Chronic aspecific chronic low back pain (CALBP) is defined as pain and/or discomfort localised below the costal margin and above the gluteal folds with possible posterior thigh irradiations not extending below the knee; symptoms have to be present for over 3 months or longer than the expected normal healing time ¹ . CALBP is one of the most common and costly syndromes of modern times, with a lifetime prevalence estimated at 70% and reaching 80% in Europe ² . Negative impacts of CALBP include for the patient: pain, reduction in activities of daily living, reduced work productivity and/or work absence; for society: increased contacts with health care providers, high demands of medical investigations and related treatments ³ . Chronicity occurs in 5% of patients suffering from low back pain (LBP) but generates up to 80% of the total costs related to this disorder ⁴ .

The multifactorial aetiology of LBP creating pain and functional limitations has been investigated with particular emphasis on physiopathological mechanisms, neuropsychosocial factors and motor control alterations. Several regional, national and international guidelines on LBP treatment exist ^{5– 8} but the levels of evidence supporting these are not always optimal, and a strong recommendation for a specific physiotherapeutic approach is currently lacking. Moreover recent studies have demonstrated an association between thickness and disorganization of the connective tissue (fascia) and CALBP ^{9– 11} but no guidelines could be found regarding the recommendations of a therapeutic approach focused on fascial tissues. For these reasons, it was decided to conduct an experimental study comparing the effectiveness of Fascial Manipulation ^® (FM) versus the physiotherapeutic recommendations provided by guidelines (standard manual therapy) (MT).

The null hypothesis was that there would be no difference in short and medium term outcomes between patients undergoing standard physiotherapy only and those where FM was added to standard care.

Thirty-two patients were recruited from April to December 2009, 24 completed the trial, 6 participants were not enrolled for organizational reasons, 2 withdrew for reasons not directly related to treatment (one had a sport injury, one had a pulmonary infection). From the 24 participants who completed the trial 2 were lost at follow-up (T16, T17). After randomization the SG was composed of 11 patients (4 males, 7 females), the CG was composed of 13 patients (4 males, 9 females). At baseline the VAS was significantly higher in the SG, all other outcomes did not reach statistical significance (see Table 1).

Figure 1a–d shows the results of the measured outcomes at T0, T15, T16 and T17. All outcomes showed improvement for both groups. It can be observed that the improvement is maintained in the medium term (at one and three month follow up) for both groups, however results in the SG were significantly better compared to the CG as seen by the inclination of the trend lines.

Table 2 shows the means, standard deviations, and statistical significance of each outcome for each group. VAS and BPI reached statistical significance for the SG. The outcome of the RMDQ did not reach statistical significance; however the percentage improvement reached significance when measured between T0-T15, T0-T16 a positive trend, not reaching significance, was observed between T0-17 for the group following FM. Table 3 shows the MCID along with the percentage of patients respecting its value for the VAS, RMDQ, BPI for T0-T15, T0-T16, T0-T17. It can be observed that both groups show clinical improvement. Strong clinical significance is reached in the SG for the VAS. In the same group both BPI and RMDQ reached higher clinical significance compared to patients following the program of MT.

The values and trend lines of the VAS prior to each treatment are seen in Figure 2. Patients of the SG had a steeper trend, showing higher reduction of pain after each FM treatment. The drop in VAS was statistically significant after the first FM treatment (z=0.0239) for the SG, and reaching significance at the 8 ^th manual therapy treatment (z=0.0405) for the CG. The mean difference in VAS value pre- and post-treatment was 2.9 during the first and 1.6 during the third session ( Figure 3). The CG did not reach clinically significant improvement in VAS during treatment sessions. Table 4 shows the mean variation in VAS for the SG was 1.7 for FM treatment and 0.69 for MT treatment. The mean variation in VAS for the CG was 0.96. The trend is more linear for the CG (0.46–1.38 variation) and is steeper for the SG (0.11–2.92 variation).

FM treatment sessions, associated to four treatments of MT over 4 weeks in chronic aspecific low back pain, resulted in significant statistical and clinical improvement compared to eight treatments of MT alone at end of therapeutic intervention, at 1 month and at 3 months follow up. Therefore the null hypothesis was rejected.

This is the first randomized trial evaluating the effectiveness of FM for CALBP. The results observed in the MT group are similar to those reported in the literature ^{24, 25} .

With regards to methodology it was decided to apply the exclusion criteria inherent to diagnostic criteria proposed in the specific guidelines for low back pain. These recommend a diagnostic triage of patients to exclude specific spinal pathologies or pain caused by compression on nerve roots. The outcome measure of pain was evaluated with the VAS and the BPI. The BPI was chosen as outcome measure because it is a complete and multidimensional test that evaluates the sensorial, qualitative and emotional aspects of pain. It initially started as a scale for oncological patients but its validity has also been proven in patients with chronic pain of non-neoplastic origin ¹⁷ . The SF-36, even though it is not specific for low back pain, was chosen as an outcome for perceived well-being.

