Introduction: Non specific low back pain (NSLBP) is a low back pain (LBP) that cannot be attributable to a known pathology. LBP has prevalence as high as 84%, making it a heavy burden for health services and society worldwide. Furthermore, LBP seems to be more prevalent as the population gets older, and is the cause of severe functional limitations. One of the treatment recommended for LBP is physical therapy. It has been shown that the trunk muscles usually maintain trunk stability by making a sequence of postural adjustment in advance of distal movement to prevent any loss of balance. However, in NSLBP, this motor control adjusting is lacking, and, furthermore, stays so even after the resolution of acute LBP, which could contribute to the recurrence of LBP. Physical therapy therefore addresses NSLBP by rehabilitating the timing and activation of trunk muscle, among other things. The importance of trunk musculature has been highlighted by many studies showing the feedforward contraction of the trunk muscles in anticipation of extremity movement. Those trunk muscles comprise the transversus abdominis (TrA) and lumbar multifidus (LM). Since those muscles create no movement, but rather an increase in abdominal pressure, it is thus often a complex contraction to teach. To help with that teaching, physical therapist may use one of the two feedback tools that exist to teach a TrA contraction, namely the pressure biofeedback unit (PBU) and rehabilitative ultrasound imaging (RUSI). Unfortunately, their efficiency to help with the teaching of TrA contraction is yet to be shown in an elderly population, and few studies compared their efficiency. Furthermore, even though the teaching of the TrA contraction must be followed by a translation of that skill in more functional position such as standing, no studies looked at the effect of the feedback in supine to the skill of TrA contraction in standing.

Objective: The principal aim of this study is thus to compare, throughout a healthy population of 60 to 80 years old, the immediate efficiency of adding PBU or RUSI to the TrA contraction teaching in supine. The secondary objectives are as follow: 1) to see if that teaching in supine can be translated by a better contraction of the TrA in standing, and if one of the two tools is superior in doing so and 2) to see if one tool favor a more specific contraction of TrA when compared to the other muscles of the lateral abdominal wall. The hypothesis is that the RUSI will prove superior to the PBU.

Method: This will be a single-blinded controlled laboratory study with randomization. The independent variables will be the group of randomization, either PBU or RUSI. The dependent variable will be the TrA contraction. To answer our objectives, forty (40) healthy people aged between 60 and 80 will be recruited. The subjects will be healthy with no current pain and no history of limiting LBP in the past year. Of those, 20 will be women. The randomization will be made in blocks to allow a good balance of the sex throughout the groups. Every subject's TrA contraction (principal outcome) will be measured before and after the intervention (immediate effect) with the RUSI (change of thickness between resting and contracting state) both in supine and standing. The intervention will consist of a 5-minute education on the trunk muscle and their role, a brief teaching of how to contract them properly and the use of the feedback tool used in their group. Will follow a training program of 15 contractions with the feedback tool (PBU vs RUSI). Descriptive analysis will be used to describe the subjects. The TRA activation ratio (AR) and preferential activation ratio (PAR) will be compared after the intervention with Wilcoxon signed rank tests for each subject first. The group's mean will then be compared, first for each age group, then for each feedback tool as a whole with the Wilcoxon rank sum test.

Impact: Age and LBP both influence the motor control and feedforward contraction of the trunk muscles. Since NSLBP is highly prevalent throughout the ages and cost a lot for the society, the identification of the best tool to help in teaching the TrA contraction is crucial. This project will provide the first steps to justify a bigger controlled study on NSLBP. Moreover, this project will familiarize the Canadian physical therapist society to the use of RUSI, a tool still seldom used in the clinic in Canada but highly promising.

