Then the VBG comes back.

Lactate: 3.1 mmol/L.

And suddenly Gary is getting a sepsis workup, a 30ml/kg fluid bolus is being contemplated, and someone is calling the ICU at a regional centre three hours away.

His chest X-ray shows a right lower lobe consolidation. He has pneumonia. He is, by the loosest technical definition, septic.

Here’s the thing. Gary has also been huffing away on his salbutamol inhaler for two hours.

What Lactate Actually Is

Lactate is produced constantly, by nearly every cell in your body. It’s not a waste product — it’s a fuel. The heart, brain, and skeletal muscle all use lactate as an energy substrate. At rest, a healthy person produces around 1500 mmol of lactate per day, clearing it mostly through the liver and to a lesser degree the kidneys and heart.

The normal serum range is up to around 2 mmol/L, though this varies by lab and method.

The classic teaching — lactate rises when tissues are underperfused, switch to anaerobic metabolism, pyruvate backs up, and lactate floods the blood — is Type A lactic acidosis. Shock. Ischaemia. Real badness.

But Type A is only a small part of the story, and arguably the less interesting part. To be honest, picking a patient with type A lactic acidosis is pretty easy in practice.

The Fallacies

Fallacy 1: Elevated lactate = tissue hypoxia = underperfusion = needs more fluid.

This is the one that causes the most harm. Type B lactic acidosis has nothing to do with oxygen delivery. Lactate rises in the absence of any perfusion problem at all. Common culprits:

• Beta-agonists — salbutamol, adrenaline. To understand why, it helps to know about the Cori cycle: the normal housekeeping loop where lactate produced in peripheral tissues — working muscle, red blood cells — is shuttled to the liver, converted back to glucose via gluconeogenesis, and returned to circulation. It’s an elegant recycling system, and under normal conditions it keeps lactate levels tightly controlled.
• Beta-2 receptor stimulation short-circuits this. Activation of beta-2 receptors on skeletal muscle cells drives Na⁺/K⁺-ATPase activity, which ramps up ATP consumption and in turn accelerates glycolysis. More pyruvate is produced than the mitochondria can handle — so lactate dehydrogenase mops up the excess, converting it to lactate. This happens with adequate oxygen delivery — it’s aerobic glycolysis, the same metabolic trick tumour cells use (the Warburg effect). The Cori cycle tries to keep up, but hepatic clearance has a ceiling. A few salbutamol nebs and a lactate of 2.5–4 mmol/L is a completely expected pharmacological consequence. It is not a distress signal from your patient’s tissues — it’s a readout of their medication.
• Infection and systemic inflammation — activated neutrophils, macrophages, and other immune cells are metabolically ravenous. They upregulate aerobic glycolysis as part of their inflammatory response — producing lactate not because they’re starved of oxygen, but because rapid ATP generation via glycolysis suits their function. Add the endogenous catecholamine surge of any acute illness and you have another beta-2 mediated push on top. A lactate of 2–3 in a patient with pneumonia who is otherwise haemodynamically stable may simply reflect an immune system doing its job.
• Thiamine deficiency — blocks pyruvate dehydrogenase, pyruvate can’t enter the Krebs cycle, and shunts instead to lactate. Relevant in your malnourished, alcohol-dependent, or post-bariatric patients.
• Metformin toxicity — inhibits complex I of the mitochondrial respiratory chain, again shunting pyruvate away from oxidative metabolism.
• Liver disease — impaired clearance. The Cori cycle depends on a functioning liver. In cirrhosis or acute hepatic failure, lactate may be elevated simply because the recycling system is broken, not because production is excessive.
• Malignancy — Warburg physiology again. Tumour cells produce lactate aerobically as a feature, not a bug.
• Seizures — transient massive rise from muscular activity, self-resolving within an hour or so.

In Gary’s case — COPD exacerbation, pneumonia, salbutamol nebs — he has at least three concurrent Type B mechanisms running simultaneously. The lactate is over 3 – the result of multiple pathways, none type A hypoxemia. Gary himself is sitting up, talking in full sentences, and asking when he can have a cup of tea.

Fallacy 2: Lactate clearance is a resuscitation target.

The Surviving Sepsis Campaign’s use of serial lactate as a guide to resuscitation adequacy has embedded this deeply. And there is an association between lactate normalisation and improved outcomes in septic shock. But association is not mechanism, and lactate clearance is almost certainly a marker of physiological recovery, not a driver of it. Treating the number — giving more fluid, more pressors — because the lactate hasn’t normalised yet is treating a proxy, not the patient.

