A recommendation in the 2009 Infectious Disease Society of America (IDSA) guidelines for diagnosing and managing intravascular catheter infections suggests obtaining blood cultures through a catheter lumen and percutaneously.Reference Mermel, Allon and Bouza¹ This has led to concern that the recommendation promotes false-positive, catheter-drawn blood cultures, resulting in unnecessary antibiotic use, inflated rates of central-line–associated bloodstream infections reported to Center for Disease Control and Prevention (CDC)’s National Healthcare Safety Network, and higher Hospital-Associated Condition scores reported to the Centers for Medicare and Medicaid (CMS). A recent survey found that 90% of responding hospitals had protocols recommending that blood cultures should be obtained percutaneously rather than through central venous catheters,Reference Garcia, Spitzer, Kranz and Barnes² a measure that reflects diagnostic stewardship efforts.Reference Morgan, Malani and Diekema³ Concern about increased risk of blood culture contamination when blood is obtained through a catheter is primarily based on studies published several years ago, prior to use of antiseptic barrier caps (so-called ‘port protectors’), which passively disinfect the surface of needleless connectors used on catheter hubs, thereby reducing the surface bioburden and risk of intraluminal microbial ingress. What was the evidence then and what is it now?

A meta-analysis of studies published between 1999 and 2011 found increased blood culture contamination with catheter-drawn cultures compared with percutaneously-drawn cultures (odds ratio [OR], 2.69; 95% CI, 2.03–3.57) with moderate study heterogeneity (Q = 19.5; P = .012).Reference Snyder, Favoretto and Baetz⁴ Some of these studies were done before widespread use of needleless connectors on catheter hubs; none were conducted in the era of antiseptic barrier cap useReference Holt, Helder and Vos⁵ nor with technologies that divert an initial aliquot of blood.Reference Rupp, Cavalieri, Marolf and Lyden^6, Reference Skoglund, Dempsey, Chen and Garey⁷ A multimodal infection prevention program involved use of (1) a blood culture collection kit, (2) a 2-nurse protocol for blood culture collection, and/or (3) limiting catheter-drawn cultures (if such cultures were needed, the needleless connector was removed and the catheter hub was scrubbed before obtaining samples for blood cultures).Reference Boyce, Nadeau and Dumigan⁸ In the first 6 months of 2010 (baseline), 2011, and 2012, contamination rates of percutaneously-drawn blood cultures were 0.9% (40 of 4,353), 0.8% (39 of 4,733), and 0.5% (21 of 4,206), respectively. For the same periods for catheter-drawn cultures, contamination rates were 4.8% (84 of 5,274), 1.3% (3 of 227), and 0% (0 of 39), respectively. That is, contamination rates combining 2011 and 2012 data of 0.67% and 1.1%, respectively, for percutaneously-drawn and catheter-drawn blood cultures [RR, 1.7; 95% CI, 0.5–5.3]). Notably, antiseptic barrier caps were not used.

Lack of compliance with routine scrubbing of catheter needleless connectors before accessing for any purpose, such as obtaining blood cultures, has led to a recommendation to change the needleless connector before blood culture collection (https://www.ins1.org/Store/ProductDetails.aspx?productId=113266). The evidence for this recommendation comes from a single publication.Reference Mathew, Gaslin, Dunning and Ying⁹ The investigators found greater concordance with percutaneously-drawn blood cultures if blood was obtained using a new needleless connector rather than an old connector. However, a recent survey found that 48% of hospitals simply remove old connectors and draw blood directly from catheter hubs rather than through a needleless connector.Reference Garcia, Spitzer, Kranz and Barnes² This practice is supported by a study that demonstrated colonization of 12% and 6.7% of central venous catheter needleless connectors and hubs after removal of the connectors, respectively.Reference Guembe, Pérez-Granda, Cruces, Martín-Rabadán and Bouza¹⁰ In this study, antiseptic barrier caps were not used, and the higher colonization rate of the connectors may, in part, reflect a more sensitive laboratory method used for the needleless connector culture.