1. Introduction Low back pain is defined as a pain in the lower back bad enough to limit your usual activities or change your daily routine for more than 1 day, that may or may not go down one or both legs and lasting at least a day. The pain must be comprised on the back, within the limits of the lower rib and the lower gluteal fold. Low back pain (LBP) can further be compartmentalized in "specific LBP", by definition a LBP that be attributable to a known cause, or in "non-specific LBP" (NSLBP), defined as a LBP that cannot be associated to a known cause. LBP is one of the leading causes of functional limitations and sick leave, carrying a heavy financial cost for both individuals and society. According the 2010 Global burden Disease study, LBP is classified the 6th in the overall burden when compared to 290 other pathologies. Global prevalence of LBP is between 11% and 84%, and would reach its peak value à 80 years old. Therefore, the impact of LBP would be more important in societies where the life expectancy is higher, such as Canada and the United States. Moreover, aging would be accompanied by an increase in the frequency of severe back pain, as well as an increase in the prevalence of pain and other symptoms and conditions associated with low back pain, occasioning higher social and functional limitations. It also is one of the main causes of physical therapy consultation, which uses therapeutic exercises, such as core muscle training to address it. Yet, it is unknown as today which method is to be privileged to teach how to contract those muscles in the elderly.
2. Literature review One of the favoured approaches in physical therapy to address LBP is the training of the core muscles, specifically the transversus abdominis (TrA). In a healthy subject, trunk muscles will maintain trunk stability by making a sequence of postural adjustment in advance of distal movement to prevent any loss of balance. This fine tuning adjusting happening before any kinesthetic or sensorial feedback may happen, it is considered as planned by the central nervous system (CNS). Yet, it has been shown that those postural adjustments are lacking in the patients with LBP, and would persist even after the resolution of the LBP episode, which might contribute to the recurrence of LBP. Furthermore, aging is also accompanied by a diminution of motor control of the trunk, constituting a part of the LBP in the elderly. Fortunately, these postural adjustments of the TrA have been shown to be trainable again.

Teaching to contract the TrA often gets to be quite a challenge, since its contraction creates no visible movement. It requires an understanding and/or a good feeling of the desired contraction from the patient. To help in the teaching of the TrA, physical therapists (PT) may use one of the few feedback tools that have been shown to be efficient in that regard. Biofeedback is a technique where the biological information otherwise unknown to the patient is transmitted in real time to him. That information goes over what is usually accessible for those patients. In physical therapy, biofeedback comes from the biomechanical measure of physiological measure of the human body. Adding biofeedback to a usual physical therapy intervention enables the patient to control physical adjustments that were thought to be controlled by the CNS, such as the postural adjustments. It also allows for a better tuning in the tasks and engages the patient in a more proactive therapy. In the literature, two non-invasive biofeedback tools used in clinics are documented for their efficiency to assist in the teaching of the TrA contraction and for their validity to measure its activation, the rehabilitative ultrasound imaging (RUSI) and the pressure biofeedback unit (PBU).

The use of RUSI by physiotherapists is growing. RUSI is a non-invasive technique to provide immediate visual feedback on changes in the thickness of the abdominal wall muscles, reflecting the contraction of these muscles, more specifically the TrA. This is done by transmitting, on a screen, an ultrasound image produced by a probe placed on the abdomen of the participant. The probe emits ultrasound waves which passes through the tissues and then return to the probe in the form of echoes. RUSI would detect recruitment of TrA at a contraction as low as 12% of the maximum voluntary contraction (MVC), and recruitment of the internal oblique (OI) when contracted to 22% a MVC. This visual feedback, coupled with the usual explanations for TrA recruitment, reduces the number of trials required for an effective recruitment of TrA in patients with and without low back pain immediately after training and in the short-term follow-up.

The PBU is a simple and inexpensive device that measures the change in pressure in a three-compartment bag filled with air placed under the patient's lumbar spine whilst lying on his back (supine position). The movement of the lumbar spine on the three-compartment bag produced by a false contraction of the TrA changes the intensity of the pressure and this change can be seen on an adjustable gauge, and therefore avoided. However, Lima et al. (2012) have shown a low association between the pressure measured with the PBU and the surface electromyography (EMG) (r = 0.2, p <0.2), with a sensitivity and a specificity of 60%. A systematic review by the same author concludes that inter- and intra-evaluative reliability ranges from medium to good (intra-class correlation (ICC) of 0.42 to 0.82 and 0.5 to 0.81, respectively) and a construct validity ranging from medium to excellent (ICC from 0.48 to 0.90). Nevertheless, it is a widely used apparatus for learning TRA in physiotherapy because of its low cost and ease of use.