The most recent (2026) Surviving Sepsis guidelines reframed lactate clearance to specifiy that it only applies in patients who are already in septic shock. So the sickest of the sick. Not Gary.

The ANDROMEDA-SHOCK trial found peripheral perfusion targets were non-inferior to lactate-guided resuscitation. The point is not that lactate is useless — it’s that it tells you something has gone wrong somewhere, not specifically that you need to pour in more saline.

Fallacy 3: Any lactate above 2 is clinically significant.

The Sepsis-3 criteria use >2 mmol/L as a criterion for septic shock in a patient with suspected infection and vasopressor requirement. That’s an appropriate clinical context with a narrow application. It is not a population-level trigger for aggressive resuscitation in any undifferentiated patient who looks vaguely unwell.

Spurious elevations from delayed sample processing, tight tourniquet application, prolonged transport in the sample tube — these are all real. Lab artefact can push a normal lactate into the “abnormal” range without a single struggling mitochondrion involved.

What to Actually Do

When you see an elevated lactate, ask yourself five questions before you reach for the fluid bag:

1. Is this patient in shock? Not the number — the patient. Mentation, skin perfusion, cap refill, pulse pressure, urine output. Treat the human.
2. What’s the clinical context? Beta-agonists on board? Liver disease? Metformin? Seizure in the last hour? Known malignancy? Active infection driving an inflammatory response?
3. Is there a perfusion problem I’m missing? Mesenteric ischaemia, massive PE, cardiogenic shock — these do cause Type A lactataemia and deserve urgent attention.
4. What does the trajectory look like? A lactate trending down in a patient who is improving clinically is reassuring. A lactate climbing despite initial resuscitation in a sick patient is alarming. Context and direction matter more than a single value.
5. Will volume actually help, or possibly harm my patient? Before embarking on a large volume resuscitation it behooves us to spend a minute thinking about where our patient sits on the spectrum of fluid tolerance and the possible downsides of “another litre”. This does require one to get off of one’s chair, but it can save a lot of time down the line if we get this right.

Back to Gary

Gary got a salbutamol cease, appropriate antibiotics for his pneumonia, a repeat lactate two hours later of 1.8, and admission for treatment of his chest infection. He did not need 3L of normal saline, a central line, or an ICU transfer.

He had pneumonia. He needed antibiotics. His lactate told us his immune system and his puffer were both working — not that his organs were failing.

Lactate is a useful tool with a narrow range of meaningful applications. It is not a substitute for clinical assessment, and a mildly elevated result should prompt thinking, not reflexive escalation.

Know what the number means. Know what it doesn’t.

Casey

Further reading

A fantastic PK presentation on Lactate in ED by Dr Laura Castle in 2013 CLICK HERE

• Garcia-Alvarez et al.: Garcia-Alvarez M, Marik P, Bellomo R. Sepsis-associated hyperlactatemia. Crit Care. 2014;18(5):503. doi: 10.1186/s13054-014-0503-3. PMID: 25394679
• Hernandez et al. (ANDROMEDA-SHOCK): Hernández G, Ospina-Tascón GA, Petri Damiani L, et al. Effect of a Resuscitation Strategy Targeting Peripheral Perfusion Status vs Serum Lactate Levels on 28-Day Mortality Among Patients With Septic Shock: The ANDROMEDA-SHOCK Randomized Clinical Trial. JAMA. 2019;321(7):654–664. doi: 10.1001/jama.2019.0071. PMID: 30772908
• Levy B. Lactate and shock state: the metabolic view. Curr Opin Crit Care. 2006;12(4):315–321. doi: 10.1097/01.ccx.0000235208.77542.39. PMID: 16810041
• Kraut & Madias: Kraut JA, Madias NE. Lactic acidosis. N Engl J Med. 2014;371(24):2309–2319. doi: 10.1056/NEJMra1309483. PMID: 25494270
• Prescott HC, Antonelli M, Alhazzani W, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2026. Crit Care Med. 2026;54(4):725–812. DOI: 10.1097/CCM.0000000000007075

Not your standard SMACC talk. This one ranges over the neurobiology of empathy, the social science and the more practical modern application of empathy in clinical medicine. You can learn a lot of history too – its a bit of a whistle-stop tour through some of the dark days of western culture.

So grab a cuppa and sit down for this 30 minute chat.

And somewhere in the room, the hesitation begins.

Not in you, necessarily. But you’ve felt it. The half-second of oh when a young woman arrives after a significant mechanism. The faintly spoken concern from a nurse. The instinct — culturally baked in, not evidence-based — to go carefully because of a possible early pregnancy. The creeping thought: what if I harm the baby?

