There are diagnoses in pediatric orthopedics where the problem isn't remembering what it could be. The problem is deciding how quickly to act.
Septic arthritis is one of them.
We all know it's an emergency. We all know, at least broadly, Kocher's criteria. We've all learned that transient synovitis of the hip can resemble septic arthritis and that a septic hip must not be missed.
Yet, in practice, the difficult cases continue to be precisely those that don't fit neatly into the box.
The child doesn't have a 39-degree fever.
The CRP isn't yet "textbook."
The X-ray is normal.
The ultrasound shows an effusion, but doesn't say why.
The pain is referred to the knee, but the hip is stiff.
Or the child is young, uncooperative, cries as soon as they're moved, and distinguishing pain, fear, and true functional impairment isn't immediate.
The point of today's update isn't to replace clinical judgment with a new test. It's to avoid the opposite error: using a single piece of data — a score, X-ray, ultrasound, CRP, white blood cell count — as if it could single-handedly conclude the reasoning.
Where Kocher helps, and where it can mislead
Kocher's criteria were developed to help distinguish between septic arthritis of the hip and transient synovitis. The four classic criteria are: fever >38.5 °C, inability to bear weight, ESR >40 mm/h, and WBC >12,000/mm³ [1]. Caird later proposed adding CRP, strengthening the weight of systemic inflammation in the predictive model [2].
These are useful criteria. The problem is when they become a shortcut.
A child with multiple positive criteria, clinically distressed, with a stiff hip and elevated markers, doesn't create much doubt: they should be treated as a high-risk patient.
The real issue lies with intermediate scores.
Even riskier is using the score in reverse: "they don't have all the criteria, so it's not septic arthritis." In pediatric orthopedics, this statement can be dangerous, especially if the child isn't bearing weight, if pain on passive mobilization is significant, or if the clinical evolution isn't convincing.
Subsequent literature has shown that the picture can be more complex, especially in the era of MRI. Some children with suspected septic arthritis actually have associated osteomyelitis, pyomyositis, abscesses, or periarticular infections that modify treatment [3].
So the question isn't just:
"Is it septic arthritis or transient synovitis?"
The more useful question is:
"Is there an infectious focus? Where is it? How extensive is it, and how much time can I afford to observe?"
A normal X-ray does not reassure
X-rays remain useful. They help rule out fractures, obvious bone lesions, differential diagnosis of tumors, and already advanced joint alterations.
But in the initial phase of pediatric septic arthritis, they can be normal.
This is one of the points where the process gets stuck. A normal X-ray is often interpreted as "there's nothing there." In reality, it only means that we don't yet see obvious radiographic bone or joint alterations.
If the child isn't bearing weight, has significant pain on passive mobilization, fever, or compatible inflammatory markers, the reasoning must continue.
Ultrasound: excellent for seeing effusion, not for saying why it's there
In the hip, ultrasound is very useful. It identifies effusion, allows comparison with the contralateral side, and can guide aspiration.
But here too, the shortcut must be avoided.
Effusion does not automatically mean septic arthritis.
The absence of effusion, if the examination is early or technically difficult, should not always close the case if the clinical picture is strong.
Ultrasound answers one question well:
"Is there joint fluid?"
It answers another question less well:
"Why is that fluid there?"
For this reason, ultrasound must be part of a pathway. In high-risk cases, it can accelerate aspiration. In doubtful cases, it can guide. In cases with suspicion of deep, multifocal, or extra-articular infection, however, it does not replace MRI.
MRI: not for everyone, but when needed, it must arrive quickly
MRI should not become the new bottleneck of the system.
If a child has a clinically septic hip, with effusion, significant pain, and compatible markers, waiting for an MRI to decide can be a mistake.
But the opposite error is to ignore how much MRI can change the course in the right cases.
MRI becomes particularly useful when the location is unclear, when pain is disproportionate to joint findings, when there is suspicion of associated osteomyelitis, when the picture does not improve as expected, or when a subperiosteal abscess, pyomyositis, or multifocal infection is suspected.
The point isn't to do an MRI on everyone. The point is not to use it too late in children where the problem isn't just intra-articular.
Joint aspiration: diagnosis, but also decision
Joint aspiration is often the step that separates suspicion from actual management.
It allows for fluid analysis, cell count, Gram stain, culture, eventual PCR/NAAT in contexts where available, and helps guide antibiotic therapy.
Variability in pediatric aspiration practice has been documented: not everyone aspirates in the same scenarios, not everyone sends the same tests, not everyone interprets synovial fluid cytology in the same way [4]. This makes it even more important to have a shared internal pathway.
A practical point: if the child is stable and aspiration can be obtained quickly, ideally microbiological samples should precede antibiotics. However, if the child is septic, unstable, or the delay risks being clinically relevant, the priority changes: blood cultures and antibiotics should not be paralyzed by waiting for the perfect sample.
