Summary: Low back pain is the leading cause of deficiency and loss of productivity worldwide. No evidence of any particular exercise was more effective than another for treating nonspecific low back pain. Objective: To evaluate the efficacy of two resistance training protocols, with different techniques for performing lower limbs exercises, in improving vertebral posture and reducing symptoms of low back pain. Methods: Randomized parallel clinical trial with two arms: Restricted Group (GR) performed all squat and Stiff exercises with neutral vertebral posture and the Complete Group (CG) performed the same exercises prioritizing the complete range of motion. Both groups had a 12-week intervention with 36 resistance training sessions. This study was conducted between November 2020 and April 2021 in Goiás (Brazil). Thirty-two participants aged 18 to 69 years with nonspecific low back pain were recruited in the extension project of the Faculty of Physical Education and Dance of the Federal University of Goiás (UFG), at the Hospital das Clínicas - UFG and at the Campos Samambaia Health Center. To ensure blindness, participants did not know why the technique of movement between them was different. The movement technique was monitored by one teacher per participant throughout the training and cannot be altered by participants at risk of compromising the results. Spinal posture was evaluated by three-dimensional reconstruction and posture quantification using dynamic posture software and pain symptoms were evaluated by the Brief Pain Inventory and Rolland Morris Questionnaire. Statistical analysis was performed in the Software SPSS and MATLAB. The Shapiro-Wilk and Bartlett tests were used to confirm the normal distribution and similar variances in the distribution of the data. The other quantitative and qualitative variables were analyzed by nonparametric statistical methods. Quantitative data with normal distribution were reported by means of means and standard deviation, minimum and maximum values, and the other data by median, interquartile range, minimum and maximum values. The pre-intervention conditions of the groups were compared by independent t-test. Two-way ANOVAs (groups X time) were used for group intervention effect comparisons for quantitative data variables with normal distribution. Significance level of 5%. The size of the effect of the results will be calculated using cohen's test.

Back Squat Exercise for Low Back Pain: Avoiding Butt Wink vs Full Range Clinical Trial.

Background The deterioration of muscle strength of the back was the most important factor to decrease the range of motion of the spine and were related to the worsening of quality of life. Back pain is the leading cause of deficiency and loss of productivity worldwide. Resistance training has become increasingly a great ally in the benefit of health and especially in the functional capacity of the human being, and reduction of injuries.

Squat is one of the most popular exercises in resistance training, widely used to achieve health, rehabilitation, performance and aesthetic goals.

Vertebral posture has been much questioned in the scientific literature whether or not it may undergo changes in its curvature associated with high overloads during the execution of the squat, believing that this may cause or increase back pain. Further studies are needed before abandoning the current standards of the squat technique that defend a neutral position of the lumbar spine during a deep squat.

OWEN et al. (2020), concluded that no evidence of any particular form of exercise was more effective than others for treating back pain. Therefore, the aim of the present study is to evaluate the efficacy of two resistance training protocols, with different techniques for performing squat exercise, in improving vertebral posture and reducing low back pain.

Methodology

Participants, Randomization and Blindness

This study is a randomized clinical trial involving 32 participants aged 18 to 69 years, all with symptoms of low back pain, controlled by participant, in the effectiveness of the movement technique with a teacher for each exercise.

The participants were recruited in the Faculty of Physical Education and Dance, Federal University of Goiás (UFG), at the Clinical Hospital and Samambaia Campus Health Center.

Research participants must watched 85% of the classes during the 12 weeks of exercise/intervention treatment.

For randomization, they were distributed in a ratio of 1:1, in order of the registration number and parallel intervention. The intervention groups were: 1) Restricted Group (RG) and 2) Complete Group (CG). All perform the same exercises, the only difference between the groups was the technique of movement of the Squats and Stiff.

To ensure blindness, participants not knew why the technique of movement between them was different. The movement technique was monitored with one teacher per participant throughout the training and cannot be altered by participants at risk of compromising the results.

For the participants of the research, a first meeting was scheduled to present and clarify the research and schedule the evaluations.

Ethical procedures All participants signed the Free and Informed Consent Form in two ways according to the Ethics Committee on Human and Animal Research of UFG, opinion number 2,458,324.

Evaluation of vertebral posture with videogrammetry

Retroreflective stickers (flat, rectangular [12x8 mm]) were positioned on the back of the participants to identify anatomical accidents (Figure 3). Markers were positioned at the point of intersection between the medial edge and the spine of the scapula (Scapula Trigone), on the left and right side; in the lower angle of the left and right scapula (SI) ; in the posterior superior iliac spine (PSIS); in the spinous processes of the second sacral vertebra (S2), fourth lumbar vertebra (L4), twelfth, sixth and first thoracic vertebrae (T12, T6 and T1). In addition to these, pairs of markers, which were used as reference points during the analysis, were positioned bilaterally and at the time of the spinous processes of L4, T12, T6 and T1, following the alignment of the PSIS. After identifying these points, the line defined by the spinous processes of the vertebrae was filled with markers regularly positioned approximately every 2.5 cm.

