Computed tomography (CT)–guided core needle biopsy (CNB) can be affected by streak artifact obscuring the needle tip. This study investigates factors that influence the occurrence and severity of streak artifact during CNB.Eight coaxial guide needles of two sizes from two manufacturers with and without stylets were imaged in a CT phantom, with CT reconstructed with adaptive statistical iterative reconstruction and filtered back projection. CNB-related streak artifact was quantified with profile analysis in an image-processing program. Differences between maximum attenuation at the needle tip and minimum attenuation in the streak artifact were compared for each variable. Diagnostic acceptability and streak artifact were subjectively assessed on each phantom image and on 40 clinical CNB procedures by three independent blinded reviewers following training case review.Artifact was significantly less with the central stylet removed versus in situ (median, 1,145 HU vs 3,390 HU; P < .001) for all needles, and less for 19-gauge needles versus 17-gauge needles (median, 1,334 HU vs 2,780 HU, respectively; P = .006). There were no differences based on manufacturer (P = .906) or reconstruction algorithm (P = .524). Independent reviews found that streak artifact was significantly reduced when the central stylet was removed (κ = 0.875–1.0; P < .001), and needle tip position was better in cases in which the stylet was removed (κ = 0.231–0.711; P < .001).Streak artifact can be reduced and needle tip visualization improved by confirming final biopsy needle position with the central stylet removed on CT and using smaller-gauge guide needles.