Salter-Harris Detail
- The Salter-Harris system classifies fractures that involve the growth plate (physis) — the cartilage band where a kid's bones get longer.
- There are five types, and the famous mnemonic SALTR maps the fracture line: Slipped, Above, Lower, Through, Rammed (crush).
- Type II is by far the most common; Type I is easy to miss because it can look completely normal on X-ray.
- Higher numbers generally mean a worse prognosis, because the more the fracture crosses both the growth plate and the joint surface, the higher the risk of growth arrest.
- The physis is invisible to fractures-only thinking: always compare to the other side and respect the cartilage you can't see.
Kids' bones come with a built-in construction zone: a ribbon of cartilage near each end called the growth plate, or physis, where new bone is laid down so the limb can get longer. It's strong, but it's also the softest link in the chain — so when a child takes a fall, the fracture often goes right through this zone instead of snapping the shaft like an adult bone would. The Salter-Harris system is just a tidy way of describing exactly which parts of that construction zone got hit.
If you want the broader fracture-description vocabulary first, it's worth a detour through how to describe a fracture; this page zooms in on the pediatric special case.
First, the anatomy you're naming around
Before the five types make sense, you need three landmarks at the end of a growing bone. Think of the bone end as an ice cream setup: the epiphysis is the rounded scoop at the very tip (the part that forms the joint), the physis is the thin layer of cone holding the scoop on, and the metaphysis is the flared top of the cone — the wider shaft just below the plate.
The physis itself is cartilage, and cartilage is nearly invisible on X-ray — it shows up as a dark, lucent band. That's the single most important thing to internalize: the growth plate is a normal black line, not a fracture. Mistaking a physis for a break is a rite of passage, and so is the opposite mistake.
The five types (and the SALTR trick)
The whole system tracks one question: where does the fracture line travel relative to the growth plate? The classic mnemonic SALTR lines the types up one through five.
| Type | SALTR | Where the fracture goes | Quick read |
|---|---|---|---|
| I | Slipped | Straight through the physis only | Epiphysis "slips" off the metaphysis; bone may look normal |
| II | Above | Through the physis, then up into the metaphysis | A metaphyseal corner fragment ("Thurston Holland" corner); most common |
| III | Lower | Through the physis, then down into the epiphysis | Reaches the joint surface — intra-articular |
| IV | Through | Through metaphysis, physis, and epiphysis | A single line crossing everything; intra-articular |
| V | Rammed | Crush/compression of the physis | The sneaky one — often invisible at first |
A way I keep "above vs. lower" straight: hold the picture so the joint is at the bottom. Type II shoots up away from the joint into the metaphysis; Type III dips lower toward the joint into the epiphysis. The letters are literally telling you which direction the line escapes the plate.
Why the numbers matter: growth arrest
The reason anyone bothers numbering these isn't trivia — it's prognosis. The higher-number types (especially III, IV, and V) either cross the joint surface or crush the growth plate, and both threaten the physis's ability to keep making bone. If part of the plate dies or bridges over with bone, that side stops growing while the rest keeps going — and you get a limb that's short, bent, or both as the child grows up. So a "small" fracture in a five-year-old can matter for the next decade.
A Salter-Harris injury is graded on a growing skeleton, so the same fracture pattern is more consequential the younger the child and the more growth they have left. The injury and the years of remaining growth both feed the worry.
Type I and Type V: the two that hide
These deserve their own spotlight because they're the ones that get missed.
Type I can leave the bone looking perfectly normal on the X-ray, because the only thing that broke is invisible cartilage. If a child has bony tenderness right over a growth plate after trauma, you treat it as a fracture even when the film is clean.
The "negative X-ray" trap. A non-displaced Salter-Harris I shows no fracture line and no displacement — the radiograph can look completely normal. Clinical tenderness over the physis is the finding. Don't let a clean film talk you out of it.
Type V is the crush injury, where the physis gets compressed straight along its axis. It's notoriously subtle or invisible up front and is often only recognized in retrospect once growth has already gone wrong. Humbling, but worth knowing it exists.
How to not get fooled
Two habits save you here. First, always get and compare the other side's anatomy when you're unsure — a kid's growth plates are symmetric, so the uninjured limb is your built-in normal. Second, anchor yourself on the physis: ask "is this dark line a normal growth plate, or is the fracture using the growth plate as part of its route?"
When you see a fracture line heading toward a growth plate, trace it to the end. Does it stop at the plate (I), kick up into the metaphysis (II), dive into the epiphysis (III), or cross everything (IV)? Naming the type is just following the line to where it exits.
For the broader pediatric trauma context — including the patterns that should make you think about non-accidental trauma — it's worth knowing where these injuries fit alongside the other fractures kids sustain.
If you remember nothing else: the growth plate is cartilage, cartilage is the weak link, and the Salter-Harris number simply tells you how badly the fracture threatened the part of the bone that still has growing to do.