Background: The use of peripherally inserted central catheters (PICCs) in cancer care is increasing over traditional central venous catheters. Nurses frequently collect blood by venipuncture when a PICC is inserted, as there is no available evidence to confirm the reliability of blood tests collected through PICCs in adults. Objective: The aim of this study was to assess the reliability of blood samples for complete blood count (CBC) obtained through PICCs as an alternative to venipuncture. Methods: A cross-sectional design was used to recruit adult hematological patients. The blood samples were collected within 5 minutes of each other by a specialist nurse. Hemoglobin, hematocrit, and platelet count were evaluated. To determine method comparison, Passing-Bablok regression, test of linearity, Pearson product-moment correlation coefficient, and Bland-Altman plots were used. Results: Thirty paired blood samples were collected in 29 hematological patients with a mean age of 66.8 years. A statistical difference was found for hemoglobin (P = .001) and hematocrit (P = .001) levels, but no clinical difference. The regression models revealed no systematic differences and no proportional differences with a linear relationship between the methods. Bland-Altman plots highlighted a good agreement between methods. Conclusion: Blood samples for CBC drawn by PICCs are as reliable