Only three studies compared the efficacy of RUSI and PBU on the recruitment of TrA in both young and asymptomatic individuals and low back pain. Bajaj et al. (2010) report that low back pain patients trained with RUSI acquire more rapidly the ability to recruit TrA (significant reduction in the number of trials required to activate the TrA and the number of days needed to consolidate learning) than those performed with the PBU. Lee et al. (2016) randomized a sample of young and asymptomatic individuals into three groups according to the type of feedback used during training: tactile, RUSI or PBU. The primary measurement was the thickness of the TrA measured with an ultrasound. An increase in TrA thickness (reflecting a better TrA activation) was observed in all three groups. When compared to the tactile feedback group, the RUSI group had significantly higher TrA thickness. Finally, Wooldridge et al. (2014) compared the immediate effect of training with different types of feedback (PBU, RUSI and PBU + RUSI) in a population similar to that of Lee et al. (mean age = 22.2 years). The authors report no significant difference in the activation of TrA, measured with different parameters extracted from ultrasound images, pre and post-training and between groups.

In summary, although two studies show that the recruitment of the TRA would be more effective following training with the RUSI, the heterogeneity of interventions and outcome measures means that the current level of evidence does not allow a decision on its superiority in young people. Moreover, the results of these studies cannot be generalized to an elderly population. Aging can disrupt the anticipated postural adjustments, and an effective modality of training in young people may not be conclusive for the elderly. Finally, no study has verified whether the training of recruitment of the TrA, which is generally carried out in supine position, allows the transfer of the learnt skills to other more functional positions, such as the standing position, for either of the tools.

Aim The main goal of this study is to compare the immediate (pre-post laboratory) effect of a TRA recruitment training with RUSI to that of training with the PBU on the thickness of the TrA, an indirect measure of recruitment measured in the supine position (SP) in a healthy elderly population. Secondary objectives are (1) to determine whether one of the feedback tools is more effective in facilitating preferential recruitment of TrA versus oblique muscles and (2) to determine whether one of the feedback tools is more effective in transferring the recruiting capacity of the TrA from the SP to the standing position. The main assumption is that in SP, the thickness of the TrA will be greater following the teaching with RUSI compared to teaching with the PBU. The secondary hypotheses are 1) the RUSI will allow a preferential recruitment of the TrA; and 2) the thickness of the TrA in the standing position will be greater following the teaching with the RUSI.

Methods Ideally, the effectiveness of an intervention is assessed by a randomized controlled trial (RCT). However, the prerequisite for an RCT are as follows: 1) plausibility of previously demonstrated effect; 2) demonstration that the research procedures are well established in real situations and 3) the possibility of recruiting a large number of participants. The plausibility of effects is still verified in a new population with valid and reliable instruments. The investigators propose an experimental "before-after" laboratory study to document the immediate effects. The independent variable will be the type of feedback used (RUSI vs PBU) during a TrA recruitment training session, and the dependent variable will be TrA thickness, an indirect measure of TrA recruitment.

Participants:

The investigators will recruit 40 people aged between 60 and 80 according to a non-probabilistic sampling method from the Nabù recruitment bank of the Center for Aging Research (CdRV) and a non-probabilistic sampling method by convenience. These participants will be randomized to two intervention groups: RUSI or PBU. The randomization will be stratified by gender (in equal proportion) using sealed envelopes and of a random number table. Participants must be between the ages of 60 and 80 years of age and be in good health (no chronic conditions restricting activities or requiring punctual care) without back pain for more than 12 months and be free from pain since less 1 week. They should be unable to contract the TrA specifically (i.e. without contraction of the oblique internal (OI) or oblique external (OE) muscles in response to a simple instruction). Those able to contract their TrA specifically will be excluded, and statistics will be kept to describe that sample. The participants will have to live in the community or in semi-autonomous residence, and will be excluded participants with a history of spinal, abdominal or lower extremity surgery, neurological or respiratory problem, structural scoliosis, body mass index greater than 30, pacemaker, lower Mini Mental State Examination (MMSE) to 24 or previous TrA training experience. The recruitment procedures will go on until the full sample size is reached with admissible participants.

