Blunt Abdominal Trauma — Laparotomy or Non-Operative Management

Adults and children are different problems, and this tool asks which you have. NICE splits several recommendations by age, and the whole non-operative-management literature treats children separately (a much higher proportion of paediatric blunt splenic and liver injury is managed without surgery, in dedicated paediatric centres). Set the age in step 1; the outputs change with it.

Where this fits. This module covers the resus-room decisions for suspected blunt abdominal injury: who needs an immediate laparotomy, imaging by haemodynamic status, non-operative management of solid-organ injury, and the hollow-viscus injury the first scan can miss. It does not replace your major haemorrhage protocol, the trauma team, or the general and interventional-radiology teams you will be calling. For the pelvis as a bleeding source see the pelvic trauma module; for the protocol engine itself, the major haemorrhage module; and for reversing an anticoagulant, the bleeding & reversal module.

Age and mechanism

Age changes the imaging and non-operative-management defaults. The abdominal pathway is written for blunt trauma; penetrating injury is a different algorithm.

AGE

MECHANISM

Haemodynamic status — and the response to resuscitation

This single axis decides the imaging route and whether the abdomen can be watched or must be opened. NICE and WSES both use the response to volume, not one blood-pressure reading.

A transient responder is an unstable patient. WSES states it explicitly: patients who respond to fluid then show ongoing loss, and those who cannot be stabilised enough to reach interventional radiology, “are to be considered as unstable patients.” A young patient with a splenic bleed can hold a normal blood pressure on a rising heart rate and then collapse. Do not let a transient response buy a trip to the CT scanner it should not have.

Is an immediate laparotomy indicated?

These override the imaging pathway. Any one of them, in the right patient, sends them to theatre — not to the scanner.

The classic combination is haemodynamic instability plus free fluid. An unstable patient with free intraperitoneal fluid on FAST, with no other explanation for the shock, is a laparotomy — the FAST here is used to direct intervention, not to decide whether to scan. Peritonitis, evisceration and impalement are laparotomy indications in their own right (WSES splenic guideline, GoR 1A).

Findings that raise suspicion of intra-abdominal injury

For the patient without an immediate operative indication, these decide how hard you look — and how long you keep looking.

The seatbelt sign is the one that catches people out. A linear bruise across the abdomen from a lap belt is associated with hollow-viscus and mesenteric injury that a first CT can under-call. It is not proof of injury, but it lowers the threshold for repeat imaging and serial examination — see the result and the reference panel below.

Context

Changes the resuscitation target, whether to reach for reversal, and where the patient goes next.

The pathway for this patient

Destination, resuscitation, imaging and what to watch for, each reported separately and recalculated as you change anything above.

Answer step 1 to begin.

This tool needs JavaScript. With it switched off, no result will be calculated. The recommendations are set out in full in the reference panels below — use those, or the source guidelines, directly.

The reference

Imaging by haemodynamic status — and what FAST can and cannot do

NICE NG39

“Limit diagnostic imaging (such as chest and pelvis X‑rays or FAST [focused assessment with sonography for trauma]) to the minimum needed to direct intervention in patients with suspected haemorrhage and haemodynamic instability who are not responding to volume resuscitation.” (1.5.29)
“Be aware that a negative FAST does not exclude intraperitoneal or retroperitoneal haemorrhage.” (1.5.30)
“Consider immediate CT for patients with suspected haemorrhage if they are responding to resuscitation or if their haemodynamic status is normal.” (1.5.31)
“Do not use FAST or other diagnostic imaging before immediate CT in patients with major trauma.” (1.5.32)
“Do not use FAST as a screening modality to determine the need for CT in patients with major trauma.” (1.5.33)

Whole-body CT

“Use whole‑body CT (consisting of a vertex‑to‑toes scanogram followed by a CT from vertex to mid‑thigh) in adults (16 or over) with blunt major trauma and suspected multiple injuries. Patients should not be repositioned during whole‑body CT.” (1.5.34)
“Do not routinely use whole‑body CT to image children (under 16s). Use clinical judgement to limit CT to the body areas where assessment is needed.” (1.5.36)

WSES, on the same axis

“The choice of diagnostic technique at admission must be based on the hemodynamic status of the patient” GoR 1A, and “CT scan with intravenous contrast is the gold standard in hemodynamically stable or stabilized trauma patients” GoR 1A (WSES splenic 2017).
“E-FAST is effective and rapid to detect free fluid” GoR 1A (WSES splenic 2017). Its job is to find free fluid in the crashing patient, not to clear the abdomen.

NICE NG39 recommendations 1.5.29–1.5.34, 1.5.36 · WSES splenic trauma guidelines 2017 (Coccolini et al., World J Emerg Surg 2017;12:40), diagnostic statements

When to open the abdomen now

NICE addresses this by physiology rather than by organ:

“Use damage control surgery in patients with haemodynamic instability who are not responding to volume resuscitation.” (NG39 1.5.37)
“Consider definitive surgery in patients with haemodynamic instability who are responding to volume resuscitation.” (1.5.38)
“Use definitive surgery in patients whose haemodynamic status is normal.” (1.5.39)

WSES names the abdominal indications directly:

“NOM in splenic injuries is contraindicated in the setting of unresponsive hemodynamic instability or other indicates for laparotomy (peritonitis, hollow organ injuries, bowel evisceration, impalement)” GoR 1A (WSES splenic 2017).
“Hemodynamically unstable and non-responder patients (WSES IV) should undergo OM” GoR 2A (WSES liver 2020).

