Age-related kinematic performance must be considered during fast head-neck rotation target task in individuals aged from 8 to 85 years old

Rationale Kinematic behavior during fast cervical rotations is a useful parameter for assessing sensorimotor control performances in neck-pain patients. However, in asymptomatic individuals from children to seniors, the influence of age still needs to be explored. Aim and method We assessed the impact of age on sensorimotor control performance of the head-neck with execution time and kinematic variables (time of task, mean speed/acceleration/deceleration, overshoots, minimum/maximum speed) during standardized fast rotation target task using the DidRen Laser test. Eighty volunteers were stratified in four different age-groups: Children [8-14y]: n=16; Young Adults [18-35y]: n=29; Old Adults [36-64y]: n=18; Seniors [65-85y]: n=17. Results To perform the test, Children were slower compared to Young Adults (p<0.001) and Old Adults (p<0.001). It was also slower in Seniors compared to Young Adults (p<0.013). Mean speed was slower in Children and Seniors compared to Young Adults (p<0.001) and Old Adults (p<0.001). Mean acceleration was slower for Children compared to Young Adults (p<0.016) and Old Adults (p<0.015). Mean deceleration was slower for Children compared to Young Adults (p<0.001) and Old Adults (p<0.003). Conclusion The DidRen Laser test allows us to discriminate age-specific performances for mean speed, acceleration and deceleration. Seniors and Children needed to be slower to become as precise as Young and Old people, no difference was observed for overshoots which assesses accuracy of movement. Age must therefore be considered as a key parameter when analyzing execution time and kinematic results during DidRen Laser test.


Introduction
The lifetime prevalence of neck pain is almost 70%, including pre-and adolescent patients, and this increases with age up to the age of 60 years 1,2 . Nevertheless, underlying causes of neck pain are poorly understood. Because head rotation is a regular movement performed during everyday activities, this movement represents one of the issues that are of interest in patients with neck pain 3 . Interestingly, previous studies supported that fast-rotational movements of the headneck complex may be applied in clinical assessment and/or therapeutic management in patients with idiopathic neck pain [4][5][6][7] . In view of these reports, fast axial rotation of the neck represents a critical motion to assess.
A target test using the DidRen Laser can be applied in response to real visual targets to standardize a head rotation motor task about 30° 8,9 . Firstly, this test is believed to rely on the integrity of neuro-musculoskeletal structures and of the sensorimotor control system 10 i.e. input from the visual, vestibular and proprioceptive systems, particularly the richly innervated cervical spine, which controls head and eye movement and postural stability 10 . Secondly, the axial head rotation about 30° will preserve the motion from the strain of the passive system (joint capsules, facets, intervertebral disks and ligament) 11 .
Target-directed head-neck complex movements are fundamental components of individuals' daily activities and usually require high levels of motion speed and accuracy. Speed-accuracy tradeoff could be considered as a "signature" of the decision process 12 . This varies according to which motion behavior is emphasized: accuracy or speed 13 during, for instance, an aiming task that encompasses amplitude of the movement, the size and position of the target 5,12-14 .
Lack of sensorimotor control of head-neck complex can be explained 15,16 by degeneration of vestibular, visual and neuromuscular functions due to aging or even neck pain in adults and due to immaturity of the central and peripheral nervous and musculoskeletal system in healthy children 17,18 . and "The Fly" 21 use a tracking system placed above the head to respectively measure the accuracy of the ability to follow a virtual target (the fly). Finally, the "DidRen Laser" is a test based on a system developed by our team in the late 2000s. It includes also both sensory and motor component by using a laser beam placed on the head of the participant with the aim to induce fast, low amplitude, and accurate rotational movements of the head-neck complex in response to real visual targets placed in front of the participant 8 . Sensorimotor performance is assessed using the time difference between two successful hits of the targets. This means that, in view of the participant's performance of speed-accuracy trade-off, the shorter the time, the faster and the more accurately the task is accomplished 12,13,22 . In 2009, we showed that the DidRen Laser was a simple and reliable device with a good reproducibility in asymptomatic and symptomatic adult individuals 8 . Compared to the other tests, the DidRen Laser is very easy to implement in a clinical setting, but it assesses only temporal variables. This does not allow us to gain insight into sensorimotor performance adopted by individuals.
In a recent study by Sarig Bahat et al. (2016), modified kinematics of the head-neck complex were reported in asymptomatic people over 60 compared to young and middle-aged adults 23 . This study did not include measurements in children, which we consider essential to provide a better understanding of head-neck complex kinematics and normative data across all stages of the lifespan and to ensure that appropriate age-matched comparisons to participants with neck pain can be made 24 .
To complete our previous results, it now appears essential to assess on asymptomatic individuals from children to seniors', in addition of the temporal variables, further detailed kinematic and accuracy analysis carried out by a motion capture system, such as the maximum/minimum rotational speed, the average speed, the acceleration, deceleration and overshoot. Further inclusion criteria were absence of neck pain episodes in the 6 months prior to the study and to ensure that we analyzed asymptomatic participants, a Neck Disability Index (NDI) score of less than or equal to 4% 25,26 . Similarly, all participants were asked to complete a Visual Analogue Scale (VAS) to confirm the absence of pain on the testing day 27 . The main characteristics of the participants are listed in Table 1.

