Journal club 9/6: Vascular access review

What’s the question we’re looking at? Why do we care in CCM?

Central lines are widely used in the ICU. However, they are associated with numerous harmful and costly complications. Some of these occur at the time of insertion (mechanical complications such as pneumothorax and bleeding), but others accumulate over time, most importantly CLABSIs (Central Line Associated Blood Stream Infections) and thrombosis.

While many critically ill patients will initially need a central venous catheter (CVC), leaving them in for prolonged periods is poor care. Often, they are not kept for administration of pressors or other vesicant medications that require central access, but merely because some access is needed and the patient is “a hard stick,” making it difficult for nursing to obtain regular peripheral IVs. What options does this leave? Other than having no vascular access (generally unacceptable for ICU patients), or intraosseous lines (so far unstudied for prolonged use), there are three options:

  1. Peripherally inserted central catheters (PICC lines)
  2. Ultrasound-guided peripheral IVs: usually placed in the upper arm using a regular but slightly longer angiocatheter, around 1.88″–2.5″
  3. Midlines: ultrasound-guided lines longer than IVs but not long enough to reach the central circulation, usually placed in the upper arm.

Here’s some of the relevant data.

PICC lines

MAGIC guidelines:

Chopra et al. The Michigan Appropriateness Guide for Intravenous Catheters (MAGIC): Results From a Multispecialty Panel Using the RAND/UCLA Appropriateness Method. Ann Intern Med. 2015 Sep 15;163(6 Suppl):S1-40

Expert panel met to develop criteria for appropriateness of PICC lines and apply them to 665 hypothetical patient scenarios

43% scenarios were deemed “inappropriate” for PICC use, 19% “questionable”

Their recommendations: PICCs are indicated for vesicant infusions that are expected to be:

  1. >6 days (but for 6-14 days midlines preferred, >15 days can consider tunneled line, >31 days consider a port). However, for critically ill patients, they should be reserved for use >14 days; you should instead use a CVC for shorter durations. (CVC for >14 days “uncertain.”)
  2. Best choice for the indication of difficult access/frequent blood draws: to be decided on a case-by-case basis


Maki et al. The risk of bloodstream infection in adults with different intravascular devices: a systematic review of 200 published prospective studies. Mayo Clin Proc. 2006 Sep;81(9):1159-71.

Systematic review of 200+ studies

Rates of line infection:

  • PICC had 2.1 bloodstream infections per 1000 catheter days
  • CVC had 2.7/1000
  • Midline had .2/1000
  • PIVs had .5/1000


Raiy et al. Peripherally inserted central venous catheters in the acute care setting: A safe alternative to high-risk short-term central venous catheters. Am J Infect Control. 2010 Mar;38(2):149-53.

638 CVC and 622 PICCs were placed and observed during a prospective campaign to reduce CVC use

  • PICCs had 2.3 BSI/1000 catheter days
  • CVCs had 2.4/1000 (although average time to infection was longer, 23 vs 13 days, probably reflecting a different route of infection — i.e. intraluminal vs infiltration of subcutaneous tract)


Bonizzolli et al. Peripherally inserted central venous catheters and central venous catheters related thrombosis in post-critical patients. Intensive Care Med. 2011 Feb;37(2):284-9

Single center study. Either PICCs or CVCs were placed as the initial access in ICU patients, using a before-and-after model, 3 months each. Dopplers were done at time of downgrade to screen for DVT.

125 CVCs and 114 PICCs placed

  • PICCs developed DVTs in 27.2% (or 7.7 per 1000 catheter-days)
  • CVCs developed DVTs in 9.6% (or 4.4 per 1000 catheter-days)

Ultrasound-guided peripheral IVs


Dargin et al. Ultrasonography-guided peripheral intravenous catheter survival in ED patients with difficult access. Am J Emerg Med. 2010 Jan;28(1):1-7

75 ED patients with difficult access

“Long” (6.35cm/2.5in) IV catheters placed using ultrasound

  • 86% success within in 1-2 attempts
  • 47% failed within 24 hours, mostly from infiltration/dislodgement
  • 63% would have otherwise required CVC placement at time of IV, but only 7% required one after the IV eventually failed (presumably the urgent need for access had passed), so it successfully avoided the central line

Midlines (practical)


Elia et al. Standard-length catheters vs long catheters in ultrasound-guided peripheral vein cannulation. Am J Emerg Med. 2012 Jun;30(5):712-6.

100 patients in a high-dependency unit (basically an IMC), all with difficult access, received either regular ultrasound-guided IVs or midlines.

Regular “long” angiocaths (5cm/1.97in)

  • 86% success
  • Mean survival 3.5 days, failure rate 45%, 42% infiltrated
  • 12.5% thrombosis (usually pericatheter)

Midline (12cm/4.7in, with wire)

  • 84% success
  • Mean survival 6.19 days, failure rate 14%, 2.3% infiltrated
  • 20.9% thrombosis (usually pericatheter)

Similar results in other studies (ScoppetuoloMeyerDeutsch)

Midlines (impact)


Moreau et al. Before-and-after study, 6 months each with initiation of a nursing-led midline program in two centers. The Journal of the Association for Vascular Access. 2015 Sept; 20(3); 179–188.

Before-and-after study in two centers, 6 months each, second period after initiation of a nursing-led midline program

Total 586 lines placed, 99.4% successful

  • PICC use declined by 58%
  • 0 BSIs documented from midlines
  • Overall CLABSI rate reduced from 1.7 BSI/1000 patient days to .2/1000 (78% reduction, with a projected cost savings of $531,570)

Note: cites varying expert sources that say midlines can be permitted to dwell for anywhere between 16-296 days. Infusion Nurses Society say 1-4 weeks. Device manufacturers usually suggest 29 days.


Pathak et al. The Incidence of Central Line–Associated Bacteremia After the Introduction of Midline Catheters in a Ventilator Unit Population. Infect Dis Clin Pract (Baltim Md). 2015 May;23(3):131-134.

Before-and-after study, 12 months each with a 3-month interim period, in a “ventilator unit”; similar patient volume before and after. For latter period, nursing and house staff trained and protocolized to switch out CVCs for midlines

  • Total catheter days for the period were reduced from 2408 to 1521
  • CLABSIs were reduced from 8 to 0 (3.32/1000 cath days to 0/1000)

Takeaway points

  • Changing out central lines to PICCs adds little to safety. In the inpatient environment, they get infected at approximately equal rates to CVCs, and develop more DVTs.
  • Ultrasound-guided peripherals can be a short-term solution, but usually last only a couple days, so in our setting they are usually only a bridge to something else. (This may, however, mean a “bridge” until the patient’s edema has decreased enough that nursing can get IVs the old-fashioned way.)
  • Maybe many of our recovering patients with central lines for “difficult access” should have midlines instead.