deletion in combination with a point mutation on the second
SMN1 allele. In order to be eligible to participate in this study, a subject had to meet all of the following additional criteria: age 8–60 years; ability to follow test instructions and no exercise restrictions. Two patients with SMA declined participation due to frequent hospital visits in the recent past and fear of increased fatigue. Patient controls were recruited from a school for special education in Utrecht. Healthy controls were recruited from the University of Applied Sciences and the University Medical Center Utrecht. The outcome measures for fatigability were pilot-tested on
‘comprehensibility’ (‘Are test instructions to participants unambiguous and well understood?’) and
‘feasibility’ (measurement completion, acceptability and perceived burden) in three consecutive pilot samples of healthy controls (pilot sample 1), paediatric patients with chronic diseases (pilot sample 2) and patients with SMA (pilot sample 3).
‘Measurement completion rate’ was defined as the number of participants able to complete the test without premature discontinuation caused by motivational issues or a-specific physical complaints [
17]. ‘Acceptability’ was defined as the willingness to perform the test again in the future and was assessed with a ‘Visual Analogue Scale’ [
29]. Perceived burden was assessed with the OMNI scale for perceived exertion [
30]. The third round table discussion was used to discuss pilot data and if necessary to make small adjustments to the protocol.
Discussion
This study aimed to provide the framework for the development of novel clinical outcome measures for fatigability in patients with SMA across the range of severity. The major strength of this study includes the use of the methodological steps as recommended by the COSMIN guidelines to systematically develop a set of endurance tests for patients with SMA with a specific emphasis on content validity [
17]. Content validity is the degree to which the content of an instrument is an adequate reflection of the construct to be measured and without it, it is difficult to select appropriate outcome measures for trials or other types of interventions [
49]. It is therefore recommended by the US Food and Drug Administration and the European Medicines Agency to establish content validity before evaluating other measurement properties [
15,
16]. The content validity of the endurance shuttle tests was established by combining evidence from scientific literature with patient reported outcome and the expertise from health care professionals and scientists, which will potentially lead to both valid and clinically meaningful outcome measures.
An important aim of this study was to develop one methodology for a broad clinical spectrum that would enable comparison between severely and mildly affected patients and with that facilitate future study trial inclusion. The methodology of the ESWT, originally validated for pulmonary disease was adjusted and applied to other motor tasks to meet with the specific disease characteristics of SMA. The ESWT speed was originally derived from a time consuming four component process including a second ISWT and a regression equation including maximal predicted oxygen uptake. Although Hill et al. simplified this method by directly using maximal walking speed it still included a second exercise test [
50]. We questioned the validity of this method because of the risk of inducing fatigability prior to the test and therefore decided to use muscle power as the parameter to determine exercise intensity in SMA. Time in which 10 m, 9 pegs or 10 blocks could be transferred were taken as maximal performance measure. It was decided not to adjust for the weight of the blocks and pegs or body weight, since both materials were very light and body weight is fixed in daily life activities as well. The convergent validity of this modified method with the original procedures and the comparability between patients with mild and severe muscle weakness need to be further analysed in future studies.
We decided to include motor tasks because we wanted to generate clinical relevant outcome measures and patients with SMA generally have normal coordinative function. The use of motor task within endurance tests potentially causes validity issues. For example, a subject might drop out because of motor coordination difficulties rather than fatigability. To confirm construct validity, it will be important to monitor other parameters of fatigability such as perceived exertion, motor behaviour and change in strength and electromyography response [
38,
40,
51,
52].
Besides the clinimetric properties, the practical application of a new measurement test is an important aspect in the development of outcome measures for clinical practice. Ideally, an outcome measure is suitable for both day-to-day clinic purposes and clinical trials. For this purpose, an instrument needs to be easy to use in a limited time period, acceptable and feasible for the individual subject, while at the same time, applicable to a large part of the study population. The endurance tests have demonstrated to be comprehensible and acceptable for both healthy subjects and patients with a wide range of severity in an age range of 10–49 years. Based on our clinical experience and an upcoming large study on validity and reliability (Bartels et al. in progress), we expect the endurance tests to be suitable for subjects aged 6 years and older for those being able to move around their dominant hand on their wheelchair table as minimal motor function. The additional burden and time consumption in the context of endurance tests as part of the already extensive trial assessments asks for a clear rationale about the efficacy in terms of function accompanied by the selection of the most appropriate tests. In order to be able to measure clinically relevant improvement in endurance, endurance tests that mimic long-term activities are required.
In the current literature, the concept of fatigability and fatigue are often used interchangeably with different terms such as fatigue [
25], fatigability [
7], neuromuscular fatigue [
53], perceived fatigability [
54], physiological fatigue [
55], physiological fatigability [
19], physical fatigue [
20], peripheral fatigue [
22], muscle fatigue [
21,
56,
57] and so on. The lack of standard definitions and the inconsistency of terminology hamper the advancement in our understanding of the pathophysiological background of fatigability in SMA and the development of appropriate outcome measures. The taxonomy used in this study was particularly suitable to standardize definitions and clarify the different concepts and means of measurements as a prelude to the development of an outcome measure for fatigability in SMA. The taxonomy makes an important distinction between ´perception´ and ´performance´, which are measured at a different level. Perception of fatigue is defined as the subjective sensations of weariness, increasing sense of effort, mismatch between effort expended and actual performance or exhaustion while fatigability is about decline in either physical or mental performance. Although there is a clear distinction in definitions and means of measure, endurance performance is regulated by an interaction between fatigability and perceptions of fatigue and influenced by physchological factors, peripheral limitations and central factors [
6,
57‐
59]. Therefore, a psychophysiological approach is needed when interpreting the outcome of endurance testing in patients with SMA. Based on both pre-clinical and clinical data it was hypothesised that fatigability would be associated with neuromuscular junction dysfunction in at least half of the patients with SMA and therefore, similar to the approach in myasthenic syndromes, best provoked with a repetitive submaximal prolonged motor task. Although Spinal Muscular Atrophy is primary characterised by loss of motor neurons, involvement of other systems such as autonomic dysfunction and altered muscle metabolism is reported and might demand additional methods of testing to capture fatigability in SMA [
60,
61]. Fatigability in SMA could also be related to an increased energy cost of movement due to progressive muscle weakness and secondary deconditioning [
20,
23]. Therefore, the individual disease course and physical activity levels should be taken in account when measuring fatigability and its change in time.
Limitations
We decided to limit our search for existing outcome measures to a scoping review in SMA although a systematic review in the entire range of neuromuscular diseases could have generated other endurance measures. Based on experience by the expert panel and the specific characteristics of SMA regarding clinical variability and complaints of fatigability, it was anticipated that a time consuming systematic review would give very limited additional information as hardly any endurance testing has been developed in neuromuscular diseases. The involvement of patients in rare disease clinical trial design is increasingly becoming a priority [
62]. Although established methods such as face-to-face meeting and focus groups were not applied yet, extensive questionnaires provided a valuable insight in the patient perspective on fatigability. In the further development of the endurance tests, patients will continue to play an important role.
Acknowledgements
The authors thank all patients and healthy controls who participated in this study.