Background: The pathogenesis of parenteral nutrition–associated cholestasis (PNAC) has not been clarified. The objective of this study was to explore the incidence of PNAC in premature infants without surgery and to identify associated risk factors. Materials and Methods: Premature neonates who received parenteral nutrition (PN) at least 14 days were included in a retrospective, dual-center study. Cholestasis was diagnosed as conjugated bilirubin ≥2 mg/dL. Infants with metabolic liver disease, cyanotic congenital heart disease, congenital syphilis, hepadnaviridae infection, and those who underwent surgery were excluded. Infants were divided into 3 groups chronologically: group A (2000–2004, n = 50), group B (2005–2009, n = 283), and group C (2010–2014, n = 741). A case-controlled study was conducted by comparing infants with PNAC to those without PNAC. Results: Of 1074 premature neonates, PNAC was confirmed in 53 infants (4.93%). There were 6.8% very low birth weight (BW) infants and 20.0% extremely low BW infants who developed PNAC. The incidence of PNAC decreased slightly during 2000–2014 (8.0%, 6.4%, and 4.2% in groups A, B, and C, respectively). Compared with those without PNAC, infants with PNAC (n = 53) had significantly younger gestational age, lower BW, longer PN duration, and higher rate of sepsis. Logistic regression showed male sex, PN duration ≥43 days, and sepsis were statistically correlated with PNAC. Conclusions: Prolonged duration (≥43 days), male sex, and sepsis are probably independent risk factors for developing PNAC in premature neonates.