Instrumentation
The DidRen Laser was used (see 8

Experimental procedure
All participants received the same instructions about how the experimental procedure was to be conducted and more specifically that of the DidRen Laser test, by watching an explanatory video on a tablet computer. The participants wore a helmet to which a laser beam is mounted and sat on a chair. They were instructed to keep their back against the backrest, the palm of hands on the thighs, feet flat on the floor with heels against a stop block placed at the feet of the chair and to refrain from talking during the test.
The procedure of one generated cycle was explained as followed: turn your head faster as you can and point the laser correctly at the target. When the laser beam is pointed correctly at the target (during at least 0.5 s to lower chance), the LED's sensor lights up and the system emits a sound signal. As soon as the central sensor has been ''hit'', the participant must, as rapidly as possible, turn his/her head to maximum 30° to the right to hit the right-hand sensor. He/she returns to the central sensor and then rotates the head to maximum 30° to the left to hit the left-hand sensor. The complete trial is composed of 5 cycles. Therefore, left and right rotation toward the target is composed by two phases. One fast rotation phase to turn the head followed by one stabilization phase to adjust the laser accurately during 0.5 second in the sensor/target (Fig.2). We used nonrandom generated cycles in order to facilitate intra and inter-subject comparisons. As in 2009 8 , two DidRen Laser tests were conducted. The first trial was considered as a short warm-up to "familiarize" the participant with the experiment by emphasizing on the adequate sitting position and speed-accuracy execution. So only the second trial was used for data collection and analysis. We did not conduct more than two trials to avoid possible fatigue which could lead to a lost precision 29 .

Outcomes measures
The DidRen Laser software calculates each time taken by the participant to go from one target's ''hit'' (that is when participant stops during at least 0.5 s on the sensor) to another target.
Only the total time (TT, in s) to complete the 5 cycles (from the first to the last target) of a trial was included in the data collection.
Head rotational and shoulders-displacement were computed using ELICLINIC software (BTS, Italy) from X, Y, and Z coordinates at each frame (Fig.1). To assess that the participants respected our theoretical calculation of 30° rotation without moving their shoulders, we calculated headrotation range of motion (in °) and shoulders displacement (in °) (Fig.3). By successive numeric finite difference (n=±5 points), we calculated speed and acceleration of head. Because of their reliability, sensitivity and specificity 4,23 , the following specific parameters were computed for each kinematics variable: maximum/minimum rotational speed (Max S, Min S, in °s -1 ); average rotational speed (Mean S, in °s -1 ) and average rotational acceleration/deceleration (Acc, Dec, in °s -2 ). To assess the accuracy, we calculated overshoot (OS, in °) (Fig.3) 30 . It was computed as the difference between peak rotation amplitude and stabilized mean rotation amplitude. All variables were calculated during 5 consecutive cycles and averaged.

