The Fast and the Furious: Intravenous Fluid Shift

Myth: When faced between a PICC / central line and a cannula for urgent fluid administration, the PICC / central line is the favoured option. 

Malady: In short, use whatever access you have – but if there are options, Pouiselle’s Law suggests using the 18G cannula if it’s reliable. Consider how badly you need that pesky bung. Clinically however, the meaningfulness of this difference is perhaps a bit unsatisfying.

---

Two vascular access devices walk into a bar, looking for a swift fill. The elaborate central venous line and the unassuming cannula. Who do you serve first when time is of the essence? It is tempting to go for the central line – they are assured paths, generally large bore, and are already there to be used. But of course, there is more than meets the lumen. 

This quick-and-dirty piece is all about how to get fluids into a patient in the most efficient way – both mathematically and practically speaking.

LES MATHÉMATIQUES

The guiding principle here – that of laminar flow – is Pouiselle’s Law. In real-life settings, it becomes more of an approximation than a law, owing to assumptions that do not hold true with the chaos of human physiology. For example, it applies to Newtonian fluids – which blood is not. Our blood vessels, despite decades of neglect and Smith’s potato chips, are never rigid tubes. It also doesn’t apply strictly to turbulent flow, which is more akin to the truth in an emergency setting. Regardless, it’s good enough as a starting point and the only fluid dynamics I can pretend to understand, so let’s keep moving.

LA PRATIQUE

Gauge / Radius

As you can see, the gauge (i.e. radius) is of critical importance to flow rates, and this is stressed endlessly from IV access principles in resuscitation to radiographers and their demands for contrast injections. Straightforward. Radius is important but not the quartic level implied from the equation both in real-life testing [1-3] and manufacturer specifications.

Length of the Catheter

A longer catheter proportionally increases resistance, and is inversely proportional to flow rate. While it may seem counter-intuitive to pick the cannula over the PICC line, the long and winding road that leads to the atrium is a semi-rigid tube that considerably increases resistance and hinders flow. This trial by EDProcedures [2] shows the magnitude of the effect – a 14G lumen of a CVC with shorter-than-PICC tubing was considerably slower than a regular 18G cannula. In Berman’s 2020 trial of various options [1], a PICC fared worse to a 24G cannula! Recent improvements with the advent of power/pressure-injectable PICCs may eventually make this point moot, but it’s hard to find clear guidance on this scenario. This is the principle behind a RIC line (rapid infusion catheter) that is generally reserved for major traumas needing aggressive blood volume replacement. 

PICCs can rupture under high flow demands, often due to operator error [4], which is an additional concern you do not want in your unwell patient. A study [5] does report it’s hard to rupture a PICC at pumpset paces however. Not to forget that nurses understandably do not appreciate it when hasty, dirty doctor hands touch these lines. The physiological consequences of the fluid ending at the SVC with these central lines, as opposed to the elbow for cannulas, and its resultant haemodynamic effects are also not well-investigated.

In essence, cannulate the largest, most proximal vein that is safe. Shorter infusion sets may also help, reducing vascular ‘dead space’ and resistance. In an emergency, should you trim the length of the cannula before insertion? Thanks to Jayanti and Dabke’s diligent work [6], we know that the promised theoretical benefits do not really translate to clinically significant superiority – turbulent flow strikes again.

Viscosity

Viscosity of the fluid influences flow. Blood is more viscous than crystalloid, and will take longer to flow in. This is a profound difference: RBCs flow in almost 5 times slower than saline [1], so you can imagine that a larger lumen is of utmost importance in haemorrhagic settings to offset this disadvantage.

Pressure Gradient

If you maximise the pressure gradient between the cannula and the bag (e.g. use gravity to position the bag higher, compression around the bag, etc.) then you attain more rapid flow. Simples.