I want to talk about that hesitation. Because it has a body count.

The problem isn’t radiation. The problem is us.

In 2024, a study of emergency department physicians found that 88.7% reported finding the management of pregnant trauma patients challenging, and expressed a preference to avoid these cases. Nearly nine in ten. Not first-year residents — physicians. And the consequence was exactly what you’d predict: delays, selective under-imaging, and practice variability that had nothing to do with the evidence and everything to do with discomfort.

Trauma is the leading non-obstetric cause of maternal death. Motor vehicle collisions account for more than half of trauma in pregnant women. The fetus does not survive maternal exsanguination. There is no version of this story where protecting the fetus means under-resuscitating and under-imaging the mother.

The best treatment for the fetus is optimal treatment of the mother. This isn’t a controversial statement. It’s in ATLS. It’s in the ACOG guidelines. It’s in the ACEM/RANZCR joint imaging guidelines. It’s the unanimous position of every relevant professional body on the planet. And yet here we are, with 88.7% of emergency physicians describing a preference to avoid these patients.

That’s not caution. That’s a knowledge gap dressed up as sensitivity.

Let’s actually talk about the radiation.

The fear is deterministic effects — the real, measurable harm that radiation causes to a developing embryo. Microcephaly. Organ malformation. Growth restriction. Death of the conceptus. These are real. They happen. And they have a threshold below which they simply do not occur.

That threshold is 100 mGy to the fetus.

Above 100 mGy, the risk curve climbs steeply. Below it, deterministic effects are not observed. This isn’t a comfortable regulatory threshold with a safety factor bolted on — it’s based on the actual biology of radiation-induced tissue damage. Below 100 mGy, there are no fetal anomalies, no increase in pregnancy loss. The ACR, ACOG, and the National Council on Radiation Protection all say so explicitly.

Now. What does a trauma CT actually deliver to the fetus?

The RANZCR, in their Diagnostic Radiology and Pregnancy position statement, cites Australian data: CT of the abdomen delivers a mean uterine dose of 7.1 mGy. Range 1–40 mGy. A whole-body trauma CT in a pregnant patient in a French multicentre study delivered a median fetal dose of 38 mGy. A single CT scan — even a full whole-body study — cannot reach 100 mGy.

Not won’t in most cases. Cannot. The ICRP states explicitly that the 100 mGy threshold cannot be reached with a single CT examination.

So the conversation about deterministic effects from a single trauma CT is over. Done. The risk does not exist at diagnostic doses.

But what about stochastic risk?

Here’s where people get tangled. Stochastic effects — principally childhood leukaemia and solid cancers — have no known safe threshold. Risk increases linearly with dose. There is no floor. And this is where the ALARA principle lives: as low as reasonably achievable, not because diagnostic doses cause deterministic harm, but because we should minimise cumulative stochastic burden where we can do so without compromising the diagnostic objective.

Here’s the important word: where we can.

The background rate of childhood leukaemia is approximately 1 in 3,000. A 10–20 mGy fetal exposure may double that risk. Which means an absolute risk of perhaps 2 in 3,000. An additional 1 in 3,000. In a child who will only exist if their mother survives long enough to deliver them.

ALARA does not mean avoid. ALARA does not mean hesitate when a woman’s spleen is haemorrhaging. ALARA means apply dose reduction techniques, involve your radiographer, document the estimated dose, and have a counselling conversation — after the scan, not instead of it.

The β-hCG question

Should you check a β-hCG before scanning? Sure, if it’s immediately available and won’t delay anything. It may be useful for counselling. It tells you nothing about whether to image.

A positive β-hCG at four weeks does not lower the CT threshold. A negative β-hCG does not raise it. The clinical question is the same regardless: does this mechanism and these clinical signs warrant cross-sectional imaging? Answer that question using standard trauma criteria, and then scan if indicated. The pregnancy status is relevant to the conversation you have afterward, not the decision you make now.

If the woman is haemodynamically unstable, you don’t have time for the conversation anyway. You have time for the scan.

So what is the risk of missing an injury if we do not CT?

This is a tricky question to answer. There is not a lot of data on pregnant women specifically. However if we look at the best, most recent Australian data (11) – the numbers are convincing. The risk of an important occult chest or abdominal injury in patients (even those with normal exam and CXR) is high – someting like 20 -25%.

Up to 19% of patients with normal abdominal examinations were later found to have evidence of injury on their CT scan, and significant injuries detected were otherwise clinically occult, including blunt aortic injury. This is corroborated by non-Australian data that nonetheless speaks directly to the clinical argument.

