Low-dose radiation for plantar fasciitis and arthritis: routine in Germany, rarely offered here.

What it actually is

The phrase "radiation for arthritis" makes most people picture cancer treatment, and the comparison is worth making because the difference is enormous. A course of radiotherapy for a tumor might deliver 60 to 70 Gy. A course of LDRT for an arthritic joint delivers a total of about 3 to 6 Gy, in fractions of 0.5 to 1 Gy spread over two to three weeks. Each session takes a couple of minutes, you feel nothing, and there is no recovery period. If a first course does not help, the standard practice is to offer a second one before giving up on it, since some people respond only after the repeat.

The interesting part is that low doses and high doses appear to do genuinely different things. High-dose radiation kills dividing cells. Low-dose radiation seems to nudge the immune system rather than destroy tissue: the laboratory work describes it pushing macrophages from their inflammatory "M1" state toward the anti-inflammatory "M2" state, triggering apoptosis in inflammatory cells, reducing matrix metalloproteinase-13, and raising type 2 collagen. Whether those bench findings translate into a patient feeling better is exactly the question the trials have been arguing about.

Why Germany and not here

This is largely an accident of medical history rather than a scientific disagreement. Radiotherapy for benign conditions stayed in continuous mainstream use in German-speaking Europe, where it is supported by national radiation-oncology guidelines and used for heel spurs, hand and knee arthritis, tennis elbow, and similar problems. Here the practice largely disappeared after mid-century concerns about radiation-induced cancer, and it stayed gone for decades. The result is that most of the good clinical data is German, and most patients on this side of the Atlantic have still never been offered it.

That said, it would be wrong to call it forgotten. Radiation oncologists here have started writing about it again, and the most recent practical review sits in an American journal with a Yale author among its writers. Renewed academic interest is not the same as availability, but the topic is no longer purely a European curiosity.

The evidence, honestly

This is where a lot of writing on LDRT gets soft, so here is the straight version. The evidence splits by body part, and it does not split the way you would expect. Three trials carry most of the weight:

	Heel spur / plantar fasciitis	Hand arthritis	Hand & knee arthritis
Trial	Niewald 2012	Minten 2018	ArthroRad 2024
Size	66 patients	56 patients	133 patients, 229 joints
Compared against	A near-useless dose (6 Gy vs 0.6 Gy)	True sham (6 Gy vs 0 Gy)	A tenth of the dose (3 Gy vs 0.3 Gy)
Result	Standard dose won, at 3 months and still at 12	29% responded vs 36% on sham	~61% improved, both arms equally
What it implies	Benefit tracked the dose	No benefit over placebo	No dose-dependence

The heel: the strongest case for it

The best foot-specific trial is a randomized multicenter study of 66 patients with painful heel spur and plantar fasciitis. It compared a standard dose (6 Gy total, in six 1 Gy fractions) against a deliberately near-useless dose (0.6 Gy total, in six 0.1 Gy fractions) that functioned as a stand-in for placebo. At three months the standard-dose group did highly significantly better on pain scores, the calcaneodynia score, and quality of life. At twelve months the advantage held, and significantly fewer standard-dose patients had needed retreatment. The authors concluded the trial confirms a superior analgesic effect of the 6 Gy dose "even for a longer time period of at least 1 year."

Why that design matters: because the benefit tracked the dose, the result is hard to explain as pure placebo. Both groups lay on the same table, under the same machine, for the same number of visits. Only the amount of radiation differed, and the outcome followed it.

The knee and hand: the strongest case against it

Now the other side, and it is genuinely damaging. A randomized, blinded, sham-controlled trial in 56 patients with hand osteoarthritis compared six 1 Gy sessions against six sessions of literally nothing (0 Gy, with the machine going through the motions). At three months, 29% of the radiation group responded versus 36% of the sham group. The authors' conclusion is blunt: they could not demonstrate a substantial beneficial effect, and given the confidence interval, "a treatment effect exceeding 20% is very unlikely." Follow-up of the parallel knee and hand trials out to a year found no delayed benefit either, and the investigators suggested the large improvements reported in everyday practice are probably explained by regression to the mean and placebo response.

A separate German trial called ArthroRad pushed on the same question from another angle, and it is the largest of the three. It enrolled 133 patients contributing 244 painful hand and knee joints; after 15 joints were excluded, 229 were randomized to either a standard 3 Gy course or an ultra-low 0.3 Gy course. About 61% improved, which sounds encouraging, except that both arms did equally well. There was no dose-dependence at all. When a treatment works just as well at a tenth of the dose, the most economical explanation is that the dose was not the active ingredient.

One wrinkle worth naming, since it affects how you read that table: ArthroRad randomized joints, not people. A single patient could contribute several arthritic fingers. So its 229 joints are not 229 independent data points the way 66 patients are 66, and the effective sample sits somewhere below the headline number. It is still the biggest trial here. It is not as much bigger as it looks.

The detail I find most telling: the ArthroRad trial and the positive heel-spur trial share the same lead investigator. This is not a case of believers versus skeptics publishing whatever suits them. It is the same group finding dose-dependence in the heel and failing to find it in the knee and hand.