Both groups were homogeneous at baseline except for the VAS value. Both followed a course of eight treatment sessions performed twice a week over 4 weeks. In both groups the therapists had the same level of education. Improvement in all outcomes was observed for both groups with the SG showing greater statistically and clinically significant improvements. The initial VAS value was higher for the SG which may have skewed the results towards a better improvement for the SG. However it has to be noted that both the VAS and the BPI reached statistical significance at all measurement times for the SG. This indicates that the FM treatment also created a significant improvement in another outcome that was similar at baseline. Therefore it may be considered that the VAS improvement reported in the SG is not solely related to the higher VAS at baseline.

The VAS values pre- and post- each treatment allow us to evaluate which intervention (MT or FM) showed the best improvement in the SG. It can be observed that the values post-FM treatment showed greater diminution compared to the CG and compared to the MT treatment performed in the SG. Indeed the first and third treatments in the SG showed a greater difference pre- and post-treatment. It can be observed that the first FM treatment lead to a statistically significant decrease in VAS compared to the CG. Such results can support the rationality of the FM method that, restoring the sliding of the fascia proximally and distally from the area of the pain, is able to recover the range of motion, decrease the stiffness and painful sensation perceived by the patient. With regards to the VAS improvement in the SG, it may be argued that, at the 8 ^th treatment, statistically significant improvement would be more difficult to reach with a VAS value pre-treatment of 0.4, compared to 1.9 in the CG.

The perceived state of well-being showed significant improvement at the end of therapeutic intervention, in the FM and MT group, compared to the MT group only. At follow-up the values did not reach statistical significance, but showed a better trend for the SG. Functionality (of the RMDQ scale) reached statistical significance at the 1 month follow-up.

Clinical significance showed higher mean of improvement for all outcomes for the SG. Also the percentage of patients having higher values of MCID was higher than the CG and reaching 100% for BPI and RMDQ at T16 and T17. This indicates that FM treatment, when added to MT treatment, leads to clinically significant improvement in the short and medium term with regards to severity of pain and disability.

The rationale for including FM treatment for patients suffering from CALBP is based on the fact that fascia is a type of dense connective tissue on which about 30% of muscle fibres have either insertions or origins ²⁶ . Fascia is therefore tensioned during any kind muscular activity in any direction of movement; it transmits tensions towards the perimysium and towards synergic muscular groups ^{27, 28} . The deep fascia is composed of layers of dense connective tissue that are dedicated to transmitting the load and loose connective tissue that allows the gliding of the collagen layers. Different authors have demonstrated the rich innervation of the fascial tissue and in particular of the thoracolumbar fascia ^{29– 32} . Langevin ^{9, 33} has demonstrated an increase of the thickness of the thoraco-lumbar fascia and a decrease of the gliding of the different layers. It is hypothesised that the layers of loose connective tissue are the ones causing the increase in thickness because fibrosis is not recognized in subject with CALBP. Recent studies have demonstrated that the major component of the loose connective tissue, the hyaluronan, by increasing concentration and/or size, begins to entangle into complex arrays, leading to a decrease in the gliding properties of the fascia ³⁴ . Due to the viscoelastic properties of the loose connective tissue, allowing it to modulate the dynamic response of the mechanoreceptor, we hypothesize that a stiff thoracolumbar fascia can alter the receptor afferents ^{35, 36} providing another mechanism to the multifactorial aetiology of chronic back pain ³⁷ .

The FM treatment utilises the non-Newtonian properties of the hyaluronan, by applying continuous shear to the fascial tissue, to increase the temperature of the fascial tissue. This leads to the destruction of the van der Waals and hydrophobic forces that hold the hyaluronan chains together. A temperature of 40°C is able to decrease the viscosity of the hyaluronan ³⁸ and increases its ability to glide ³⁹ . A less viscous loose connective tissue allows fibroblasts to perceive the lines of tensions of the fascial layers and may therefore lead to the remodelling of their dense connective tissue with a deposition of collagen fibres in the correct lines of force ⁴⁰ .

This study shows that the implementation of FM, within a course of MT following established guidelines, reaches statistically and clinically significant improvement in the outcomes of pain, function and quality of life in patients suffering CALBP both at the end of care as at one and three months follow-up compared to MT alone. Considering the costs of CALBP to health care systems it may be considered to implement FM treatment as part of routine care in existing physiotherapy programs. Considering the significant decrease in pain especially following the first two FM treatment sessions, it may be hypothesized that FM may reduce the number of treatments required for patients with CALBP and therefore reduced the overall costs to health care systems and patients.

The data referenced by this article are under copyright with the following copyright statement: Copyright: � 2016 Branchini M et al.

Data associated with the article are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication).

F1000Research: Dataset 1. Experimental dataset from Branchini et al., ‘Fascial Manipulation ^® for chronic aspecific low back pain: a single blinded randomized controlled trial’, 10.5256/f1000research.6890.d100559 ⁴¹

Written informed consent for publication of clinical details was obtained from the patients.