BackgroundDue to limitations of conventional medicine for atopic eczema (AE), complementary and alternative medicine (CAM) is widely used as an alternative, maintaining, or simultaneous treatment for AE. We aimed to evaluate the beneficial and harmful effects of CAM for children with AE under 14 years old.MethodsWe searched for randomized trials on CAM in 12 Chinese and English databases from their inception to May 2018. We included children (< 14 years) diagnosed with AE, who received CAM therapy alone or combined with conventional medicine. We extracted data, and used the Cochrane “Risk of bias” tool to assess methodological quality. Effect was presented as relative risk (RR) or mean difference (MD) with 95% confidence interval (CI) using RevMan 5.3.ResultsTwenty-four randomized controlled trials involving 2233 children with AE were included. Methodological quality was of unclear or high risk of bias in general. The trials tested 5 different types of CAM therapies, including probiotics, diet, biofilm, borage oil, and swimming. Compared to placebo, probiotics showed improved effect for the SCORAD index (MD 9.01, 95% CI 7.12–10.90; n = 5). For symptoms and signs such as itching, skin lesions, CAM combined with usual care was more effective for symptom relief ≥95% (RR 1.47, 95% CI 1.30–1.68; n = 8), and for ≥50% symptoms improvement (RR 1.34, 1.25–1.45; n = 9) compared to usual care. There was no statistic significant difference between CAM and usual care on ≥95% improvement or ≥ 50% improvement of symptoms. However, swimming, diet and biofilm showed improvement of clinical symptoms compared with usual care. At follow-up of 8 weeks to 3 years, CAM alone or combined with usual care showed lower relapse rate (RR 0.38, 0.28–0.51, n = 2; RR 0.31, 0.24–0.40, n = 7; respectively) compared to usual care. Twelve out of 24 trials reported no occurrence of severe adverse events.ConclusionsLow evidence demonstrates that some CAM modalities may improve symptoms of childhood AE and reduce relapse rate. Safety remains unclear due to insufficient reporting. Further well-designed randomized trials are needed to confirm the potential beneficial effect and to establish safety use.Electronic supplementary materialThe online version of this article (10.1186/s12906-018-2306-6) contains supplementary material, which is available to authorized users.