Objective: To determine the effectiveness of MDT management provided by trained therapists in relation to comparator interventions in terms of pain and disability in acute and chronic low back pain patients.

Study Design: Review with meta-analysis.

Methods:

Eligibility Criteria - English or French RCTs examining the effectiveness of MDT for pain and disability in low back pain; clinicians needed to have taken at least one McKenzie Institute course focusing on applying MDT to the lumbar spine. Studies were excluded if they duplicated data from other studies, other interventions were combined with MDT, were published in non peer-reviewed journals, and if no a priori classification was done prior to MDT treatment.
Search - six electronic databases were searched for relevant studies up to September 2017. References of identified studies and from the McKenzie Institute website were also manually searched.
Study Characteristics - when studies had a mix of acute and chronic patients, the average duration of symptoms dictated the group they fit into. Chronic LBP was defined as pain greater than 12 weeks, and acute less than 12 weeks.
Risk of Bias & Strength of Evidence - the methodological quality of included studies was rated on the PEDro scale. The GRADE approach was used to assess the quality of the body of evidence for each outcome (pain and disability). The quality of evidence was initially set at “high” and could be downgraded due to limitation of design (< 5 on PEDro), indirectness of evidence (differences in populations, interventions, outcome measures, indirect comparisons), inconsistency of results (heterogeneous effect estimates), imprecision (studies with < 400 participants).
Statistical Analysis - Analyses were done separately for acute and chronic LBP. The effectiveness of MDT compared to other interventions were examined using random-effects models with statistical significance set at P < 0.05. Standardized mean differences (SMD) and 95% confidence intervals were calculated for each analysis.

Results:
After conducting the search, 11 studies were ultimately included for meta-analysis: 4 were considered for acute LBP, and 7 were considered for chronic LBP.

Acute LBP: Primary Analysis – there was moderate quality evidence in 3 studies of no significant difference in pain between interventions (SMD = -0.45, 95% CI -0.99, 0.10). There was high quality evidence in 4 studies of no significant difference in disability between interventions (SMD = -0.07, 95% CI -0.34, 0.20).

Acute LBP: MDT vs manual therapy + exercise – there was moderate evidence in 2 studies of a significant difference in pain favouring MDT (SMD = -0.74, 95% CI -1.45, -0.03). There was moderate evidence in 3 studies of no significant difference in disability between interventions (SMD = -0.24, 95% CI -0.77, 0.28).

Acute LBP: MDT vs education – 1 study found a significant improvement in pain (p = 0.02) in MDT + first-line care compared to first-line care alone. There was high quality evidence in 2 studies of no significant difference in disability between interventions (SMD = -0.09, 95% CI -0.31, 0.14).

Chronic LBP: Primary Analysis – there was moderate evidence in 6 studies of a significant difference in pain in favour of MDT (SMD = -0.33, 95% CI -0.63, -0.03). There was high quality evidence in 7 studies of a significant difference in disability in favour of MDT (SMD = -0.28, 95% CI -0.44, -0.12).

Chronic LBP: MDT vs manual therapy + exercise – there was moderate evidence in 2 studies of no significant difference in pain between interventions (SMD = -0.26, 95% CI -0.73, 0.22). There was high quality evidence of no significant difference in disability between interventions (SMD = -0.11, 95% CI -0.29, 0.07).

Chronic LBP: MDT vs exercise – there was moderate evidence in 4 studies of no significant difference in pain between interventions (SMD = -0.38, 95% CI -0.82, 0.05). There was high quality evidence in 4 studies of a significant difference in disability favouring MDT (SMD = -0.45, 95% CI -0.64, -0.25).

Chronic LBP: MDT vs education – 1 study found no significant difference between interventions in pain or disability 3 months after starting treatment.

Chronic LBP: MDT vs placebo – 1 study found a significant difference in pain favouring MDT.

Conclusion: The authors concluded that there is moderate to high quality evidence that MDT is not superior to other rehabilitation interventions for reducing pain and disability in acute LBP. There is moderate to high quality evidence that MDT is superior to other rehabilitation interventions for reducing pain and disability in chronic LBP; however, because effect sizes were small to moderate, the authors suggested that clinical significance needs to be determined. Also, they posited that therapists should be careful when using this approach exclusively, because other treatments have shown similar effectiveness, and patients’ values and preferences should be considered.

Implications for the MDT clinician:
This review was well designed in the sense of utilizing appropriate study inclusion criteria. Only studies with therapists who had taken at least Part A were included; inspection of the 11 studies included for meta-analysis revealed 10 out of 11 used Credentialed therapists. Also, only studies in which therapists performed a mechanical assessment prior to beginning treatment were included. Taken together, the findings can be interpreted in the context of MDT as it is meant to be used.

Significantly, we now have contemporary evidence of moderate to high quality that MDT is superior to other interventions for chronic lower back pain. While it can be somewhat reductive to call MDT management an “intervention” as opposed to a complete clinical reasoning, assessment, and management system, clinicians can rest assured as to the effectiveness of the approach. Interestingly, this point seems to have been downplayed by the authors; any approach shown to have significantly better outcomes in chronic lower back pain than other commonly used interventions would, in most cases, be heralded as a promising finding.

Several points need to be taken into consideration when interpreting the findings of the acute analysis. Firstly, only 4 studies were included, with 2 of the 4 having sample sizes under 35 patients. These latter two showed the greatest treatment effect in favour of MDT for pain and disability but were weighted relatively lower in the meta-analysis compared to the other larger studies. Secondly, 3 of the 4 studies included heterogeneous cohorts of patients with acute low back pain. It is likely that many people in each group would have responded due to natural history and the presence of positive prognostic variables (e.g. centralization more common in acute lumbar patients), regardless of the treatment given. Thus, the response to specific treatment might have been washed out, as has been hypothesized in many previous RCTs.

The authors’ assertion that clinicians should be cautious to use MDT exclusively in lumbar patients when other interventions have been shown to have similar effects is curious, especially in light of their own findings in the chronic population. Additionally, no other intervention in this review was shown to be superior to MDT. The authors’ urge to take patient values and preferences into account is something inherent to MDT as a comprehensive, patient-centred, biopsychosocial assessment and management system. With this in mind, MDT clinicians should remain confident in exclusively employing this approach with all patients they see.

Lumbar: Reviews