Low-intensity pulsed ultrasound shows a 0.79 healing rate in nonunion fractures across observational studiesTrial results show mixed success for ultrasound on nonunion fractures

This meta-analysis evaluated the clinical efficacy of low-intensity pulsed ultrasound (LIPUS) specifically for patients with established nonunion fractures. The analysis synthesized data from two distinct types of study designs: observational studies and randomized controlled trials (RCTs). The total sample size across all included studies was 3,439 patients in the observational cohort and 264 patients in the RCT cohort. The primary outcome measured across both groups was the healing rate of the fracture.

The intervention analyzed was low-intensity pulsed ultrasound (LIPUS), which is often proposed as a non-invasive method to stimulate bone healing. In the observational study group, the comparator was not reported. In the RCT group, LIPUS was compared against a sham treatment. Secondary outcomes included the time required for fracture healing.

Results from the observational studies indicated a pooled healing rate of 0.79 among the n=3439 nonunions (95% CI: 0.73-0.85; P <.01). However, the quality of evidence for these results was noted as very low due to significant methodological concerns. In contrast, the meta-analysis of RCTs involving n=264 nonunions showed a nonsignificant risk ratio in favor of LIPUS compared to sham (RR 1.13; 95% CI: 0.93-1.37; P =.23). Furthermore, regarding the secondary outcome of healing time in RCTs involving n=135 nonunions, there was a nonsignificant mean difference between LIPUS and sham treatment of 14.99 days (95% CI: -120.37 to 150.35; P =.83).

Safety and tolerability data were not reported for this meta-analysis, meaning specific adverse event rates or discontinuation rates for LIPUS could not be quantified from the available evidence. The study identified several significant methodological limitations that impact the strength of the findings. The observational studies were noted to have a low to serious risk of bias and exhibited high heterogeneity (I2=92%). The RCT meta-analyses also showed substantial heterogeneity, with I2 values of 60% and 74%. Consequently, the certainty of evidence for all meta-analyses was categorized as low or very low.

When compared to historical clinical practices, these results suggest that while observational data may show a positive association between LIPUS and healing, the controlled trials do not provide sufficient evidence of superiority over sham treatments. The high heterogeneity and risk of bias in the larger observational datasets likely inflate the perceived benefit of the intervention. Therefore, the findings do not support the use of LIPUS as a first-line treatment for nonunion fractures at this time.

Clinical implications suggest that practitioners should exercise caution when recommending LIPUS as a primary intervention for established nonunions. Because the RCT data did not reach statistical significance and the observational data are of low certainty, LIPUS may currently only be considered an adjunctive therapy rather than a primary treatment. Questions remain regarding the specific mechanisms of action in different fracture types and whether more homogeneous, large-scale trials could provide clearer evidence on the efficacy of LIPUS compared to standard care.