To record the back of these volunteers, three OptiTrack Flex 13 cameras with 100Hz acquisition frequency were used and were regulated before postural recording. The measurement of the geometric curvature of the spine was performed with the method described in Campos et al., 2015). The method consists of automatic tracking and three-dimensional reconstruction of retroreflective markers with video cameras. The movement of markers on the back of volunteers is tracked and analyzed.

The processing of images to measure the spatial position of the markers was performed in the Dynamic Posture software (CAMPOS, 2010) developed in Matlab® (The MathWorks, Natick, Massachusetts,USA). In each frame of each video collected, for all cameras, the bidimensionais coordinates of the centroid of the markers were calculated from their respective barycenter (GRUEN, 1997). The calibration of the system was performed through the point s register with a known location, which allowed the three-dimensional reconstruction using the method of linear transformation direct (ABDEL-AZIZ; KARARA, 1971). The system global reference of the laboratory was defined as: vertical axis Z (para up), horizontal back-row axis X(forward) and Y horizontal lateral axis (to left).

During locomotion the trunk presents a regular oscillatory behavior along the vertical axis, with one cycle per step, reaching a maximum peak during each simple support phase and a minimum peak during the double support phases of the lower limbs. Thus, the beginning and end of each step were defined at the moments when the minimum vertical oscillation peaks of the column markers occurred (we calculated the average of the Z coordinate of all column markers). Thus, every two minimum peaks we had a complete stride. Through a function based on finite differences, all minimum peaks of vertical trunk oscillation were identified and consequently passed. We emphasize that we did not claim to identify the events and phases of each pass with this procedure. It was only identified the stride as a whole.

After three-dimensional reconstruction, for locomotion, the data were smoothed with a spline filter adjusted so that the residues were below 1 mm. Each stride cycle was normalized at 101 points in time, representing positions from 0 to 100% of the complete movement cycle. In the analysis of the locomotion of each participant, an average cycle of 10 stride cycles was used, called in this standard cycle work of the stride. For static posture analysis, each marker had its position represented by the average three-dimensional position in a stretch of one second (100 frames).

At each moment in which the posture was recorded, the 3D s of all spinal markers between S2 and T1 were described in a local reference system in the trunk (CAMPOS et al., 2015) with or from T12. The vector with origin in L4 and end in T6 defined the orientation of the longitudinal axis z (upward). An auxiliary vector y' was defined with origin at the midpoint of the reference points to the right of L4 and T6 and with an end at the midpoint of the reference points to the left of L4 and T6. The vector product between y' and z defined the local sagittal axis of trunk x (forward) and the vector product between z and x defined the local transverse axis of trunk y (left).

The method proposed in Brenzikofer et al. (2000), is adopted to measure the geometric shape of the column. By means of the minimum squares method, a polynomial adjustment was made on the position of the markers projected in the sagittal and frontal plane. The lower and upper extremities of the polynomial curve were discarded in the analyses because they could present in very robust adjustments. Eighth-degree polynomials defined by the X²red, chi-square-reduced test (VUOLO, 1992) were used. The geometric curvature of the column vertebral, K(z), was calculated (Figure 4 - left; Figure 3) with the first and second derivatives, P'(z)and P"(z),by means of equation (1): K(z)= P"(z) / [1 + P'(z)2 ]3/2.

Geometric curvature can be interpreted as being the inverse of the radius of the circumference that adjusts and touches the curve at each height of the longitudinal axis of the column. The unit of measurement of geometric curvature is "m-1". In the sagittal plane, v positive curvature allocated previous (kyphosis) and posterior negative (lordosis) concavities. In the frontal plane, positive values indicated left bending and negative for the right.

For the analysis of locomotion, both in the sagittal and frontal plane, the mean posture of the standard cycle of the stride was calculated, called the neutral curve (CAMPOS et al., 2015). We also analyzed the Oscillatory Component, defined by the difference between the postures presented at each instant of the standard stride cycle and the Neutral Curve.

In the analysis of static posture and neutral gait curve, para the sagittal plane, the peak of absolute curvature in the lumbar region (z < 0 cm) was identified as a representative variable of lumbar lordosis and the peak of absolute curvature in the thoracic region (z > 0 cm), as a representative variable of thoracic kyphosis. Also in the sagittal plane, the sagittal geometric curvature at the level of L4, T12 and T6 was analyzed. Frontal geometric curvature was analyzed at the level of L4, T12 and T6 as well as the absolute peak of lateral flexion (higher absolute value of curvature) in the lumbar and thoracic regions.

