What acupuncture actually does in the body
A patient rolled up her sleeve for the needles last month and said, "I'll be honest with you, I think this is placebo. But my knee hurts and I'm out of other ideas." I told her that's a fair place to start. I'd rather treat a skeptic who shows up than a believer who doesn't.
So here is what I actually tell people about what the needles do. Not the version with soft music and vague talk about energy. The version I'd defend to her orthopedist.
What the needle actually touches
An acupuncture needle is a filiform needle, thinner than a hair, solid rather than hollow. I place it through the skin into muscle and connective tissue at points that sit, almost always, over nerve-rich territory. Then I move it a little. That small twist is doing more than it looks like it's doing.
When the needle goes in and I rotate it, the tissue around it releases adenosine, a molecule that quiets local pain signaling through A1 receptors on nearby nerves. This isn't a theory I'm attached to. Goldman and colleagues showed it directly in 2010: adenosine jumped roughly 24-fold around the needle during treatment, and in animals bred without the A1 receptor, the pain relief vanished (Goldman N, et al. Nature Neuroscience, 2010).
There's a mechanical piece too. The needle lightly grabs the collagen in connective tissue and tugs, and the cells respond to that pull. So part of what you feel, that dull heaviness practitioners call de qi, is real tissue mechanics, not suggestion.
The nervous-system part most people miss
Here's where acupuncture gets interesting to me, and where the knee-pain conversation opens up into everything else I use it for.
Those nerve-dense points send a signal up the spinal cord to the brainstem, and the brainstem answers by shifting the autonomic balance. In plain terms, it nudges you out of fight-or-flight and toward the parasympathetic side, the rest-and-digest side.
I watch it happen on the table. Breathing slows. Shoulders drop off the ears. People who swore they were "too wired to relax" are asleep in nine minutes. That drop is a vagal shift, and in research it shows up as measurable change in heart rate variability.
The mechanism isn't hand-waving anymore. In 2021, a Harvard team mapped a specific population of sensory neurons that lets electroacupuncture at certain body points drive a vagal-adrenal anti-inflammatory response, and they showed that where you needle and how strongly both change the result (Liu S, et al. Nature, 2021). That's a long way from meridians as magic. It's neuroanatomy.
This is why I reach for acupuncture in patients who never came in for pain at all. The stressed-flat executive. The woman whose sleep breaks at 3 a.m. The gut that seizes up every time work does. They share an autonomic pattern, and needles speak that language well.
Where the evidence is strongest, and where it isn't
I try to be straight with patients about this, because overselling it does the field no favors.
The strongest evidence is for chronic pain. The largest analysis we have pooled individual data from about 20,000 patients across high-quality trials and found acupuncture more effective than both sham needling and usual care for chronic back and neck pain, osteoarthritis, and chronic headache, with benefits that held at a year (Vickers AJ, et al. The Journal of Pain, 2018). The effect over sham is modest but real, and the effect over "carry on as you were" is substantial.
For migraine, a Cochrane review of the prevention trials found acupuncture at least as effective as the standard preventive medications, with fewer people dropping out from side effects (Linde K, et al. Cochrane Database of Systematic Reviews, 2016). I've watched that play out in clinic more times than I can count.
Knee osteoarthritis sits in the same solid tier, which brings me back to my skeptic. It's one of the conditions in that 20,000-patient analysis where acupuncture outperformed usual care, and it's a joint where I'd rather try needles and herbs for a few weeks than start someone down the road of standing anti-inflammatory use at 58. We can always escalate. We can't un-ring some of those bells.
And then there are conditions where the evidence is thin or mixed, and I say so. Acupuncture is a tool with a shape. It's very good at pain and at downshifting an overcranked nervous system. It is not a cure for everything, and anyone who tells you it is should make you nervous.
Acupuncture is not dry needling, and I keep saying so for a reason
Patients ask me all the time whether the dry needling their physical therapist offers is the same thing. It uses the same filiform needle, so I understand the confusion. The training behind the needle is where they part ways.
A licensed acupuncturist completes thousands of supervised clinical hours, national board examinations, and formal certification in clean-needle technique, point location, and pattern diagnosis before practicing independently. Dry needling, as it's commonly offered, is a short add-on course, sometimes a weekend, bolted onto another profession's scope.
That gap matters for your safety. Needling near the ribcage carries a real risk of puncturing a lung if you don't know the anatomy cold, and those case reports exist. Patients deserve to know how many hours of needle training the person holding the needle actually has. I'm not disparaging physical therapists, whose work I refer to and respect. I'm saying the two things are not equivalent, and treating them as interchangeable is how people get hurt.
What I pair it with in clinic
Needles are rarely the whole plan. They move something; the rest of the work keeps it moved.
For musculoskeletal pain, I'll often send a patient home with an herbal formula to hold the gains between sessions. Corydalin is one I use a lot for circulation and muscle discomfort, and for stubborn cases I'll point people to the broader topical relief options too.
For the stress-and-sleep crowd, the ones whose nervous systems forgot how to stand down, I lean on calming formulas like Shine alongside the needling, and I'll build the rest of the plan from our sleep and stress shelf depending on the pattern. The acupuncture opens the door. The herbs and the habits keep it from swinging shut.
So, should you try it
If you have chronic pain that hasn't budged, tension headaches or migraines, or a nervous system stuck in high gear, acupuncture is one of the better-supported things I can offer you. A first visit is mostly talking, then a short, calm treatment. Most people feel the autonomic shift before they feel anything change in the pain, and the pain follows over a few sessions.
On timing, I'm honest that this is not a one-and-done. For a chronic problem I usually want to see you weekly for about six visits before we judge whether it's working, then we taper. If nothing has moved by the sixth session, I'll tell you, and we'll change course rather than keep you coming back out of hope. That's the deal I make with every patient on the table.
Come in as a skeptic. That knee-pain patient did. She's still skeptical about the word "energy," and her knee is a lot quieter. I can work with that.
These statements have not been evaluated by the Food and Drug Administration. This information is not intended to diagnose, treat, cure, or prevent any disease. Consult your healthcare provider before starting any new supplement regimen.
References
Goldman N, Chen M, Fujita T, et al. Adenosine A1 receptors mediate local anti-nociceptive effects of acupuncture. Nature Neuroscience. 2010;13(7):883-888.
Liu S, Wang Z, Su Y, et al. A neuroanatomical basis for electroacupuncture to drive the vagal-adrenal axis. Nature. 2021;598(7882):641-645.
Vickers AJ, Vertosick EA, Lewith G, et al. Acupuncture for chronic pain: update of an individual patient data meta-analysis. The Journal of Pain. 2018;19(5):455-474.
Linde K, Allais G, Brinkhaus B, et al. Acupuncture for the prevention of episodic migraine. Cochrane Database of Systematic Reviews. 2016;(6):CD001218.