Periprosthetic joint infection (PJI) is one of the most devastating complications after total joint replacement (TJA). Up to now, the diagnosis of PJI is still in a dilemma. As a novel biomarker, whether D-dimer is valuable in the diagnosis of PJI remains controversial. This meta-analysis attempts to determine the diagnostic accuracy of D-dimer in PJI. Relevant literature was retrieved from PubMed, Embase, Web of Science, and Cochrane Library (from database establishment to April 2020). Literature quality was evaluated using Revman (version 5.3). The random effect model was used in the Stata version 14.0 software to combine sensitivity, specificity, likelihood ratio (LR), diagnostic odds ratio (DOR), summary receiver operating characteristic (SROC) curve, and area under SROC (AUC) to evaluate the diagnostic value of overall D-dimer for PJI. Meta regression and subgroup analysis were performed according to the threshold, the study design, the sample size, the diagnostic gold standard, the country of study, and the type of sample. A total of 9 studies were included in this study, including 1592 patients. The pooled sensitivity and specificity of D-dimer for PJI diagnosis are 0.82 (95% CI, 0.72~0.89) and 0.73 (95% CI, 0.58~0.83), respectively. The pooled positive likelihood ratio (PLR) and negative likelihood ratio (NLR) were 2.99 (95% CI, 1.84~4.88) and 0.25 (95% CI, 0.15~0.41), respectively. The pooled AUC and diagnostic odds ratios were 0.85 (95% CI, 0.82~0.88) and 12.20 (95% CI, 4.98~29.86), respectively. D-dimer is a promising biomarker for the diagnosis of PJI, which should be used in conjunction with other biomarkers or as an adjunct to other diagnostic methods to enhance diagnostic performance.