In the analysis of the Oscillatory Component, in the sagittal and frontal plane, for the lumbar and thoracic regions, the place where the greatest range of motion presented in the standard cycle was identified. The range of motion was analyzed.

The trunk inclination was calculated similarly to CAMPOS et al. (2015). The local longitudinal axis of the z trunk was projected in the sagittal plane of the laboratory (normal to Y) in such a way that 0º indicated that the trunk was completely vertical, and 90º indicated that the trunk was inclined to previously, completely horizontal. The local longitudinal axis z of the trunk was also projected in the frontal plane of the laboratory (normal to X) in such a way that 0º indicated that the trunk was completely vertical and 90º indicated that the trunk was tilted to the left, completely horizontal. The cross-sectional and local axis of the trunk was projected in the transverse plane of the laboratory (normal to Z) in such a way that 0º indicated that the trunk was completely aligned with the Y axis and positive values indicated that the trunk was rotated to the left and negative to the right.

Six angular variables were calculated for the two regions of the spine studied (lumbar and thoracic). For the calculation of lumbar angles (Figure 3), a local system was constructed in this region. The vector with origin in S2 and end in T12 defined the orientation of the lumbar longitudinal axis (upward). A lumbar auxiliary vector was defined with origin in the right superior superior iliac spine and extremity in the left upper superior iliac spine. The vector product between the auxiliary vector and the longitudinal lumbar axis defined the lumbar sagittal axis. The vector product between the longitudinal lumbar axis and the sagittal lumbar axis defined the transverse lumbar axis. The lower lumbar segment was defined as the straight segment starting at S2 and end in L4. The upper lumbar segment was defined as the straight segment with beginning at L4 and end in T12. The upper transverse lumbar segment was defined as the straight segment starting at the bilateral point on the right side of T12 and the lower transverse lumbar segment was defined as the straight segment with beginning in the right upper inferior iliac spine and extremity in the upper superior iliac spine. The lower and upper lumbar segments were projected in the sagittal and frontal planes, defining respectively the sagittal lumbar angle(α)and frontal lumbar angle(β)(Figure 3a and 3b). The transverse lumbar angle (θ) was defined by the projection ofthe transverse lumbar segments (Figure 3c).

For the thoracic region, the same procedures used for the lumbar region angles were adopted, replacing in the formulas S2, L4, T12 and the respective bilateral scans with markers of T12, T6, T1 and the respective bilateral ones. Thus, theSagittal Thoracic Angle, TheFrontal Thoracic Angle andTransverse Thoracic Angle were calculated.

Interventions

The participants were frequently trained three times a week, being 2x for lower limbs (Smith Machine Squat, Free Squat, Unilateral Squat and Deadlift with Rigid Legs, on the 3rd and 6th and 1x for upper limbs, (Lat Pulldown, Low Rowing Machine, Inverted Crucifix with Dumbbells, Complete and Inverse Abdominal, Supine with Bar and Supine with Bar).

The selected exercises were the same for the two protocols, but what differentiated one group from the other was the technique of movement of squat and ground lifting exercises with rigid legs. The evaluations were made in two moments, at baseline and at the end of the intervention, after 12 weeks.

In exercise, the ground survey with rigid legs was also controlled by neutral vertebral posture for the participants, and the CG prioritize or range of motion. Foot positioning was also controlled in the Stiff Legged Deadlift exercise in the two protocols respectively in the positions explained for the squat. For the CG the orientation was to go down as much as possible seeking the best posture I could without.

During the execution of both protocols, the knees exceed ed the foot line so that the positioning of the trunk was as straight as possible and did not provide the greatest range of motion of the knees.

The intensity of the exercises see you adjusting load according to the volitive tolerance for 10 to 12 maximum repetitions. The loads and repetitions achieved in each series of each exercise per training session of each participant were noted to control the evolution of the training load. The rest interval was only one minute timed between all series. The cadence of repetitions was 2 seconds in concentric contraction and 4 seconds in eccentric contraction. All 36 training sessions were followed from 14:30 to 16:30 with guidance and supervision of a teacher graduated in Physical Education and Physiotherapy, who perform so this research and one more teacher and three trainees.

Adhesive markers were fixed on the ground, which were measured with goniometer and measuring tape at the squat site, to control the positioning of the feet in the two protocols in all training sessions.

Inclusion Criteria:

Adults and older adults with back pain and able to practice regular physical exercise.

Exclusion Criteria:

• Recent myocardial infarction (less than three months)
• Recent (less than three months) or disabling stroke
• Congestive heart failure
• Chronic kidney failure
• Decompensated Diabetes
• Any other disease or limitation that could compromise the performance of physical exercise protocols.
• Participants who perform other resistance exercise activities

There is not a plan to make IPD available.