J Vasc Access 2016; 17(4): e92 - e92
Article Type: CORRESPONDENCE
AuthorsJona V. Caparas, Jian-Ping Hu
- • Accepted on 28/03/2016
- • Available online on 01/06/2016
- • Published online on 12/07/2016
This article is available as full text PDF.
The authors appreciate the thoughtful letter from Spencer and Bardin (1). In addressing their concerns, it is important to bear in mind two facts: (i) Because of the risks of central line-associated blood-stream infection (CLABSI) (1%-3%) and peripherally inserted central catheter (PICC)-associated deep-vein thrombosis (DVT) (silent DVT = 71.9%, clinical DVT = 5%-7%), PICCs can no longer be assumed to deliver risk-free central venous access; and (ii) not all midlines perform the same (2).
At the time of our study, agents with pH of <5 or >9 were wrongly considered an indication for central venous access. This false indication has been removed from the 2016 Infusion Therapy Standards of Practice. As shown by Gorski et al, the pH of intermittently delivered medications categorically does not cause thrombophlebitis (3).
We demonstrated that vancomycin (4 mg/mL) can be safely given short-term via the study midline (POWERWAND™, Access Scientific, San Diego, CA). Our data that demonstrate equivalent safety for long-term vancomycin delivery via the same midline, is pending publication. We have placed over 5000 of these midlines, and our outcomes support the observations of Warrington et al, that this midline, with its unique material and insertion method, produces the lowest total complication rate and highest completion of therapy rate of any vascular access device (VAD) studied (4).
Spencer states, “Central line associated bloodstream infection should not be combated with increased midline use….” We respectfully reject this notion. As Pathak et al have shown, implementation of a robust midline program results in a sustained reduction of CLABSI (5). In this connection, the study midline has been shown to allow clinicians to reduce total central line days by 37% (by removing existing lines earlier and avoiding unnecessary central venous catheters [CVCs]), resulting in 100% reduction of CLABSI (ibid) (5).
As to Spencer’s specific concerns regarding phlebitis scales, we assert that regardless of which scale is used, zero is zero. As to leakage, midlines placed using the modified Seldinger technique are often associated with back-leaking and DVT. One study showed that a different 12 cm midline (Arrow International, Reading, PA) was associated with a DVT rate of 20.9% (6). However, our study midline does not require placement through a peelable sheath and, therefore, tends not to engender these complications. The one incident of leakage we reported was between the catheter hub and the extension set. Finally, we welcome the in-press information alluded to by Spencer. However, we caution that with the PICC-associated risks of CLABSI and DVT looming, the advice to “…use a central venous catheter for vesicant or irritant medication” violates the new Standards which clearly endorse midlines, provided clinicians “use caution with intermittent vesicant administration.”
The myth of pH-induced thrombophlebitis has been debunked. It therefore behooves clinicians to weigh the real risks of PICC-related CLABSI and DVT against the often-phantom risk of thrombophlebitis. The right midline, as our study proved, can often be used to administer appropriately diluted medications safer and less expensively.