Objectives Systematic review and meta-analysis on shortening antibiotic therapy for Lyme borreliosis (LB) patients. Methods Data sources: Medline, Google, and Google Scholar (queried from January 2022-February 2024), following the PRISMA method and the Cochrane Handbook. Eligibility criteria: Randomized clinical trials, comparative studies; clear definitions of LB, duration of antibiotics and outcome; follow-up >= 6-12 months. Meta-analysis included studies that examined three outcomes: treatment failure; residual symptoms; adverse events. Intervention: Short vs. extended antibiotic therapy for erythema migrans (<= 10 days vs. > 10 days) and disseminated LB (<= 21 days vs. > 21 days). Assessment of risk of bias. Independently, using the Cochrane Tools. Methods of data synthesis. Estimation of treatment effects based on a fixed-effect model (Mantel-Haenszel or Peto method), with odds ratio (OR) and 95% confidence intervals (CI). Results Thirty-eight full-text articles were examined (850 patients): 29 were included in the qualitative analysis; six in the meta-analysis. Heterogeneity was low (I2 = 0%). At 12 months, short-term treatment did not differ from long-term treatment in terms of failures (OR1.50, 95%CI[0.43-5.22]) and residual symptoms (OR0.95, 95%CI[0.66-1.37]), albeit with small samples. Conclusion This meta-analysis was underpowered to prove non-inferiority of shorter treatment, but suggests its safety for EM. Studies focusing on antibiotics duration, with sufficient sample sizes and clear outcomes, are warranted.