Skip to content
Arthritis & Joint Pain

Do You Need an MRI for Hip or Knee Replacement — or Is an X-Ray Enough?

Medically reviewed by Matthew Harb, M.D.Updated August 12, 20267 min read

One of the most common questions I get is whether an MRI is needed before deciding on a hip or knee replacement. For the large majority of patients the answer is no. A properly taken X-ray shows the space between the bones — which is the cartilage — and that is what tells me how advanced the arthritis is. An MRI is excellent at soft tissue, and soft tissue is rarely what decides a joint replacement.

Key takeaways

  • For most patients considering hip or knee replacement, a weight-bearing X-ray is all the imaging that is needed.
  • The space between the bones on an X-ray is the cartilage. Narrowing of that space is how arthritis severity is graded.
  • How the X-ray is taken matters enormously — standing, weight-bearing, and the right specialty views. A film taken lying down can make an arthritic joint look far better than it is.
  • MRI is the better study for soft tissue: muscles, tendons, ligaments, and labrum. Those structures rarely decide whether a joint gets replaced.
  • MRI has a role in specific situations — early cartilage damage in patches, suspected avascular necrosis, or pain that does not match the X-ray.
  • An MRI almost never changes the decision in a patient with clear arthritis on X-ray and symptoms to match.

The short answer

For the large majority of patients considering a hip or knee replacement, you do not need an MRI. A good X-ray is enough. This is one of the questions I am asked most often, usually by someone who assumes the more advanced scan must be the more useful one. For this particular decision, it generally is not.

The two studies answer different questions. An X-ray tells me how much cartilage is left. An MRI tells me about soft tissue — muscles, tendons, ligaments, the labrum. When the question is whether a joint is arthritic enough to justify replacing it, the X-ray is the study that answers it.

What an X-ray actually shows

Cartilage does not appear on an X-ray. What you see is bone — and, critically, the space between the bones. That space is the cartilage. It is there by inference: the gap exists because something is holding those two bone surfaces apart.

So when that space narrows on the film, it means the cartilage has worn. When it disappears entirely and the bone surfaces meet, that is what “bone-on-bone” means. That single measurement is the backbone of how arthritis severity is graded, and it is visible on a study that takes a few minutes and costs a fraction of an MRI.

What I am looking at on your X-ray

Joint space narrowing, bone spurs (osteophytes), hardening of the bone just beneath the surface, cysts, and the overall alignment of the limb. Together those tell me how advanced the arthritis is and what an operation would involve.

Why how the X-ray is taken matters

This is the part patients almost never hear about, and it matters more than whether you also had an MRI.

An X-ray taken lying down can make an arthritic joint look considerably better than it is. Without your body weight pressing through the joint, the remaining space is not being loaded, and the gap on the film looks wider than it functionally is. I have had patients told their arthritis was “mild” on the basis of a film taken on a table, whose standing views showed bone touching bone.

  • Weight-bearing — taken standing, so the joint is loaded the way it is when it hurts
  • The right views for the joint, including specialty views that show compartments a standard film can miss
  • Recent enough to reflect where the arthritis is now, not a year ago
  • Both sides where it is useful, for comparison

With the right views, a plain X-ray shows essentially everything needed for surgical decision-making in an arthritic joint. When patients tell me their imaging was inconclusive, the problem is more often how the films were taken than a limitation of X-ray itself.

What an MRI shows

MRI is an excellent study, and I order them when they will change something. It shows soft tissue in detail that X-ray cannot: meniscus, ligaments, tendons, muscle, labrum, and cartilage directly rather than by inference. It also shows the bone itself — marrow oedema, stress reactions, and early avascular necrosis.

The catch is that those structures are rarely the ones that decide a joint replacement. A torn meniscus in a knee that is already bone-on-bone does not change the recommendation, because the arthritis is the problem and treating the tear in isolation tends to disappoint.

When I do order an MRI

There are genuine situations where it earns its place:

  • The X-ray looks better than the patient does — significant pain without the joint space narrowing to explain it
  • Suspected avascular necrosis, where the bone can be dying well before the X-ray shows anything
  • Focal cartilage damage in a younger patient — wear in patches rather than across the whole joint, which an X-ray can genuinely miss
  • A mechanical symptom such as true locking or giving way that suggests a soft-tissue problem alongside the arthritis
  • A younger patient where a joint-preserving option is still realistically on the table and I need to see the labrum or the cartilage surfaces
  • Suspicion that the pain is not coming from the joint at all

The one thing an MRI cannot do

It cannot tell you whether to have surgery. No image can. Imaging tells me what the joint looks like; only you can tell me what it is costing you to live with.