Sample size :

The sample size (n=40) was calculated to respect an effect size of 0.7 with an alpha error of 5%.

Procedure for data collection:

Those interested in participating will contact the Project Coordinator to verify their eligibility and schedule a meeting. Upon arrival at the Research Center of the Centre hospitalier universitaire de Sherbrooke (CRCHUS), a member of the research team will welcome them and verify their ability to recruit the TrA to confirm their eligibility to participate. They will be asked to contract the TrA following a standardized verbal instruction and the presence or absence of contraction (i.e. movement of muscle fibers) will be checked with ultrasound. Individuals who will be excluded because of their ability to contract the TRA will receive, from a physiotherapist, the advice that is provided in current practice to improve their performance in contracting the TRA. The consent form will be presented and explained to those who will meet this last inclusion criterion. Once their free and informed consent is obtained, participants will be randomized to the two intervention groups. They will provide their age and they will be measured and weighed. This data will be used to describe the sample. They will also be asked to complete a questionnaire to diagnose the presence of urinary incontinence (QUID), which will be used as a covariance in the analyzes.

Interventions (independent variables):

RUSI Group: Participants will be taught about the anatomy of abdominal wall muscles and the general functioning of RUSI. Then, they will be invited to take the SP position, knees flexed. The feedback will consist of an ultrasound image transmitted to brightness (B)-mode. In order to obtain an image of the TrA, the probe will be placed halfway between the 12th rib and the iliac crest. The probe will be moved laterally to have the rim of the thoraco-lumbar fascia at the right edge of the screen, so that the 3 muscles of the abdominal wall are clearly seen, i.e. the oblique external (OE), oblique (OI) and TrA, according to the method described by Teyhen et al. ( 2005). Participants will receive the necessary education to understand the image and the desired recruitment according to a standardized protocol. They will then be asked to perform 15 contractions held 10 seconds of the TrA with a visible thickening of the TrA on the RUSI monitor exceeding the measurement at rest (visible on the monitor). One-minute rest will be allowed between each contraction to avoid fatigue. Participants will have to watch the RUSI monitor during the 15 tests. These parameters were determined from previous studies.

PBU Group: Participants will receive the same instruction on the anatomy of abdominal wall muscles and general information about the PBU device. They will be invited to take the same position as for the RUSI group (SP, knees flexed). The feedback will be to visualize the intensity of pressure on the dial of a STABILIZER PBU. The PBU will be placed under the lumbar spine of the participants and inflated to 40 mmHg, as recommended by the manufacturer. Participants will receive the education necessary to understand the tool and the desired recruitment according to a standardized protocol. They will then be asked to perform 15 contractions of the TrA with a duration of 10 seconds while maintaining the constant pressure at 40 mmHg ± 2 mmHg. A minute of rest will be allocated between each contraction. Participants will have to watch the PBU dial during the 15 tests.

For both groups, interventions will be given by the same physiotherapist (PT treating) experienced with the use of both tools, and the duration of the lessons (anatomy and general information) will be the same.