Other findings that mandate laparotomy once identified: free intraperitoneal air (pneumoperitoneum), enteric contrast extravasation or a discontinuous bowel wall on CT, and diaphragmatic rupture. These are CT diagnoses — the resus-room gate is instability with free fluid, peritonitis, evisceration or impalement.

NICE NG39 recommendations 1.5.37–1.5.39 · WSES splenic 2017 statement (GoR 1A) · WSES liver 2020 statement (GoR 2A)

Non-operative management of the spleen and liver

Non-operative management (NOM) is the default for the haemodynamically stable patient, and — this is the counter-intuitive part — it does not depend on the injury grade.

Spleen (WSES 2017)

“Patients with hemodynamic stability and absence of other abdominal organ injuries requiring surgery should undergo an initial attempt of NOM irrespective of injury grade” GoR 2A.
“NOM of moderate or severe spleen injuries should be considered only in an environment that provides capability for patient intensive monitoring, AG/AE, an immediately available OR and immediate access to blood and blood product … and only in patients with stable or stabilized hemodynamic and absence of other internal injuries requiring surgery” GoR 2A.
“AG/AE may be considered the first-line intervention in patients with hemodynamic stability and arterial blush on CT scan irrespective from injury grade” GoR 2B.
“Injury grade on CT scan, extent of free fluid, and the presence of PSA do not predict NOM failure or the need of OM” GoR 1B.

Liver (WSES 2020)

“NOM should be the treatment of choice for all hemodynamically stable minor (WSES I) (AAST I–II), moderate (WSES II) (AAST III), and severe (WSES III) (AAST IV–V) injuries in the absence of other internal injuries requiring surgery” GoR 2A.
“Serial clinical evaluations (physical exams and laboratory testing) must be performed to detect a change in clinical status during NOM” GoR 2A.
“AG/AE may be considered as a first-line intervention in hemodynamically stable patients with arterial blush on CT scan” GoR 2B. In children, “the presence of contrast blush on CT scan is not an absolute indication for AG/AE” GoR 2B.
“NOM should be attempted in the setting of concomitant head trauma and/or spinal cord injuries with reliable clinical exam, unless the patient could not achieve specific hemodynamic goals for the neurotrauma and the instability might be due to intra-abdominal bleeding” GoR 2B.

NICE, on the interventional-radiology side

“Consider interventional radiology techniques in patients with solid‑organ (spleen, liver or kidney) arterial haemorrhage.” (NG39 1.5.41)

Who is more likely to fail NOM

Not an absolute contraindication, but a reason for a higher index of suspicion and closer monitoring: “age above 55-years old, high ISS, and moderate to severe splenic injuries are prognostic factors for NOM failure” GoR 2B (WSES splenic 2017). The spleen has a second trap the liver has less of: delayed rupture of a subcapsular haematoma or a pseudoaneurysm, days later.

WSES splenic trauma 2017 statements (GoR 1B–2B) · WSES liver trauma 2020 statements (GoR 2A–2B) · NICE NG39 recommendation 1.5.41

The hollow-viscus trap — when the first CT is normal

Solid-organ injury bleeds and shows on CT. Hollow-viscus and mesenteric injury, and the retroperitoneal duodenum and pancreas, can do neither early — and this is where blunt abdominal trauma is missed.

The seatbelt sign — a linear ecchymosis across the lower abdomen from a lap belt — is a recognised marker of bowel and mesenteric injury and should lower your threshold for repeat imaging and serial examination. (Association from standard trauma teaching and the trauma literature, not from NICE.)
WSES on the retroperitoneal organs: “A repeat CT-scan within 12–24 h from the initial injury should be considered in hemodynamically stable patients with high clinical suspicion for duodeno-pancreatic injury … with negative CT-scan or non-specific CT findings on admission imaging, and/or elevated amylase and lipase, or persistent abdominal pain” GoR 2A.
“Repeated and combined measurement of serum amylase and lipase levels, starting from 3 to 6 h after the initial injury, is a useful tool to support clinical evaluation in suspicion of pancreatic injury” GoR 1B.
“Serial clinical examination is an important part of follow-up after biliary and pancreatic-duodenal trauma” GoR 2A.
“In selected cases where an intra-abdominal injury is suspected in the days after the initial trauma, interval laparoscopic exploration may be considered” GoR 2C (WSES liver 2020).

The unifying point: a normal early CT does not clear the abdomen in a patient with the wrong mechanism, a seatbelt sign or persistent pain. Admit, examine serially, re-image, and do not discharge on a single reassuring scan.