The rebuttal, and the problem underneath all of it

Radiation oncologists who use this treatment do not simply ignore the negative trials, and their counter-argument deserves a fair hearing. The 2026 review in Seminars in Radiation Oncology makes it directly: the trials showing no benefit have been criticized for small sample sizes, short follow-up, questionable patient selection, and for deviating from standard LDRT protocols, while other randomized trials and large retrospective series do report clinically meaningful benefit in well-chosen patients.

But look again at the table above, because there is a problem that cuts through this entire debate, and it cuts in both directions. The sham trial that found nothing had 56 patients. The heel trial I have just spent three paragraphs praising had 66. If 56 is too small to kill a treatment, 66 is too small to prove one. You cannot dismiss the negative evidence for being underpowered and then lean on positive evidence that is barely bigger. I would be doing exactly that if I let the "small sample size" criticism land only on the trials whose answer I did not like.

So here is the fair version. Every trial in this field is small. What makes the heel study interesting is not its size, which is unimpressive, but its design: benefit tracked the dose, and dose-response is genuinely hard to fake with placebo. That is a real point about study architecture, not about statistical muscle. It earns the treatment the benefit of the doubt in the heel. It does not earn it a verdict.

What would settle this is a large, properly powered, sham-controlled trial using a standard protocol in well-selected patients. That trial has not been published. Until it is, anyone telling you this definitely works, or definitely does not, is reading more into a few dozen patients than a few dozen patients can carry.

So does it work?

My honest read: the heel is where this treatment has its best evidence, and that is a genuinely interesting fact for anyone treating feet. A dose-response relationship that persists at a year is a meaningful signal. But it is one trial of 66 patients, and the sham-controlled work in other joints is a real warning that a lot of what looks like response to LDRT is response to attention, time, and the natural tendency of a painful flare to settle.

What that means practically, if you have foot arthritis or a stubborn heel: this is not a treatment you need to go chasing, but it is no longer one you cannot get. A growing number of centers here offer it, and if you are on Medicare there may be a payable path (see below). The things that reliably help plantar fasciitis, though, are unchanged: the stretching, the supportive shoes worn indoors, the offloading, the patience. Those remain unglamorous and effective, and they should be genuinely exhausted first. LDRT is worth knowing about, reasonable to ask a radiation oncologist about if you have run out of road, and not the missing answer.

Can you actually get it here, and will it be paid for?

This is probably the question you actually came here with, and the answer has shifted recently enough that most write-ups are out of date. A growing number of radiation oncology practices now offer LDRT for arthritis and plantar fasciitis, mostly to patients over 65, and they generally state that Medicare covers it. That claim is plausible on its face: a practice that was not getting paid would stop advertising the service.

It is worth knowing where that claim comes from, though. Essentially every source asserting "Medicare covers this" is a clinic that bills Medicare for it. I could not find a Medicare policy document that affirmatively names osteoarthritis or plantar fasciitis as a covered indication, and there is no CPT code specific to this treatment; it is billed using general radiation delivery codes.

Commercial insurers, meanwhile, have often looked at the same evidence and said no. Aetna's policy classifies radiotherapy for plantar fasciitis as experimental, investigational, or unproven, which is a denial. Several other large plans maintain dedicated policies on low-dose radiotherapy for non-oncologic indications, which is usually the paperwork of a considered and restrictive position rather than an enthusiastic one.

So the practical shape of it: if you are on Medicare, there may well be a payable path, and practices offering this are counting on exactly that. If you are on commercial insurance, expect resistance and read your plan's policy before anyone schedules anything. And the point worth holding onto: coverage is not evidence. Medicare pays for plenty of things whose evidence base is thinner than anyone would like, and a treatment being billable tells you about billing codes, not about whether it will help your heel.

What about the radiation risk?

This is the reasonable first question and it deserves a real answer rather than reassurance. The theoretical concern with any radiation exposure is inducing a cancer years later. Two things are usually offered in response: the doses involved are far below the range where secondary malignancy becomes a meaningful worry (the review literature puts the threshold for high risk above 2.5 Gy, and a typical LDRT course sits at or under that), and radiation-induced cancers take decades to appear, while the typical LDRT patient is an older adult with arthritis.

Both points are fair, and both are also the reason this treatment is generally reserved for older patients and not offered to young people with sore joints. The risk is small, not zero, and it has to be weighed against a benefit that, as above, is well supported in the heel and poorly supported elsewhere. That is a conversation for a radiation oncologist who can look at your specific situation, not a decision to make from a web page.

The bottom line

Low-dose radiation therapy is a real, established, unglamorous treatment that most Americans have never heard of and most Germans consider routine. The mechanism is anti-inflammatory rather than destructive, the dose is a tiny fraction of cancer treatment, and the sessions are quick and painless. The evidence is honestly mixed, and it is mixed in a direction that happens to matter here: the painful heel is where it looks best, and the knee and hand are where sham-controlled trials have undercut it. If you are in the US with plantar fasciitis, the boring plan still wins. If you are somewhere it is offered and you have genuinely run out of options, it is a defensible thing to ask about. This page is general educational information, not medical advice, and LDRT decisions belong with a radiation oncologist.