Dependent variable and measuring tools:

The dependent variable is the change in thickness of the TrA muscle between the rest state and the recruitment state measured from an ultrasound image (LOGIQ E, GE Healthcare, 13 Mega Hertz (MHz) linear probe, at B-mode). Before and after the intervention, three images will be taken in the SP, knee-flexed (main objective) position at rest and during a TrA recruitment maneuver carried out according to the following instruction: "inhale then exhale, then bring your navel back to your vertebral column ". Three other images will be recorded in standing position (secondary objective 1), both at rest and when recruiting the TrA according to the same instruction. During these measurements, the ultrasound device will be arranged so that the participants cannot see the screen. The image capture will be taken with the probe in transverse plan and positioned midway between the 12th rib and the iliac crest. It will be moved laterally to have the rim of the thoraco-lumbar fascia at the right edge of the screen, allowing a clear image of the 3 muscles of the abdominal wall (i.e. OE, OI and the TrA). The angle on the skin will be adjusted to have the best image quality by keeping the same position perpendicular to the skin. The use of anatomical markers rather than superficial markers on the skin will improve reliability and reproduce clinical reality. Resting and activation measures will be taken at the end of the expiration to standardize the effect of breathing. This method has been demonstrated to be valid and reliable for indirectly measuring the activation of abdominal wall muscles in asymptomatic individuals. This measure has also been validated in an elderly population . In addition to the change in thickness of the TrA (main variable), other parameters will be used for secondary analysis purposes, such as the activation ratio (AR) and preferential activation ratio (PAR). Image recording and the intervention will be performed by a physiotherapist (PT measurement) with this expertise. She will be blinded the participant's group.

Analysis :

Descriptive analysis will be used to describe the sample. The investigators will compare the change in thickness of the pre and post-intervention TrA using Wilcoxon's signed row tests for dependent groups. Then, the averages for each group will be compared (RUSI vs PBU, young and elderly) using tests of the sum of the Wilcoxon ranks for independent groups. Analyzes of covariances (ANCOVA) will be used to study the effect of the presence of urinary incontinence (categorical variables) on changes in TrA thickness. In the event of missing or invalid data, a sensibility analysis will be made post hoc. The analysis will be made using the Statistical package for the social sciences (SPSS v.25).

5. Potential Bias Participants and the attending physiotherapist will not be blinded for the intervention, but the physiotherapist responsible for the analysis and image capture will be unaware of the groups of randomization. The fact that the main variable measurement tool (TrA thickness) is also used as a feedback tool for one of the groups could induce an intervention bias. Note, however, that for the measurement of the main variable, no participant will see the monitor; they will therefore have no visual feedback. However, contact of the probe on the abdomen during ultrasound measurement will provide a tactile stimulation similar to that which the RUSI group was exposed during surgery. It is possible that these favours the realization of the task by the participants of this group during the measurement of the main variable. However, the investigators believe that this effect will have a trivial impact. The limitations of the other measurement tools would further compromise the validity or feasibility of our study (eg noise in the surface EMG or implanted EMG which is an invasive procedure).

6. Clinical Benefits Age and NSLBP have been shown to affect motor control of trunk stabilizing muscles. Knowledge is limited and non-consensual for the young population and absent for the elderly. It is therefore the first study to focus specifically on the training of the TrA in a population of elderly people. This research project will allow us to realize the first steps (plausibility of the effect of RUSI and PBU in healthy elderly people, tuning of the procedures, testing of the recruitment and data collection procedures) to justify a larger RCT comparing the two feedback tools to promote recruitment of the TrA, reduce pain and improve function in a population suffering from NSLBP. Due to the high prevalence of NSLBP in the general population and particularly in the elderly, it is important to identify the most effective tool to optimize rehabilitation (better efficiency and less time).

Inclusion Criteria:

• Good health
• without back pain for more than 12 month
• free from pain for 1 week
• live in community,

Exclusion Criteria:

• No chronic health condition restricting activities or requiring punctual care
• a capacity to specifically contract the TrA
• surgery to the thorax, abdomen or spine in the last 3 years
• pacemaker
• BMI>30
• previous training of TrA
• structural scoliosis
• lower than 24 on the MMSE

No current study undertaken in the principal investigator's lab would require such data; this being said, this study could be the preliminary of a bigger, broader study comparing both tools in low back (LBP) pain participants, and if this happens, comparison between raw data of healthy and LBP participants might be of interest.

A single-blind study where the assessor will be taking measurement before and after the intervention, albeit without knowing which intervention was given.