WSES-AAST duodeno-pancreatic and extrahepatic biliary tree trauma guidelines 2019 (Coccolini et al., World J Emerg Surg 2019;14:56), diagnostic statements · WSES liver 2020 statement (GoR 2C) · seatbelt-sign association: standard trauma teaching, not NICE

Resuscitation, tranexamic acid and blood (NG39 1.5)

Restrictive volume resuscitation

“For patients with active bleeding use a restrictive approach to volume resuscitation until definitive early control of bleeding has been achieved.” (1.5.18)
In hospital, “move rapidly to haemorrhage control, titrating volume resuscitation to maintain central circulation until control is achieved.” (1.5.20)
“In hospital settings do not use crystalloids for patients with active bleeding.” (1.5.23)

Blood component ratio and the protocol trigger

Adults (16 or over): “use a ratio of 1 unit of plasma to 1 unit of red blood cells to replace fluid volume.” (1.5.24) Children: “use a ratio of 1 part plasma to 1 part red blood cells, and base the volume on the child’s weight.” (1.5.25)
“For patients with active bleeding, start with a fixed‑ratio protocol for blood components and change to a protocol guided by laboratory coagulation results at the earliest opportunity.” (1.5.27)
“Use physiological criteria that include the patient’s haemodynamic status and their response to immediate volume resuscitation to activate the major haemorrhage protocol.” (1.5.13) “Do not rely on a haemorrhagic risk tool applied at a single time point” (1.5.14).

Tranexamic acid

“Use intravenous tranexamic acid as soon as possible in patients with major trauma and active or suspected active bleeding.” (1.5.4)
“Do not use intravenous tranexamic acid more than 3 hours after injury in patients with major trauma unless there is evidence of hyperfibrinolysis.” (1.5.5)

In February 2016 this was an off-label use of tranexamic acid. NG39 gives no dose; the widely used CRASH-2 regimen is 1 g IV over 10 minutes then 1 g over 8 hours — check your local major haemorrhage protocol and see the drug monographs. The dedicated major haemorrhage module runs this in full.

NICE NG39 recommendations 1.5.4–1.5.5, 1.5.13–1.5.14, 1.5.18–1.5.27

Anticoagulation, heat and pain

Anticoagulation (NG39 1.5.6–1.5.12)

“Rapidly reverse anticoagulation in patients who have major trauma with haemorrhage” (1.5.6), and “Do not reverse anticoagulation in patients who do not have active or suspected bleeding” (1.5.12).
Adults on a vitamin K antagonist with active bleeding: “Use prothrombin complex concentrate immediately” (1.5.8); “Do not use plasma to reverse a vitamin K antagonist” (1.5.9).
Any other anticoagulant in an adult (1.5.10), or any anticoagulant in a child (1.5.11): consult a haematologist immediately.

The dose calculation and the DOAC agents are in the bleeding & reversal module.

Heat (NG39 1.6.1)

“Minimise ongoing heat loss in patients with major trauma.” Hypothermia, acidosis and coagulopathy are the lethal triad; the exposed, wet, transfused trauma patient is the one who gets cold.

Pain (NG39 1.7.4–1.7.6)

“For patients with major trauma, use intravenous morphine as the first‑line analgesic and adjust the dose as needed to achieve adequate pain relief.” (1.7.4) If no IV access, consider intranasal atomised diamorphine or ketamine (1.7.5, off-label in 2016); “Consider ketamine in analgesic doses as a second‑line agent” (1.7.6). No doses in NG39; use the drug monographs and, for children, the paediatric tool. Analgesia should not be withheld for fear of masking the abdominal examination — but it is a reason to examine serially and document.

NICE NG39 recommendations 1.5.6–1.5.12, 1.6.1, 1.7.4–1.7.6

Pelvic trauma · Major haemorrhage protocol · Bleeding & anticoagulation reversal · Chest trauma · Injury classifications (AAST-OIS)

The decision tool tells you what to do. This chapter is about why — why the blood pressure decides more than the CT, why a normal early scan does not clear the abdomen, why the injury grade does not decide the treatment, and where blunt abdominal trauma is actually missed. It is meant to be read before the shift, not during it.

NICE NG39, verbatim WSES World Society of Emergency Surgery, with its grade RCT randomised trial Observational cohort, series or review Teaching standard trauma teaching, not guidance

Who this is about. NICE NG39 applies to adults and children unless it names an age, and the decision tool branches on age. But the whole non-operative-management and imaging literature treats children differently — higher rates of successful non-operative management, no routine whole-body CT, poorer FAST performance — and much of it comes from adult series. For a child, use the tool’s paediatric branches and your paediatric major trauma centre; do not carry the adult numbers across.

A case to hold in mind · illustrative, not a real patient

A 24-year-old is brought in after a high-speed car crash, restrained. GCS 15, heart rate 105, blood pressure 112/70. There is a livid linear bruise across the lower abdomen from the lap belt, and the abdomen is mildly, diffusely tender. The trauma CT is reported as no solid-organ injury, a small volume of free fluid, no active contrast extravasation. The registrar says: “CT’s clear, obs are fine — admit overnight for pain control and home in the morning.”

By the end of this chapter you should be able to say why that plan is dangerous, and what one physical sign in the paragraph should change it. And, separately: what would have been the error if this patient had arrived hypotensive and not responding. Both are answered at the bottom.

The whole pathway

Why we do what we do, in one table

Every step in the decision tool, and the reason it is there.

