Objectives: To address the diagnostic accuracy of endoscopic ultrasound guided through-the-needle-biopsies (TTNBs) and simultaneously obtained cytology samples from pancreatic cysts compared to the final histopathological diagnosis of the surgical specimen, and to give an overview of ancillary tests performed on TTNBs. Methods: A literature search was conducted in MEDLINE, Embase and Scopus. Studies were included in the meta-analysis, if they had data for TTNB, cytology and a surgical specimen of pancreatic cysts as reference standard. The assessment of the risk of bias and quality of the included studies was conducted using the modified QUADAS-2 tool. Results: Ten studies with 99 patients were included in the meta-analysis. Data regarding study design and clinicopathological features were extracted systematically. For TTNB, pooled sensitivity was 0.86 (95 % CI 0.62−0.96), specificity 0.95 (95 % CI 0.79−0.99) and area under the curve (AUC) 0.86 for the diagnosis of a mucinous cyst and pooled sensitivity was 0.78 (95 % CI 0.61−0.89), specificity 0.99 (95 % CI 0.90−0.99) and AUC 0.92 for the diagnosis of a high-risk cyst. For a specific diagnosis, pooled sensitivity was 0.69 (95 % CI 0.50−0.83), specificity 0.47 (95 % CI 0.28−0.68) and AUC 0.49. For cytology performed simultaneously, pooled sensitivity was 0.46 (95 % CI 0.35−0.57), specificity 0.90 (95 % CI 0.46−0.99) and AUC 0.64 for the diagnosis of mucinous cysts, and pooled sensitivity was 0.38 (95 % CI 0.23−0.55), specificity 0.99 (95 % CI 0.90−0.99) and AUC 0.84 for the diagnosis of a high-risk cyst. For a specific diagnosis, pooled sensitivity was 0.29 (95 % CI 0.21−0.39), specificity 0.45 (95 % CI 0.25−0.66) and AUC 0.30. Furthermore, immunohistochemical stains can be useful to establish the specific cyst subtype. Conclusions: TTNBs have a higher sensitivity and specificity than cytology for the diagnosis of mucinous cyst and high- risk cysts of the pancreas.