Network meta-analysis finds acupuncture therapies may improve glycemic parameters in type 2 diabetes

This systematic review and network meta-analysis evaluated the comparative efficacy of different acupuncture therapies for glycemic control in patients with type 2 diabetes mellitus (T2DM). The analysis synthesized data from a total of 12,231 patients, though the specific number of included studies, their individual designs, and the geographic or clinical setting were not reported. The population was defined broadly as patients with T2DM, with no further demographic or clinical characteristics provided. The study aimed to rank the effectiveness of various acupuncture modalities against each other and versus conventional treatment.

The intervention of interest was a class of acupuncture therapies, specifically body acupuncture, auricular acupuncture, and laser acupuncture. The comparator was conventional treatment (CT), which was not defined in detail but presumably represents standard medical management for T2DM. The analysis did not report specific dosing, frequency, duration of acupuncture sessions, or the precise protocols used for the different acupuncture techniques. The follow-up duration for the included studies was also not reported.

The analysis did not specify a single primary outcome. Instead, it presented network meta-analysis results for several key glycemic and metabolic parameters, ranking the 'best effect among acupuncture regimens.' For fasting blood glucose (FBG), the best acupuncture regimen showed a mean difference (MD) of -0.84 mmol/L compared to conventional treatment (95% CI: -1.01 to -0.67). For HbA1c, the MD was -1.08% (95% CI: -1.42 to -0.76). For 2-hour postprandial blood glucose (2hBG), the MD was -1.56 mmol/L (95% CI: -2.05 to -1.06). All confidence intervals for these three outcomes did not cross zero, suggesting a statistically significant association.

For other secondary outcomes, the point estimates favored acupuncture but with wider, inconclusive confidence intervals. The best acupuncture regimen for homeostasis model assessment for insulin resistance (HOMA-IR) had an MD of -2.46 (95% CI: -5.85 to 0.96). For fasting insulin (FINS), the MD was -7.21 μIU/mL (95% CI: -19.00 to 4.54). The overall response rate, a composite measure, had a risk ratio (RR) of 1.45 for the best acupuncture regimen versus CT (95% CI: 0.97 to 2.31). The analysis did not provide the absolute event rates needed to calculate the number needed to treat.

The review provided no data on safety, tolerability, or adverse events. The authors did not report the incidence of adverse events, serious adverse events, or treatment discontinuations related to acupuncture or conventional treatment. This represents a critical gap in the evidence, as the risk-benefit profile of adding acupuncture to diabetes management cannot be fully assessed without this information.

Compared to prior landmark pharmacological trials in T2DM, which typically demonstrate HbA1c reductions of 0.5% to 1.5% with various drug classes, the point estimate for acupuncture in this analysis (-1.08%) appears numerically similar. However, direct comparison is inappropriate due to fundamental differences in study design, patient populations, and the nature of the interventions. This analysis aggregates various acupuncture techniques, whereas drug trials test specific, standardized molecules. Furthermore, the conventional treatment comparator in this analysis is poorly defined and may not represent contemporary, guideline-directed medical therapy.

The authors explicitly note key methodological limitations, primarily the low quality of the included randomized controlled trials. This low quality likely stems from issues with blinding, allocation concealment, and high risk of bias, which are common challenges in trials of physical interventions like acupuncture. The network meta-analysis inherits these limitations. The authors correctly conclude that more evidence is needed to verify the conclusions. Additional limitations not explicitly stated but inherent in the report include the lack of primary outcome specification, no safety data, an undefined conventional treatment comparator, and no patient-level characteristics or subgroup analyses.

The clinical implications are tentative. The analysis suggests a signal that certain acupuncture techniques might be associated with improvements in standard glycemic parameters. For clinicians, this could inform discussions with patients interested in complementary therapies, emphasizing that while some low-quality evidence exists for potential benefit, the safety profile is unknown and the evidence is not robust enough to recommend acupuncture as a standard adjunct to evidence-based medical therapy. It should not replace proven pharmacological or lifestyle interventions.

Significant questions remain unanswered. The optimal acupuncture technique, treatment frequency, and duration are unknown. The safety profile and specific adverse event rates are unreported. It is unclear if effects are sustained long-term or if they apply to all T2DM patient subtypes. Crucially, it is unknown how acupuncture interacts with or compares to specific, modern glucose-lowering drug classes when used in combination. High-quality, double-blind, sham-controlled trials with rigorous methodology and comprehensive safety reporting are needed to move this field forward.