What we do	Why	Basis
Ask the haemodynamic status first, not the organ	Response to resuscitation decides destination — theatre, scanner or watch — more reliably than any single injury does.	NG39 1.5.29–1.5.39 WSES 1A
Unstable, not responding → minimum imaging, then theatre	The scanner is not a place of safety; the treatment for uncontrolled bleeding is damage-control surgery.	NG39 1.5.29, 1.5.37
A negative FAST does not exclude bleeding	FAST misses retroperitoneal blood and has modest sensitivity for abdominal injury; it rules in, not out.	NG39 1.5.30 Stengel 2018
Immediate CT if responding or normal	The stable patient can be characterised fully; CT is the gold standard once physiology allows it.	NG39 1.5.31 WSES 1A
Non-operative management by physiology, not grade	Injury grade on CT does not predict who fails; a stable patient earns a trial of NOM irrespective of grade.	WSES 1B–2A
Angioembolisation for a blush in a stable patient	It treats the arterial component and can preserve the organ without a laparotomy.	WSES 2B NG39 1.5.41
A normal early CT does not clear the abdomen	Hollow-viscus and pancreatic injury can be invisible early; seatbelt sign and persistent pain demand serial exam and repeat imaging.	WSES 2A Teaching
Blood not crystalloid; 1:1; restrictive; TXA early	Replace blood and clotting capacity while buying time for haemorrhage control — not normalise a pressure.	NG39 1.5.4–1.5.27 CRASH-2
No routine whole-body CT in children	WBCT did not lower mortality even in adults, and children take a real radiation cost.	NG39 1.5.36 REACT-2

Chapter 1

The blood pressure decides, not the organ

Why the tool asks the haemodynamic question before it asks anything about the abdomen.

The instinct in abdominal trauma is to reason from the injury: which organ, what grade, how bad on the CT. NICE and WSES both invert that. The first question is physiological — is the patient stable, transiently responding, or not responding to resuscitation — and it decides the destination before the diagnosis is complete.

Not responding. “Limit diagnostic imaging … to the minimum needed to direct intervention” NG39 1.5.29, and “Use damage control surgery in patients with haemodynamic instability who are not responding to volume resuscitation.” NG39 1.5.37 This patient goes to theatre, not the scanner.
Transiently responding. Treat as ongoing bleeding: “Consider definitive surgery in patients with haemodynamic instability who are responding to volume resuscitation.” NG39 1.5.38 The response bought time, not a diagnosis.
Stable or normal. “Consider immediate CT for patients with suspected haemorrhage if they are responding to resuscitation or if their haemodynamic status is normal.” NG39 1.5.31 Now the CT can do its job.

WSES says the same in one line: “The choice of diagnostic technique at admission must be based on the hemodynamic status of the patient” GoR 1A. The whole architecture of the tool is that sentence.

Why the unstable patient must not go to CT

The CT scanner is a corridor away, and a diagnosis feels safer than a decision. But an unstable, non-responding patient with a bleeding abdomen needs the bleeding stopped, and that happens in theatre or the interventional-radiology suite — not on a scanner table where they can arrest away from the team and the equipment. “Minimum imaging to direct intervention” is a deliberate limit, not an oversight.

Chapter 2

What FAST is for, and what it is not

The single most misused investigation in the trauma bay.

FAST answers exactly one question well: is there free fluid I can see right now in the crashing patient? It is fast, radiation-free and can be done during resuscitation. What it cannot do is exclude injury, and NICE states this in the bluntest recommendation on the page: “Be aware that a negative FAST does not exclude intraperitoneal or retroperitoneal haemorrhage.” NG39 1.5.30

The number behind that sentence: in the Cochrane review of point-of-care ultrasound in blunt trauma (34 studies, 8635 patients), pooled sensitivity for abdominal injury was only 0.68 (95% CI 0.59–0.75) — with a specificity of 0.95. Stengel 2018 In other words, a positive FAST is informative, but roughly a third of abdominal injuries are missed by a negative one. In children it is worse: pooled sensitivity 0.63. The review’s conclusion is explicit — “a negative POCS exam does not rule out injuries and must be verified by a reference test such as CT.”

So NICE forbids two tempting misuses: “Do not use FAST or other diagnostic imaging before immediate CT in patients with major trauma” 1.5.32, and “Do not use FAST as a screening modality to determine the need for CT.” 1.5.33 FAST is not the gatekeeper to the scanner; in the stable patient the scanner is the investigation, and in the unstable patient FAST tells you where to take them, not whether to.

Chapter 3

The laparotomy-now gate

The findings that end the decision tree before imaging.

Some patients do not need a scan to decide the destination: the resus-room gate to immediate laparotomy is haemodynamic instability with free fluid, peritonitis, evisceration, or impalement. WSES states the contraindication to leaving the abdomen alone directly: “NOM in splenic injuries is contraindicated in the setting of unresponsive hemodynamic instability or other indicates for laparotomy (peritonitis, hollow organ injuries, bowel evisceration, impalement)” GoR 1A, and “Hemodynamically unstable and non-responder patients (WSES IV) should undergo OM” GoR 2A.

Beyond those clinical gates, a handful of CT findings mandate operation once identified: free intraperitoneal air (pneumoperitoneum), enteric contrast extravasation or a discontinuous bowel wall, and diaphragmatic rupture. Those are scan diagnoses. The point of the gate is that the resus-room decision — instability plus free fluid, or peritonitis, or evisceration, or an object still in situ — does not wait for them.

Why “impalement” means do not remove the object

An impaling object may be tamponading the very vessel it has injured. Removing it in the ED can convert a contained injury into catastrophic haemorrhage. It is left in place, immobilised, and removed in theatre with vascular control available — which is precisely why impalement is on the straight-to-laparotomy list.

Chapter 4

Leaving the spleen and liver alone

The counter-intuitive rule: the grade does not decide it.