Statistical analysis
To assess the effect of age on the average variables, a one-way ANOVA with post hoc Holm-

Results
Total sample size consisted of 87 participants from which 7 participants were excluded due to an NDI score >4%. The anthropometric characteristics of the participants are listed in Table 1.
All results are showed in Table 2 and Figure 4. A significant effect (p<0.05) of age was observed for 4 kinematic variables in Children and Seniors. The Total Time (TT (s)) was longer in Children compared to Young Adults (p<0.001) and Old Adults (p<0.001). It was also longer in Seniors compared to Young Adults (p<0.013). The average rotational speed (Mean S (° s -1 )) was slower in Children and Seniors compared to Young Adults (p<0.001) and Old Adults (p<0.001).

Discussion
The aim of this study was to analyze the effect of age from 8 to 85 years old on sensorimotor control performance adopted by asymptomatic individuals using kinematic and accuracy variables derived from rotational motion of head-neck complex during the execution of the DidRen Laser test.
The results of the study reveal significant effects in asymptomatic Children and Seniors groups for four rotational kinematic variables: TT, Mean S, Acc and Dec. These kinematic differences were observed during the DidRen Laser test, showing its capacity to discriminate agerelated differences. Children-age limit of 14 years was based on need to ensure of an incomplete maturation of the sensorimotor representation 17 . Moreover, we decided to split adults participants into two adults groups because neck pain increase with age and is most common around the fifth decade of life 1 .
Velocity and its time derivate forms were chosen to analyze the performance of age-group participants in accordance with previous studies 4, 9 . Sarig Bahat et al. (2016) showed that the most powerful age-groups differences were Velocity Peaks and Number of Velocity Peaks 23 . With a different task and method of calculation, our kinematic variables such as TT, Mean S, Acc and Dec appeared to be very significant especially with the groups of Children and Seniors.
To reach the speed-accuracy required by our protocol: "turn your head faster as you can and point the laser beam correctly at the target", participants needed to trade-off with their sensorimotor ability to perform first the fast-rotational movement and then to enable the participant to stabilize the motion and to adjust the laser into the sensor of the target 12 . The task reflects an orchestrated pattern of neural activation to select a minimize time to reach target 31 , but participants needed to trade accuracy for speed because they were instructed to respond as fast as possible to a constrained target-aiming task 13,16,32 . In light of the foregoing, we showed that Seniors and Children needed to be slower (TT, Mean S, Acc, Dec) 5,14,32,33 to become as precise as Young and Old people (no significant differences for OS in all age-groups). The validity and reliability of the OS seems to be good when comparing patients with WAD with controls 21  This study presents some limitations. First, neck ROM was limited to 30° that allowed us to avoid the strain of the neck passive system (joint capsules, facet joints, intervertebral disks and ligaments) and to improve input from the upper cervical proprioceptive system which is highly developed in the sub-occipital region upper neck 48 and which corresponds to the spinal muscles that provides dynamic stability during the first degrees of rotation 11 . Second, only head-neck rotational movement was assessed but rotation seems to be a regular movement during daily activities, the assessment of other motion directions (e.g. flexion/extension) appeared to have limited interest. Third, we used for the test regular generating head rotation cycles. We acknowledge that the use of randomly generated cycles might reduce an induced anticipatory motions and predictions of participants but that has not yet been investigated for the DidRen test.
In conclusion, DidRen Laser test allows us to discriminate age-specific performances for mean speed, acceleration and deceleration. Empirically we showed that Seniors and Children needed to be slower to become as precise as Young and Old people by showing no difference for overshoots which assesses accuracy of movement. Age must therefore be considered as a key parameter when analyzing execution time and kinematic results during DidRen Laser test but not for accuracy.