In real life, this is the least "textbook" part: it requires coordination among the orthopedist, internist, anesthesiologist, radiologist, and infectious disease specialist.
Kingella, negative cultures, and false microbiological security
In young children, especially under 4 years old, Kingella kingae has changed the way we think about pediatric osteoarticular infections.
It can present with more subtle pictures, less dramatic markers, and negative traditional cultures. Culturing joint fluid in blood culture bottles and molecular techniques can increase diagnostic yield in centers that use them.
The clinical point is simple: a negative culture does not always equate to a wrong diagnosis.
Sometimes it means antibiotics have already been started.
Sometimes a poor sample.
Sometimes a difficult microorganism.
Sometimes suboptimal technique.
This doesn't mean treating every effusion as septic. It means that microbiology should be interpreted within the overall picture, not as an isolated verdict.
Practical Pathway 1 — Child with painful limp or refusal to bear weight
| Clinical Junction | What to assess immediately | If risk is high | If picture is intermediate/low |
|---|---|---|---|
| General condition | distressed, toxic, lethargic, unstable child, high or persistent fever | blood cultures, urgent tests, empirical antibiotics without unjustified delays, activation of orthopedics/anesthesia | close observation, scheduled clinical re-evaluation, tests if the picture is not clearly benign |
| Weight-bearing | walks, limps, or completely refuses to bear weight | refusal to bear weight carries significant weight, especially if associated with fever or joint pain | if walking and improving, more calm reasoning is possible, but not if worsening or if pain remains significant |
| Pain on passive mobilization | hip, knee, ankle, shoulder, elbow | significant pain on passive mobilization = high joint suspicion | mild pain or predominantly muscular/tendinous pain suggests a different direction |
| Laboratory | CBC, CRP, ESR, blood cultures | elevated or rising CRP/ESR increase risk and aid monitoring | initial normal values do not always rule out an early infection |
| Initial Imaging | X-ray if trauma/bone doubt/differential diagnosis; ultrasound if suspected joint effusion | targeted imaging, without delaying aspiration or drainage if the picture is clear | ultrasound/MRI according to location, evolution, and suspicion of osteomyelitis or abscess |
Practical message: In a child who isn't bearing weight and has significant pain, there's no need to wait for "all criteria" to become positive. Re-evaluation must be active, not passive.
Practical Pathway 2 — Suspected septic arthritis of the hip
| Situation | Interpretation | Next Step |
|---|---|---|
| Painful hip + refusal to bear weight + fever/elevated markers | high risk of septic arthritis | rapid ultrasound; if effusion, urgent joint aspiration ± drainage; microbiological samples before antibiotics if this does not delay treatment |
| Ultrasound effusion + intermediate clinical picture | effusion alone is not enough to distinguish synovitis and septic arthritis | integrate Kocher/Caird, CRP/ESR, pain, age, and trend; consider aspiration if risk is not low |
| Significant pain but negative or doubtful ultrasound | do not close the case based solely on ultrasound | re-evaluate location, repeat exam if necessary, consider MRI if suspicion of osteomyelitis, pyomyositis, abscess, or unclear location |
| Pain referred to knee/thigh but hip limited | possible misleading hip presentation | always examine the hip in a child with a painful limp, even if they report pain elsewhere |
| Picture not improving within 12–24 hours | initial diagnosis to reconsider | clinical re-evaluation, serial CRP, advanced imaging, eventual aspiration/drainage if not already performed |
Practical message: Kocher and Caird help stratify risk, but they should not become an automatic traffic light. The "intermediate" child is where the pathway makes a difference.
Practical Pathway 3 — After aspiration or drainage
| Phase | Objective | What to check |
|---|---|---|
| Microbiological samples | identify pathogen when possible | joint fluid for Gram stain, culture, cell count; blood cultures; eventual PCR/NAAT if available |
| Empirical antibiotics | cover probable pathogens without delay | age, local epidemiology, MRSA if relevant, Kingella in young children, comorbidities |
| First 24–48 hours | understand if the trajectory is correct | pain, fever, weight-bearing, joint range, serial CRP |
| Clinical improvement and decreasing CRP | continue the pathway | targeted therapy when antibiogram available; oral switch according to clinical/laboratory criteria |
| Persistent pain, fever, or CRP not decreasing | suspect unresolved problem | insufficient drainage, associated osteomyelitis, abscess, uncovered pathogen, alternative diagnosis |
| Orthopedic follow-up | prevent and recognize sequelae | mobility, residual pain, return to weight-bearing, stiffness, joint damage, growth alterations in at-risk sites |
Practical message: Microbiological cure does not always coincide with the end of the orthopedic problem. In pediatric joint infections, follow-up serves to intercept stiffness, persistent pain, and growth sequelae.
When to truly drain?