Non-operative management is the default for the stable patient — and this is the part that surprises people — it does not depend on the injury grade. WSES: “Patients with hemodynamic stability and absence of other abdominal organ injuries requiring surgery should undergo an initial attempt of NOM irrespective of injury grade” GoR 2A, and, flatly, “Injury grade on CT scan, extent of free fluid, and the presence of PSA do not predict NOM failure or the need of OM” GoR 1B.

What NOM requires is not a low grade but a setting: intensive monitoring, immediate access to angioembolisation, an available operating theatre and blood, and serial clinical evaluation. For an arterial blush on CT in a stable patient, “AG/AE may be considered the first-line intervention … irrespective from injury grade” GoR 2B — NICE agrees, “Consider interventional radiology techniques in patients with solid-organ (spleen, liver or kidney) arterial haemorrhage.” NG39 1.5.41

Who is more likely to fail

Grade does not predict failure; some patient factors do. WSES names “age above 55-years old, high ISS, and moderate to severe splenic injuries” as prognostic factors for NOM failure GoR 2B. The evidence behind the age factor is the EAST multi-institutional splenic study: patients aged 55 and over failed NOM more often (19% vs 10%) and died more often across every treatment route (43% vs 23%). Harbrecht 2001 Older age is not a contraindication to NOM — it is a reason for a lower threshold to escalate and closer watching. The spleen has a second trap the liver has less of: delayed rupture of a subcapsular haematoma or pseudoaneurysm, days later. Teaching

Chapter 5

The scan that clears nothing

Where blunt abdominal trauma is actually missed.

Solid-organ injury bleeds, and bleeding shows on CT. Hollow-viscus and mesenteric injury, and the retroperitoneal duodenum and pancreas, can do neither early — and this is where blunt abdominal trauma is missed. Hollow-organ and mesenteric injuries occur in up to 7% of abdominal trauma and carry high morbidity and mortality when the diagnosis is delayed; the early symptoms are non-specific, and CT — the gold standard — has limited sensitivity despite high specificity. Schmitt 2026

The single most useful external clue is the seatbelt sign — a linear ecchymosis across the lower abdomen from a lap belt — which is a recognised marker of bowel and mesenteric injury and should lower the threshold for serial examination and repeat imaging. Teaching It is not, on its own, a reason to operate: in a cohort of 239 patients with an abdominal seatbelt sign, a hollow-viscus-injury prediction score of 5 or less had a 98.3% negative predictive value, while 2.1% still had a delayed diagnosis of hollow-viscus injury. Noori 2025 The sign raises suspicion; it does not settle it either way.

WSES puts the retroperitoneal organs on an explicit re-imaging clock: “A repeat CT-scan within 12–24 h from the initial injury should be considered in hemodynamically stable patients with high clinical suspicion for duodeno-pancreatic injury … with negative CT-scan or non-specific CT findings on admission imaging” GoR 2A, supported by serial amylase and lipase “starting from 3 to 6 h after the initial injury” GoR 1B.

The unifying rule. A normal early CT does not clear the abdomen in a patient with the wrong mechanism, a seatbelt sign or persistent pain. Admit, examine serially, re-image — and do not discharge on a single reassuring scan.

Chapter 6

Resuscitating the bleed

The same haemorrhage spine as the rest of major trauma — with two abdomen-specific traps.

The resuscitation is NICE’s major-haemorrhage spine: a restrictive approach to volume until bleeding is controlled 1.5.18, no crystalloid for active bleeding in hospital 1.5.23, a 1:1 plasma-to-red-cell ratio to start 1.5.24 moving to lab-guided as soon as possible 1.5.27, and tranexamic acid as soon as possible, never beyond 3 hours without hyperfibrinolysis 1.5.4–1.5.5. Reverse anticoagulation only if there is active or suspected bleeding 1.5.6, 1.5.12; minimise heat loss 1.6.1; IV morphine first-line for pain 1.7.4. The major haemorrhage and reversal modules run these in full.

The two abdomen-specific traps

Analgesia does not mask the abdomen away. Withholding morphine to “keep the examination clean” is not supported — give adequate analgesia, and manage the risk of a changing abdomen by examining serially and documenting, not by leaving the patient in pain. And the physiological trigger, not a score, activates the protocol: “Use physiological criteria … and their response to immediate volume resuscitation to activate the major haemorrhage protocol” 1.5.13; “Do not rely on a haemorrhagic risk tool applied at a single time point” 1.5.14.

Pitfalls

Ten ways this goes wrong

Each is a real pattern, not a hypothetical.

01Taking the unstable patient to CT

Why it happens: The scanner is next door and a diagnosis feels safer than a decision.
Instead: Minimum imaging to direct intervention, then theatre or IR. NG39 1.5.29, 1.5.37

02Clearing the abdomen on a negative FAST

Why it happens: It is quick, negative, and feels like an answer.
Instead: A negative FAST excludes nothing; abdominal sensitivity is ~0.68. Verify with CT in the stable patient. NG39 1.5.30 Stengel 2018

03Discharging on a normal early CT

Why it happens: The scan is reported normal and the patient looks well.
Instead: Hollow-viscus and pancreatic injury can be occult early. Admit, examine serially, re-image if suspicion is high. WSES 2A

04Ignoring the seatbelt sign

Why it happens: It is “just a bruise” and the CT was clear.
Instead: It is a recognised marker of bowel and mesenteric injury; lower your threshold for serial exam and repeat imaging. Teaching Noori 2025

