Author: Mark Winstanley, Becky Maxwell, Chris Connolly, Andy Neill, Dave McCreary/ Codes: NeuP2, RP4, RP7, SLO3 / Published: 05/04/2022



Andy Neill
Dave McCreary

Clinical Question

– can we put IV lines in the arms of patients who have had prior axillary node clearance.

Paper Title

– Ipsilateral Intravenous Catheter Placement in Breast Cancer Surgery Patients


– Naranjo 2021, Anaesthesia Analgesia


– Breast cancer is common and commonly patients will have axillary node clearance as part of surgical treatment. Given the lack of lymph nodes patients are advised to be careful with the affective arm to the extent of not letting venepuncture or cannulation or even BP checks in that arm.
– the risk of a cellulitis or lymphoedema due to IV cannulation is unclear and given that preventing IV access in that arm effectively reduces our access by 50% it would be nice to know if the patient actually does explode if we put a cannula in the arm where there has been axillary node clearance.
– the astute among you will note that we have looked at this before in New in EM way back in June 2016 but that paper just looked at venepuncture rather than IV lines.


– this was a single centre retrospective study (read: chart review) in the states. Not a great start
– they looked at their electronic record for all people who had surgery with a prior history of breast cancer over a 3 year period
– they excluded people who already had lymphoedema (which seems very reasonable)
– they divided patients into those with a history of node clearance and those without and they then looked for a variety of ICD codes during their stay to see whether there were complications potentially related to the IV line.
– so you could be having a hip replacement and have a history of prior mastectomy and ANC and get into this study


– they found 3700 such patients
– most surgeries were occuring withint 2 years of the original breast cancer surgery
– 2/3 of patients had an ipsilateral IV line and 1/3 contralateral. Initially that stuck me as “wow these guys are crushing the dogma” but the vast majority of the ipsilateral placements were patients with a bilateral ANC so they had no real option!
– they found 4 complications out of roughly 8000 IV placements split between ipsilateral and contralateral IV placements. Details are given and they are what seems like minor lymphoedema complaints rather than necrotising fascitits and awfulness


– methodology is obviously severely limited by its retrospective nature but it seems that complications are really very rare.
– this is actually not that controversial. There is an expert consensus guideline from a group in the US who have (somewhat confusingly said) that IV access and BP measurements “are not contraindicated”. The message just doesn’t seem to have gotten through quite yet.

Clinical Question

Is saline as evil as we’ve been taught in trauma resuscitation?

Title of Paper

Resuscitation with blood products in patients with trauma-related haemorrhagic shock receiving prehospital care (RePHILL): a multicentre, open-label, randomised, controlled, phase 3 trial

Journal and Year

Lancet. 2022.

Lead Author

Nicholas Crombie.


– We’ve been taught for the last 20 years or so that normal saline is bad for trauma patients: it can’t carry oxygen, it’s acidotic (pH 5.5) and it’s diluting the red cells and clotting factors that the patient has left.
– We’ve had studies looking at what ratio of products we should be giving in haemostatic resuscitation, we’ve had studies looking at which of our colourful products we should be giving first (with the PAMPer trial showing an impressive advantage to plasma-first), but we’ve not really had a randomised study looking at products vs saline.

Study Design

– Multicentre, open-label RCT across several air-ambulance services in the UK

Patients Studied

– Over 16 years
– Traumatic injury
– Hypotension (SBP <90) or absent radial pulse
– Believed by the clinician to be due to traumatic haemorrhage


– Blood products: 1xPRBC→1xPlasma→1xPRBC→1xPlasma
– To maximum of 4 units after which further fluid if indicated was non-trial 0.9% saline


– 0.9% Saline: 250ml → 250ml → 250ml → 250ml
– To maximum of 4 units after which further fluid if indicated was non-trial 0.9% saline


– Primary outcome (powered for 10% difference) was episode of:
– Death from time of injury to discharge from acute care
– OR failure to clear lactate by >20% in first 2 hours
– OR both
– Secondary included the individual components of primary and all-cause mortality 3h & 30 days, amongst many others

Summary of Results

– Stopped short of sample size (COVID 🙄), reaching 432 of the 580 target
– Most were white (62%) males (82%) median age 38
– They were sick:
– average BP 73/46
– median ISS 36 [IQR 25-50]
– Primary outcome occurred in 64% of intervention and 65% of the control group
– Adjusted risk ratio 1.01 [0.88-1.17]
– Secondary outcome:
– Trend towards benefit of blood products for mortality at 3 hours (16 v 22%, adjusted OR 0.75 [0.5-1.13]).

Authors Conclusion

the RePHILL trial did not demonstrate that PRBC-Lyoplas improved episode mortality or lactate clearance when compared with 0.9% sodium chloride for participants with trauma-related hemorrhagic shock. Based on current evidence, the decision to commit to routine prehospital transfusion in civilian practice will require careful consideration by all stakeholders.

Clinical Bottom Line

This is an impressive trail that required some serious work to get done.

There are a couple of issues though:

– The use of the composite end-point with one clinical and one lab-based – I certainly don’t consider lactate clearance to be on power with patient mortality.
– It is underpowered to show a difference of <10%. In this patient group I would argue a difference of even up to 5% would be considerable (though I accept the sample size required would be prohibitive).
– This is a study of prehospital patients who are not at the same point in their journey as those in our trauma bays.

This study certainly isn’t telling you to stop using blood products in your resuscitation of the shocked trauma patient, but it should perhaps be reassuring for those who don’t have immediate access to blood products that normal saline may not be quite as detrimental as we’ve previously thought. As with many studies, it should us that there is no one-size-fits-all approach to trauma resuscitation – more research is certainly needed to narrow down exactly which patients benefit the most from which products.

Other #FOAMed Resources / References

The recorded livestream from critical care reviews with the authors talking through the study, an excellent editorial by Simon Carly and an interesting panel discussion is really worth watching.