A meta-analysis that excluded low-quality methodology studies revealed that antiseptic barrier caps reduce central-line–associated bloodstream infections (incidence rate ratio [IRR], 0.66; 95% confidence interval [CI], 0.53–0.83).Reference Holt, Helder and Vos⁵ Cluster-randomized trials involving hemodialysis catheters have also demonstrated significant reductions in these infections when unique antiseptic barrier caps are used.Reference Hymes, Mooney, Van Zandt, Lynch, Ziebol and Killion^11, Reference Brunelli, Van Wyck, Njord, Ziebol, Lynch and Killion¹² Thus, one would expect a commensurate reduction in blood culture contamination. In the peer-reviewed literature, 3 studies assessed this issue and found a reduction in intraluminal colonization or blood culture contamination (Table 1).Reference Sweet, Cumpston, Briggs, Craig and Hamadani^13–Reference Kamboj, Blair and Bell¹⁵ In these studies, no data were provided regarding contamination of simultaneously obtained, percutaneously-drawn, blood cultures. Notably, the use of antiseptic barrier caps is not without complications; pediatric patients have swallowed these devices.Reference Tawfik, Myer, Shikary and Goldschneider¹⁶

Table 1.

Blood Culture or Intraluminal Catheter Colonization With and Without Use of Antiseptic Barrier Caps

Note. RR, risk ratio; CI, confidence interval.

^a RR, 0.18; 95% CI, 0.02–1.37; P = 0.10.

^b P < .001.

^c P = .01 when comparing before barrier cap use and with barrier cap use; P = .003 when comparing with barrier cap use and after discontinued barrier cap use.

^d P < .001.

Blood cultures collected through a short-term peripheral intravenous catheter during insertion in pediatricReference Ramsook, Childers, Cron and Nirken¹⁷ and adult patientsReference Self, Speroff and McNaughton¹⁸ have a significantly greater risk of contamination compared to percutaneously-drawn cultures. Blood cultures drawn during central venous catheter insertion in an intensive care unit have a significantly higher risk of contamination compared to percutaneously-drawn cultures; however, contamination of blood cultures drawn during arterial catheter insertion was not significantly different than percutaneously-drawn cultures.Reference Stohl, Benenson and Sviri¹⁹

From this review, it is evident that (1) a multimodal program can significantly reduce catheter-drawn blood culture contamination; (2) the use of antiseptic barrier caps can reduce the risk of catheter-drawn blood culture contamination; and (3) blood cultures should not be obtained during insertion of short-term peripheral intravenous catheters or central venous catheters. No published studies have assessed contamination rates when percutaneously-drawn blood cultures are done with technologies that divert an initial aliquot of blood compared with catheter-drawn cultures when antiseptic barrier caps are used and cultures are obtained from the catheter hub or from a new needleless connector after the old connector is discarded. It is important to remember that catheter colonization is not static—it is a continuous process. Catheter lumens can become colonized over time,Reference Mermel²⁰ which can lead to a bloodstream infection. As such, catheter-drawn blood cultures may reveal such colonization before a true bloodstream infection is evident or may reveal colonization in a patient who had recent or transient bacteremia from the colonized catheter that was cleared and undetected by percutaneously-drawn cultures. In situ diagnosis of catheter-related bloodstream infection, in contradistinction to the less-specific central-line–associated bloodstream infection, requires catheter-drawn and percutaneously-drawn cultures.Reference Mermel, Allon and Bouza¹

Based on the published literature, blood cultures should be obtained percutaneously when possible. Definitive in situ diagnosis of catheter-related bloodstream infection requires catheter-drawn cultures, which should be obtained from a catheter hub after removal of the old needleless connector that was protected with an antiseptic barrier cap or through a new connector after the old connector protected by a barrier cap is discarded. The best way forward at the present time is for institutions to measure the frequency of percutaneously-drawn and catheter-drawn blood cultures, to measure the contamination rates of each, and then to use these data to guide policies regarding recommended sites for obtaining cultures.