05Deciding NOM by the injury grade

Why it happens: A high grade looks like it must mean surgery.
Instead: Grade does not predict NOM failure; physiology does. A stable patient earns a trial of NOM irrespective of grade. WSES 1B–2A

06Attempting NOM without the wrap-around

Why it happens: The patient is stable now, so watching feels low-cost.
Instead: NOM needs monitoring, immediate AG/AE, an available theatre and blood, and serial evaluation — or it is not NOM, it is neglect. WSES 2A

07Crystalloid to chase a blood pressure

Why it happens: Fluid is immediate and blood takes a phone call.
Instead: No crystalloid for active bleeding in hospital; restrictive, blood-based resuscitation. NG39 1.5.18, 1.5.23

08Withholding analgesia to keep the exam clean

Why it happens: Fear of masking peritonism.
Instead: Give adequate analgesia; manage the changing abdomen by serial examination and documentation. NG39 1.7.4

09Routine whole-body CT in a child

Why it happens: It is the adult reflex and feels thorough.
Instead: Do not routinely WBCT under-16s; limit CT to the areas that need it. WBCT did not lower mortality even in adults. NG39 1.5.36 REACT-2

10Treating the abdomen as the only source

Why it happens: The seatbelt sign and the tender abdomen anchor attention.
Instead: Shock in blunt trauma has several sources — chest, pelvis, long bones, external. Find the physiology’s cause, not the first plausible one. Teaching

Red herrings

Seven findings that look reassuring and are not

None of these is wrong information. Each answers a different question from the one you are asking.

Looks reassuring

“FAST was negative.”

It misses the retroperitoneum and roughly a third of abdominal injuries. A negative FAST rules nothing out. NG39 1.5.30 Stengel 2018

Looks reassuring

“The CT was clear.”

An early CT can be normal in bowel, mesenteric and pancreatic injury. With the wrong mechanism or a seatbelt sign, it does not clear the abdomen. Schmitt 2026

Looks reassuring

“The blood pressure is normal.”

Young patients compensate until they suddenly cannot. Watch the trend and the response to resuscitation, not the first reading. Teaching

Looks reassuring

“It’s only a low-grade splenic injury.”

Grade does not predict NOM failure or the need for surgery; physiology does. WSES 1B

Looks reassuring

“No contrast blush on CT.”

Absence of a blush does not exclude bleeding, and venous or hollow-viscus injury may not blush at all. It informs, it does not clear.

Looks reassuring

“The lactate is normal.”

A single normal value does not exclude significant injury; NICE warns against a risk tool at a single time point. Trend it. NG39 1.5.14

Looks reassuring

“Amylase is normal.”

Early amylase and lipase can be normal after pancreatic injury; WSES repeats them from 3–6 h and serially, precisely because a single value misleads. WSES 1B

The case, revisited

What was wrong with the plan

The seatbelt sign is the sign that changes it. With a lap-belt ecchymosis, diffuse tenderness and free fluid on CT, “clear scan, home in the morning” is exactly the missed-hollow-viscus pattern. Admit, examine serially, and re-image if suspicion persists. (Chapter 5)
“No solid-organ injury” is not “no injury”. The free fluid without a solid-organ source should raise bowel or mesenteric injury, not be dismissed. (Chapter 5)
The abdomen was not cleared, it was scanned once. A single early CT does not exclude the injuries that declare themselves late. (Chapter 2, 5)
And the counterfactual: had this patient been hypotensive and not responding, the error would have been the opposite one — taking them to CT at all. That patient goes to theatre on minimum imaging. (Chapters 1, 3)

The pattern behind both halves: the CT is a powerful test that answers one question at one moment. It does not decide the destination in the unstable patient, and it does not clear the abdomen in the stable one.

How much of blunt abdominal trauma care rests on trials? The resuscitation spine has large RCTs (CRASH-2, PROPPR); the abdomen-specific decisions — non-operative management, imaging strategy, the hollow-viscus trap — rest almost entirely on observational data and one imaging RCT. What follows is the evidence behind the main decisions, what each study actually showed, and how far it can carry the weight put on it.

At a glance

The evidence map

The strength column is this module’s appraisal, not a published grade.

Question	Best evidence	Design	Strength
Does immediate whole-body CT save lives?	REACT-2 2016	Multicentre RCT (n=1403)	Moderate–high — randomised; no mortality benefit
Can a negative FAST exclude injury?	Stengel Cochrane 2018	Diagnostic meta-analysis (34 studies, 8635)	Moderate — heterogeneous, but consistent direction
Does injury grade decide treatment?	WSES 2017/2020; Harbrecht 2001	Guideline synthesis; multicentre cohort	Low–moderate — observational, but large and concordant
Who fails non-operative management?	Harbrecht/EAST 2001	Retrospective multicentre (n=1488)	Low–moderate
Is a normal early CT reassuring?	Schmitt 2026; Noori 2025	Narrative review; single-centre cohort	Low — but the direction is not in doubt

Does immediate whole-body CT save lives?

REACT-2 2016 — total-body CT vs selective imaging

RCTn = 14035 centres

An international multicentre RCT randomising 1403 severely injured adults (≥18) to immediate total-body CT or a standard work-up with conventional imaging and selective CT. Sierink 2016

In-hospital mortality	16% vs 16%, p=0.92 (86/541 vs 85/542)
Polytrauma subgroup	22% vs 25%, p=0.46
Traumatic brain injury subgroup	38% vs 44%, p=0.31

So what. Immediate total-body CT did not reduce mortality against a selective work-up, and the authors flagged the radiation cost. This is the level-1 evidence behind NICE’s refusal to make WBCT routine in children (1.5.36) and the general principle that imaging should be targeted, not reflexive. It enrolled adults, so it does not directly answer the paediatric question — but it removes the mortality argument for scanning everything.