There is no single answer here for all joints.
The pediatric hip is different from a small peripheral joint. A shoulder is different from an ankle. A knee can be managed arthroscopically or with lavage depending on the setting, experience, and clinical picture. In some joints and selected cases, repeated aspiration and antibiotics can be discussed; in others, especially the hip and shoulder, surgical drainage often remains the safest choice when suspicion is high or the fluid is purulent.
The criterion should not be solely "how much fluid is there."
It should be: involved joint, pain, general condition, laboratory results, quality of aspirate, age, anesthetic risk, availability of close follow-up, and probability of associated bone infection.
Follow-up is not an administrative detail
Once the acute phase is over, follow-up isn't just about saying "they're better."
It serves to verify recovery of joint range, resumption of weight-bearing, residual pain, normalization or clear decrease of CRP, any signs of associated osteomyelitis, deformity, stiffness, or late joint damage.
In younger children, and in joints near the physis, it also serves to avoid missing growth sequelae.
The 2026 meta-analysis on neonatal septic arthritis specifically highlighted the issue of sequelae: in neonates, the problem doesn't end with microbiological cure, because the hip and growing joints can suffer long-term consequences [5]. Although a different chapter from older children, the message remains useful: in pediatric joint infections, the acute phase and orthopedic follow-up are part of the same pathway.
The practical point
Pediatric septic arthritis is not well managed with a phrase like: "let's do Kocher."
Kocher helps.
CRP helps.
Ultrasound helps.
MRI helps.
Aspiration helps.
But none of these elements, alone, replaces the pathway.
The most frequent error is not failing to know the diagnosis. It's losing time in intermediate cases: those where the child isn't "textbook," tests haven't yet exploded, the X-ray is normal, and one is convinced that observation without a true re-evaluation strategy is sufficient.
In the hospital, the goal should be to have a shared flowchart rather than a single dogma: who evaluates, which tests start immediately, when to call anesthesia, when to aspirate, when not to wait for MRI, when MRI truly changes treatment, and who checks the child after the first 12-24 hours.
Because in pediatric septic arthritis, the correct diagnosis is important.
But the time it takes to get there is often even more so.
Bibliography
[1] Kocher MS, Zurakowski D, Kasser JR. Differentiating between septic arthritis and transient synovitis of the hip in children: an evidence-based clinical prediction algorithm. Journal of Bone and Joint Surgery American. 1999;81(12):1662-1670.
PubMed: https://pubmed.ncbi.nlm.nih.gov/10608376/
[2] Caird MS, Flynn JM, Leung YL, Millman JE, D’Italia JG, Dormans JP. Factors distinguishing septic arthritis from transient synovitis of the hip in children. A prospective study. Journal of Bone and Joint Surgery American. 2006;88(6):1251-1257.
PubMed: https://pubmed.ncbi.nlm.nih.gov/16757758/
[3] Nguyen A, Kan JH, Bisset GS, Rosenfeld S. Kocher Criteria Revisited in the Era of MRI: How Often Does the Kocher Criteria Identify Underlying Osteomyelitis? Journal of Pediatric Orthopaedics. 2017;37(2):e114-e119.
PubMed: https://pubmed.ncbi.nlm.nih.gov/27455184/
[4] Shaw KA, Sanborn R, Shore B, Truong W, Murphy JS. Current Variation in Joint Aspiration Practice for the Evaluation of Pediatric Septic Arthritis. JAAOS Global Research & Reviews. 2020;4(9):e20.00142.
DOI: https://doi.org/10.5435/JAAOSGlobal-D-20-00142
PubMed: https://pubmed.ncbi.nlm.nih.gov/32804837/
[5] Zhang K, Hu J. Prognostic outcomes of neonatal septic arthritis: a systematic review and meta-analysis. Journal of Orthopaedic Surgery and Research. 2026.
DOI: https://doi.org/10.1186/s13018-026-06662-1
[6] El-Sobky TA, Mahmoud S. Acute osteoarticular infections in children are frequently forgotten multidiscipline emergencies: beyond the technical skills. EFORT Open Reviews. 2021;6(7):584-592.
DOI: https://doi.org/10.1302/2058-5241.6.200155
PubMed: https://pubmed.ncbi.nlm.nih.gov/34377549/
[7] Woods CR, Bradley JS, Chatterjee A, et al. Clinical Practice Guideline by the Pediatric Infectious Diseases Society and the Infectious Diseases Society of America: 2021 Guideline on Diagnosis and Management of Acute Hematogenous Osteomyelitis in Pediatrics. Journal of the Pediatric Infectious Diseases Society. 2021;10(8):801-844.
DOI: https://doi.org/10.1093/jpids/piab027
PubMed: https://pubmed.ncbi.nlm.nih.gov/34350458/
Disclaimer: content for general informational purposes only. It does not replace medical evaluation.
