The nursing delirium screening scale (Nu-DESC) was developed for the rapid assessment of delirium, but little is known regarding the performance of the Nu-DESC in assessing postoperative delirium. The purpose of this systematic review and meta-analysis was to review and summarize the evidence of the diagnostic accuracy of the Nu-DESC in assessing postoperative delirium. The EMBASE, MEDLINE, Cochrane Library, CINAHL, and a Chinese e-Journal database were searched from the period of the inception of the Nu-DESC to June 18, 2020. Participants were adult (age ≥ 18 years) postoperative patients who received any type of surgery and any method of anesthesia. The quality assessment of diagnostic accuracy studies-2 was employed to assess the risk of bias among the selected studies, and meta-regression analyses were used to detect sources of between-study heterogeneity. Eleven studies involving 2062 postoperative patients in surgical settings were included. The Nu-DESC revealed a pooled sensitivity of 0.73 (95% confidence interval, 0.44–0.90) and a specificity of 0.93 (0.87–0.96). The area under the summary receiver operating characteristics curve was 0.94 (0.91–0.96), which also confirmed the accuracy of the Nu-DESC. Covariates, including the region of study and the standard reference used, were identified as possible sources of heterogeneity in the meta-regression. The findings of this review can update existing clinical guidelines for postoperative delirium. On the basis of its satisfactory diagnostic performance, we suggest that the Nu-DESC could be considered for nurses and allied health professionals for assessing postoperative delirium. However, because of the heterogeneity, the result of this meta-analysis should be considered with caution.