Resistance Multipliers

Consider the infusion kit you are using as they are rated for different flow rates. And the hidden demon – the BUNG! This unassuming intermediary, often all that stands between cannula and the outside world, actually slows flow rates quite significantly (quoted rates vary from 40% to 85%!) [1-2, 7-8].

The bigger the gauge, the more the bung is holding you back. This is likely due to the slit-like labyrinth within a bung that fluid must navigate to emerge on the other end [2]. But of course, the bung is also holding back extravasation when nothing is connected, so take that with a pinch of salt in non-urgent settings to bypass becoming a hospital pariah.

Lumens

More the merrier? Yes, for instance, two 20Gs beat one 18G [9].

So in summary, to maximise your flow rate in a critical emergency – more lumens, larger lumens, shorter lines and pump sets, and do away with the bung. A less viscous solution flows faster, but this is a poor reason to replace blood loss with crystalloid. For those wondering about IO – I know, cover your eyes and ears educated ortho bros – flow rate is bone dependent, but seems to cap out at 100ml/min [10]. Not all solutions are suitable for IO administration, which further complicates this.

LA RÉALITÉ

After all of this talk about the HOW, there must be a mention about the WHY. Is marginally faster fluid resuscitation absolutely critical or a pre-terminal sign of a faltering blog trying to find things to talk about? It’s a great question, one you should look into and report back on tomorrow’s rounds. 

In seriousness, a consensus of intensivists determined that a fluid bolus is a relatively tame 250mls of crystalloid within 30 minutes [11]. This is achievable with a 22G that slots into a small vein, infused through gravity without needing pressure. 
DerangedPhysiology [12] provides a treatise about this topic, as does Alves et al [13], but there is some inconsistent evidence about this. An assertion is that slightly slower rates may be more effective at staying within the intravascular compartment. But it is tricky to bring into reality, when dealing with vasoplegic blood pressures of 60mmHg. All bets are off in haemorrhaging patients such as in trauma or massive GI bleeding – for settings where you are squeezing the bag of red cells yourself or squeezing it through the IV hand pump set, it is worthwhile to not shoot yourself in the foot by infusing it through a PICC – or a bung for that matter!

Giant’s Shoulders:
[1] Berman, D. J., Schiavi, A., Frank, S. M., Duarte, S., Schwengel, D. A., & Miller, C. R. (2020). Factors that influence flow through intravascular catheters.
[2] Check out this detailed Aussie attempt at a real-world face-off (no PICCs, but has a CVC): Large bore IV access showdown: does short and thick really do the trick? – ED Procedures and https://edprocedures.etmcourse.com/large-bore-iv-access-showdown-round-2/ 
[12] Applied physiology of intravenous fluid replacement 

Also cited above:
[3] Fluid administration device flow rates
[4] Power PICCs for CT https://www.jvir.org/article/S1051-0443(07)60412-6/abstract
[5] Hard to rupture PICCs under infusion pump pressures: Royer, T. (2012). Maximum Flow Rates Achievable Through Peripherally Inserted Central Catheters Using Standard Hospital Infusion Pumps. Journal of the Association for Vascular Access
[6] A delightful study of trimming cannulas The effect of IV cannula length on the rate of infusion – PubMed
[7] Bungs & slowing flow rate part 1: Reddick, A. D., Ronald, J., & Morrison, W. G. (2010). Intravenous fluid resuscitation: was Poiseuille right? Emergency Medicine Journal, 28(3), 201–202. doi:10.1136/emj.2009.083485 
[8] Bungs & slowing flow rate part 2: https://pubmed.ncbi.nlm.nih.gov/25695674/
[9] 2 cannulas are better than one, duh https://pubmed.ncbi.nlm.nih.gov/20637391/
[10] LITFL re: IO access Intraosseous access
[11] Intensivists debating what constitutes a fluid bolus: https://www.sciencedirect.com/science/article/pii/S0883944116300892
[13] Alves’ et al’s in-depth delve into fluid bolus physiology: https://www.sciencedirect.com/science/article/pii/S0883944123000448