STUDY: Real laser therapy no better than a red LED for back pain

Photobiomodulation therapy (PBMT) is the use of light as a medical treatment. This is popularly known as laser therapy, although it’s not all about true laser light (coherent, monochromatic, directional); sometimes it’s red and infrared light (more tissue-penetrating), and sometimes a mix.

The big idea is that electromagnetic waves might “stimulate healing,” a regenerative medicine effect, some kind of cellular jiggery pokery. We know some cells respond to light (duh, that’s how eyes work), but with a huge upgrade? The evidence has been perpetually mixed and meh — one of the great examples of premature hype about a futuristic, “high-tech,” unproven therapy, where disappointing results are always dismissed with “wrong parameters!” So why doesn’t someone study the right parameters? Parameters the PBMT proponents can’t gripe about? Or not as easily, anyway?

Well, someone finally did!

Premium PBMT vs premium placebo

The journal Pain published a high-quality trial in 2021, which got a couple things particularly right:

1. Their placebo control was an ideal fake.
2. The treatment was a specific radiation cocktail endorsed by WALT. That is, they made a point of using frequencies and intensities recommended by the World Association of PhotobiomoduLation Therapy.

They pitted their perfect PBMT against a premium placebo in 148 Brazilians with chronic unexplained low back pain. Pain and disability were tracked for a year, plus some other secondary measures. Patients got a dozen treatments over a month. Half received genuine PBMT, and the other half got… some wimpy red light.

The sham device was the same device as the real one, just with the business end disabled, not emitting a therapeutic dose — just one low-powered red light. Laser therapy doesn’t feel like anything, so there was simply no way for patients to know whether they were getting real lasers or not. The quality of the placebo is so important to this study that I’ll quote their description in full:

In the PBMT device used, the only visible diode is the red one. For the placebo treatment, 905 nm laser diodes and the 875 nm LED diodes were deactivated (turned off), and the power of the 640-nm LED diodes were turned down to 1 mW (mean power for each diode) to keep the visual aspect of red light, but not to deliver an effective therapeutic or considerable dose according the current available evidence.

(Funny that the authors felt it was necessary to clarify “deactivated” with the parenthetical “turned off.” Of all the jargon in this paper, that was the term that needed translation? 😜)

Tommology, CC BY-SA 4.0 | cropped, coloured

A laser shining through a cloud of particles.

The illuminating results

So what happened? What any reasonably savvy cynic would expect: there was no important difference between legitimate lasers and lame ones.

More specifically, there was no clinically important difference in pain intensity, general disability, or any secondary outcome after one, three, six, or twelve months.

Yes, there was a slight boost in “global perceived effects” at one month for the lasered folks, and PBMT advocates will point to this single data point and try to snatch victory from the jaws of defeat. But it’s just a moderate, temporary, short-term benefit for a rather squishy secondary way of measuring the effect of therapy — and it’s might well vanish if you did the whole study all over again.

Consider the source

This is a rare case when I think the quality and credibility of the journal is noteworthy. Most studies of laser therapy have been published in obscure and junky little journals. This one stands out… and, as I hoped it might, the trial itself really ticks all the right boxes. Unlike so many studies in this field, it was designed to actually answer the question — big enough and good enough. No trial is perfect, but this one has a lot going for it.

One good example of the quality of the trial is that, delightfully, it was “prospectively registered” — which means that they declared in advance what they were up to, which makes it harder to fudge things later. “P-hacking” — torturing the data until it tells you what you want to hear — is harder to get away with when you’ve put all your cards on the table in advance.

This isn’t the last word, of course

It’s always reasonable to ask if different parameters might work better, and the authors — who have some vested interests in laser therapy — urge further investigation of “the optimization of PBMT parameters.” And it’s hard to disagree with that.

But they did use lasers the WALT way. Their parameters were WALT-recommended. One of the authors is a past WALT president.

And yet the results show that the WALT way is probably the wrong way. And what are the odds that some other way is going to work well? Even if different parameters are eventually validated, that would also mean that the experts at WALT got it hopelessly wrong up to now… which strongly suggests that their recommended PBMT parameters have been guesses, not based on good evidence.

Which is, in think, precisely the case.

WALT’s mission? “To stimulate high quality research in photobiomodulation…” They got that in this case — arguably for the first time. Be careful what you wish for, WALT!

This post is excerpted from a more detailed guide to laser therapy. It’s also a re-run of a summer 2021 post … but significantly revised and refined. I also added the audio version, and did a bunch of fact checking, and factored in some complaints and criticisms I’ve gotten about this topic. Unsurprisingly, PBMT proponents can and do argue that the study still got their parameters wrong — as if they can actually know without a rigorous test just like this one.

For instance, one critic claimed (without evidence) that the device tested is not a “real” laser therapy device. I don’t take that very seriously, because its importance relies on the second, implied claim that lasers are the One True kind of light therapy — a claim that doesn’t hold up to scrutiny very well (see Heiskanen). Lasers have always been fetishized by proponents of PBMT, often to an absurd degree. It’s almost a cliché of PBMT crankery to get overheated about lasers over mere light sources, which further reduces the credibility of the criticism. But, in any case, Guimarães et al was clearly a test of lasers and less coherent light.