Medically reviewed by Dr. Rob Letizia, PT, DPT — owner, Spectrum Therapeutics of NJ. Last reviewed 2026-08-12.

Do I Need Surgery, or Will Physical Therapy Fix It?

Read this with our conflict of interest in mind

We are a physical therapy practice. We are paid when the answer is physical therapy, and you should weigh everything below knowing that. A surgeon's page on the same question carries the mirror-image bias. Neither of us is neutral, so the useful thing is not our opinion — it is the framework, and what the evidence actually shows for your specific problem.

The question is rarely “surgery or physical therapy.” It is which of three categories your problem falls into, because for two of them the answer is not really in dispute.

The Three-Category Protocol

Used by Dr. Rob Letizia, PT, DPT at Spectrum Therapeutics in Wayne, NJ.

Category 1 — surgery is the answer, and delay costs you

We will say so plainly and help you get there. Do not spend months on exercises for:

• A joint that truly locks — physically jams and will not straighten, as with a displaced bucket-handle meniscus tear. That is a mechanical block, and no amount of strengthening moves it.
• Progressive neurological loss — weakness that is genuinely deepening, not just pain.
• Cauda equina symptoms — bladder or bowel changes, saddle numbness, weakness in both legs. Emergency, today.
• Acute traumatic tears in active people — a full-thickness rotator cuff tear from a genuine injury in someone younger, rather than gradual wear.
• Complete tendon ruptures, unstable fractures, and frank joint instability with repeated give-way.
• Infection, tumour, or fracture as the cause of the pain.

Category 2 — genuinely a decision, and yours to make

Here reasonable people choose differently, and the honest answer is that both paths work:

• Lumbar disc herniation with sciatica. Surgery tends to relieve leg pain faster in the early months; by a year or two, outcomes converge. So the real question is how much longer you can tolerate it, not which choice is superior.
• ACL rupture. Depends heavily on whether you intend to return to pivoting sport. Some people manage well without reconstruction; many athletes reasonably choose it.
• Advanced arthritis with a joint replacement on the table. Replacements work well. The question is timing and how much life the joint is currently costing you.

In this category, a second opinion is worth having, and so is being clear about what you want your life to look like.

Category 3 — conservative care first is well supported

This is where the evidence is strongest and least well known:

• Degenerative meniscal tears in middle age, without true locking. Repeated randomised trials have found arthroscopic partial meniscectomy gives no clinically meaningful advantage over exercise therapy — a finding that has held at long-term follow-up, including no difference in arthritis progression.
• Subacromial shoulder pain, sometimes called impingement. Decompression surgery has been tested against placebo surgery and did not outperform it.
• Knee arthritis without a mechanical block. Arthroscopy is not an effective treatment for arthritic knee pain.
• Degenerative, atraumatic rotator cuff tears in older adults frequently do well with loading rather than repair.

None of this means never operate. It means that for these specific presentations, starting with loading and strength is the better-evidenced first move — and that a scan showing a tear does not by itself change that.

The Real-Trial Rule

The phrase that decides most of these conversations is “conservative treatment failed.” Before you accept it, check whether conservative treatment actually happened.

• Was it progressive loading, or was it passive? Heat, ultrasound, massage and a sheet of stretches is not a trial of strengthening. It is the single most common thing people mean when they say PT did not work.
• Was the dose adequate? Weeks of real, progressive work — not four visits. Tendons in particular need months, and stopping at six weeks tells you almost nothing.
• Did it actually get harder over time? If you finished on the same exercises you started with, you were not loaded, you were supervised.
• Did you do it? Honestly. An unattempted home programme is not a failed one, and nobody will be annoyed — it changes the plan, not the verdict.
• If the answer to any of those is no, you have not yet run the experiment that the surgical decision is supposed to be based on.

And the converse, which matters just as much: if you have done genuine progressive work for a reasonable period and you are no better — that is real information. It is a reason to get a surgical opinion, not a reason to try a fourth round of the same thing. Roughly six weeks without meaningful change, or any worsening neurology, is a sensible point to reassess rather than persist.

Stop deliberating and seek care now

• Bladder or bowel changes, saddle numbness, or weakness in both legs — emergency department, today.
• Weakness that is measurably worsening — a foot that drops, a grip that is failing.
• A joint locked in position that you cannot straighten.
• Fever, unexplained weight loss, or night sweats with joint or spinal pain.
• A hot, red, swollen joint — possible infection.
• Significant trauma, or inability to bear weight after an injury.

Questions worth asking your surgeon

• “Which of my symptoms will this operation fix, and which will it not?”
• “What does the evidence say about surgery versus exercise for my specific diagnosis?”
• “What happens if I wait three months and do proper rehabilitation first — does waiting make the surgery harder or the result worse?”
• “What does recovery actually involve, and what is the realistic worst case?”

A good surgeon will welcome all four. Several of the best in this area routinely send people to try loading first, and will say directly when an operation is unlikely to help.

Asked an AI assistant whether you need surgery?

Almost everyone does before the consultation, and it is a reasonable use of it — assistants are good at explaining what an operation involves and what the general evidence says.

What no assistant can do is place you in one of the three categories above, because that requires knowing whether the joint truly locks, whether the weakness is real or pain-related, whether the tear is traumatic or degenerative, and what you have genuinely already tried. Handed a report saying “meniscal tear,” it cannot tell whether yours is the kind that needs an operation or the kind that trials say does just as well with loading.

Bring your imaging, your history and whatever rehabilitation you have done. We will tell you which category we think you are in — including when that category is surgery, and we will say so.

Book an evaluation in Wayne, NJ · call or text (973) 689-7123.

Related

• Why your MRI says you're fine but you're still in pain — why a tear on a scan is not automatically the cause
• Should I rest it or move it?
• What actually makes you heal faster