Effect of High-Intensity Laser Acupuncture on Muscle Performance, Pain, and Quality of Life in Patients with Chronic Nonspecific Low Back Pain: A Randomized Placebo-Controlled Trial [Letter]

Dear editor

We read with the great interest the randomized, placebo-controlled trial by Elimy et al reporting the effects of high-intensity laser acupuncture (HILA), combined with physical therapy exercises, on trunk muscular performance, pain, and quality of life (QoL) in patients with chronic nonspecific low back pain (CNSLBP).¹ The study addresses a genuine gap in the literature regarding objective, isokinetically measured muscular outcomes of laser acupuncture, and the authors are to be commended for prospective trial registration and adherence to a CONSORT-based reporting structure. However, on evaluating the report against the CONSORT 2010 Statement² and its 2017 extension for trials of nonpharmacologic treatments (NPTs),³ several methodological issues warrant clarification before the reported effect sizes are extrapolated into clinical practice.

Integrity of the Reported Double-Blinding

The manuscript is described throughout as a double-blinded trial, yet the Intervention section states that the physiotherapist who administered both the active HILA and the sham laser was “not blinded to the group allocation”.¹ Because this same clinician selected the treatment sequence, positioned the probe over the acupoints, and could directly observe cues unique to active irradiation (eg, skin warming), performance bias introduced by an unblinded care provider cannot be excluded even though allocation concealment and assessor blinding were otherwise well described. The CONSORT extension for NPTs specifically recommends that, when blinding of the care provider is not feasible, this should be explicitly acknowledged along with any measures taken to limit the resulting bias.³ We would welcome the authors’ clarification of why the study is labelled double-blind rather than participant-and-assessor-blind, and of any procedural safeguards used to standardize the unblinded therapist’s conduct across the two arms.

Absence of Any Participant Attrition

The CONSORT flow diagram (Figure 1) and the outcome tables report exactly 30 participants retained in each group at the week-4 and 1-month follow-up assessments, with no losses to follow-up, protocol deviations, or missing isokinetic or questionnaire data over an approximately eight-week observation period.¹ While the sample-size calculation explicitly budgeted for a 10% dropout rate,¹ zero attrition across 60 participants undergoing 12 supervised sessions plus follow-up testing is atypical for a musculoskeletal rehabilitation trial of this duration. Per CONSORT item 13, we would ask the authors to confirm whether any participant temporarily withdrawn for adverse-event evaluation (as described in the Adverse Event Monitoring section) was retained in the final analysis, and how any such instance was handled statistically.

Multiplicity of Statistical Comparisons

Six continuous outcomes (peak torque and average power for trunk extensors and flexors, pain, and two WHOQOL-BREF domains) were each analyzed with a mixed repeated-measures ANOVA, a main effect of group, a group-by-time interaction, and three Bonferroni-adjusted pairwise contrasts, but no correction was applied for the family-wise error rate across the six outcome domains themselves.¹ Because multiplicity across outcomes, and not only within an outcome’s repeated measurements, inflates the probability of at least one false-positive conclusion,⁴ we ask the authors to clarify whether a global alpha adjustment across the outcome domains was considered, and whether a primary versus secondary outcome hierarchy was pre-specified at trial registration.

Selective Reporting of WHOQOL-BREF Domains

The WHOQOL-BREF instrument used in this trial comprises four domains: physical, psychological, social relationships, and environmental.¹ However, only the physical and psychological domain scores are reported in Table 3 and discussed in the Results and Discussion; the social relationships and environmental domains are not mentioned. CONSORT item 6a calls for complete reporting of all pre-specified outcomes, and omission of two of four domains of a validated composite instrument raises the possibility of outcome-reporting bias.² We would appreciate confirmation of whether these two domains were measured and, if so, why their results were not presented, and whether the reported domains match those pre-specified on ClinicalTrials.gov (NCT06612918).

Sample Size Calculated for a Single Endpoint, Applied to Multiple Endpoints

The a priori sample-size calculation was based solely on a pilot effect size for peak torque of the trunk extensors,¹ yet the trial subsequently reports and clinically interprets statistically significant between-group effects across six further muscular, pain, and QoL variables. As the study was not powered for these additional comparisons, we suggest the corresponding effect sizes for the non-primary endpoints be framed as hypothesis-generating rather than confirmatory, consistent with CONSORT guidance on distinguishing primary from secondary outcome inference.²

Minor Numerical Inconsistency

In Table 2, the reported mean difference for average power of the trunk extensors (HILA group, Pre vs Follow-up) is listed as “9.52 (7.01, + 12.03)”; the unexplained plus sign preceding the upper confidence bound appears to be a typographical artifact. We ask the authors to confirm the correct 95% confidence interval for this comparison.¹

Generalizability and Durability of the Treatment Effect

Eligibility was restricted to a narrow age band (20–40 years) and a normal body-mass index (18.5–24.9 kg/m²), with obese individuals explicitly excluded.¹ Because both the prevalence and the muscular consequences of nonspecific low back pain increase with age and adiposity,⁵ the magnitude of benefit reported here may not generalize to the broader population that typically presents with CNSLBP, a point the authors partly acknowledge. In addition, every outcome in both study arms declined significantly between the 4-week post-intervention assessment and the 1-month follow-up,¹ and with only a single post-treatment follow-up point, the durability of the reported HILA-associated gains beyond eight weeks cannot be established. Longer follow-up would better align this line of research with recommendations for evaluating nonpharmacologic treatments of chronic low back pain.⁶

We commend the authors for undertaking a prospectively registered trial in an under-studied area of rehabilitation and believe that addressing the points raised above, ideally in an author response or corrigendum, would strengthen confidence in the reported findings and better inform clinicians seeking to interpret the true magnitude of HILA’s adjunctive benefit in CNSLBP.

Funding Statement

This letter received no specific funding.

Disclosure

The author declares no conflicts of interest in this communication.