Can a negative FAST exclude injury?

Stengel 2018 — Cochrane review of point-of-care ultrasound

Diagnostic meta-analysis34 studiesn = 8635

A Cochrane systematic review of point-of-care ultrasonography for thoracoabdominal injury in blunt trauma. Stengel 2018

Overall sensitivity / specificity	0.74 (0.65–0.81) / 0.96 (0.94–0.98)
Abdominal injury	sensitivity 0.68 (0.59–0.75), specificity 0.95 (0.92–0.97)
Children	sensitivity 0.63 (0.46–0.77) — poorer
In 1000 patients (28% prevalence)	FAST would miss 73 with injuries

So what. A positive FAST is useful (high specificity), but a negative FAST misses roughly a third of abdominal injuries and is worse in children. The review’s conclusion — “a negative POCS exam does not rule out injuries and must be verified by a reference test such as CT” — is the evidence base for NG39 1.5.30 and for never using FAST to decide whether a stable patient needs a scan.

Who fails non-operative management?

Harbrecht 2001 — EAST multi-institutional splenic study

Retrospective multicentren = 148827 centres

1488 adults with blunt splenic injury across 27 US trauma centres, examining the contribution of age to outcome. Harbrecht 2001

NOM failure, ≥55 vs <55	19% vs 10%
Overall mortality, ≥55 vs <55	43% vs 23%
Mortality after successful NOM	8% vs 4% (p<0.05)

So what. This is the evidence behind WSES naming age >55 as a prognostic factor for NOM failure. It does not make older age a contraindication — most older patients still succeed with NOM — but it justifies a lower threshold to escalate and closer monitoring. It is retrospective 1997 data, so absolute rates have moved with modern angioembolisation; the age effect is the durable finding.

Is a normal early CT reassuring in a seatbelt-sign patient?

Schmitt 2026 & Noori 2025 — the hollow-viscus problem

Narrative review + cohortn = 239 (SBS)

Hollow-organ and mesenteric injuries occur in up to 7% of abdominal trauma, with high morbidity when the diagnosis is delayed; early symptoms are non-specific and CT has limited sensitivity despite high specificity. Schmitt 2026 In a single-centre cohort of 239 patients with an abdominal seatbelt sign, a hollow-viscus-injury prediction score of ≤5 had a 98.3% negative predictive value, while 2.1% of the whole cohort still had a delayed diagnosis of hollow-viscus injury. Noori 2025

So what. These are not high-level studies — one is a narrative imaging review, the other a single centre — but the direction is not in doubt and is consistent with WSES’s re-imaging clock: a normal or non-specific early CT does not clear the abdomen when the mechanism or the seatbelt sign says otherwise. The management answer is serial examination, serial amylase/lipase and repeat imaging, not discharge.

Reading it honestly

Critical appraisal

The abdomen-specific evidence is observational. Non-operative management, angioembolisation and the hollow-viscus trap rest on cohorts and case series, not trials. WSES grades most of its own statements 1B–2C for that reason. The recommendations are strong and concordant, but they are consensus on observational data.
The imaging evidence is stronger and points to restraint. REACT-2 is a genuine RCT and found no mortality benefit from scanning everything; the FAST meta-analysis is large and consistent. Both argue for targeted, physiology-led imaging rather than reflex whole-body CT or FAST-as-gatekeeper.
The age-and-NOM data are old. Harbrecht is 1997 practice; modern angioembolisation has changed absolute failure and mortality rates. The durable message is the direction — older patients fail NOM more and die more — not the specific percentages.
The seatbelt-sign and hollow-viscus literature is weak but clinically decisive. The studies are small or narrative, yet the failure mode they describe — the missed bowel injury after a reassuring early scan — is one of the classic causes of preventable death in blunt abdominal trauma. Here, the weakness of the evidence is not a reason to relax; it is a reason to over-observe.
WSES is international, not UK guidance. It is used here for the abdomen-specific questions NICE does not address. Where WSES and NICE overlap — imaging by physiology, damage control, IR for arterial haemorrhage — they agree.

A balanced position

Blunt abdominal trauma is a field where the strongest evidence argues for less imaging (REACT-2, the FAST meta-analysis) while the weakest evidence argues for more caution (the hollow-viscus trap). The two are not in conflict: the lesson of both is that a single test at a single moment does not answer the question. The physiology decides the destination; the CT characterises the stable patient; and neither the negative FAST nor the normal early scan discharges the patient with the wrong mechanism. The recurring preventable death here is the missed hollow-viscus injury sent home on a reassuring scan — which is why the module’s loudest sentence is that a normal CT clears nothing.

About this module

What this is, and what it is not

The Decision tool implements NICE NG39 and the WSES guidelines and quotes them verbatim. The Teaching and Evidence tabs are the author’s synthesis of those guidelines and the primary studies behind them, written to explain the tool, not to extend it. Nothing in them changes what the tool tells you to do. Where they go beyond NICE, the source is named and tagged.

Teaching and Evidence were added in September 2026 and have had no clinical review by anyone other than their author.

