The scan is normal. The injury is not.

WHEN THE IMAGING DOES NOT MATCH THE PATIENT

A normal scan after an injury means the imaging did its job. It ruled out fracture, dislocation, cord compression and other emergencies. It was never built to find the small structures packed with nerves, such as a facet joint, that cause lasting pain. Those are found by numbing them and watching the response.

A certain silence follows a normal report. Nobody says you are making it up. The file just stops moving. You are left holding a study that answered a question you were not asking.

What the study was ordered to do

Imaging after a crash is ordered to rule out disaster: fracture, dislocation, cord compression, bleeding, a surgical emergency. It is very good at that, and a negative result is truly good news. But it is a different tool from the one that finds which small, nerve-rich part is causing pain six months later. The radiologist who read it never claimed it was that tool.

Two errors that look opposite and are the same error

Both come from treating a picture as a diagnosis.

• The normal scan, blamed on the patient. Nothing shows, so nothing is there, so this must be deconditioning, or anxiety, or something to do with a claim. That conclusion comes from the limits of a tool.
• The abnormal scan, blamed for the injury. Something shows — disc degeneration, a bulge, a cuff tear — and it gets blamed for everything. Findings that common usually cause no symptoms.

The data on the second error is not subtle. A systematic review of 3,110 asymptomatic people (people with no pain) across 33 studies looked at their spines. It found disc degeneration in 37% of 20-year-olds, rising to 96% of 80-year-olds. Disc bulges ran from 30% to 84% over the same span. The Chingford general population cohort followed 1,003 women aged 64 to 87. Full-thickness rotator cuff tears were present in 22.2%. And 48.4% of those tears caused no symptoms at all.

A finding present in most pain-free people cannot, on its own, explain one person's pain. It also cannot date it. Yet dating the injury is the use it most often gets put to after a crash.

The structures that are diagnosed by response, not by picture

Several of the most common sources of lasting pain after an injury barely show on imaging. They are found by numbing them and watching what happens.

• Cervical facet joints. In chronic neck pain after whiplash, controlled diagnostic blocks found a painful facet joint in 54% of 50 consecutive patients. Placebo-controlled blocks put the overall rate at 60% in 68 consecutive referrals. Details are on the whiplash page.
• Lumbar facet and sacroiliac joints. Same logic, same discipline. See the facet page and the sacroiliac page.
• The disc that hurts versus the disc that looks bad. This is a separate problem with its own page: a painful disc and an ugly disc are not the same disc.

What a dynamic examination adds that a static image cannot

A magnetic resonance study takes pictures of a limb lying still. Ultrasound can watch a tendon move. It can load a joint and watch it, and compare the hurt side with the other side in real time. It shows the tendon that subluxes (slips partly out of place) only at the end of its range. It shows the ligament that gapes only under stress. Those findings are invisible when the part rests inside a magnet.

Guidance also decides whether treatment reaches its target. Systematic reviews and meta-analyses found ultrasound-guided injections more accurate than landmark-guided injections at the hip. At the shoulder girdle, they were both more accurate and more effective. An injection that missed the structure is not a treatment that failed. It is a treatment that never happened. That is one reason published results in this field vary so much. The wider version of that argument is on why the trials disagree.

What we write down instead

A record that says “MRI unremarkable, continue therapy” carries nothing. The record we build states the mechanism and the date. It lists the structures examined and how. It gives the specific findings that provoked or relieved pain. It states what was injected, including its preparation, and how placement was confirmed. And it measures function the same way at each visit, so the numbers can be compared. Reporting in this field is poor. A systematic review found randomized trials of platelet-rich plasma for knee osteoarthritis follow the MIBO reporting guideline badly. If published trials do not describe what they injected, a clinic note surely will not, unless someone decides it must.

The same record carries the terrain: the inflammatory and metabolic markers measured at the start and again later. They are not background detail. They are there because they are the conditions the repair had to run in, and a result read without them is a result nobody can learn from. What we measure is set out separately.

Questions about imaging after an injury

If my MRI is normal, what is causing the pain?

Most often it is a structure the study is not built to show. That may be a joint capsule, a ligament under load, a small nerve, or a sensitized pain system. In the neck, the most common source found is a facet joint, which is set out on the whiplash page.

Should I get a better scan?

Sometimes. More often the answer is a different kind of exam, not a stronger magnet. That may be a dynamic ultrasound exam that watches the part move. Or it may be a diagnostic block that tests one structure directly. What an exam here involves is on the what-to-expect page.

The report mentions degeneration. Does that mean my injury did not cause this?

It means a common age-related finding is present. That is true of most people without pain — 96% of asymptomatic 80-year-olds have disc degeneration. A degenerated structure has less reserve, and an event can exceed what was left. What that means for a joint that was injured, not worn out, is on the post-traumatic knee page.

Does ultrasound guidance actually change the result of an injection?

At the hip and the shoulder girdle, the pooled evidence says yes on accuracy. At the shoulder, it says yes on effectiveness too. Why we treat placement and preparation as part of the treatment, not just technique, is on the preparation page.

Nobody has been able to tell me what is wrong. Where do I start?

Start with a mechanism, a date and an exam aimed at likely structures, not at the existing report. Then take a diagnostic step that tests the leading suspect. That sequence is described on the tissue timeline page.

The diagnostic thread

• The neck injury a scanner cannot see
• A disc that hurts and a disc that looks bad
• Why the trials disagree
• The tissue clock after an injury
• The shoulder after a collision
• Why we ask for bloodwork