Pediatric Trauma II: Solid Organ Injury
With Dr. David Notrica · hosted by Dr. Todd Ponsky & Dr. Mark McCollum
Part of
Blunt Liver Injury 4 items
Chapter 1 of 13 · Fundamentals
ATOMIC origins
Introduction and ATOMIC Consortium Overview
Expert statements on this page
No expert statements were drawn from this page.
Host summaries · secondary, not cited in answers
Educational content from recorded physician discussions — not medical advice. Talk to your (or your child's) care team about your situation.
What the experts said
The ATOMIC consortium started around 2010 as a multi-center collaborative to prospectively study solid organ injury management, initially with 5 hospitals expanding to 10.
Hypotension is a late finding in pediatric hemorrhagic shock; almost half of patients requiring early transfusion are not hypotensive.
Defining hemodynamic stability proved difficult because concurrent head injury may prevent tachycardia, some unstable pediatric patients are not hypotensive, and tachycardia can be due to pain rather than bleeding.
The terminology was changed from 'stable/unstable' to 'bleeding or having bled recently' because when a hypovolemic pediatric patient arrives, you know they have bled but often cannot confirm ongoing bleeding until after blood transfusion.
Children have an amazing ability to stop bleeding, far superior to adults; a patient may present in shock but have already stopped bleeding.
An elevated shock index has high sensitivity (rarely misses patients in trouble) but lower specificity (many elevated patients are not actively bleeding). If shock index is not elevated, the patient is very unlikely to be hypovolemic.
After 20 mL/kg of crystalloid, blood transfusion should be initiated if non-operative management and good outcomes are desired. Evidence from adults is strong; pediatric evidence (Denver) shows children are somewhat more resistant to crystalloid's negative effects but still experience harm (longer ventilator time).
Massive transfusion in children is rare, making complication studies difficult. Adult series show 1:1:1 transfusion (packed red cells:FFP:platelets) appears safe and effective.
Real-time viscoelastic assays (TEG/ROTEM) for directed component therapy make theoretical sense and are used at Phoenix Children's, but definitive pediatric RCT evidence comparing 1:1:1 to TEG-guided therapy is lacking.
Angioembolization is safe in pediatric blunt liver and spleen injury.
Many patients with contrast extravasation on CT stop bleeding spontaneously without angioembolization.
In the ATOMIC failure cohort, no patients who underwent angioembolization for splenic bleeding went on to fail non-operative management.
Many patients who had angioembolization for liver bleeding still required laparotomy or washout, often for bile complications or large hemoperitoneum rather than continued bleeding.
Not every patient with contrast extravasation needs angioembolization; the indication is ongoing bleeding (continued transfusion requirement), similar to failure criteria.
A patient with grade 5 splenic injury and active extravasation on CT had no active bleeding 45 minutes later in the angio suite, illustrating children's ability to spontaneously stop bleeding.
For grade 1-4 injuries, ICU admission should be based on hemodynamic status (whether the patient is actively bleeding or has bled recently), not injury grade.
Grade 5 injuries should be admitted to the ICU regardless of hemodynamic stability, based on literature evidence of higher risk.
Bed rest was never included in the original APSA protocols but became part of the culture of solid organ injury management.
Bed rest was eliminated in the ATOMIC renal injury protocol because research showed it made no difference whether patients walked to the bathroom or remained on strict bed rest.
Patients who were transfused and those who failed non-operative management all failed early, supporting the 18-hour observation threshold.
Serial hemoglobins are unreliable for detecting ongoing bleeding; they decrease due to lab error, incorrect draw technique (upstream from IV), or dilution from excess fluid. Physical examination and vital signs are superior indicators.
Increasing heart rate, poor capillary refill, pallor, and cold extremities are clinical signs that indicate ongoing bleeding, not serial hemoglobin drops.
Half of patients who fail non-operative management do so because of peritonitis, not continued bleeding.
Many patients with low-grade injuries (1-2) never returned for follow-up appointments, prompting a switch to telephone follow-up to confirm patient well-being without requiring missed school days.
In-person follow-up visits for solid organ injury rarely provide clinical benefit; patients who need additional imaging or therapy present to the emergency room with symptoms (pain, problems).
Complications of non-operative management were converted into symptoms (abdominal pain, respiratory problems, jaundice) and listed in ATOMIC discharge instructions to guide patients on when to return to the emergency room.
Jaundice is fairly common with large hematomas (mobilization of blood products) and does not always indicate a complication; these patients can be triaged by phone rather than automatic emergency room referral.
Routine follow-up ultrasound is not needed for most spleen injuries. For liver injuries, there is a possibility of benefit, but the evidence is unclear.
Studies performing routine ultrasounds found many pseudoaneurysms, but the number was discordant with the rate of delayed bleeding or complications, indicating that asymptomatic pseudoaneurysms rarely lead to clinically significant problems (likely <10%, possibly much less).
It is extremely rare for a patient discharged home doing well to develop a finding that puts them at risk for life-threatening bleeding.
Activity restrictions follow the APSA recommendation of grade+2 weeks (e.g., grade 4 injury = 6 weeks restriction). This is expert opinion, not evidence-based.
Children can return to school but should avoid gym/PE and may need to change classes 5 minutes early to avoid rough contact in hallways.
Recurrent hypotension after initial stabilization (e.g., patient stabilizes, goes to CT, becomes unstable again) is an early failure criterion requiring operative intervention or angioembolization, not continued non-operative management. This differs from gradual hemoglobin drift in the ICU.
Early research by Sam Smith (Arkansas) and Sean St. Peter showed that patients could be managed based on hemodynamic status rather than CT injury grade, challenging the original grade-based consensus approach developed by Steve Stilanos.
The Arkansas group published 'Throwing Out the Grade Book,' demonstrating that hemodynamic status, not CT grade, should guide management decisions.
The Shock Index Pediatric Adjusted (SIPPA) was first described by Acker et al. from Denver, adjusting the adult shock index (heart rate/systolic BP) for pediatric physiology.
Pediatric shock index cutoffs: ages 4-6 >1.2, ages 7-12 >1.0, ages 13+ >0.9 indicate potential instability.
ATLS previously recommended 60 mL/kg of crystalloid before blood transfusion, but adult literature showed excessive crystalloid dilutes coagulation factors and worsens outcomes in 1:1:1 resuscitation (Duchesney, New Orleans).
John Holcomb's work showed that excessive crystalloid reduces the success rate of non-operative management.
One massive transfusion trial found that early and frequent blood transfusion reduced the number of patients progressing to require massive transfusion, suggesting early blood prevents DIC.
Emerging data suggest directed component therapy (TEG/ROTEM) saves money, conserves blood products, and may resuscitate more efficiently than ratio-driven massive transfusion.
The threshold for failure of non-operative management is 40 mL/kg of all blood products, based on data from NEF and military experience showing increased mortality and operative need above this threshold.
There is no literature to support mandatory bed rest for solid organ injury. One adult study comparing bed rest to no bed rest found no increased bleeding.
Sean St. Peter's prospective study and a follow-up study showed that patients do not need to be hospitalized for a minimum time period (grade+1 days); 18 hours of observation is sufficient if no bleeding occurs.