The knee specifically

For a knee, standing views — including a flexed weight-bearing view that shows the back of the joint where wear often starts — usually settle the question. Alignment matters too: whether you are bow-legged or knock-kneed tells me which compartment is taking the load, and that is visible on a full-length standing film.

Where MRI helps in a knee is a younger patient with a focal cartilage lesion, or someone whose symptoms suggest a meniscus problem in a joint that is not yet arthritic. If you have been told your knee is bone-on-bone, an MRI is unlikely to add anything to that conversation.

The hip specifically

Hips are, if anything, more X-ray-driven than knees. A standing pelvis film shows the joint space, the shape of the socket and the femoral head, and any deformity from dysplasia or impingement. That is also what the operation is planned from.

The main reason I order a hip MRI is suspected avascular necrosis, where bone can be failing long before the X-ray reflects it — and in that situation the MRI genuinely changes what I recommend and how urgently. The other is a younger patient where a labral or joint-preserving question is live.

Ordering imaging you do not need has a cost

An unnecessary MRI is not harmless. It costs money, it adds weeks, and it frequently finds something incidental — a small tear, a bit of tendinopathy — that has nothing to do with why you hurt but is now on a report, worrying you and inviting a procedure that will not help.

If you have a good weight-bearing X-ray and symptoms that match it, you already have what a surgeon needs to give you a real answer. Bring the films to your visit — and if you are gathering records for a second opinion, those X-rays are the single most valuable thing you can send.

Frequently asked questions

Do you need an MRI before a knee replacement?

Almost never. A weight-bearing X-ray shows the joint space — the cartilage — and that is what determines how advanced the arthritis is. If the X-ray shows significant narrowing and your symptoms match, an MRI will not change the plan. I order one when something does not add up: pain out of proportion to the X-ray, a suspicion of avascular necrosis, or a mechanical symptom that suggests a soft-tissue problem alongside the arthritis.

Do you need an MRI before a hip replacement?

Rarely. A standing X-ray of the pelvis and hip shows the joint space and the shape of the bone, which is what a hip replacement is planned from. MRI becomes useful when I am concerned about avascular necrosis, when the X-ray looks better than the patient does, or when I need to see the labrum and surrounding soft tissue in a younger patient where a joint-preserving option might still be on the table.

Can an X-ray show if you need a knee replacement?

An X-ray shows how much cartilage is left; it cannot by itself tell you whether to have surgery. That decision comes from what the joint is costing you day to day. I have seen bone-on-bone X-rays in patients walking several miles a day who need nothing, and far milder films in patients who cannot get through a workday. The X-ray tells me what I would be operating on. Your symptoms and function tell me whether I should.

Does an MRI show arthritis?

Yes, and often in more detail than an X-ray — but more detail is not the same as more useful. MRI can pick up cartilage thinning and bone marrow changes earlier than plain films, which matters in a young patient with a focal cartilage problem. For someone with established arthritis who is weighing a replacement, that extra detail generally does not change what I would recommend.

Why did another doctor order an MRI for my arthritis?

It happens often, and it is not necessarily wrong — MRI is the right study when a soft-tissue problem is genuinely in question, such as a suspected meniscus tear or rotator-cuff problem. What I would push back on is the idea that a replacement decision requires one. If you already have clear arthritis on a good weight-bearing X-ray, the MRI is unlikely to change the recommendation, and it adds cost and delay.

Do I need a new X-ray if mine is a year old?

Usually yes, if we are actively deciding about surgery. Arthritis changes over time, and I want to see where the joint is now rather than where it was. It is also worth making sure the films are weight-bearing, because a study taken lying down is not comparable and can understate the narrowing considerably.

References

  1. American Academy of Orthopaedic Surgeons — Management of Osteoarthritis of the Knee (Non-Arthroplasty), Evidence-Based Clinical Practice Guideline

This article is for general education and is not a substitute for personalized medical advice. Recovery timelines vary by patient, procedure, medical history, and surgeon-specific protocol. Please consult Matthew Harb, M.D. about your specific condition.

Patient experiences

What patients say

“A really smooth operation — I was discharged the same day and basically able to walk easily within a day.”
Mark T.Hip replacement
“I walked into the surgical center in great pain and walked out with a new knee and a renewed person.”
Brian K.Knee replacement
“My full knee replacement is a big success — six months after surgery I’m hiking and kayaking again.”
Lynn H.Knee replacement

5.0 rating based on 524 verified patient reviews

Read reviews on Google: Washington, D.C.Germantown

Have questions about your hip or knee?

Schedule a consultation with Dr. Harb to discuss your options and build a plan to get you back to an active life.