Do You Really Need Surgery? An Orthopedic Surgeon's Honest Framework

Do you really need surgery? Sometimes yes, and sometimes it's recommended too soon. You likely need an operation when there's a clear structural failure that won't heal on its own and is robbing you of strength or function. You can usually try non-surgical care first when you have pain without that kind of failure and you haven't yet done a fair trial of physical therapy, bracing, and time. The only way to know which camp you're in is an exam and the right imaging.

I spent more than two decades as an orthopedic surgeon and performed over 3,500 procedures. I believe in surgery. When it's the right call, it's one of the most reliable things we do in medicine. More than 90% of modern total knee replacements are still working well 15 years later (OrthoInfo, AAOS). I'd never talk you out of an operation you truly need.

But I now also offer non-surgical, regenerative care, and that changes the conversation in your favor. A surgeon who only operates is paid to operate. A surgeon who can also offer you physical therapy guidance, bracing, time, or a regenerative injection has no reason to push you toward an OR you don't need. That's the whole point of this practice, and it's why I'd rather refer you to surgery than sell you something that won't work. This article is the honest framework I use, written so you can apply it to your own joint.

Here's my bias, stated plainly: I'd rather under-treat than over-treat. A surgery you didn't need can't be un-done. A few extra weeks of conservative care, in the right case, almost never hurts you.

Signs surgery is likely needed soon

These are the situations where, after an exam and imaging, I'm most likely to say "this is a surgical problem, and waiting probably won't help." Not all of them are emergencies, but they share one thing: a real structure has failed, and your body can't repair it on its own.

• A large, full-thickness rotator cuff tear with real weakness. When the tendon is torn all the way through and you've genuinely lost strength or the ability to lift the arm, repair gives the best long-term result, and tears can enlarge with time while the muscle atrophies or fills with fat, which lowers the odds of a good repair later (OrthoInfo, AAOS). See our honest breakdown of the shoulder and rotator cuff.
• True, end-stage bone-on-bone arthritis with deformity. When the cartilage is genuinely worn away across the joint and the bone is rubbing on bone, often with bowing or angular deformity, no injection regrows that cartilage. Joint replacement is the treatment with no viable regenerative substitute at that stage (StatPearls, 2023). More on knee arthritis.
• Mechanical locking from a displaced meniscus or loose body. If a torn flap of meniscus or a loose fragment is physically blocking the joint, so the knee locks or won't fully straighten, that's a mechanical problem an injection can't release. Displaced bucket-handle tears that cause true locking are typically treated arthroscopically (PMC, case literature).
• Certain acute traumatic injuries. Displaced fractures, dislocations that won't stay reduced, complete tendon ruptures in active people, and similar acute structural failures often need timely surgical treatment. These are usually decided in the days after the injury, not after months of waiting.
• You've already done appropriate conservative care, and it failed. If you've completed a real course of physical therapy, activity changes, bracing, time, and reasonable injections for a structural problem and you're still limited, that's a legitimate reason to move toward surgery. "Failed appropriate conservative care" is the phrase that matters, and the word appropriate is doing a lot of work.

Signs it's reasonable to try conservative or regenerative options first

On the other side, here's where I most often say "let's not operate yet." The common thread: you have pain, but no major mechanical or structural failure, and you haven't yet given the non-surgical path a fair shot.

• Mild-to-moderate arthritis that isn't yet bone-on-bone. There's still cartilage to protect, and the AAOS recommends conservative measures, low-impact exercise, strengthening, weight management where relevant, first for symptomatic knee arthritis (AAOS, 2021).
• Tendinopathy with the tendon still intact. An irritated, degenerated tendon that hasn't fully torn often responds to progressive loading rehab and, for the right condition, a regenerative injection. Rotator cuff and elbow tendinopathy are typically managed conservatively first, surgery only if that fails (Medicina, 2025).
• Pain without major mechanical or structural failure. No locking, no giving-way from instability, no clear loss of strength from a torn structure. Pain alone, while real and worth treating, is rarely a reason to rush to surgery.
• You simply haven't done the non-surgical work yet. If you haven't completed physical therapy, tried bracing or activity changes, or given it time, you don't yet know what your body can do on its own. Many conditions improve without an operation. Roughly 90% of low-back pain, for instance, settles without surgery (AANS).

The honest part: some problems do need surgery, and waiting can cost you

I don't want the message to be "avoid surgery." That would be its own kind of dishonesty. The truth is balanced: many conditions improve without surgery, and some genuinely require it, and for a few time-sensitive problems, delaying truly-needed surgery can make it harder to fix.

The rotator cuff is the clearest example. A torn cuff tendon can enlarge over time, and the muscle can atrophy or fill with fat, all of which lower the chances of a durable repair (OrthoInfo, AAOS). End-stage arthritis is different, it won't get medically dangerous if you wait, but it also won't get better, and no injection will rebuild cartilage that's already gone. The skill isn't "always wait" or "always operate." It's matching the decision to the actual tissue in front of us, which is why the exam and imaging matter more than any rule of thumb.

Where PRP fits in this picture

Because I offer it, I want to be precise about what PRP (platelet-rich plasma) can and can't do. It is one tool among several, not a surgery-eraser. The evidence is genuinely strong for some conditions and weak for others, and an honest framework has to say so.

So PRP can be a reasonable step before surgery for the right knee, elbow, or heel. It is not a substitute for repairing a significant full-thickness tear or replacing an end-stage joint, and I'll tell you that to your face rather than book an injection that the research doesn't support. The deeper dive lives on our PRP therapy page and across our conditions hub. The honest bottom line: the only way to know whether PRP makes sense for you is an exam plus imaging, not a marketing promise.

Questions to ask before you agree to surgery

Print this, screenshot it, bring it to any consult. A surgeon who's confident in the recommendation will welcome every one of these questions.

1. What exactly is structurally wrong, and how do you know? Ask to see the imaging and have it explained in plain language.
2. What happens if I do nothing for three months? If the answer is "it stays the same" or "it slowly improves," there's usually time to think.
3. Have I done a real trial of non-surgical care? Physical therapy, bracing, activity changes, time, and where appropriate, injections, done properly, not just mentioned.
4. Is this surgery time-sensitive? Some problems get harder to fix if you wait. Ask directly whether yours is one of them.
5. What are the realistic odds this surgery fixes my specific problem? Not the average, your case, your tissue, your activity level.
6. What does recovery actually look like, week by week? Time off work, rehab, and what "good" looks like at six and twelve months.
7. What are the risks and the failure rate? Every operation has both. A straight answer is a good sign.
8. Would you support me getting a second opinion? The right answer is an enthusiastic yes.

A second opinion isn't an insult to your surgeon. In published studies of patients told they needed elective orthopedic surgery, a large share had not had the operation six months later once a second opinion suggested a brace, physical therapy, or that symptoms weren't severe enough yet (PMC, second-opinion study). Good surgery survives a second look.

The bottom line

Do you really need surgery? The honest answer is "it depends on what's actually broken, and how much of the non-surgical path you've genuinely walked." If you have a true structural failure with weakness or locking, surgery may be exactly right, and waiting may not help. If you have pain without that kind of failure and haven't yet tried physical therapy, bracing, time, or a well-matched regenerative option, it's usually reasonable to try those first.

I built this practice so you can get that read from someone who isn't paid only to operate. If surgery is what you need, I'll tell you, and help you get to the right surgeon. If it isn't, I'll tell you that too. Find out in two minutes whether non-surgical care is a reasonable first step for you, or see our open pricing and HSA/FSA details if cost is on your mind.

Frequently asked questions

← Back to all articles