• Izumi Nakayama, Junichi Izawa, Koichiro Gibo, Sara Murakami, Taisuke Akiyama, Yuki Kotani, Rie Katsurai, Yuki Kishihara, Takahiro Tsuchida, Shunichi Takakura, Yoshihiro Takayama, Masashi Narita, and Soichi Shiiki.
• Division of Intensive Care Medicine, Department of Internal Medicine, Okinawa Prefectural Chubu Hospital, Uruma, Japan; Department of Public Health, School of Medicine, Yokohama City University, Yokohama, Japan; Department of Health Data Science, Graduate School of Data Science, Yokohama City University, Yokohama, Japan. Electronic address: nakayama.izu.vd@yokohama-cu.ac.jp.
• Chest. 2023 Jul 1; 164 (1): 9010090-100.

BackgroundCollecting blood cultures from indwelling arterial catheters is an attractive option in critically ill adult patients when peripheral venipuncture is difficult. However, whether the contamination proportion of blood cultures from arterial catheters is acceptable compared with that from venipuncture is inconclusive.Research QuestionIs contamination of blood cultures from arterial catheters noninferior to that from venipuncture in critically ill adult patients with suspected bloodstream infection?Study Design And MethodsIn this multicenter prospective diagnostic study conducted at five hospitals, we enrolled episodes of paired blood culture collection, each set consisting of blood drawn from an arterial catheter and another by venipuncture, were obtained from critically ill adult patients with cilinical indication. The primary measure was the proportion of contamination, defined as the number of false-positive results relative to the total number of procedures done. The reference standard for true bloodstream infection was blinded assessment by infectious disease specialists. We examined the noninferiority hypothesis that the contamination proportion of blood cultures from arterial catheters did not exceed that from venipuncture by 2.0%.ResultsOf 1,655 episodes of blood culture from December 2018 to July 2021, 590 paired blood culture episodes were enrolled, and 41 of the 590 episodes (6.9%) produced a true bloodstream infection. In blood cultures from arterial catheters, 33 of 590 (6.0%) were positive, and two of 590 (0.3%) were contaminated; in venipuncture, 36 of 590 (6.1%) were positive, and four of 590 (0.7%) were contaminated. The estimated difference in contamination proportion (arterial catheter - venipuncture) was -0.3% (upper limit of one-sided 95% CI, +0.3%). The upper limit of the 95% CI did not exceed the predefined margin of +2.0%, establishing noninferiority (P for noninferiority < .001).InterpretationObtaining blood cultures from arterial catheters is an acceptable alternative to venipuncture in critically ill patients.Clinical Trial RegistrationUniversity Hospital Medical Information Network Center (UMIN-CTR); No.: UMIN000035392; URL: https://center6.umin.ac.jp/.Copyright © 2023 American College of Chest Physicians. Published by Elsevier Inc. All rights reserved.

22 keywords...

hide…