In population-level studies of pregnant women involved in motor-vehicle collision, maternal death occurred in 3.6 per 1,000 women and fetal death or stillbirth in 6.6 per 1,000. These numbers are about 10-20 times higher than the theoretical risk of CT-induced future childhood malignancy.

In blunt trauma patients with normal sensorium and normal physical examination, 15% were found to have an occult chest injury on CT despite a normal chest exam, 35% had occult injuries on chest CT despite a negative chest exam and negative chest X-ray, and 16% had occult abdominal injuries despite a negative abdominal exam. Overall, 25% of patients with normal chest, abdomen, and pelvis examinations had occult injuries detected on CT.

So, keep that in mind when thinking about risk. We are comparing a possible 1:3000 risk of future badness to a 1:4 or 1:5 risk of missing current real badness. Not even close.

Why not Ultrasound?

Ultrasound is awesome.  No radiation, no risk right?  Unfortunately, even in the best hands ultrasound cannot give us the answers that we need in women suffering traumatic injury.  We know that US is inferior to CT for injuries such as solid organ and visceral bleeding.  However, it is also very insensitive for uterine injury and abruption – which are common (30-40%) of major trauma in the 3rd trimester.  CT actually has a very high pick up rate for uterine injury and placental abruption.  Obviously we do not use CT to diagnose abruption in non-traumatic scenarios as we can use clinical assessment, CTG and US to guide the management.  However, when trauma happens in later pregnancy the risk:benefit ratio is definitely towards CT as it can pick all the other injuries as well as abruption.  So it is a no-brainer really

The rural reality

I want to be direct about something that doesn’t appear in the guidelines, because guidelines are written for trauma centres with consultant radiologists on call, obstetric teams available, and someone to second-guess you in a collegial way.

In remote practice, you are the protocol.

You are the trauma team, the intensivist, the obstetrician, and the surgeon. There is no second opinion arriving in twenty minutes. The RFDS is on its way and the window for definitive diagnosis is the window you create right now, with the tools you have.

Withholding a CT because of radiation anxiety in that setting doesn’t make you cautious. It makes you blind. An occult solid organ injury, an aortic haematoma, a mesenteric tear — missed because someone was worried about a stochastic risk of 1 in 3,000 in a fetus that may not even implant — is an entirely avoidable tragedy.

The senior emergency physicians in the study above were the ones most likely to scan regardless of gestational age. The hesitancy was concentrated in those without specific training or experience in this area. That’s important information about where the educational gap is.

A note on shielding

The lead apron on the abdomen of a pregnant trauma patient is well-intentioned and often unhelpful. Most of the fetal dose in trauma CT comes from internal scatter — radiation that has already passed through the maternal body and is bouncing around internally. External shielding does not significantly reduce that scatter. More importantly, some modern CT scanners use automated dose modulation — if the detector senses lead shielding, it may increase tube current to compensate, actually raising the dose.

Shielding is not harmful, and if it reassures the patient and clinician without delaying or compromising the scan, fine. But it is not a mitigation strategy. It’s a ritual comfort.

The synthesis

The RANZCR position statement says it plainly: most diagnostic radiological procedures pose no substantial risk to the fetus. The ACEM and RANZCR joint imaging guidelines support CT as the investigation of choice in the trauma context. The WA statewide cohort data — from Fiona Stanley and Royal Perth — found that even in trauma populations receiving multiple CTs, the excess cancer signal was modest against the background of trauma mortality itself.

The literature is not ambiguous. The guidelines are not equivocal. The professional bodies are aligned. And the 88.7% of emergency physicians who describe this as challenging are not reflecting on a genuinely difficult evidence question — they are describing a gut reaction that is costing lives.

Scan her.

Reference List

Australian and New Zealand Sources

1. Royal Australian and New Zealand College of Radiologists. Diagnostic Radiology and Pregnancy — Clinical Radiology Position Statement. Version 2.0. November 2017. [Under review May 2026.] Available: https://www.ranzcr.com/college/document-library/diagnostic-radiology-and-pregnancy-policy (PDF direct link: https://www.ranzcr.com/wp-content/uploads/edocman/professional-documents/policies/Diagnostic%20Radiology%20and%20Pregnancy%20Position%20Statement.pdf)

2. Australasian College for Emergency Medicine (ACEM); Royal Australian and New Zealand College of Radiologists (RANZCR). Guidelines on Diagnostic Imaging. NSW Agency for Clinical Innovation. Available: https://aci.health.nsw.gov.au/__data/assets/pdf_file/0007/273715/ACEM-Guidelines-diagnostic-imaging.pdf