What this module deliberately does not do

It does not grade the injury. AAST-OIS grading is a radiological classification made on the CT; the ED decisions above turn on physiology, not grade. The AAST grades are in the classifications tool.
It does not give drug doses. NG39 gives none for TXA, morphine or reversal; the drug and paediatric tools do.
It does not run the whole major haemorrhage protocol. It points to the physiological trigger NICE names; the major haemorrhage module and your trust’s protocol are the operational documents.
It does not cover the pelvis as a separate bleeding source — that has its own module — nor the operative technique of laparotomy, packing or resection, which are surgical decisions.
It does not extend to penetrating abdominal trauma. The operative logic there is different; the resuscitation recommendations still apply.

Sources in full — decision tool

NICE NG39. Major trauma: assessment and initial management. Published 17 February 2016. The full Recommendations chapter was parsed for this module; the haemorrhage, imaging, damage-control, heat and pain recommendations used here are in sections 1.5–1.7. Recommendations apply to both children (under 16s) and adults (16 or over) unless otherwise specified.
WSES splenic trauma guidelines 2017. Coccolini F, Montori G, Catena F, et al. Splenic trauma: WSES classification and guidelines for adult and pediatric patients. World J Emerg Surg 2017;12:40. Diagnostic and management statements quoted with their grade of recommendation (GoR).
WSES liver trauma guidelines 2020. Coccolini F, Coimbra R, Ordonez C, et al. Liver trauma: WSES 2020 guidelines. World J Emerg Surg 2020;15:24. Statements quoted with their GoR.
WSES-AAST duodeno-pancreatic and biliary trauma guidelines 2019. Coccolini F, Kobayashi L, Kluger Y, et al. Duodeno-pancreatic and extrahepatic biliary tree trauma: WSES-AAST guidelines. World J Emerg Surg 2019;14:56. Diagnostic statements quoted with their GoR.

Every NICE quotation on this page is verbatim from the recommendation named beside it. Every WSES statement is quoted verbatim and carries its published grade of recommendation. The principal element attributed to neither — the seatbelt-sign association with hollow-viscus injury — is labelled as standard teaching wherever it appears; a few background didactic points (delayed splenic rupture, the lethal triad, the CRASH-2 TXA regimen) are likewise flagged as standard teaching rather than guideline recommendations. The WSES guidelines are international, not UK national guidance, and are used here for the abdomen-specific questions NICE does not address; where they and NICE overlap, they agree.

References — teaching and evidence

Sierink JC, Treskes K, Edwards MJ, et al. Immediate total-body CT scanning versus conventional imaging and selective CT scanning in patients with severe trauma (REACT-2): a randomised controlled trial. Lancet 2016;388(10045):673–83. PMID 27371185.
Stengel D, Leisterer J, Ferrada P, Ekkernkamp A, Mutze S, Hoenning A. Point-of-care ultrasonography for diagnosing thoracoabdominal injuries in patients with blunt trauma. Cochrane Database Syst Rev 2018;12:CD012669. PMID 30548249.
Harbrecht BG, Peitzman AB, Rivera L, et al. Contribution of age and gender to outcome of blunt splenic injury in adults: multicenter study of the Eastern Association for the Surgery of Trauma. J Trauma 2001;51(5):887–95. PMID 11706335.
Schmitt JMM, Lock JF, Anger F. Detection and emergency management of visceral injuries after abdominal trauma: hollow organ and mesenteric injuries. Unfallchirurgie (Heidelb) 2026;129(2):104–10. PMID 41513814.
Noori E, Lee AT, Santos JW, et al. A single-center evaluation of a novel hollow viscus injury prediction score for abdominal seatbelt sign. Am Surg 2025;91(10):1658–64. PMID 40377004.

Every figure in the Evidence tab is taken from the abstract of the study cited beside it. Verbatim quotations in Teaching are from NICE NG39 or the WSES guidelines, as tagged.

The reference

NICE NG39

Whole-body CT

WSES, on the same axis

Spleen (WSES 2017)

Liver (WSES 2020)

NICE, on the interventional-radiology side

Who is more likely to fail NOM

Restrictive volume resuscitation

Blood component ratio and the protocol trigger

Tranexamic acid

Anticoagulation (NG39 1.5.6–1.5.12)

Heat (NG39 1.6.1)

Pain (NG39 1.7.4–1.7.6)

Related

Why we do what we do, in one table

The blood pressure decides, not the organ

What FAST is for, and what it is not

The laparotomy-now gate

Leaving the spleen and liver alone

Who is more likely to fail

The scan that clears nothing

Resuscitating the bleed

Ten ways this goes wrong

01Taking the unstable patient to CT

02Clearing the abdomen on a negative FAST

03Discharging on a normal early CT

04Ignoring the seatbelt sign

05Deciding NOM by the injury grade

06Attempting NOM without the wrap-around

07Crystalloid to chase a blood pressure

08Withholding analgesia to keep the exam clean

09Routine whole-body CT in a child

10Treating the abdomen as the only source

Seven findings that look reassuring and are not

What was wrong with the plan

The evidence map

REACT-2 2016 — total-body CT vs selective imaging

Stengel 2018 — Cochrane review of point-of-care ultrasound

Harbrecht 2001 — EAST multi-institutional splenic study

Schmitt 2026 & Noori 2025 — the hollow-viscus problem

Critical appraisal

What this is, and what it is not