What it is treating

The knee joint capsule is supplied by a cluster of small sensory branches — the superolateral, superomedial and inferomedial genicular nerves, most commonly — that carry pain signals up to the spinal cord. They do not move or stabilize the joint. Their only job is reporting, and in an arthritic knee they report constantly, because worn cartilage, thickened synovium and bone-on-bone contact all generate the irritation those nerves exist to detect.

A genicular nerve block anesthetizes those specific branches under fluoroscopic or ultrasound guidance. If the reporting stops, the pain stops, at least for as long as the local anesthetic lasts. That is useful information on its own, and it is also the audition for a longer procedure.

Who this is for

Three groups make up most of the patients we do this for: knee osteoarthritis in someone who is not a surgical candidate — cardiac or metabolic risk, a BMI a surgeon wants lowered first, or a job that cannot absorb the recovery window right now; someone who is a candidate but is choosing to wait, and wants the joint quiet rather than tolerated in the meantime; and pain that persists after a replacement, where imaging shows a sound implant and the generator turns out to be the same peri-articular nerves that reported on the native joint.

It is a poor fit where the pain is coming from somewhere else — hip and lumbar pathology both refer into the knee often enough that we examine both before reaching for a needle at the joint itself.

The two-step process

What it does not do

It does not change the joint. Cartilage that is gone stays gone, alignment does not correct, and the X-ray taken the week after ablation looks exactly like the one before it. What changes is the reporting, not the structure — which is why this fits a patient who has already decided replacement is not happening yet, and does not fit one hoping to avoid ever having that conversation.

The joint got here through more than mechanics, and the terrain around it decides how well a quiet nerve holds. Cartilage is living tissue with its own metabolic demand, and insulin resistance and metainflammation degrade cartilage matrix and slow the low-grade repair that keeps a joint tolerable well before an X-ray calls it severe. Add the load of carrying more weight than the joint was built for, and a job that keeps you on concrete floors eight or ten hours a day where getting to a gym is a scheduling problem before it is a motivation problem, and the social piece is doing as much work as the biology. A nerve block turns down the alarm. It does not touch any of that — see the practice’s approach to metabolic health and pain and movement as part of the plan if the alarm keeps returning faster than it should.

The evidence, honestly

A 2024 systematic review and meta-analysis of cooled and pulsed genicular ablation found consistent gains in pain and function, better in studies that required a positive block first. A 2023 placebo-controlled trial of the block alone found real but shorter-lived relief than sham — proof the block works, and a reminder it is not the durable answer ablation is built to be. A 2022 trial of ultrasound-guided block against physical therapy favored the block at short-term follow-up, though the comparison group was rarely people who had already exhausted therapy, which is most of who walks into this office.

What the trials mostly exclude: uncontrolled diabetes, active joint infection, significant vascular disease in the limb, or the deconditioning of someone who has avoided the knee for years. That does not mean the procedure fails there — it means the published numbers were not measured on it, and we say so rather than borrow a trial’s confidence for a patient it did not enroll. Our position: a positive block is the gate, not the guess, and when relief fades early the honest reason is usually the terrain the joint sits in, not the technique.

What happens on the day

Risks, in plain language

Bruising and soreness at the injection sites are common and short-lived; infection is rare with sterile technique. A small number of patients get a patch of numb skin near the site, usually temporary. Ablation carries a small added risk of a burning sensation in the days after, which we tell you to expect rather than call to report unless it is severe. Nerve tissue regenerates, which is why relief is not permanent — and why repeating it later does not carry the cumulative risk repeated steroid exposure would.

Common questions

Related reading

• Knee pain
• Image-guided joint injections
• When injections stop working
• Metabolic health and pain
• Hip pain