3. Yaw LK, Song S, Ho KM. Dose-related association between radiation exposure from computed tomography (CT) scans during trauma hospitalisations and subsequent risk of developing new-onset cancers: statewide cohort study, Western Australia 2004–2020. Communications Medicine 2026 Jan 5. doi: 10.1038/s43856-025-01354-z PMC: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12881511/ (Fiona Stanley Hospital / Royal Perth Hospital, WA)

International Guidelines and Position Statements

4. American College of Obstetricians and Gynecologists. Committee Opinion No. 723: Guidelines for Diagnostic Imaging During Pregnancy and Lactation. Obstet Gynecol 2017;130(4):e210–e216. doi: 10.1097/AOG.0000000000002355 PMID: 28937575 Available: https://www.acog.org/clinical/clinical-guidance/committee-opinion/articles/2017/10/guidelines-for-diagnostic-imaging-during-pregnancy-and-lactation

5. Qamar SR, Green CR, Ghandehari H, Holmes S, Hurley S, Khumalo Z, Mohammed MF, Ziesmann M, Jain V, Thavanathan R, Berger FH. CETARS/CAR Practice Guideline on Imaging the Pregnant Trauma Patient. Can Assoc Radiol J 2024 Nov;75(4):743–750. doi: 10.1177/08465371241254966 PMID: 38813997 PubMed: https://pubmed.ncbi.nlm.nih.gov/38813997/

6. McCollough CH, Schueler BA, Atwell TD, Braun NN, Regner DM, Brown DL, LeRoy AJ. Radiation exposure and pregnancy: when should we be concerned? RadioGraphics 2007 Jul–Aug;27(4):909–917. doi: 10.1148/rg.274065149 PMID: 17620458 Available: https://pubs.rsna.org/doi/10.1148/rg.274065149

Epidemiology and Clinical Practice Data

7. Abback PS, Benchetrit A, Delhaye N, Daire JL, James A, Neuschwander A, Boutonnet M, Cook F, Vinour H, Hanouz JL, Cotte J, Pastene B, Jouffroy V, Gauss T; Traumabase Group. Multiple trauma in pregnant women: injury assessment, fetal radiation exposure and mortality. A multicentre observational study. Scand J Trauma Resusc Emerg Med 2023 May 2;31(1):22. doi: 10.1186/s13049-023-01084-y PMID: 37131266 PMC: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10152762/

8. Karaaslan A, Ipekci A, Karaaslan Y. A Flowchart to Guide Emergency Physicians to Order Radiological Imaging in Pregnant Patients: Findings from an Emergency Department Questionnaire. Healthcare 2025 Dec 2;13(23):3138. doi: 10.3390/healthcare13233138 PMC: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12692375/ (Source of the 88.7% hesitancy figure)

9. MacDermott R, Berger FH, Phillips A, Robins JA, O’Keeffe ME, Abu Mughli R, MacLean DB, Liu G, Heipel H, Nathens AB, Qamar SR. Initial Imaging of Pregnant Patients in the Trauma Bay — Discussion and Review of Presentations at a Level-1 Trauma Centre. Diagnostics 2024 Jan 26;14(3):276. doi: 10.3390/diagnostics14030276 PMID: 38337792 PMC: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10855036/

10. Lucas AN, Tay-Lasso E, Zezoff DC, Fierro N, Dhillon NK, Ley EJ, et al. Significant variation in computed tomography imaging of pregnant trauma patients: a retrospective multicenter study. Emerg Radiol 2024;31:69–77. doi: 10.1007/s10140-023-02195-w PMID: 38148383 PMC: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10830714/

11. Salam Findakly, Adil Zia, Helen Kavnoudias, Joseph Mathew, Dinesh Varma, Bruno Di Muzio, Robin Lee, Heather K Moriarty, Tim Joseph, Warren Clements.
The use of whole-body trauma CT should be based on mechanism of injury: A risk analysis of 3920 patients at a tertiary trauma centre, Injury, Volume 54, Issue 7, 2023, 110828,
https://doi.org/10.1016/j.injury.2023.05.059.

Justin also has a great schema for discussing radiation risk with women over at First10EM

If this interests you then you should also check out Justin’s post on “When Norepinepherine isn’t working…”

The new Docnomo / Fagan nomogram interactive tool on the website is unveiled and demonstrated with discussions around PE, small bowel obstruction and paediatric appendicitis.

Enjoy the nerdiness!