We finally put the whole decision on one page.
So a group of us just published something I have wanted in print for a long time. It is a consensus statement in the International Journal of Sports Physical Therapy, nine clinicians who needle for a living, and it lays out a decision tree for dry needling. Who is a candidate, how much to needle, when to go to the spine, and what to do after. Three focuses this issue, and they are the three I will run every time: the framework and the safety inside it, the model underneath it, and one real clinical spot where you can use it Monday.
So here is the part that matters. This was a modified Delphi, nine of us, four rounds, and we hit unanimous agreement on all of it. And what did we agree on? Not a magic point or a fancy technique. We agreed that the needle is the end of the decision, not the start. You screen first. Contraindications, any present, you stop. Precautions, any present, you slow down and do your risk-benefit out loud. Then you dose by irritability and tolerance, not by habit. High irritability, low tolerance, keep it small, one to two pistons or stop at the first twitch. Low irritability, high tolerance, now you open it up, ten to twenty passes, multiple sites, e-stim. Same diagnosis, three different doses. And a few regions stay off the table unless you are trained in them, the thorax, the pelvic floor, the neurovascular bundles, the jaw, the front of the neck. This is Level V, expert consensus, so I will not dress it up as more than it is. But it is the first time this whole decision lives on one page.
Dose the patient before you dose the needle. Grade the irritability first, then pick your needles and pistons off that. The irritable patient who gets a low-irritability dose leaves worse, and usually never comes back to tell you why.
So this is the move that makes the whole thing work, and it is right there in the consensus. Dry needling goes to the spine when your neurological assessment shows a deficit, in the dermatomes, the myotomes, the deep tendon reflexes, or neurodynamics. If the segment is driving the symptom, you treat the segment first, the multifidus at that nerve root level, and you reassess before you ever chase the painful periphery. That is the radiculopathic model in one sentence. Think of the nerve like a garden hose. If it gets smooshed up at the root, it does not feed the garden downstream, and you can needle the calf or the elbow all day and it keeps flaring. There is even new literature this month leaning the same way, a review using the herniated disc as the poster child for the scan and the pain not lining up.
Before you needle the painful calf, elbow, or shoulder, screen the segment above it. Dermatomes, myotomes, reflexes, neurodynamics. If the segment is driving it, needle the segment first and reassess. The calf that will not settle is often a back nobody assessed.
So let me put the framework on a real diagnosis, because TMD is a great test of it. A blinded trial needled the masseter and the temporalis in 40 patients with myofascial TMD against a sham, and one session dropped the pain, a real drop. But then they put EMG on those same muscles, at rest and at a hard clench, and the electrical activity basically did not change once you corrected the numbers. Sit with that. The pain moved and the muscle reading did not, which is exactly the point from focus one. We are not fixing a knot in the meat, we are talking to the nervous system. And notice what they needled, the muscles, not the joint, because the jaw is one of those trained-only regions in our consensus. So you respect that line. Why does the needle do that? Honestly, I still cannot tell you the full mechanism, and neither can anybody else yet. But my gut, and now this, says stop grading it by the muscle.
If you treat jaw pain, needle the muscles you are trained to needle and stay off the joint unless you have the reps. Then judge it by the reassessment, not by how the muscle feels, and load and educate before they leave. Verify, then reinforce.
You have the reps. The growth edge is not your hands.
So if you have been needling for years, none of this is news, and I want to say that out loud. You already feel the tissue, you already get the twitch, your hands are not the problem. And honestly, that is the trap. When you get good, the dosing quietly goes on autopilot, and you start grading the treatment by the twitch you felt instead of the reassessment you ran. So here is the Monday challenge, and for a lot of you it is really an affirmation. Take your three hardest chronic patients this week, the ones who keep flaring no matter how clean your technique is, and before you needle the site, screen the segment. Dermatomes, myotomes, reflexes. My gut says you will find a central driver you have been needling around. And if you are already doing that, already dosing by irritability, already loading them before they walk out the door, then good. You were doing the consensus before we wrote it down. Keep going, and go teach somebody else to do it.
We put the whole decision on one page, and the needle still is not the treatment. Screening is. Dosing is. The reassessment is. The needle is just how you deliver the change, and reinforcement is how you keep it.
If you train or supervise people who needle, forward this one. The decision tree is worth a hallway conversation, and it is open access, free to read.
